Leibniz and Kant 9780199606368, 0199606366

Table of contents :
Cover
Title Pages
Acknowledgments
Contents
List of Abbreviations
List of Contributors
1. Kant’s Leibniz: A Historical and Philosophical Study
2. How Kant was Never a Wolffian, or, Estimating Forces to Enforce Influxus Physicus
3. Breaking with Rationalism: Kant, Crusius, and the Priority of Existence
4. Leibniz on the Ideality of Space
5. Leibniz and the Transcendental Deduction
6. Bodies, Matter, Monads, and Things in Themselves
7. Kant and the (Alleged) Leibnizian Misconception of the Difference between Sensible and Intellectual Representations
8. Kant’s Amphiboly as Critique of Leibniz
9. The Teleologies of Leibniz and Kant: So Close Yet So Far Apart
10. Kant’s Theory of Divine and Secondary Causation
11. The Development of Kant’s Conception of Divine Freedom
12. Leibniz and Kant on Empirical Miracles: Rationalism, Freedom, and the Laws
Bibliography
Index
Пустая страница

Citation preview

Leibniz and Kant

Leibniz and Kant Edited by

B R A N D O N  C .  L O O K

1

1 Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © Oxford University Press 2021 The moral rights of the authors have been asserted First Edition published in 2021 Impression: 1 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, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2020945751 ISBN 978–0–19–960636–8 DOI:10.1093/oso/9780199606368.001.0001 Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

Acknowledgments The majority of chapters in this volume were originally delivered at a conference of the Leibniz Society of North America in conjunction with the North American Kant Society and held at the University of Kentucky in September 2009. I wish to thank the College of Arts & Sciences and the Department of Philosophy as well as the Consulate General of the Federal Republic of Germany in Chicago for their financial support of that conference. I would also like to thank the participants of that conference, the contributors to this volume, and Peter Momtchiloff for their extraordinary patience while this volume was completed.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 30/07/21, SPi

Contents List of Abbreviations List of Contributors

1. Kant’s Leibniz: A Historical and Philosophical Study Brandon C. Look (University of Kentucky)

ix xiii

1

2. How Kant was Never a Wolffian, or, Estimating Forces to Enforce Influxus Physicus 27 Ursula Goldenbaum (Emory University) 3. Breaking with Rationalism: Kant, Crusius, and the Priority of Existence Eric Watkins (University of California, San Diego) 4. Leibniz on the Ideality of Space Donald Rutherford (University of California, San Diego)

57 79

5. Leibniz and the Transcendental Deduction Alison Laywine (McGill University)

112

6. Bodies, Matter, Monads, and Things in Themselves Nicholas F. Stang (University of Toronto)

142

7. Kant and the (Alleged) Leibnizian Misconception of the Difference between Sensible and Intellectual Representations Anja Jauernig (New York University)

177

8. Kant’s Amphiboly as Critique of Leibniz Martha Brandt Bolton (Rutgers University)

211

9. The Teleologies of Leibniz and Kant: So Close Yet So Far Apart Paul Guyer (Brown University)

233

10. Kant’s Theory of Divine and Secondary Causation Des Hogan (Princeton University)

265

11. The Development of Kant’s Conception of Divine Freedom Patrick Kain (Purdue University)

295

12. Leibniz and Kant on Empirical Miracles: Rationalism, Freedom, and the Laws Andrew Chignell (Princeton University)

320

Bibliography Index

355 375

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 30/07/21, SPi

List of Abbreviations Unless otherwise noted translations of the works of Leibniz, Kant and other historical ­figures are from the standard English-­language editions. A

A/B

AG Ak

Anon-­K2 Anth

AT

BDG

C

CSM

D

Leibniz, G.  W.  1923–. Gottfried Wilhelm Leibniz: Sämtliche Schriften und Briefe. Ed. Deutsche Akademie der Wissenschaften. Darmstadt and Berlin: Akademie Verlag. Cited by series, volume and page number. Kant, Immanuel. 1998. Kritik der reinen Vernunft (in Ak. 3 and 4). Translations from Critique of Pure Reason. Ed. and trans. Paul Guyer and Allen  W.  Wood. Cambridge: Cambridge University Press. Cited according to the first (A) and second (B) edition page numbers. Leibniz, Gottfried Wilhelm. 1989. Philosophical Essays. Ed. and trans. Roger Ariew and Dan Garber. Indianapolis: Hackett. [= AG] Kant, Immanuel. 1902– . Kants gesammelte Schriften. Berlin: Königlich preussischen Akademie der Wissenschaften (now Walter de Gruyter). Cited by volume and page number. Transcriptions of metaphysics lectures from early 1790s (in Ak. 28). Anthropologie in pragmatischer Hinsicht (in Ak. 7). “Anthropology from a Pragmatic Point of View.” In: Anthropology, History and Education. Ed. Günter Zöller and Robert Loudent. Trans. Robert Loudent. Cambridge: Cambridge University Press, 2007, pp. 231–429. Descartes, René. 1897–1913. Oeuvres de Descartes. Ed. Charles Adam and Paul Tannery. 12 vols. Paris: L.  Cerf. Cited by volume and page number. Der einzig mögliche Beweisgrund zu einer Demonstration des Daseins Gottes (in Ak. 2). “The Only Possible Argument in Support of a Demonstration the Existence of God.” In: Immanuel Kant, Theoretical Philosophy, 1755–1770. Ed. and trans. D.  Walford. Cambridge: Cambridge University Press, 1992. Leibniz, Gottfried Wilhelm. 1903. Opuscules et Fragments Inédits de Leibniz. Ed. Louis Couturat.) Paris: Félix Alcan, 1903. (Repr. Hildesheim: Olms, 1988.) Descartes, René. 1985. The Philosophical Writings of Descartes. 2 vols. Ed. John Cottingham, Robert Stoothoff and Dugald Murdoch. Cambridge: Cambridge University Press. Cited by volume and page number. Leibniz, Gottfried Wilhelm. 1768. Opera omnia, nunc primum collecta . . . . 6 vols. Ed. Louis Dutens. Geneva. (Repr. Hildesheim: Olms, 1989.) Cited by volume and page number.

x  List of Abbreviations DM EEKU

FM

GM

GMS

GP

GW H KpV

KrV

KU

L

LA

Leibniz, Gottfried Wilhelm. Discourse on Metaphysics (in A VI 4, 1629–1688; in AG 35–68). Cited by section number. Erste Einleitung in die Kritik der Urteilskraft (in Ak. 20). First Introduction to the Critique of the Power of Judgment. In: Critique of the Power of Judgment. Ed. and trans. Paul Guyer and Eric Matthews. Cambridge: Cambridge University Press, 2000. Welches sind die wirklichen Fortschritte, die die Metaphysik seit Leibnitzens und Wolf ’s Zeiten in Deutschland gemacht hat? (in Ak. 20). “What Real Progress has Metaphysics Made in Germany Since the Time of Leibniz and Wolff?” In Theoretical Philosophy after 1781. Ed. Henry Allison and Peter Heath. Trans. Peter Heath. Cambridge: Cambridge University Press, 2002, pp. 349–412. Leibniz, G. W. 1849–1863. Leibnizens mathematische Schriften. 7 vols. Ed. C.  I.  Gerhardt. Halle: Schmidt. (Repr. Hildesheim: Olms, 1971.) Cited by volume and page number. Grundlegung zur Metaphysik der Sitten (Ak. 4). “Groundwork for the Metaphysics of Morals.” In: Immanuel Kant, Practical Philosophy. Ed. and trans. Mary Gregor. Cambridge: Cambridge University Press, 1996, pp. 37–108. Leibniz, G. W. 1875–1890. G. W. Leibniz: Die philosophischen Schriften. Ed. C. I. Gerhardt. 7 vols. Berlin: Weidmann. (Repr. Hildesheim: Olms, 1965.) Cited by volume and page number. Wolff, Christian. 1962– . Gesammelte Werke. Ed. Jean École et al. Hildesheim: Olms. Cited by division, volume and page number. Leibniz, Gottfried Wilhelm. 1985. Theodicy. Ed. and trans. E. M. Huggard. La Salle, IL: Open Court. Kritik der praktischen Vernunft (Ak. 5). Immanuel Kant, Critique of Practical Reason. In: Immanuel Kant, Practical Philosophy. Ed. and trans. Mary Gregor. Cambridge: Cambridge University Press, 1996, pp. 133–271. Kritik der reinen Vernunft (Ak. 3 and 4). Immanuel Kant. English translations from Critique of Pure Reason. Ed. and trans. Paul Guyer and Allen  W.  Wood. Cambridge: Cambridge University Press. Cited according to first (A) and second (B) edition page numbers noted as stated above. Kritik der Urteilskraft (Ak. 5). Critique of the Power of Judgment. Ed. and trans. Paul Guyer and Eric Matthews. Cambridge: Cambridge University Press, 2000. Leibniz, Gottfried Wilhelm. 1969. Philosophical Papers and Letters. Ed. and trans. Leroy  E.  Loemker. 2nd ed. Dordrecht and Boston: Reidel. Leibniz, Gottfried Wilhelm. 1967. The Leibniz–Arnauld Correspondence. Ed. and trans. H.  T.  Mason. Introduction by G.  H.  R.  Parkinson. Manchester: Manchester University Press.

List of Abbreviations  xi LC

LDB

LH

Log

MAN

MKTI Mon MP MSI

OCM Pölitz PNG Prol

R

Leibniz’s letters to Samuel Clarke. From GP VII 352–420. Cited by letter and section number. Translations from The Leibniz-­Clarke ­ Correspondence. Ed. H.  G.  Alexander. Manchester: Manchester University Press, 1956. Leibniz, Gottfried Wilhelm. 2007. The Leibniz–Des Bosses Correspondence. Ed. and trans. Brandon  C.  Look and Donald Rutherford. New Haven: Yale University Press. Die Leibniz-­Handschriften der königlichen öffentlichen Bibliothek zu Hannover. Ed. Eduard Bodemann. Hanover and Leipzig: Hann’sche Buchhandlung. Cited according to Bodemann’s classification. Logik (in Ak. 9). “The Jäsche Logic.” In: Immanuel Kant, Lectures on Logic. Ed. and trans. Michael Young. Cambridge: Cambridge University Press, 1992, pp. 527–640. Metaphysische Anfangsgründe der Naturwissenschaften (Ak. 4). Immanuel Kant, Metaphysical Foundations of Natural Science. In Theoretical Philosophy after 1781. Ed. Henry Allison and Peter Heath. Trans. Michael Friedman. Cambridge: Cambridge University Press, 2002, pp. 181–270. Leibniz, Gottfried Wilhelm. Meditations on Knowledge, Truth and Ideas (in A VI 4, 585–592; in AG 23–27). Leibniz, Gottfried Wilhelm. Monadology (in GP VI 607–623; in AG 213–225). Cited by section number. Leibniz, Gottfried Wilhelm. Philosophical Writings. Ed. Mary Morris and G. H. R. Parkinson. London: Everyman, 1995. De Mundi sensibilis atque intelligibilis forma et Principiis (Ak. 2). “On the Form and Principles of the Sensible and Intelligible World” (Inaugural Dissertation). In: Immanuel Kant, Theoretical Philosophy, 1755–1770. Ed. and trans. D.  Walford. Cambridge: Cambridge University Press, 1992, pp. 373–426. Malebranche, Nicolas. 1958–1984. Oeuvres complètes de Malebranche. Ed. André Robinet. Paris: Vrin. Cited by volume and page number. Religionsphilosophie Pölitz. Lectures on Religious Philosophy from 1780s (in Ak. 28). Leibniz, Gottfried Wilhelm. Principles of Nature and Grace (in GP VI 598–606; in AG 206–213). Cited by section number. Prolegomena zu einer jeden künfigen Metaphysik, die als Wissenschaft wird auftreten können (in Ak. 4). Prolegomena to Any Future Metaphysics that Will Be Able to Come Forward as a Science. In Theoretical Philosophy after 1781. Ed. Henry Allison and Peter Heath. Trans. Gary Hatfield. Cambridge: Cambridge University Press, 2002, pp. 49–169. Reflexionen (Ak. 15–19). Immanuel Kant, Notes and Fragments. Ed. Paul Guyer. Trans. Curtis Bowman, Paul Guyer, and Frederick Rauscher. Cambridge: Cambridge University Press, 2005.

xii  List of Abbreviations RB

Rel

ÜE

V-­Lo/Busolt V-­Lo/Dohna

V-­Lo/Pölitz V-­Lo/Wiener

V-­ Met/Dohna

V-­Met-­L1 V-­Met-­L2/Pölitz V-­ Met-­ Vron

V-­Th/Baumbach WF

Leibniz, Gottfried Wilhelm. 1996. New Essays on Human Understanding. Ed. Peter Remnant and Jonathan Bennett. Cambridge: Cambridge University Press. Cited by page numbers, which correspond to the Akademie edition in A VI 6. Religion innerhalb der Grenzen der bloßen Vernunft (1794) (in Ak. 6). Immanuel Kant, Religion within the Boundaries of Mere Reason, and Other Writings. Ed. Allen  W.  Wood and George di Giovanni. Cambridge: Cambridge University Press, 1998. Über eine Entdeckung, nach der alle neue Kritik der reinen Vernunft durch eine ältere entbehrlich gemacht warden soll (in Ak. 8). “On a Discovery Whereby Any New Critique of Pure Reason Is to Be Made Superfluous by an Older One.” In Theoretical Philosophy after 1781. Ed. Henry Allison and Peter Heath. Trans. Henry Allison. Cambridge: Cambridge University Press, 2002, pp. 283–336. Logik Busolt (Ak. 24). Logik Dohna (Ak. 24). “The Dohna-­Wundlacken Logic.” In: Immanuel Kant, Lectures on Logic. Ed. and trans. Michael Young. Cambridge: Cambridge University Press, 1992, pp. 379–423. Logik Pölitz (Ak. 24). Wiener Logik (Ak. 24). “The Vienna Logic.” In: Immanuel Kant, Lectures on Logic. Ed. and trans. Michael Young. Cambridge: Cambridge University Press, 1992, pp. 249–377. Metaphysik Dohna (Ak. 28). “Metaphysik Dohna.” In: Immanuel Kant, Lectures on Metaphysics Ed. Karl Ameriks and Steve Naragon. Cambridge: Cambridge University Press, 1997, pp. 355–391. Metaphysik L1. (Ak. 28). “Metaphysik L1.” In: Immanuel Kant, Lectures on Metaphysics Ed. Karl Ameriks and Steve Naragon. Cambridge: Cambridge University Press, 1997, pp. 19–107. Metaphysik L2. (Ak. 28). “Metaphysik L2.” In: Immanuel Kant, Lectures on Metaphysics Ed. Karl Ameriks and Steve Naragon. Cambridge: Cambridge University Press, 1997, pp. 297–354. Metaphysik Mrongovius. (Ak. 29). “Metaphysik Mrongovius.” In: Immanuel Kant, Lectures on Metaphysics Ed. Karl Ameriks and Steve Naragon. Cambridge: Cambridge University Press, 1997, pp. 109–288. Danziger Rationaltheologie nach Baumbach (Ak. 28). Leibniz, Gottfried Wilhelm. 1998. Philosophical Texts. Ed. Roger S. Woolhouse and Robert Francks. New York: Oxford University Press.

List of Contributors Martha Brandt Bolton (Rutgers University) Andrew Chignell (Princeton University) Ursula Goldenbaum (Emory University) Paul Guyer (Brown University) Des Hogan (Princeton University) Anja Jauernig (New York University) Patrick Kain (Purdue University) Alison Laywine (McGill University) Brandon C. Look (University of Kentucky) Donald Rutherford (University of California, San Diego) Nicholas F. Stang (University of Toronto) Eric Watkins (University of California, San Diego)

1

Kant’s Leibniz A Historical and Philosophical Study Brandon C. Look (University of Kentucky)*1

1 Introduction The essays in this volume concern the relation between the philosophers Gottfried Wilhelm Leibniz (1646–1716) and Immanuel Kant (1724–1804), the two giants of eighteenth-­century German philosophy. While Kant refers often to Leibniz in his writings, sometimes praising him but more often criticizing him, it is nevertheless difficult for the historian of philosophy to know how well Kant understood Leibniz and what Kant actually understood under the name “Leibniz.” What could Kant have known, after all? What were the texts available to him? What works do we know that he read and studied? What shaped his picture of Leibniz’s thought? These questions are certainly not of merely antiquarian interest either, for if it can be shown that Kant did not know Leibniz’s philosophy well, then his claims of victory over his silent opponent might need to be moderated. Even if Kant’s general criticism of Leibniz should be valid, even if he understood the historical Leibniz in sufficient detail, it is still important to understand what Leibnizian theses were at issue in Kant’s critique of dogmatic metaphysics. In attempting to understand the Rezeptionsgeschichte of Leibniz’s philosophy in the eighteenth century and, in particular, how Kant could have seen Leibniz, the following facts should be borne in mind. First, Leibniz never wrote a magnum opus comparable to Descartes’s Meditations on First Philosophy, Spinoza’s Ethics, or Locke’s Essay Concerning Human Understanding that could serve to express his considered thought. Thus, the philosophical and historical position that Spinoza and Locke were in with respect to Descartes or that Hume was in with respect to Malebranche, Locke, and Berkeley differs vastly from the position of Kant vis-­à-­ vis Leibniz. Moreover, the typical form of Leibniz’s philosophical expression differed from that of his great early modern counterparts. With the exception of his Theodicy and book-­length commentary on Locke’s Essay, Leibniz’s philosophical writings tended to be short, occasional essays, letters, and private notes and *  Thanks to Ursula Goldenbaum and to the members of the Early Modern History Workshop at the Institute for Advanced Study, who read an earlier draft of this chapter and gave valuable feedback.

Brandon C. Look, Kant’s Leibniz: A Historical and Philosophical Study. In: Leibniz and Kant. Edited by Brandon C. Look, Oxford University Press. © Oxford University Press 2021. DOI: 10.1093/oso/9780199606368.003.0001

2  Kant’s Leibniz: Historical and Philosophical Study memoranda. Yet, of Leibniz’s essays, only a relatively small number were published during his lifetime. And while there was a small cottage industry of publishing Leibniz’s works posthumously in the eighteenth century, it is actually surprising how few new works appeared in the first five decades after Leibniz’s death in 1716. It was not until the great editions of Raspe (Leibniz  1765) and Dutens (Leibniz  1768) that the philosophical public had anything approaching a detailed picture of the depth and breadth of Leibniz’s work. And many of Leibniz’s reflections on the nature of logic and even metaphysics were not published until well into the nineteenth century. Unsurprisingly, historians of phil­ oso­ phy from the mid-­ 1800s on have been in a much better position to understand Leibniz’s philosophy in all its variety and complexity than Kant ever could have done. Further, much of Leibniz’s thought was filtered through the writings of others—most notably, Christian Wolff and his disciples—and, to some degree, altered in the process. One might add to this list of facts about the history of the reception of Leibniz’s philosophy that Kant was not always terribly concerned about the details of the history of philosophy. In fact, he had a rather negative view of those more interested in studying the texts of others than in thinking through a philosophical problem for themselves. As he remarks in the preface to the Prolegomena to Any Future Metaphysics (1783), “There are scholars for whom the history of philosophy (ancient as well as modern) is itself their philosophy; the present prolegomena have not been written for them. They must wait until those who endeavor to draw from the wellsprings of reason itself have finished their business, and then it will be their turn to bring news of these events to the world” (Ak. 4:255/Kant 2002, 53). In a similar spirit, and against an opponent, Eberhard, whom Kant sees as far more devoted to the “truths” uttered by another philosopher, Leibniz, he writes, “what is philosophically correct neither can nor should be learned from Leibniz; rather the touchstone, which lies equally to hand for one man as for another, is common human reason, and there are no classical authors in philosophy” (Ak. 8:219 n./Kant 2002, 309). Kant lived this position too, for he took the thought of his predecessors as positions against which to argue and was less concerned to understand their views in all their historical and textual complexity.1 It would seem unlikely, then, that Kant should hunt down various Leibnizian texts published several generations earlier in order to work out the nuances of Leibniz’s system. At the same time, this means that the Leibniz with whom Kant argues may not exactly be the Leibniz who lived, thought, and wrote before him.

1  In his Early German Philosophy, Lewis White Beck claims that much of Kant’s information and misinformation concerning the history of philosophy came from Brucker’s Historia critica philosophiae [Critical History of Philosophy] (1742–1744) (Beck 1969, 277). For example, Kant falsely asserts that, on Plato’s view, punishment would be unnecessary in the ideal state (A 317/B 373).

Brandon C. Look   3

2  Leibniz’s Philosophy: The Unfolding of a System Leibniz had already been established as one of the great philosophers of Europe by the middle of the eighteenth century. Writing in the Encyclopédie in the 1760s, Denis Diderot effuses, “Perhaps never has a man read as much, studied as much, meditated more, and written more than Leibniz . . . What he has composed on the world, God, nature, and the soul is of the most sublime eloquence. If his ideas had been expressed with the flair of Plato, the philosopher of Leipzig would cede nothing to the philosopher of Athens” (1975, 7:709). While Diderot’s enthusiasm for Leibniz is remarkable, what is more remarkable is that he marvels at the sheer quantity of Leibniz’s writings when, in fact, only a small portion were even known to him and his contemporaries. At present, the Leibniz-­Archives in Hanover contain over 200,000 manuscript pages, including over 15,000 letters to his more than 1,000 correspondents and 50,000 distinct essays, sketches, and exposés. Had Diderot known this he might really have followed up on his lament elsewhere: “when one compares the paltry talents one has been given with those of a Leibniz, one is tempted to throw away one’s books and go die quietly in the dark of some forgotten corner” (1975, 7:678). Thankfully he was ignorant of Leibniz’s enormous philosophical output. Whether we should be thankful that others in the eighteenth century were ignorant of so many of Leibniz’s works is another matter. Although Leibniz published several early works in philosophy from his student days—Metaphysical Disputation on the Principle of Individuation (1663) and Dissertation on the Combinatorial Art (1666)—and in natural philosophy prior to his Paris sojourn of 1672–1676—New Physical Hypothesis (1671) and Theory of Abstract Motion (1671)—his first mature philosophical publication was not until November 1684, when the Acta Eruditorum printed his short piece Meditations on Knowledge, Truth and Ideas just a month after his first statement of the calculus in the same journal.2 Almost ten years would pass before Leibniz published another work in philosophy, even though he was working diligently on difficult issues in logic and metaphysics, composing among, other things, the Discourse on Metaphysics (1686), General Inquiries about the Analysis of Concepts and of Truths (1686), and Primary Truths (1689)—none of which would become known to philosophical readers until almost two centuries later. When Leibniz began to make his metaphysical views known to the public in the 1690s, he was already celebrated as one of the most important mathematicians and physicists of his day, having published further on differential and integral calculus as well as articles critical of Cartesian and Newtonian physics. His notion of force, in particular, became one of the crucial aspects of his new metaphysics, first announced in the 1694 essay On the Correction of First Philosophy and the Notion of Substance again 2  Titles of well-­known works by Leibniz, Kant, and other canonical authors are given in English; titles of other works are given in the original language with English translation in brackets.

4  Kant’s Leibniz: Historical and Philosophical Study published in the Leipzig Acta Eruditorum. But it was not until Leibniz presented his theory of pre-­established harmony in the New System of Nature and of the Communication of Substances, as well as the Union of Mind and Body the following year in the Parisian Journal des Sçavans that he became recognized as a metaphysician of the first order. This essay occasioned a lively debate about mind–body causation throughout the republic of letters, and Leibniz went to considerable pains to respond to his various critics in different journals. Three years later Leibniz published On Nature Itself, again in the Acta Eruditorum, in which the philosophical public first encountered in connection with Leibniz’s thought the term “monad.” The beginning of the eighteenth century saw Leibniz engaged in the project of critically responding to Locke’s Essay Concerning Human Understanding which, though first published in 1690, was not studied in detail by Leibniz until Pierre Coste’s French translation in 1700. When Locke died in 1704, however, Leibniz withheld publication of this, his most detailed work in epistemology, and it remained unknown until 1765, when Rudolf Erich Raspe printed it with a selection of other works that he had discovered while working at the Royal Library of Hanover. At the urging of Queen Sophie Charlotte of Prussia, Leibniz turned to natural religion and the classic problem of evil, which had been given new life by the writings of Pierre Bayle, and composed his Theodicy, published in Amsterdam in 1710. While in Vienna in 1714, Leibniz decided to write two short treatises that would present his metaphysics in a condensed form and thus make his views accessible to a wider audience. The first, the Principles of Nature and Grace, was sent to both Prince Eugene of Savoy, then in Vienna, and Nicolas Remond in Paris. The other treatise, the text that we have come to call the Monadology (a title that Leibniz never used), expands upon some issues in the Principles of Nature and Grace and the Theodicy. In his final years, Leibniz also engaged in his important correspondence with Clarke. Even on the continent, there was a tremendous interest in the so-­called priority dispute between Leibniz and Newton, and it was clear that, in disputing with Samuel Clarke, Leibniz was really engaged in a kind of proxy war with Newton. Moreover, as was common in the day, Leibniz wrote his letters to Clarke with an eye to their eventual publication. While somewhat limited in their scope, these do in fact represent a fairly sophisticated presentation of some of Leibniz’s views. One of the crucial difficulties for anyone trying to understand Leibniz’s views is that he wrote differently for different audiences. And what he intended for a general audience is, unsurprisingly, not as straightforward as what he wrote to his philosophical correspondents or for himself. Thus, in a passage from a letter to Placcius in 1696, famous to all Leibniz scholars, he writes, “he who knows me only from my published works does not know me” (D VI 1, 65). And in a similar vein to Jacob Bernoulli one year later, Leibniz says, “I have written countless things on countless subjects; but I have published only a few things on a few subjects” (GM III 61). The situation changes somewhat in the last decades of his

Brandon C. Look   5 life—but only somewhat. Leibniz claims in a letter to Nicolas Remond that “It is true that my Theodicy does not suffice to present my system as a whole. But if it is joined with what I have published in various learned journals, those of Leipzig, Paris, and those of Mr. Bayle and Mr. Basnage, it will not fall far short of doing so, at least for the principles” (GP III 618). But we should note that Leibniz’s claim is that this is enough to get the attentive reader the basic principles of his philosophy—hardly the deep foundations of the system. For example, many of Leibniz’s metaphysical theses are closely connected to logical theses and theses about the nature of the infinite, and his thoughts about these topics were largely unknown until well into the nineteenth century. In fact, Leibniz is simply reluctant to share some of his views; as he writes to Des Bosses: I do not think those things we have discussed in letters concerning philosophical matters are suited for communication in any sort of public way, for they are unorganized and not gathered together in a system, such as I was hoping for from you. I have written these things for you, namely for the wise, not for any one at all; thus, they are hardly appropriate for the Memoires de Trevoux, which is intended more for a popular audience. I hope that you, by virtue of your goodwill toward me, would not allow them to appear in such an unsuitable place.  [LDB 83]

Leibniz’s reluctance to publish his more daring and difficult philosophy has invited cynicism and skepticism over the years. Bertrand Russell, for example, famously remarked that Leibniz’s published work “was optimistic, orthodox, fantastic, and shallow; [his unpublished work], which has been slowly unearthed from his manuscripts by fairly recent editors, was profound, coherent, largely Spinozistic, and amazingly logical” (1945, 581). The cynicism, on the part of Russell, is that Leibniz purposefully withheld his deeper and more dangerous philosophy and published his orthodox views in order to curry favor with the rich and powerful. That is going too far. But Russell is right that the works published and intended for a wider public are conciliatory in tone and content and that what is often most interesting in Leibniz’s thought is to be found in material that came to light long after his death. Ernst Cassirer indirectly suggests that we should be wary of claims of eighteenth-­ century scholars with respect to their knowledge of Leibniz’s philosophy, for they simply could not know Leibniz’s esoteric metaphysics (1998, 34).3 For example, Leibniz’s important correspondence with Des Bosses, which deepens our understanding of so much of Leibniz’s metaphysics, only started to become widely available to scholars and philosophers in Dutens’s edition of 1768; and, while 3  The distinction between an exoteric and esoteric Leibniz can be seen even in the late eighteenth century, with Johann August Eberhard’s Neue Apologie des Sokrates [New Apology for Socrates] (Eberhard 1787) and the critical response of Gotthold Ephraim Lessing, Leibniz von den ewigen Straften [Leibniz on Eternal Punishment] (Lessing 1886, 11:461–487). See also Wilson (1995, 460–462). I shall discuss this issue in some more detail in §4 of this chapter.

6  Kant’s Leibniz: Historical and Philosophical Study some important letters trickled out in the mid-­eighteenth century, Leibniz’s cor­res­pond­ences with Arnauld and De Volder only became known in the nineteenth century through the editions of Grotefend and Gerhardt. Likewise, Leibniz’s tremendously important and complex writings on logic and language from the 1680s, which seem to fill the three-­part volume 4 of the Akademie’s edition of Leibniz’s philosophical writings, only began to be published at the turn of the twentieth century in Couturat’s collection of texts. In the years following his death, Leibniz’s works were slowly published, but not enough and not fast enough to mitigate the judgment of Russell and Cassirer. Aside from the Theodicy, the best-­ known texts of Leibniz’s philosophy in eighteenth-­ century Germany were certainly the Specimen Dynamicum, the Monadology, and the correspondence with Samuel Clarke. While little hangs on the matter of translation, it is of some interest that the text of the Monadology was never published in its original French until Erdmann’s Leibnitii opera philosopica [Leibniz’s Philosophical Works] of 1840. Instead, it was published in 1720 in a German translation by Heinrich Köhler as Lehrsätze über die Monadologie [Theorems on the Monadology]4 and in a Latin translation by Michael Gottlieb Hansch in 1721 in the Acta Eruditorum as Principia Philosophiae, Autore G.G.  Leibnitio [Leibniz’s Principles of Philosophy].5 Moreover, the Principles of Nature and Grace was published in France in 1718 and then in German translation along with Gottsched’s German translation of the Theodicy in 1744. Likewise, the correspondence between Leibniz and Clarke was published by Clarke in 1717 and translated into German by Köhler in 1740. And it was not until Raspe’s edition of the New Essays in 1765 and Dutens’s Leibnitii opera omnia [The Collected Works of Leibniz] of 1768 that scholars and philosophers had a significant trove of Leibnizian texts to draw a more sophisticated interpretation of Leibniz’s views. By that time, however, a certain picture had already been drawn. As Cassirer puts it, “The influence of Leibniz’s thoughts is therefore indirect: they are efficacious only in the reformulation [Umbildung] that they underwent in the system of Wolff ” (1998, 34). Interestingly, a more refined and subtle account of Leibniz’s phil­oso­ phy then began to be seen as a reaction to Kant’s philosophy and other developments in the last decades of the eighteenth century through writings of Jacobi, Eberhard, Maaß, and Herder.6

3  The Origins of the “Leibniz-­Wolffian Philosophy” Leibniz’s final years were certainly difficult. His health was failing; he was embroiled in the bitter priority dispute with Newton and his partisans over the 4  See Leibniz (1720). 5  See Lamarra et al. (2001). 6  My thanks to a referee for suggesting this point.

Brandon C. Look   7 discovery of the calculus; and, when his employer, Duke and Elector Georg Ludwig of Hanover, became King George I of Great Britain and Ireland in 1714, he was told that his services were not required in London. At the same time, a young philosopher, whom Leibniz had personally helped to secure a professorship at the University of Halle, was gaining ascendance: Christian Wolff (1679–1754). From a distance of three centuries, Wolff ’s fate now seems quite remarkable. No philosopher was as dominant on the philosophical scene in Germany in the first half of the eighteenth century. Johann Christian Edelmann has the simpleton character in his controversial Moses mit aufgedeckten Angesichte [The Revealed Face of Moses] complain of a “veritable lycanthropy” among the educated public— everyone was becoming a wolf.7 By many, Wolff was even considered dangerous.8 Yet, in what is still the standard history of pre-­Kantian German philosophy in English, Lewis White Beck offers the following picture of Wolff and his works: [Wolff] illustrates what needs no illustration. He proves (though often by proofs so invalid that the fastidious reader may squirm) what needs no proof and what admits of no proof. He defines what needs no definition. He cites, by elaborate cross-­references, his other works, which all too often are found not to elucidate the passage in question but to be almost equivalent to it. He recommends his other books. He boasts of what he has accomplished. He moves with glacial celerity. He ruthlessly bores.  [1969, 258]

To complain that Wolff ’s books are boring is, however, to ignore their purpose: these works are first and foremost textbooks with a definite pedagogical goal; they are not the freestanding philosophical essays or books written for other philosophers. Moreover, Beck’s extremely negative judgment should not blind us to what the situation was surely like in the German academic milieu in the early part of the century. For example, in the first decades of the eighteenth century at the University of Königsberg there was relative diversity in the curriculum, with the philosophy of the moderns, specifically Cartesianism, and Aristotelian philosophy both on offer. Johann Christoph Gottsched (1700–1766) tells us that in his first years at the university he read Descartes, Locke, Christian Thomasius, Samuel Pufendorf, Jean Le Clerc, and others as well. But, in the end, he felt at sea—that is, until he came upon Leibniz’s Theodicy and the metaphysics of Wolff: “Then it seemed to me like one who, from a wild sea of competing opinions, entered a safe harbor and, after the topsy-­turvy, came again to stand on solid

7  See Edelmann (1740 III 108), facsimile edition republished in Edelmann (1969, vol. 7.1). 8 As we shall see, the Pietists of Halle successfully campaigned to have him exiled from Brandenburg-­Prussia. And over forty years later, Leonhard Euler claimed in his Letters to a German Princess that “les monadistes sont des gens bien dangereux,” by which he clearly meant Wolff and his followers (Euler 2003, 266 = Letter 132).

8  Kant’s Leibniz: Historical and Philosophical Study ground. I found here that certainty that I had earlier sought all over in vain.”9 Wolff ’s popularity and importance rested precisely in the systematicity of his presentation; that is, in the application of the mathematical method in phil­oso­ phy, something that was unusual in the universities of Central Europe.10 At the same time, the resulting philosophical books bore a superficial resemblance to the philosophical textbooks of Scholasticism they were replacing. Despite the troubles that Wolff would eventually have in Halle with conservative Pietist ­thinkers, his works in metaphysics, ethics, law, and natural theology came to be taught throughout Central Europe and Scandinavia in both Catholic and Protestant lands. Moreover, his textbooks in mathematics were of tremendous importance for most of the century. Indeed, even Kant used the Elementa matheseos universae [Elements of Universal Mathematics] for his own lectures in math­ em­at­ics and natural philosophy. Wolff adopted many of Leibniz’s theses, advocating the doctrine of pre-­ established harmony, asserting the importance of the Principles of Contradiction and Sufficient Reason, and arguing that metaphysics could be expressed in a mathematical or geometrical manner.11 And the proximity of their views in some regards led to their assimilation in the mind of the philosophical public. This fact, combined with the prominence of Wolff and his students, also had the effect of essentially drowning out Leibniz’s original voice. As Giorgio Tonelli once remarked, the works of Wolff and his students in Germany “crushingly outnumbered” the works of Leibniz (1974, 444). Indeed, from the time of Wolff ’s troubles in Halle, a common story arose, repeated by both sympathizers and opponents alike: that Wolff essentially systematized Leibniz’s philosophy. Gottsched, for example, claimed that Wolff only expressed more clearly what Leibniz had already said: One would err terribly if, after reading through the writings of both of them, one did not find that Leibniz and his follower had the same system in mind— though expressed in different ways. The former presented it piecemeal and in an exoteric way . . . The latter, however, presented everything systematically, coherently and esoterically, with many gaps filled, and countless truths added, which could be deduced as valid consequences of what went before.12

And Gottsched’s distinguished wife, Victoria, wrote the following verse to the Marquise du Châtelet in 1742: “what half the world from Leibniz came to know,/

9  Gottsched (1733–1734, 1: preface n.p.) also quoted in Wundt (1945, 121). 10  This is a point that Ursula Goldenbaum makes nicely in Chapter 2 of this volume. 11  To this end, Wolff (and Baumgarten) presented arguments that purported to demonstrate the Principle of Sufficient Reason. See Look (2011b). 12 Quoted in Döring (1999, 62–63), Preface to the 3rd volume of the translation of Bayle’s Dictionary.

Brandon C. Look   9 our great man Wolff did better still winnow.”13 Writing to Meiran in 1741, Voltaire, whose sympathies were always with Newton, Locke and other empiricist philosophers, runs the thought of Leibniz together with Wolff, damning both: Frankly, Leibniz just came to muddle the sciences. His insufficient reason, his continuity, his plenum, his monads, etc. are the seeds [germes] of confusion from which M. Volf [Wolff] has methodically drawn [a fait éclore méthodiquement] 15 quarto volumes, which will more than ever give German minds the taste for reading much and understanding little.14

By the end of the century, the close relation between Leibniz and Wolff had become standard. As the Baron von Eberstein put it: Leibniz’s excellent thoughts were still little used, and they were just scattered throughout his writings, not formed into a cohesive whole and brought together with other truths. In short, other than in Leibniz’s mind, there did not exist a metaphysics. But the builder of a system worthy of the name was Christian Wolff.  [Eberstein 1794, 1:123]

The story became official when Hegel presented his lectures on the history of philosophy and wrote, “On the whole Wolff ’s Philosophy is in its core Leibniz’s philosophy, except that he systematized it” (1971, 20:259). Kant himself, on numerous occasions, likewise spoke of a close family resemblance between the philosophy of Leibniz and the philosophy of Wolff. For example, in the opening of his polemic against Johann August Eberhard, who had claimed that the Critique of Pure Reason contained nothing that could not already be found in the writings of Leibniz, Kant countered, “How it came to pass that these things were not long ago already seen in the great man’s philosophy and in its daughter, the Wolffian, he does not, to be sure, explain” (Ak. 8:187/Kant 2002, 283). But Kant was aware of some important differences. His account of the Leibnizian monadology, for example, in the “Amphiboly” chapter of the Critique

13  “Und was die halbe Welt vom Leibnitz neu gelernet,/Hat unser großer Wolf noch besser ausgekörnet” (Gottsched 1763, 122). 14  Quoted in Cassirer (1998, 35); see original citation to Meiran 5 Mai 1741 (Voltaire 1820–1822, 58:119). Cf. his letter to Maupertuis, August 10, 1741 (ibid., 150–152). “That man [Wolff] brings back to Germany all the horrors of scholasticism overloaded with sufficient reasons, monads, indiscernibles and all the scientific absurdities that Leibniz brought into the world out of vanity and that the Germans study because they are Germans.” Voltaire’s friend and lover, the Marquise du Châtelet, also saw Wolff as following Leibniz, but she clearly had a much higher opinion of Leibniz than did Voltaire—advocating essentially a Leibnizian explanation of living forces rather than the solution offered by British and continental Newtonians. See Châtelet (1740, 12–14 and chs. 20 and 21). Voltaire and Châtelet eventually separated, having irreconcilable differences over Leibniz, Wolff, and presumably other matters.

10  Kant’s Leibniz: Historical and Philosophical Study of Pure Reason shows that he was careful not to conflate the two monadologies.15 Nevertheless, Kant did group Leibniz and Wolff together in crucial respects. He did not employ the term “rationalist”; in fact, the term “rationalist” and its cognates are rarely used by Kant.16 Rather, Leibniz and Wolff were both guilty of engaging in “dogmatic metaphysics,” by which Kant meant attempting to make claims about the supersensible without having first engaged in a critique of the powers of the human intellect. And the contrast that Kant draws throughout his Critical writings, the opposing schools that his philosophical system is designed to resolve, is between “dogmatism” and “empiricism.” The term “Leibniz-­Wolffian Philosophy” was commonplace in the century, thus solidifying the connection between the two thinkers. For his part, however, Wolff was uncomfortable with it and in his autobiography charged his own student, Georg Bernhard Bilfinger (1693–1750), with creating it (Wolff 1841, 142). But this is certainly not true. Bilfinger did discuss the commonality between Wolff and Leibniz on the issue of pre-­established harmony in his early works, but he also recognized differences between the two thinkers in a score of other areas.17 It is much more likely that the term derived from Wolff ’s opponents: Joachim Lange, Franz Budde, and Andreas Rüdiger.18 Whether or not Wolff ’s philosophy should be seen as the systematization of Leibniz’s philosophy, it is indisputable that anti-­Wolffians correctly saw similarities between Leibniz and Wolff on the crucial matters of faith and freedom. What first drew the ire of Pietist thinkers in Halle were Wolff ’s lectures on the natural religion of the Chinese, in which he argued that Chinese culture proved that it was possible to lead moral lives without a revealed religion.19 Such a view, then as now, was greatly disturbing to conservative religious thinkers, and the Halle Pietists argued that the philosophy of Wolff was opposed to Christian orthodoxy. Wolff tried to defend his view from such charge, but to little avail.20 To some degree, however, Wolff was merely continuing along the path that Leibniz had already established several decades earlier.21 For Leibniz played an important role in creating an interest in Chinese culture and the interaction between Jesuit

15  See Rutherford (2004, 215f.). 16  The Paralogisms of Pure Reason is directed against theses in the “rational psychology” of Wolff and Baumgarten—theses that Leibniz, too, endorsed. 17  See Bilfinger (1723), Bilfinger (1741), and Wundt (1945, 150 n.). 18  See Wundt (1945, 150 n.). See also Ludovici (1737 vol. 1, §136). That a group should be labeled by its opponents is, of course, all too common. The Pietists themselves owed their sobriquet to their opponents at the end of the seventeenth century (Hinrichs 1971, 1). 19  An excellent modern edition with a very helpful introduction can be found in Wolff (1985). 20  Wolff claimed, for example, in lectures and in his Theologia naturalis that there was no conflict between reason and revealed religion. But Joachim Lange took great pains to argue the contrary pos­ ition before his theology students (Hartmann 1737, 385–387). 21  It is not clear that Wolff ever read Leibniz’s Novissima Sinica, however; instead, it is likely that his interest in Chinese philosophy was piqued by the publication in 1711 of translations of six classical Chinese philosophical texts by the Jesuit François Noël (Noël 1711). See Wolff (1985, xxi ff.).

Brandon C. Look   11 missionaries and the Chinese people, publishing in 1697 the Novissima Sinica [The Latest on China], which concerned the Chinese rites controversy. His reflections on the nature of the Chinese religion led him to argue that Chinese culture demonstrated the independence of morality from the teachings of revealed religion. Put differently, Leibniz held that the practical moral dictates of Confucianism were identical to the practical dictates of Christianity.22 In the Theodicy, he also explicitly took up the issue of the relation between faith and reason, arguing that reason can never conflict with the word of the Bible when properly understood. Pietist thinkers, however, saw the issue differently and regarded Wolff ’s lectures as an affront to the Christian religion and the moral order. The more interesting philosophical issues concerned the scope of the principle of sufficient reason, freedom, and determinism and the relation between mind and body. And this philosophical problem-­complex certainly played as great a role in animating the dispute between Pietists and Wolffians.23 The Pietists demanded that the soul freely exercise causal influence on the body, and they saw the doctrine of pre-­ established harmony as inconsistent with any kind of orthodox libertarian doctrine. Indeed, in his writings against Wolff, Lange put Leibniz, Wolff, and Spinoza in the same camp and argued that all three philosophers embraced a kind of fatalism or necessitarianism (Lange 1723, 65–66).24 Spinoza, of course, would happily have accepted this claim. Leibniz, however, developed a rather involved metaphysical system that allowed him to endorse a kind of compatibilism; yet much of the philosophical underpinning of this more complex defense was unknown in this period. For example, no philosophers of the eighteenth century could have been familiar with the kind of reasoning that Leibniz employed in his correspondence with Arnauld or in the Discourse on Metaphysics which appealed to a “complete individual concept” for each substance.25 Wolff, for his part, rejected any hint of Spinozism, affirming that the world came about through an act of divine will; he further sought to argue that the doctrine of pre-­ established harmony (with substance dualism) was the best defense against the kind Spinozistic fatalism the Pietists so feared, for the soul always acted according to its own laws.26

22  For more on this issue see Li and Poser (2000), Riley (1999), and Leibniz (1994). 23  See Watkins (1998) and Goldenbaum in Chapter 2 of this volume. 24  But as Wundt has remarked, while Lange railed against Leibniz’s system in his Modesta disquisitio, he cited almost exclusively the work of Wolff as evidence of Leibniz’s pernicious doctrine (Wundt 1945, 150 n.). 25  In the Theodicy, Leibniz did present an account of freedom, according to which an agent is free if and only if the contrary of a particular action does not entail a contradiction (that is, if and only if there is another possible world in which the agent (or likeness) does otherwise). See the final sections, §§410ff., in which Leibniz discusses the different Sextuses in different possible worlds. This view is directly related to Leibniz’s earlier view, but the logical conception of substance is notably absent in this popular work. 26  See Wolff (1737a II:3); Wolff (1737b, 14–17).

12  Kant’s Leibniz: Historical and Philosophical Study Despite the common narrative about the “Leibniz-­Wolffian philosophy,” the differences between the two thinkers on a number of issues were known to many in the eighteenth century, at least to those who cared to look. Wolff highlighted many of these differences himself, and other scholars at the time were keen to point out these differences as well.27 Ludovici, for example, who attempted to write one of the first histories of Leibniz’s philosophy, Ausführlicher Entwurff einer vollständigen Historie der Leibnitzischen Philosophie [Detailed Outline of a Complete History of the Leibnizian Philosophy], urged readers to distinguish between Wolff and Leibniz: “one will find the true philosophical system [Lehrgebäude] of Leibniz not at all in the writings of Wolff, or at least only in pieces and very imperfectly expressed.”28 And as Formey put it in his Histoire Abrégée de la Philosophie, Wolff “profited from the ideas of Leibniz; but he did not completely follow them, and he put many of his own thoughts into his system.”29 It might serve to give a concrete example of the way in which Leibniz’s mature philosophical views were obscured from the view of the philosophically minded reading public by Wolff and his followers. While it is by no means a settled issue in Leibniz scholarship, it is fair to say that, according to the standard in­ter­pret­ ation, Leibniz was an idealist.30 That is, he believed that the world was grounded in the mind, or that the mental or the formal was alone fully real. As he expressed it in a letter to Burcher De Volder in 1704, “considering the matter carefully, we must say that there is nothing in things but simple substances, and in them, perception and appetite” (G II 270, AG 181). In other words, only simple substances, or minds, truly exist; bodies are phenomena, though they are grounded in the simple substances. The simple substances are themselves unities endowed with forces, which in turn can be understood in terms of representational activity and their capacity to express or mirror the entire world from a unique point of view. For Leibniz, however, not all representative or expressive states are conscious sensations; indeed, individual substances simply express the world, though often, even mostly, unconsciously. While Wolff also claimed that the ultimate constituents of the world were simple substances or monads, it is clear that his monads did not have the mental, dynamic, or representational features that Leibniz’s did; he denied the thorough-­going and universal mirroring of simple substances, down to the least petite perception, and in so doing rejected Leibniz’s monadology and a more sophisticated version of pre-­established harmony.31 In the end, Wolff ’s 27  The differences between Wolff and Leibniz are very important to bear in mind in trying to understand the reception of Leibniz and the history of Leibnizianism in the eighteenth century. This issue has been discussed by many scholars of the period—e.g. Tonelli (1963, 1966, 1974, 1987); Beck (1969, 1993); Wundt (1924, 1945); École (1979, 1998) and, more recently, by Jauernig (2008, 2011). See also Goldenbaum (2004, 27–28). 28  Ludovici (1737 II 427). 29  Formey (1760, 297). 30  The literature on this subject is extensive. I present an introduction to the debate and defense and interpretation of a kind of Leibnizian idealism in Look (2010). 31  See e.g. Wolff (1737c, §243 and 1720, §§598ff.). See also Erdmann (1876, 63).

Brandon C. Look   13 simple substances were not so much metaphysical points as they were physical monads, the least ingredients in material beings. And there was thus a great deal of truth in his comment to his benefactor Ernst Christoph von Manteuffel that Leibniz’s system of monads “begins where mine ends” (Wolff 1841, 82). Wolff and his students, however, largely prevented the Leibnizian monadology from being interpreted in what we might now think of as an idealistic manner. In his Dilucidationes philosphicae [Philosophical Explanations] (1725), Bilfinger defined idealism as the thesis that there is an infinite spirit and that finite spirits are dependent upon it, but that nothing else exists beside spirit. Further, for Bilfinger, spirits are simple beings endowed with intellect and will; they are immortal and capable of reward and punishment. Bodies, on the other hand, have no real existence outside of us; they are simply thought by us to have real existence because they are represented as existing outside of us and because these representations follow a constant order (§115). Now, according to Bilfinger, Leibniz did not advocate idealism for two obvious reasons. First, while Leibniz’s simple substances are endowed with perception, they are not all endowed with intellect and will, because intellect and will require distinct cognition (§110). Thus, for Leibniz, it is false that all beings are spirits—even if all simples are essentially mind-­like. Second, Bilfinger claims that idealism entails the view that there is nothing real and independent to which our mental representations correspond. But since Leibniz’s view is that each mind represents other monads— that is, other real existents, albeit confusedly and as bodies—it is nevertheless the case that Leibniz’s representations correspond to something real. But it should be clear from this account that one could be both an idealist in our sense and a non-­ idealist in Bilfinger’s sense—as, I would argue, Leibniz in fact was. That is, one could claim that the ultimate beings are mind-­like, incorporeal beings, endowed with representations, while also holding that representations do correspond to things independent of a perceiving mind. Indeed, on my view, the correct interpretation of Leibniz’s metaphysics—from 1679 on—is one in which minds or the mental or the formal ground all other beings.32 Wolff also argued that Leibniz was not an idealist—though he did so perhaps for self-­serving reasons. For Wolff, dogmatic thinkers can be classified either as monists or dualists, the former in turn either as idealists as materialists (which is what Leibniz said as well). But, insofar as Leibniz’s theory of pre-­established harmony requires that bodies be recognized as real and distinct from minds, then Leibniz must be a kind of dualist.33 As has been pointed out already, Wolff himself disliked the expression “Leibniz-­Wolffian philosophy,” and he correctly saw that the crucial aspect of 32  I argue for this view in detail in Look (2010 and 2017). 33  See the Preface to the Vernünftige Gedanken von Gott, der Welt, der Seele des Menschen, auch allen Dingen überhaupt (usually called the German Metaphysics) (that is, Wolff [1720]). Of course, neither Bilfinger nor Wolff had the texts that make the arguments most forcefully for Leibnizian idealism (in our sense of the term), e.g. the correspondences with De Volder and Des Bosses.

14  Kant’s Leibniz: Historical and Philosophical Study Leibniz’s system was the monadology, which, to the extent that he understood its idealistic underpinnings, he rejected.34 At the same time, however, Wolff did advocate the system of pre-­established harmony as a solution to the problem of mind–body causation. Wolff ’s solution to this metaphysical dilemma was to de-­ emphasize the idealistic tendencies in Leibniz’s thought, seeking to make Leibniz into a dualist like himself. Further, since he advocated a substance dualism of mind and body, Wolff was left to endorse a system of pre-­established harmony that resembled Leibniz’s popular presentations of his own theory. And the same was true of his students. For example, in his De harmonia animi et corporis humani, maxime praestabilita, ex mente illustris Leibnitii, commentatio hypothetica [Hypothetical Commentary on Leibniz’s Pre-­ established Harmony of the Soul and Human Body] (1723), Bilfinger demonstrates a level of knowledge of the actual historical debate about the doctrine that Wolff rarely showed himself, but he nevertheless cannot get behind the Leibnizian façade. He gives an argument for the “metaphysical union” of mind and body that Leibniz had claimed to exist in his discussion with Father René-­Joseph de Tournemine, and in doing so, offers a perfectly respectable Leibnizian defense of pre-­established harmony. For example, Bilfinger says the following in a footnote to his discussion of Tournemine’s objection that the “harmony” of mind and body can never constitute a real union: It is impossible that a true physical relation or union exist between our soul and body, unless the soul is also a body. The physical does not pertain to anything but body. You say, what is physical? But whatever is common to soul and body is itself metaphysical. Therefore, the union too, which is common to them, is metaphysical. For metaphysics is just that which is goes beyond the spirit and the body.35

Bilfinger’s explicit argument is that, for the mind and body to form a “real, physical union,” the mind would have to be corporeal. Since it is not, the mind and body can have at most a metaphysical relation between them. And the relation is metaphysical, he claims, because it extends beyond the spiritual and corporeal realms. If one takes the Leibnizian language of minds and bodies at face value—as Tournemine, Wolff and Bilfinger all did—then this is as good a response to Tournemine as one can give. What Bilfinger could not have known is that the problem that Tournemine highlighted is one that Leibniz took quite seriously— though only in its extension to the realm of monads. In a draft of a letter to Des Bosses in 1706, just as his debate with Tournemine was going to press, Leibniz 34  See Rutherford (2004). Wolff was also no advocate of Leibnizian optimism as expressed in the Theodicy. 35  Bilfinger (1741, 218), reprinted in GW III 21).

Brandon C. Look   15 writes, “The union I find some difficulty explaining is that which joins the different simple substances or monads existing in our body with us, such that it makes one thing from them” (LDB 23). But, as I have argued elsewhere, this is a problem that plagues Leibniz for the last decade of his life.36 German philosophy in the first decades after Leibniz’s death became a partisan affair. By 1736 there were 126 polemical attacks on the “Leibniz-­ Wolffian” philosophy; Wolff countered with 14 separate defenses of his views, and his students and followers contributed another 68 works.37 Whatever differences there may have been between Leibniz and Wolff, they were clearly on the same side, fought most vociferously by Pietist theologians and philosophers, who denied the principle of sufficient reason, pre-­established harmony, and the conformity of faith and reason. This dispute between rationalists and Pietists is central to the rest of the century in German intellectual life, and the later culture war between Enlightenment and Counter-­Enlightenment thinkers is in some ways presaged in it.38 The philosophical landscape in Central Europe was also becoming even more complex as the empiricism of Locke and scientific methodology of Newton were making headway in Germany. Thus, Wolff and his supporters found themselves fighting opponents on several different fields of battle.39 This complexity also meant that Kant could eventually carve out a position as both an Enlightenment philosopher and an opponent of Leibnizian and Wolffian rationalism.

4  The Philosophical Scene in Mid-­century Germany As a matter of institutional power, Wolff ’s philosophy predominated in most German universities. Although Wolff ’s works were officially banned in Brandenburg-­Prussia from the time of his expulsion from Halle, they were widely used throughout the rest of the German-­ speaking world. And his students occupied positions at nearly all major universities. Moreover, by the 1730s Frederick William I evidently had grown tired of the Pietist attacks on Wolff, for he removed the ban on Wolff ’s philosophy and even tried to lure the most prominent German philosopher back to his lands from the Landgraviate of 36  See Look (1999). 37 Hartmann (1737, 835ff.). Of these anti-­ Wolffian polemical attacks, many were by non-­ philosophers, and they lacked a certain degree of philosophical sophistication. This sophistication will come, however, in the works of Crusius and others in the next generation. 38  At the same time, insofar as Pietism was individualistic and anti-­authoritarian, it can also be seen as also contributing to demise of ancien régime Europe. 39  Christian Thomasius and his followers were also important in the eighteenth-­century German philosophical world. (See e.g. Beck [1969] and Hunter [2001]; Wundt [1945], on the other hand, claims that Thomasius had relatively little impact on German philosophy until his ideas were taken up by counter Enlightenment philosophers in the 1780s.) However, as their views are not directly relevant to the way that Kant came to understand the philosophy of Leibniz, I shall ignore this strand of history here.

16  Kant’s Leibniz: Historical and Philosophical Study Hesse-­ Kassel, where Wolff had been for over a decade. Wolff refused the invitations, citing his gratitude to the university in Marburg and the difficulties of such a move, but upon the accession of Frederick II in 1740 Wolff was successfully coaxed back to Halle to take up the chancellorship of his old university. By all accounts (including Wolff ’s own), crowds greeted him like a hero upon his arrival in the city.40 There then began what can best be thought of as a second wave of Wolffianism. In many works of academic philosophy that would become central to Kant’s own studies and teaching, Wolff ’s philosophy was spread and further developed. A short list of such works will be familiar to any Kant scholar: Gottsched’s Erste Gründe der gesamten Weltweisheit [Foundations of a Complete Philosophy] (1733); Baumeister’s Institutiones philosophiae rationalis [Principles of Rational Philosophy] (1735) and Institutiones metaphysicae [Metaphysical Principles] (1738); Baumgarten’s Metaphysica (1739) and Aesthetica (1750); and Meier’s Vernunftlehre [Doctrine of Reason] (1752). In these and other writings of the period Wolff ’s influence is on clear display. Comparatively few academic philosophers used as a springboard for further reflection the actual writings and thought of Leibniz.41 At the same time, some philosophers started to push back against this school, advancing arguments far more thoughtful and interesting than those given by the majority of Pietist polemicists. Certainly the deepest of these authors was Christian August Crusius (1715–1775), whose Dissertatio de usu et limitibus principii rationis determinatis, vulgo sufficientis [Dissertation on the Use and Limits of the Principle of Determining Reason, Commonly known as the Principle of Sufficient Reason] (1743) and Entwurf der nothwendigen Vernunftwahrheiten wiefern sie den zufälligen entgegengesetzt werden [Outline of the Necessary Truths of Reason, insofar as they are Opposed to Contingent Truths] (1745) among other works opposed doctrines central to both Leibniz and Wolff and had a great impact on the young Kant.42 Indeed, in several important respects Crusius laid the groundwork for many of Kant’s central views, in asserting the fundamentality 40  Hinrichs (1971, 440); Wolff (1841, 167–170). 41  There are also, however, differences between these philosophers and Wolff, which are sometimes important. In a fascinating article, École (1991) argues that many of Kant’s claims about Wolff ’s phil­ oso­phy come from his reading of Baumgarten, Baumeister, and Gottsched and not directly from Wolff ’s texts themselves; for when Kant misrepresents Wolff, it is usually in a manner and on a topic that can be found in one of these other authors. As should be clear, I believe the same is sometimes true of Kant’s claims about Leibniz’s philosophy. It should also be pointed out that, while his Metaphysica was clearly written in the spirit of Wolffian philosophy, Baumgarten was not simply a Wolffian rationalist. First, he actually presented views that were closer to Leibniz’s than Wolff ’s. Second, he came from a strongly Pietist background and at the end of his life seemed to return to the fold, claiming on his deathbed that only Christ comforted his soul and “neither the philosopher nor the theologian could help, only faith alone.” “My old faith, with this I depart, is the demonstratio demonstrationum . . . ” Mendelssohn saw this turn as a “misological death” and inexcusable. (Quoted in Baumgarten [2011, xxviii–xxix].) I examine Baumgarten in more detail in Look (2018). 42  In Chapter 3 of this volume, Eric Watkins addresses the role of Crusius’s thought in Kant’s philosophical development.

Brandon C. Look   17 and power of the will, in limiting the scope of the principle of sufficient reason, and in defending a theory of real causal interaction between things. While Wolffianism was generally dominant throughout Germany, the situation in Königsberg was more complicated. The university had strong Pietist elements from the turn of the century on, and even had an eclectic mix of Scholasticism and modern philosophy as attested to by Gottsched as we saw above. While a Wolffian professor, Christian Gabriel Fischer (1686–1751), was forced to leave the university under Pietist pressure in 1725, less than a decade later the university came to tolerate Wolff ’s philosophy (Erdmann  1876, 19). In fact, the Pietist professor of theology, Georg Friedrich Rogall (1701–1733) required his own students to pass the cursum philosophicum and, in so doing, ultimately encouraged Wolffianism among the Pietists (Erdmann  1876, 21). On the other hand, the philosophers in Königsberg, who in many ways stood close to Wolff, also sought to correct and improve his philosophy. A clear case in point is Martin Knutzen, Kant’s influential teacher, who offered one of the most important critiques of the doctrine of pre-­established harmony in his Systema causarum efficientium [System of Efficient Causes] (1745) while at the same time remaining committed to central tenets of Leibniz and Wolff. Given Knutzen’s commitment to Newton, his philosophy was genuinely eclectic. Universities were not the only prominent institutions for philosophy, science, and the arts. Frederick II also decided to revive the Royal Academy of Sciences, which had been in decline since Leibniz’s death under his less-­than-­intellectual father, Frederick William I, the “Soldier King.”43 At the suggestion of Voltaire, Frederick invited Pierre Louis Maupertuis (1698–1759), one of the most famous natural philosophers of the day, to serve as President of the Academy. Frederick then turned to a younger man to aid Maupertuis in the operation and development of the Royal Academy, Leonhard Euler (1707–1783), who at 33 was already known as the foremost mathematician on the continent. Both Maupertuis and Euler were highly critical of Leibniz and Wolff, and they used the power of the 43  While a common story is that Frederick William I named his court fool to be Leibniz’s successor as head of the Royal Academy, this is not exactly right. Jacob Paul von Gundling (1673–1731), the second president, was a trained jurist and historian and former professor of the Ritterakademie in Berlin, who had developed a reputation in Berlin for his amusing stories and who in 1713 was made a court councilor. While not entirely a figure held in deep respect by the king, he nevertheless was given a number of other posts along with the presidency of the Academy. And Gundling served the Academy well by requiring that one copy of every book published in the Kingdom of Prussia be housed in the library of the Royal Academy. On the other hand, it has been suggested that Gundling, whose brother was a Pietist professor at the University of Halle, played a role in Christian Wolff ’s banishment from Brandenburg-­Prussia, since it was he who had joked that Wolff ’s determinism would give soldiers an excuse for desertion (Zeller 1862, 65–66). In the end, however, the fact that he came to be so ridiculed at the court probably had as much to do with Gundling’s alcoholism as with Frederick William’s anti-­intellectualism. When Gundling died, the king had him buried in a wine barrel. See Harnack (1900, 220ff.). Hinrichs (1971, 417) rejects the idea that Gundling played a role in the Wolff affair on the grounds that Gundling said he did not do so, which can hardly count as probative.

18  Kant’s Leibniz: Historical and Philosophical Study Berlin Academy to advance their philosophical views. The reasons for Euler’s opposition to Leibniz and Wolff were quite complex. On the one hand, Euler certainly saw a danger in the doctrine of pre-­established harmony, for the very reasons that the Pietists had done. On the other hand, he was also a committed Newtonian, advocating the laws of motion as expressed in the Principia as well as absolute space and time.44 Maupertuis also had deep misgivings about the way in which Leibniz’s natural philosophy and mathematics—specifically his dynamics— were employed by his supporters. But Maupertuis himself may have been just as much caught up in the anti-­Leibnizian spirit of the day as actively contributing to it. Indeed, there is a certain irony to Maupertuis’s position, for in many crucial aspects his own thought was very much in harmony with that of Leibniz. Like Leibniz, Maupertuis endorsed a principle of continuity in physics and also, against the Newtonians, favored a relational or phenomenalistic conception of space and time. Moreover, his debate with Samuel König on the principle of least action reveals his complex relation to Leibniz quite well.45 The Academy played a central role in the philosophical and scientific world in the middle decades of the eighteenth century by sponsoring regular essay competitions. But these essay competitions not only reflected current consensus about the important philosophical issues through their choice of topics, they also arguably sought to delegitimize Wolffian philosophy through the selection of winners. In the first three competitions of the philosophical class46—1747, 1751, and 1755, when Maupertuis and Euler had the most influence over the Academy— the topics clearly confronted the philosophy of Leibniz and Wolff.47 The 1747 essay question concerned the nature of monads; authors were asked to prove or refute the “doctrine of monads,” and in the case of a positive view of monads “to deduce an intelligible explication of the principal phenomena of the universe, and in particular of the origin of the movement of bodies.” (Harnack 1900, 2:305) The winner was Johann Heinrich Gottlob Justi, who submitted a work critical of the theory of monads in which he largely rehearsed the anti-­monadist arguments published the previous year by Euler in his Gedancken von den Elementen der Cörper [Considerations on the Elements of Bodies].48 In 1751, the Academy essay question focused on free will and determinism, and in this case Abraham Gotthelf Kästner, a professor of mathematics from Leipzig and supporter of Wolff, was surprisingly declared the winner. Four years later the Academy turned to the issue of philosophical optimism. While the official question explicitly cited 44  Euler’s views on the nature of space and time are complex, and so it is probably too simplistic to simply call them Newtonian. But he certainly rejected Leibnizian relationalism. 45  See Cassirer (1998, 89); Harnack (1900, 252ff.); Beck (1969, 317–319). 46  The Academy was divided into four classes—medical/physical, mathematical, philosophical, and philological—and each class took a turn posing an essay question. 47  Maupertuis left his position as President of the Academy in 1753, but after the prize essay question was posed. 48  This work can be found in Euler (1911– , III, 2, 349–366).

Brandon C. Look   19 Alexander Pope’s Essay on Man, it was clear that Leibniz’s Theodicy was the target. Gottsched called out the Academy immediately on the ruse in a short essay, De optimismi macula [Optimism Stained] (1753). And the winning submission by Adolf Friedrich von Reinhard made the connection in its title, Le Système de Mr. Pope sur la perfection du monde, comparé à celui de Mr. de Leibnitz [Pope’s System of the Perfection of the World compared with that of Leibniz], and savaged the optimism of both Pope and Leibniz. Ultimately, this ploy of the Royal Academy backfired, for it also occasioned a biting work critical of the question and the only work still read in connection with the essay question: Lessing and Mendelssohn’s Pope ein Metaphysiker! In it they argue that the work of the poet and the metaphysician fundamentally differed and that a real poet neither wished to develop a metaphysical system nor could he do so even if he so wished. Thus, they claimed it was clear that what the Academy really wanted was a rejection of Leibnizian optimism. More important, their analysis of Leibniz’s system also displays a penetrating reading of the Theodicy. As a young intellectual with scholarly and philosophical ambitions, Kant could not have been ignorant of these matters, nor could he have been ignorant of the general anti-­Leibnizian and anti-­Wolffian sentiment issuing from the capital.49 The period from the end of Leibniz’s life to the zenith of Kant’s philosophical career also witnessed a profound change in German intellectual and cultural life. Christian Wolff played an important positive role in this change. Not only was he one of the first professors to lecture in German, he was also one of the first to publish significant works in the vernacular.50 Moreover, he was largely re­spon­ sible for the development of German philosophical vocabulary. Wolff ’s influence was perhaps also so strong because, after a century in which the dominant philo­ sophers in Europe worked outside of the university milieu, he was a professor who successfully appealed to the insights of the modern philosophers and thereby attracted a stream of students who went on to occupy many other professorships. Thus, for both linguistic and institutional reasons, Hegel was correct to call Wolff “the teacher of the Germans.”51 At the same time, there began to arise in German-­speaking lands a cultivated middle class, interested in literature, philosophy, and the sciences. At the Leipzig Book Fair in 1701, for example, there were nearly 1,000 titles listed in the catalog, nearly half of which dealt with theological matters. Eighty years later the number of titles had increased to 2,600, with a far smaller proportion concerning religion and theology. Moreover, the ratio of books in German to those in Latin increased

49  Indeed, Kant’s writings on optimism from the 1750s were occasioned by the essay competition, though he never submitted a work for consideration. See Ak. 2:27–35 and Ak. 17:229–239, also in Kant (1992b, 67–83). 50  Christian Thomasius was also very important in this respect. 51  Hegel (1986, 20:258).

20  Kant’s Leibniz: Historical and Philosophical Study from approximately 2:1 to 10:1.52 By mid-­century there were in Germany, as throughout the continent, ever more intellectual journals aimed at a popular and not necessarily academic audience. “Popularphilosophen” both created and responded to a desire for a critique of religion and the political power structures of the day.53 While many popular philosophers advocated doctrines close to the empiricism of Locke and skepticism of Hume, many more advocated a kind of rationalism close to that of Leibniz, Wolff, and Wolff ’s followers. While the topics of academic philosophy—for example, concerning pre-­established harmony and physical influx or concerning innate ideas and empiricism—were not at the forefront of these discussions, all shared a deep commitment to the power of human reason and its ultimately liberating effects. Moses Mendelssohn (1729–1786) was the best of such Popularphilosophen. Lacking a university position, he made his living outside of professional, academic philosophy as a bookkeeper and, more importantly, as a writer of essays, reviews, and books.54 But to call Mendelssohn a “popular philosopher” should not give the impression that his work was somehow shallow—on the contrary. Mendelssohn’s writings were penetrating and interesting, and he made important advances in the rationalist positions of Leibniz and Wolff, especially in aesthetics.55 Unfortunately, he resides now mainly in the shadow of Kant. While his Phaedon oder über die Unsterblichkeit der Seele [Phaedo, or On the Immortality of the Soul] (1767) achieved great success and was one of Goethe’s favorite books,56 it is now best known to students and historians of philosophy as the work refuted in the Paralogisms chapter of the Critique of Pure Reason. His essay Abhandlung über die Evidenz in metaphysischen Wissenschaften [Essay Concerning Evidence in the Metaphysical Sciences] (1764) won the Academy’s prize essay after its anti-­ Wolffian phase had ended and defeated Kant’s submission Untersuchung über die Deutlichkeit der Grundlage der natürlichen Theologie und der Moral [Inquiry Concerning the Distinctness of the Principles of Natural Theology and Morality].57 Of Mendelssohn’s essay, Beck writes, “No other single work gives so perspicuous a presentation of the Leibniz-­ Wolffian epistemology; every strength of that tradition is persuasively presented, every fault in it inadvertently revealed” (1969, 332). Kant certainly read this work, but if he saw in it what Beck saw, then he might have been led astray, for Mendelssohn composed this essay prior to the publication of Leibniz’s Nouveaux Essais, which reveals a difference between the epistemological positions of Leibniz and Wolff. Indeed, the publication in 1765 of Raspe’s edition of Leibniz’s New Essays along with Dutens’s six-­volume collection of Leibniz’s writings three years later led to a 52  Ward (1974, 29ff.). 53  The period after the publication of the first edition of Kant’s Critique of Pure Reason is masterfully discussed in Beiser (1987). 54 Mendelssohn was nominated several times to become a member of the Royal Academy of Sciences in Berlin, but his membership was blocked every time by Frederick II. 55  See Beiser (2009). 56  See Lewes (1855, 1:213). 57  See Ak. 2:273–301.

Brandon C. Look   21 wave of interest in Leibniz and a deeper understanding of his philosophy. As Max Wundt notes, “the true Leibniz was to this point not really known” (1945, 317). Not only Mendelssohn, but also Lessing and others, came to a greater appreciation of the depth and breadth of Leibniz’s metaphysics, epistemology, and philosophical theology. For example, Lessing’s beautiful essay, Leibniz von den ewigen Strafen [Leibniz on Eternal Punishment] (1773), shows that he had studied Leibniz’s phil­ oso­phy carefully, most likely from Dutens’s edition.58 Lessing argues that Leibniz always sought to lead the reader to the truth from the path he was already on and that the distinction between Leibniz’s “exoteric” and “esoteric” philosophy had to do with his manner of presentation and not with the content. But this line of argument required that Lessing know what Leibniz said in his popular publications as well as what he wrote in more guarded moments and that Lessing be able to discern the connections between the two sets of writings. It is also meant that Lessing saw that a superficial reading of the popular presentations of Leibniz’s writings could lead readers astray. Another careful reader of Leibniz was Herder, who was able to see in these new editions something importantly different between Leibniz and Wolff in both style and substance: Since the lively mind of this man [Leibniz] liked to see everything as a hypothesis and present it half as a poem, even his monads, which Wolff himself seems not to have fully grasped, were considered just a funny tale. But I am convinced that . . . this hypothesis is the most rigorous and certainly will someday triumph.59

Herder’s prediction for the triumph of Leibnizian monads, of course, proved wrong, but his description of the hypothetical and nearly lyrical manner of Leibniz’s writings was right. By the time the Romantics came upon the scene in Germany, Leibniz was revered precisely because he had never presented a ponderous system of the sort that Wolff and his students had done.

5  Kant as a Reader of Leibniz Ideally, scholars interested in Kant’s relation to Leibniz would have a set of Leibniz’s works with detailed marginalia in Kant’s hand or simply extended reading notes on Leibniz’s philosophy—precisely what we have in the case of

58  For this essay, see Lessing (1886, 11:461–487). It should also be noted that, in it, Lessing took to task none other than Johann August Eberhard, Leibniz’s later advocate against Kant, for his poor understanding of Leibniz in his Neue Apologie des Sokrates (1772). 59  See the second dialogue of his Gott. Einige Gespräche (1787) in Herder (1994, 4:709). Somewhat later in the same text, Herder also writes of the New Essays that they are “just about the most in­struct­ ive of all of the writings of Leibniz, from whom incidentally every line is instructive” (1985, 4:731–732 n.). Herder, of course, had been a student of Kant’s, and they maintained a relatively active cor­re­ spond­ence. I have not been able to find any writings in which Leibniz is discussed, however.

22  Kant’s Leibniz: Historical and Philosophical Study Leibniz’s study of Spinoza. We do have, however, a catalog of Kant’s library, which offers some tantalizing clues about the sources of his knowledge of his philosophical predecessors (Warda  1922). Kant owned, for example, works by many of great thinkers of the seventeenth and eighteenth centuries: Berkeley, Boyle, Cudworth, D’Alembert, Descartes, Galileo, Gassendi, Hume, Hutcheson, Maupertuis, More, Newton, Tschirnhaus, and Voltaire.60 He also had a good collection of works from the prominent philosophers of eighteenth-­ century Germany, many of which he used in his own university teaching: Baumgarten, Crusius, Euler, Gottsched, Lambert, Meier, Mendelssohn, Tetens, and Wolff. Kant also owned many books that dealt with mathematics, natural philosophy and natural history, which is again unsurprising given his teaching duties. Absent from his library, however, are works by Leibniz, Locke, and Spinoza. Of course, as all scholars and bibliophiles know, one often has books on one’s shelves that one has never read, and one often has read many books that one never owned.61 Indeed, on Kant’s death, it was remarked that his library was comparatively small for an academic. It is likely also that Kant’s relation to books differed from that of Leibniz’s.62 Leibniz was not only born into a family prominent in Leipzig’s academic and legal circles and had free rein in his father’s extensive library from the age of 8, but his official duties in Hanover included serving as the court historian and court librarian. For close to forty years, he kept an apartment in whatever building housed the court library, and for the last eighteen years of his life, after the library had been moved to Schmiedestraße 10, one could say that the royal library was his house. Leibniz inherited in 1676 a good library, with over 3,000 volumes, and he used his position over the years to accumulate many more philosophical and scholarly titles (Antognazza 2009, 201 and 383). Leibniz clearly felt possessive of these volumes too, for many contained his notes written in the margins. Kant, on the other hand, came from a much more modest background, with few books in the household of staunchly Pietist parents, and as a student Kant is said to have bought few books. Unsurprisingly, though, he was also known as a voracious reader. In order to supplement his income from teaching, he took on the position of sub-­librarian in the Schloßbibliothek, essentially the university library, in 1766 and worked there for seven years, even after he was appointed Professor at the University of Königsberg. There he had easy access to many 60  The contents of Kant’s library also strongly suggest that, although he knew French, Kant preferred to read texts in German translation rather than in the original. Indeed, Waschkies has argued that Kant never refers to an original French-­language text when there is a German translation (Waschkies 1987, 541f.). On Kant’s limitations with French, one of his contemporaries recalled, “Of the modern languages, he understood French, but did not speak it” (Jachmann 1804, 41). 61  A case in point: as has been mentioned, there is no doubt that Mendelssohn’s Phaedon is a target in the second edition of the Paralogisms chapter (B 395ff.), but this book was not in his library at auction. 62  There is a great deal of truth in T.S. Eliot’s observation that “Leibniz’s originality is in direct, not inverse ratio to his erudition” (Eliot 1916, 568).

Brandon C. Look   23 books in all subjects. The added income from this new job also allowed Kant to take up new rooms in the house of his publisher, Kanter, a house that was the location of one of the best bookstores in Prussia, where it was common practice to peruse, read, and discuss the latest works of science, history, literature, and philosophy. Kant was allowed to borrow all the books he wished and read them in his own apartment (Kuehn 2001, 159–160). Even after he was relatively financially secure and had his own house, it was not Kant’s practice to acquire books as it was Leibniz’s. Therefore, evidence for what Kant knew and of how he understood Leibniz will largely have to come internally, from determining what topics he discussed and how and by following up on his relatively rare references to other philosophers. Kant was surely exposed to the standard Leibnizian texts known in mid-­ century German philosophy: the Specimen Dynamicum; the New System of Nature; On Nature Itself; the Theodicy; the Principles of Nature and Grace, which was published by Gottsched along with the Theodicy in German translation in 1744; the Monadology; and the Leibniz–Clarke correspondence.63 Nevertheless, his view of the Leibniz’s philosophy was also shaped by the work of Wolff, Baumgarten, and Meier, philosophers Kant used in his teaching, and this is often reflected in his writings. Thus, disentangling Kant’s comments about Leibniz and Leibnizianism often proves to be quite difficult. Kant’s pre-­Critical writings unsurprisingly demonstrate an engagement with many of the pressing philosophical issues of the day. And, as we have seen, these issues often related to the philosophy of Leibniz, the philosophy of Wolff and tensions between growing Enlightenment and counter-­Enlightenment thought. A few examples here might suffice to show which texts Kant knew and knew well. Kant’s very first publication Von der wahren Schätzung der lebendigen Kräfte [On the True Estimation of Living Forces] from 1747 addressed the vis viva dispute between Leibnizians and their opponents, and in the opening section, Kant quotes from Leibniz’s Specimen Dynamicum, the locus classicus for Leibniz’s account of vis viva, on the nature of extension (Ak. 1:17).64 Likewise, Kant’s work 63  While the Meditations on Knowledge, Truth and Ideas is a very important text in the Leibnizian corpus, a work published in Leibniz’s lifetime and to which he makes frequent references in his later writings, it was not republished until the edition of Dutens and was known principally through Wolff ’s reworking of Leibniz’s theses in his German Logic. I am tempted to bracket the New System of Nature as well, for Kant shows no signs of having studied this text; his knowledge of the pre-­established harmony more likely come from his reading of the Theodicy and the presentations of Leibniz’s phil­oso­phy by Wolff, Hansch, Bilfinger, and Baumgarten. This fact is of some importance because, as Leibniz scholars know, the New System also foreshadows the monadology to come and contains hints of some of Leibniz’s more esoteric doctrines. While Daniel Garber (2009) has argued that the New System does not contain a proto-­monadology, if one looks at the work of Hansch (1728), for example, who wrote shortly after Leibniz’s death, one finds monad, substantial form, soul, and entelechy all used largely synonymously. Leibniz himself treats these terms as largely equivalent as late as 1710 in the Theodicy. 64  However, Ursula Goldenbaum argues in Chapter 2 of this volume that Kant’s arguments about living forces—inspired by the debate about Leibnizian vis viva—are directed less against actual Leibnizian texts and more against the Institutions de Physique (1740) of Émilie du Châtelet.

24  Kant’s Leibniz: Historical and Philosophical Study on optimism intended for the Prize Essay question of the Royal Academy clearly indicates that Kant knew Leibniz’s Theodicy, by then available in German translation. Yet, many of Kant’s reflections on Leibnizian issues can just as easily be traced to works of Wolff and Wolffians. For example, the early Nova Dilucidatio [New Elucidation] (1755) contains a criticism of the Principle of the Identity of Indiscernibles (Ak. 2:277), which could come either from the Monadology or from Leibniz’s correspondence with Clarke or from any number of Wolff ’s books. While the brief mention of the ars combinatoria suggests an acquaintance with Leibniz’s texts, it is far more likely that Kant’s knowledge derives from the reference in Christian Wolff ’s German Metaphysics (§324). Similarly, Kant’s essay Regions of Space (1768) clearly picks up on issues from the Leibniz–Clarke correspondence, but his references to the Leibnizian idea of analysis situs in the opening paragraph (and in his Inquiry [1763]) should not lead us to conclude that he was drawing on Leibniz. Few (if any) of Leibniz’s writings on this topic were extant at this time, and it is thus much more likely that he knew of this from Wolff ’s Elements of Universal Mathematics, the text that Kant used in his teaching of mathematics. There Wolff briefly mentions the doctrine and gives a few indications of its content.65 Moreover, even in the Physical Monadology (1759), Kant makes no statements about ontology that show any kind of deep knowledge of Leibniz’s metaphysics beyond what one could glean from the presentations of Wolff and Baumgarten. Kant’s Dreams of a Spirit-­Seer (1766) gives another clue to the sources of Kant’s understanding of Leibniz. In this work, Kant refers to an anecdote from Hansch’s Godefridi Guilielmi Leibnitii Principia Philosophiae More Geometrico Demonstrata [Leibniz’s Principles of Philosophy Demonstrated in a Geometrical Manner], in which, while drinking coffee with Hansch in Leipzig, Leibniz says that he is not sure if the monads in the coffee might not someday be human souls (Hansch 1728, 135) (Ak. 2:327). Since, to the best of my knowledge, this story is only found in Hansch’s Principia, it is likely that Kant’s picture of Leibniz was informed by his reading to Hansch’s interpretation. Certainly, the Specimen Dynamicum, New System of Nature, Theodicy, Principles of Nature and Grace, Monadology, and the correspondence with Clarke can give a good picture of Leibniz’s philosophy, but as all students and scholars of Leibniz know, a great deal is missing from this picture. As mentioned earlier, the editions of Raspe (Leibniz 1765) and Dutens (Leibniz 1768) went a long way to filling in the details of Leibniz’s philosophy and providing the philosophical community with a true resource for study. This raises the obvious question whether Kant took the time to study these works carefully. It has become common to assume, for example, that Kant read the New Essays in 1769, four years after the publication of Raspe’s edition. This date corresponds to the time when “a great light” went on for 65  GW II 29, 296. According to De Risi, Wolff and others are in fact responsible for the transmission and transmogrification of Leibniz’s actual views (De Risi 2007, 101–102).

Brandon C. Look   25 Kant, and reading Leibniz at this time is taken to explain the distinction in the Inaugural Dissertation between the sensible and the intelligible.66 Whether or not Kant read Leibniz’s New Essays in the 1760s and this was the spark of the great light that went on for him in 1769, it is very likely that he read the work while thinking through the issues central to the Critical project in his “silent decade,” the 1770s. While the “Amphiboly” chapter of the Critique of Pure Reason, for example, contains an analysis of the weaknesses of Leibniz’s philosophy, Kant’s charge is ultimately that both Leibniz and Locke are guilty of not having properly engaged in transcendental reflection. And Kant’s presentation seems clearly to have arisen through an engagement with Leibniz’s dialogue with Locke. Moreover, Kant was certainly in a position to know of Leibniz’s account of innate ideas as found in the New Essays through an article by Michael Hißman, which appeared in the Teutsche Merkur, a journal that he would certainly have read in Kanter’s bookstore.67 On the other hand, there is little evidence in the Critique of Pure Reason that Kant studied the edition of Dutens, which might have led him to a more sophisticated understanding of Leibniz’s system, comparable to that of Lessing. In his Critical period, Kant presents a fairly consistent picture of Leibniz’s philosophy. In the Critique of Pure Reason, On a Discovery, and What Progress? Kant claims that the following theses typify Leibniz’s metaphysics: pre-­established harmony, the principle of the identity of indiscernibles, the monadology, the real opposition of forces, the principle of sufficient reason, and relationalism of space and time. There is nothing wrong with this list, but, as any Leibniz scholar will attest, missing are some crucial Leibnizian views, most obviously, the in-­esse notion of truth and the complete individual concept theory of substance— metaphysical theses that actually ground those theses identified by Kant. Of course, neither Kant nor anyone writing in the latter part of the eighteenth century could have known of these views, for the texts were simply not available. Despite Kant’s important arguments against Leibniz’s philosophy, he is not dismissive of it. For example, not only does he claim in On a Discovery that “the Critique of Pure Reason might well be the true apology for Leibniz,” he also speaks in the Metaphysical Foundations of Natural Science of the “intrinsically correct platonic concept of the world devised by Leibniz” (Ak. 8:251/Kant 2002, 336; Ak. 4:507/Kant 2002, 219). This latter claim can easily be read as saying that, if one could have cognition of the sort that Leibniz claimed—that is, if a human being could cognize the world solely through concepts alone—then it would be revealed to be just like that described by Leibnizian metaphysics. In Kant’s view, however, 66  See Refl. 5037 (Ak. 18:69). This view is suggested by Vaihinger (1881 I:48), Tonelli (1974, 437) and many others; against this view, see e.g. Wundt (1924, 160). For a very enlightening article on “das große Licht” of 1769, see Schmucker (1976). 67  Kant refers to the article in his polemic On a Discovery against the Leibnizian Eberhard in 1790 (Ak. 8:244).

26  Kant’s Leibniz: Historical and Philosophical Study Leibniz’s philosophy is based on a fundamentally flawed view of the nature of the mind and its relation to the world; Leibniz, according to Kant, fails to recognize that sensibility is a separate source of mental content and that sensibility works in conjunction with the understanding in the formation of our judgments.68 One way to reconcile the seeming discrepancy between opposition and support is to see in Kant’s philosophy a rejection of Leibniz’s claims to know the supersensible objects of metaphysics—God, freedom, and the soul—through the use of theoretical reason and an affirmation of those same supersensible objects obtained through practical reason. This is, indeed, the core of the Critical project.69

6 Conclusion It has become a commonplace in the history of philosophy to see Kant’s work as a response to “Hume’s problem”—as a response to the skeptical challenge to the possibility of (synthetic) a priori claims to knowledge. Kant’s famous claim from the opening of the Prolegomena forces this picture upon us: “the remembrance of David Hume was the very thing that many years ago first interrupted my dogmatic slumber and gave a completely different direction to my researches in the field of speculative philosophy” (Ak. 4:260/Kant  2002, 57). Yet, the very same passage also suggests that the Critical philosophy is a response to dogmatic philosophy, for Kant locates himself to some degree within this tradition.70 Understanding Kant’s relation to Leibniz, the greatest of the “dogmatic” metaphysicians, is therefore crucial to understanding his Critical philosophy. But, as I have tried to indicate, Kant’s knowledge of Leibniz was limited in many important respects: he did not have access to many of Leibniz’s writings, particularly his works in logic which, on the view of many Leibniz scholars, ground his metaphysical views or his important correspondences with Arnauld and De Volder, which present much more sophisticated treatments of ontology, modality, and natural philosophy; and a somewhat modified and caricatured Leibniz was created through mid-­century advocates and opponents who could not but influence Kant’s Leibnizbild. Be that as it may, the story of Kant’s reaction to the real and imagined Leibnizian philosophy is an integral part of the story of philosophy itself.

68  Both Anja Jauernig and Martha Bolton show in their contributions to this volume (respectively Chapters 7 and 8) that Kant’s criticism of Leibniz on this matter, as it appears in the “Amphiboly” chapter of the Critique of Pure Reason, is overstated and that Kant’s understanding of the details of Leibniz’s position flawed. 69  Paul Guyer makes this case in Chapter 9 of this volume. 70  The extent to which Kant adhered to dogmatism is discussed by Ursula Goldenbaum and Eric Watkins in, respectively, Chapters 2 and 3 of this volume.

2

How Kant was Never a Wolffian, or Estimating Forces to Enforce Influxus Physicus Ursula Goldenbaum (Emory University)

Kant ventures quite an enterprise, To educate the world. He estimates the living forces, But fails to weigh his own.

(Lessing 1751b, 32, trans. UG)

No researcher of true scientific mind can today and could around 1750 ignore the consequences Kant denies to acknowledge. (Adickes 1924a, 117, trans. UG) Within Kant scholarship, everybody seems to agree that Kant started his ­philosophical journey as a Wolffian, although as an independent and critical (or eclectic) partisan of Christian Wolff.1 Evidence of his pro-­Wolffian stance is usually seen in his use of Wolffian terminology (monads), his emphasis on the harmony of the universe and the law of continuity, and moreover in his explicit statements of respect for the great Leibniz and the honorable Wolff. His independence is seen, above all, in his early turn to Newton, mediated by Kant’s teacher at the university at Königsberg, Martin Knutzen (who is considered to be such an independent or eclectic Wolffian himself).2 1 “During the last years at university and presumably even during the first years as a tutor [Hauslehrer], Kant was a definite Wolffian in the way Knutzen was one” (Erdmann 1876, 141, trans. UG). This view became more or less the standard in Kant scholarship. See also Zeller (1875, 328–335), and this view can of course be found throughout Beck’s work (Beck 1996). Although more differentiated, Kuehn also shares this position (Kuehn 2003, 91–94). Notwithstanding his pointed description of the Pietist ruling at Königsberg and its university and without giving evidence, Kuehn still claims that Kant had rather inclined to Wolffian than Pietist teachers at Königsberg. But Kant mentions and sends regards from the Pietist Kypke and the earliest biographers emphasize Kant’s interest in Teske and Knutzen whereas the Wolff-­sympathetic professors Flottwell, Quandt, Rappolt, or Marquardt are never ever mentioned, neither by Kant nor his early biographers. 2  Whereas Erdmann emphasizes the influence of the Wolffian Knutzen on Kant, esp. in ch. 8, in agreement with Kant’s early biographers, Kuehn tries to argue that Kant had fallen out with his teacher

Ursula Goldenbaum, How Kant was Never a Wolffian, or Estimating Forces to Enforce Influxus Physicus. In: Leibniz and Kant. Edited by Brandon C. Look, Oxford University Press. © Oxford University Press 2021. DOI: 10.1093/oso/9780199606368.003.0002

28  How Kant was Never a Wolffian It is the aim of this chapter to question both of these opinions in order to show that the young philosopher is actually setting out from a strongly Pietist question—just as it was under discussion at the university at Königsberg. The city was not only geographically miles away from the intellectual centers of the German Empire.3 It certainly could not compete with Hamburg, Leipzig, or even Berlin where European publications, books, as well as journals and newspapers arrived within days; where journals and newspapers were edited, based on regular correspondences with members of the European intellectual community; where theaters and concerts attracted the citizens and coffee shops and reading societies flourished, providing the ground for an ever increasing intellectual life.4 You will hardly find Königsberg mentioned in any of the journals reporting academic events while even smaller German cities come up regularly. In Königsberg, even theater and poetry were still considered impious during the rule of the Pietists Rogall and Schulz.5 The same can be said about the significance of the University of Königsberg. It could not hold a candle to those in Leipzig, Jena, Halle, Frankfurt (Oder), or Göttingen where mathematics and modern sciences were taught by teachers who knew the latest results even if they did not themselves make contributions in the field.6 In Königsberg, not even Wolff ’s books about higher mathematics were taught, to be silent about more recent publications by the second generation of the Bernoullis or Leonhard Euler.

about philosophical differences and—as a result—had been let down. I am not sure though that there is sufficient evidence to make such a strong claim (cf. 87–89). Cf. also Watkins (1998). 3  Kuehn says so explicitly (Kuehn 2003, 84–85) but without providing any evidence. The common judgment about Germany’s general backward development in the seventeenth and eighteenth cen­tur­ ies, comparing it with London and Paris (although not with smaller cities in England or France), should finally give way to a more differentiated judgment and take into account the very different ­levels of culture and education in different areas of the German Empire, especially between the Catholic and Protestant areas, university cities and residence cities, and above all book trade cities such as Leipzig. It depended on very particular political, theological, and economic conditions whether a university flourished and could attract students or teachers. 4  A good indicator for my claim is the absolutely rare appearance of Königsberg in the Gelehrte Zeitungen. Scientific news had been reported according to the origin of the correspondent who reported them. Even smaller cities and universities in the Empire and Europe reported regularly to the journals at Leipzig, Hamburg, Göttingen, and Berlin. Not so Königsberg! 5  About the Pietists’ political networking to conquer Königsberg and thereby East Prussia and their success since the arrival of Rogall and Schulz, Kant’s teachers, see Hinrichs (1971, 243–300). It is still the most instructive and comprehensive history of Pietism, in spite of the more recent Geschichte des Pietismus in three volumes (Brecht  1995). The latter is rather a celebration of Pietism, completely omitting the public debate on the Wertheim Bible and the political persecution of the Wolffian translator initiated by the Pietists at Halle. See also the complaining letters from Flottwell and Marquardt at Königsberg to Gottsched in Gottsched (2007 ongoing). The relevant volumes for the battle between Wolffians and Lutheran theologians are published by now: vol. 1 (1722–1730) 2007; vol. 2 (1731–1733) 2008; vol. 3 (1734–1735) 2009; vol. 4 (1736–1737) 2010. 6  Cf. Hinrichs (1971, 397 and 132–136); Clemm states “that not only beginners but some professors of mathematics as well need to spell out [Vorbuchstabieren]; there are only very few who can read a Newton right away. Among those who can do so are they to whose mathematical insights Wolff ’s Elementa are still the Asymptotes” (Clemm 1764 X 4 recto, trans. UG). We get a similar statement from Kästner in 1754, see Waschkies (1987, 158 n. 135).

Ursula Goldenbaum  29 Obviously, Kant was a gifted, ambitious, and somewhat independent philosophy student—but he set out from a Pietist discourse. He began his studies with mediocre philosophical speculations within the framework of Pietist philosophical attempts to adapt their Aristotelian doctrines to modern science thereby using elements of Leibniz-­Wolffian philosophy as done by Martin Knutzen, the Baumgarten brothers, and Georg Meier but also by Crusius. Therefore, they are often called eclectics, but that only muddies the water. It hides their deep disagreement with Leibniz’s and Wolff ’s denial of an influence between body and soul and their distance from modern mechanics in clear contrast to Leibniz, Wolff, and their true partisans. Kant’s independence (as well as that of Baumgarten and Meier) within Pietism can be found in his strong intention to reconcile Christian religion with modern science instead of simply subordinating science to religion as the leading Pietists like Francke and Lange did. This intention, although not yet capability, would allow him to overcome his weak beginnings in the future and even to approach rationalist positions of Leibniz and Wolff at a later age. That Kant’s theoretical roots in Pietist discourse are so often missed and even denied7 is above all a result of our ignorance of the then-­current German intellectual discourse. The first half of the eighteenth century is still a blank area on the map of German intellectual history, including philosophy. Ironically, this is itself due to the success of Kant’s revolution in thinking and his dismissal of all earlier philosophy as dogmatic. With the rise of German Idealism and, in particular, the Hegelian school of history of philosophy and literature, Hegel’s judgment that German philosophy prior to Kant had been weak and meaningless became so prevalent that this early philosophy was no longer studied (with the exception of Leibniz).8 It became a self-­fulfilling story: Because we already know that Wolff and his disciples are dogmatic and boring thinkers, we do not read them anymore—except to find the confirming passages for our prejudice.9 As a result, even new studies about eighteenth-­century literature, philosophy, the­ ology, law, music, etc. are usually shaped by the standard judgments of the last 200 years that Wolffianism would neglect experience, history, art, emotions, etc. and would be one-­sided rationalistic.

7  Kuehn even argues that Kant, in spite of his upbringing in a Pietist family, his education at Pietist schools, and his studying at the Pietist university Königsberg, and moreover, his known closeness to the Pietists Knutzen and Kypke, inclined rather to Wolffianism and to Leibniz (Kuehn 2003, 39–40, see also 76–85). 8  One common attitude of the widespread Hegelian narrative of German pre-­Kantian philosophy is the splitting of Leibniz and Wolff. Whereas Wolff is considered the exemplary dogmatic thinker, such a judgment can hardly be found about Leibniz, in order to hold on to Leibniz as a great philosopher while continuing in our contempt for Wolff. But as much as Wolff and Leibniz may differ in their metaphysical groundwork, they certainly shared just those positions that were under Pietist attack. 9  The more Frederick Beiser’s new book has to be appreciated, in taking a fresh new look at Wolff and some Wolffians and discussing their arguments from their own perspective, with surprising results (Beiser 2009, esp. 2–4, and 45–71).

30  How Kant was Never a Wolffian The situation has slightly improved during the last decades, due to a new interest of Kant scholars in the early writings and the teachers of Kant, both clearly belonging to the discourse of early eighteenth-­century German philosophy.10 However, recent authors still look at these earlier philosophers through the lenses of the mature Kant, Hegel, and their followers, without studying Christian Wolff or the leading Wolffians on their own and within their own discourse; that is, without taking the questions to be discussed from Wolff and his partisans instead from Kant.11 We hardly know the philosophical views of Wolff ’s close disciples and friends—such as Johann Bernoulli, Georg Bernhard Bilfinger, and, above all, Samuel König—because Hegel and the Hegelian historians of philosophy did not even mention them. Moreover, we even misconceive the third generation of Pietists like Baumgarten and Meier as Wolffians because Hegel and Hegelian historians said so. But who ever asked about Christian Wolff ’s judgment about their philosophy? And yet, he was their contemporary.12 Such an ignorance of the German discourse can easily lead to misunderstandings about and even an overestimation of Kant’s early achievements.13 In this chapter, I will first present the battle of Pietist and other Lutheran theologians against Wolffianism as a theological-­political battle, which explains its extension as well as its fierceness. Then I will explain how an abstract metaphysical hypothesis such as Leibniz’s pre-­established harmony could become the subject of such a theological-­political debate in the Protestant area of the Empire, and how it could last for decades. Only in the third section will I then situate Kant’s very first, but quite lengthy book in this context and discuss his metaphysical approach to the question about the estimation of forces. While this question was much under discussion between Leibnizians and Cartesians, in close connection with modern mechanics, Kant’s approach lacks any familiarity with modern mechanics and is instead rooted in the Pietist intention to secure the interaction of body and soul. Finally, I will point to the contemporary reception of Kant’s first book to show how it confirms my evaluation. In this way, I hope to show how the 10  In the United States, see esp. the excellent analysis of Laywine (1993); Watkins (1998, 136–203). Most recently, we got an edition of the English translation of some important sources for Kant, among them writings of Wolff, Knutzen, Crusius, Baumgarten, and Euler (Watkins 2009). See also Watkins (2006) and Dyck (2011). 11  This is also true for Dyck’s (2011) where he starts from Kant’s questions to check how Wolff approached the same questions. Watkins, who gives a good survey about a variety of German thinkers concerning the mind–body problem, considers these various versions as steps to the final success of the doctrine of influxus physicus thus eventually overcoming pre-­established harmony, as becomes already clear by his title From Pre-­established Harmony to Physical Influx. I cannot see such a development outside of Pietism though. True Wolffians such as Bilfinger or König do not embrace the doctrine of influxus physicus; others rather—under pressure—prefer to make agnostic statements. 12  In fact, Wolff qualified the writings of Baumgarten and his circle as “elendes Zeug” (Oelrich 1782, 62–63). We do not even have yet an edition of Wolff ’s correspondence and it is only recently that the work on the comprehensive Gottsched correspondence began. 13  A full-­fledged example of such a naïve overestimation of the results of the young Kant can be found recently in Schönfeld (2000).

Ursula Goldenbaum  31 philosopher started his career precisely from the core problem of the Pietists in the abovementioned theological-­political debate of Pietism against Wolffianism even if he enriched his argument by borrowing from the Leibniz-­ Wolffian terminology.

1  The Two Camps of Pietism and Wolffianism Pietism was initiated by Jakob Spener, who rejected dogmatic controversies and preached Christian love of neighbor instead. Although it began as a religious, anti-­orthodox movement within German Lutheranism in the last decades of the seventeenth century, it transformed into a powerful empire within a few decades. It had its headquarters at the Orphanage in Halle in Prussia (Hallisches Waisenhaus), being closely related to the Theology Department at the University of Halle. However, Pietist leaders there soon came up with their own strong dogmatic doctrine and were ready to persecute heretics, first fighting Orthodox Lutherans. Their remarkable and fast rise to power—that is, political power—was the result of forging a strong political alliance with the Prussian king. Although being a Calvinist, Frederick William I soon understood the political advantage of the Pietist’s project of a well-­ organized orphanage transforming potential troublemakers into skilled and useful and obedient Lutheran subjects.14 Originally, the Pietists did not have at hand any particular philosophy distinct from other Lutherans and stuck with common Aristotelian scholastic philosophy. At Halle however, Christian Thomasius had supported the Pietists as a lawyer in their struggles with the Orthodox Lutheran theologians (cf. Thomasius (2007)). He also provided arguments against rationalism, raising skeptical doubts against philosophical systems in general. While he soon fell out with Pietists over the issue of freedom of thought, his disciple, Andreas Rüdiger,15 a doctor at Halle, developed an empiricist philosophy, including logic, that would increasingly be adopted by the Pietists after 1723. It was just in those years when, due to their struggle with Wolff and his partisans, Pietists urgently needed philosophical ammunition against their philosophical enemies. Rüdiger was the “philosophical grandfather” of the better-­known Christian August Crusius (the “father” being Adolph Friedrich Hoffmann).16 Rüdiger had himself been considered a problematic philosopher by Lutheran theologians due to his deviation from Aristotle and his adoption of modern ideas (e.g. from John Locke). But in the course of the lasting controversy between Pietists and Wolffians, he and his

14  In fact, Frederick William I understood Pietist theologians as a political tool to enforce his absolutist power against the local estates, cf. Hinrichs (1971, 216–300 and 432.) 15  Cf. Schepers (1959, 47–50 and 72–80). See also Watkins (1998, 156–167). 16  Cf. Döring (1999, 59); see also Goldenbaum (2004, 53–55, 259, 336–337).

32  How Kant was Never a Wolffian disciples were increasingly valued because of their helpful arguments against Leibniz’s and Wolff ’s rationalism and against the geometrical method. Pietists also appreciated the empirical approach, and the natural sciences that flourished at the University Halle were indeed purely empirical sciences like medicine and pharmacy (Duchesneau 2010, 135–158)). In contrast, Christian Wolff started his career as a professor of Mathematics at the University of Halle in 1706, on the recommendation of Leibniz. He became a member of the Royal Society in 1710, of the Berlin Academy in 1711, and of the Academy at Paris in 1733 (Gottsched  1755, 33–34 and 87–88). He stayed in mathematical correspondence with Leibniz, the Bernoulli brothers, Jakob Hermann, Varignon, and Samuel König.17 His mathematical textbooks and his mathematical dictionary were the only mathematical textbooks in use at German universities for decades to come.18 It was only in 1710 that Wolff began to teach philosophy. Although tensions between him and the department of theology arose early on, it was Wolff ’s lecture on the occasion of the end of his term as prorector in 1721 that caused an open confrontation. Wolff had chosen the rational foundation of morals as the topic for his lecture, using the example of the Chinese people as evidence. Although the Chinese lacked Christian faith, he argued, they were able to live a moral life based on Confucianism; that is, a rational doctrine of moral life without supernatural revelation.19 Wolff did not deny that Christian faith alone would provide an afterlife. He also acknowledged Christian religion as a great additional incentive for morals. However, his lecture was an extreme challenge to the Pietists because, according to Wolff, even an atheist could live as a moral being (something Pierre Bayle had said before). As a result, Joachim Lange, the Pietists’ leader used all of his close political connections to the Prussian court in order to achieve a ban of Wolff ’s philosophy in Prussia. He succeeded in 1723 and Christian Wolff was banned from Prussia under the threat of being hanged if he did not leave Prussia within forty-­eight hours.20 His philosophy was banned at all Prussian universities, including Königsberg. Wolff was lucky enough to be offered a position by the ruler of Hessen-­Cassel at the University at Marburg. However, Joachim Lange did not tire in his efforts to secure a ban of Wolff ’s philosophy in other German states of the Empire as well. And he succeeded to an extent. 17 Cf. Briefwechsel zwischen Leibniz und Christian Wolff (1860). The edition of Bernoulli’s cor­re­ spond­ence is still on its way (ed. by the Naturforschende Gesellschaft in Basel, vol. 1, Birkhäuser: Basel, 1955) and does not yet include the letters between Wolff and Bernoulli. 18  Cf. Waschkies (1987, 120–171). Cf. also n. 6 above. 19  Cf. The excellent introduction of the editor Michael Albrecht to Wolff (1985, ix–lxxxix). 20  For the text of the Royal rescript see Gottsched (1755, Beilagen, 33). It is worth noticing that historians of philosophy not only often neglected the political persecution of Wolff (and his partisans) but minimized it or even ascribed the responsibility of Wolff ’s clash with the Pietists to his behavior, very different from the treatment of other persecuted philosophers like Spinoza, Diderot, or Locke. Did they not provoke their persecutors too? Cf. esp. Bianco (1989, 111–155). Unfortunately, Watkins follows this unfair view (Watkins 1998, 146).

Ursula Goldenbaum  33 The open battle between Pietism and Wolffianism began by 1721 and continued after Wolff ’s death in 1754. It had its peaks though. The first, of course, was around 1721, due to the Chinese lecture with the following wave of persecution. The second peak, less recognized, followed in 1735, after the publication of the Wertheim Bible, a Wolffian translation of the Pentateuch by Johann Lorenz Schmidt (1735). This serious and careful translation into then-­modern German was produced in great awareness of the hermeneutical problems that beset such a project. Schmidt discussed these problems in more than 1,600 footnotes and resolved them in the spirit of Wolff ’s logic and, silently though, in awareness of Tyndal or Spinoza.21 However, there is no mention of Jesus Christ in the translation. All the passages traditionally read as prophecy of the Christian savior in the Old Testament had disappeared because they simply could not be found in the original text. This, of course, was a provocation for Orthodox and Pietist theologians alike, and it is from this time that the old opponents, the Pietist Joachim Lange and the Orthodox Ernst Valentin Löscher, united against the Wolffians. The publication of the Wertheim Bible offered a new opportunity to Joachim Lange to renew his battle against Wolff and to eventually secure a ban of his philosophy in the whole Empire. Arguing that the Wertheim Bible, with its elimination of all the prophecies of Jesus Christ in the Pentateuch, was a necessary product of Wolff ’s philosophy, he hoped to succeed once and for all. Accordingly, the Wolffians and Wolff himself made the greatest efforts to show that there was no such a necessary connection between Wolffianism and the Wertheim Bible in order to avoid a new and more general ban. This theological-­political battle is the reason for the enormous increase of anti-­Wolffian literature after 1735, often noticed but hardly understood by academic historians of philosophy. In this battle, one did not have to come up with novel arguments at all but simply to nail one’s colors to the mast. Silence meant abstention not serving well a career. That alone explains the enormous redundancy of these writings. Wolff was not the only victim of this wave of political persecution. In a little-­ known episode the well-­known Wolffian philosopher and professor at the university at Leipzig Christoph Gottsched was subjected to a trial at the court of his Saxonian ruler in Dresden. He had to stand in front of the tribunal of theologians an entire day, being interrogated about single sentences of his writings. In the end, he was given an ultimatum—to no longer teach Wolffian philosophy or to be fired.22 The same happened to Steinwehr, mentioned later in this chapter. The leading Wolffians Georg Bernhard Bilfinger, Jacob Hermann, and Samuel König found positions in Russia, Switzerland or Venice rather than within the German Empire. They were European philosophers and mathematicians, but due to 21  Cf. Goldenbaum (2004, 195–209). Cf. Wolff (1996, 167–173 (§§ 142–173)). 22  For Gottsched’s trial see Goldenbaum (2004, 366–367). The protocol is published in Döring (1999, 141–152).

34  How Kant was Never a Wolffian Hegelian historians they are barely known today. The translator of the Wertheim Bible, Johann Lorenz Schmidt (Spalding 1998), was sued and threatened with lifelong imprisonment in the casemates of the Bamberg castle, before he escaped to Altona, a city near Hamburg but belonging to Denmark. Schmidt died in Wolfenbüttel, as a teacher of mathematics at the court, still living under a false name. He was persecuted for his solid and still readable translation of the Pentateuch, or rather for using Wolffian hermeneutics. In spite of this massive wave of persecution, Wolffianism gained increasing influence. The obvious attraction of Wolff ’s philosophy for the academic youth during these decades lay in its clarity and coherence, in its systematic character, in its methodical attitude, in its agreement with modern science, and, moreover, in its practicability (i.e. its applicability in engineering, agriculture, and jurisprudence). He rejected blind belief and studying without understanding and gave reasons for his claims. Kant’s “sapere aude,” famously quoted in 1783, had long before been the Wolffian Aletophiles’s motto.23 Especially the philosophy department at Jena, outside Prussia, having incomparably more students than any other university in Germany, became the Mecca of Wolffianism during the 1720s.24 But when it came to a vote of the professors about a possible ban of Wolff ’s philosophy, at the request of the ruler of Sachsen-­Weimar, the majority decided in favor of a ban. Two professors, though, voted against the ban, and the students organized a sit-­in to protest the ban (Schmidt  1983, 93–100). Moreover, the teaching graduate students and adjuncts (magistri legentes), who covered the larger part of teaching math­em­ at­ics and philosophy, continued to teach Wolffianism. Above all, Wolff must have been an outstanding teacher. Within the first five years of his teaching at Marburg, the number of students increased by 50 percent.25 In contrast, Prussia lost students in such great numbers that the decrease in tax revenues caused worries at the Prussian court. As a result, Lange was warned by the king to stop his attacks on Wolffianism (Hinrichs  1971, 434 and 439). But Wolff also attracted advanced students in mathematics and law, such as Samuel König from Switzerland who had studied before with the famous Bernoullis in Basel. At the request of the Petersburg Academy, Wolff also became the teacher of Lomonossow, who later continued with Leonhard Euler in Berlin (Wolff 1860). In fact, Wolff became a European celebrity and his recommendations of scholars to the Academy in Petersburg were taken as his orders. Moreover, after the 1730s, even Pietist theology students like Martin Knutzen, and the entire circle of Pietist 23  Cf. Gottsched (1755, 104). Of course, the line was originally produced by Horaz. It is however quoted as Kant’s particular motto whereas the Wolffian use of this very same epitaph is almost unknown. 24  Cf. Schmidt (1983); Steinmetz (1958); Goldenbaum (2004, 210–214). 25  When we feel bored today while reading Wolff, due to his redundant and school didactic style, we should keep in mind that he had to—and wanted to—teach students who had hardly any idea about methodical thinking, not to mention their complete lacking of knowledge in science or mathematics.

Ursula Goldenbaum  35 students and friends at Halle—Alexander Baumgarten, Georg Friedrich Meier, Immanuel Pyra, and Samuel Gotthold Lange (the son of the infamous Joachim Lange) (Goldenbaum  2008, 267–281)—were won over to a more methodical phil­oso­phiz­ing and were even denounced as Wolffians by the older generation. They did not embrace Wolffian metaphysics and epistemology though. Rather they used pieces of Leibniz-­ Wolffian philosophy to modernize their Pietist discourse. Wolff ’s immense influence threatening traditional Lutheran theologians, both Pietists and Orthodox, can be seen most clearly in the series “Quo ruitis?” of Valentin Ernst Löscher, the leader of the Orthodox Lutherans in Germany. He addressed this series in his influential theological journal Frühaufgelesene Früchte der Theologischen Sammlung to his “beloved sons,” “the noble students of philosophy at the Evangelisch-­Lutheran universities” (Löscher 1735–1742, 1735, 71). It began in 1735 and continued until 1742, thus paralleling the ongoing public debate of the Wertheim Bible and Kant’s first years at the Albertina. It was Löscher’s aim to show the enormous damage to Christian faith that had been done by Wolff and his disciples. He was particularly concerned about the principle of sufficient reason (ibid., 1735, 119–128), the pre-­established harmony (ibid., 1740, 236–252), the geometrical method (ibid., 1735, 128–131; 234–236), and above all—the mechanism (ibid.,1735, 227–244) as the model of explanation in natural science. Interestingly, he did not recommend rejection of modern philosophy or science altogether. It was only mechanical philosophy, and thus Wolffianism, that was presented as the enemy of Christian faith. In contrast, he praised Newton’s and Boyle’s philosophy as being in best agreement with Christian religion because they, together with theologians, rejected the universal demand for mechanical explanations and defended final causes (ibid., 1740, 247). But neither Joachim Lange with all his political network and propaganda machine nor Ernst Valentin Löscher with his journal could stop the growing enthusiasm for Wolff ’s philosophy among the academic youth in the Protestant states of the Empire. Eventually, Wolff ’s continuously increasing influence was even crowned by a change of mind on the part of the Prussian king, influenced by his advisors, among them the king’s new father of confession Johann Gustav Reinbeck. As a result, Frederick William I, who had banned Wolff in 1723, was ready to call him back to Prussia in 1736.26 Although this would actually happen only after Frederick the Great’s accession to the throne in 1740, it was already on the order of Frederick William I that Wolff ’s German Logic became the official textbook at Prussian universities (Hinrichs 1971, 434). That was the end of the Prussian ban of Wolffian philosophy. In 1740, when Wolff approached Halle, he got a welcome like that of a king (Gottsched 1755, 113–114). 26 About this change of mind at the Prussian court and its promotors, cf. Buschmann (1989, 73–105, esp. 81–101). See also Goldenbaum (2004, 310–330).

36  How Kant was Never a Wolffian However, Provost Reinbeck, Wolff ’s new protector at the Prussian court, s­upported Wolffian philosophy but—not pre-­established harmony. Given the extremely precarious political situation of the long decade between 1723 and 1736, Wolff needed any possible political support he could get in order to keep his philosophy legal. Theologians had not only the power to decide who was a heretic but also a strong political network to provide jobs, and at times even the ear of the king.27 Thus in order to keep his political alliance with the influential Provost Reinbeck at the Prussian court, Wolff and his disciples were ready to play down the significance of pre-­established harmony for his philosophy as has been often noticed but not truly understood.28 However, the variety of statements about pre-­ established harmony is not arbitrary. Wolff and his partisans clearly differed from the young Pietists who only adopted single pieces of methodology from Wolff. Wolff, Bilfinger, Hermann, and König all stuck with pre-­established harmony while admitting that its truth could not be strictly demonstrated. However, they understated its fundamental significance for the Wolffian system. But Wolff was famous and known for holding the doctrine and his best true disciples lived outside the Empire—at the Academy at St. Petersburg, and at the Universities of Padua, Basel, or Groningen—and thus could afford to stick with an unweakened pre-­established harmony. Gottsched though, and other then less well-­known Wolffians within the Empire, took an agnostic stance, arguing that neither pre-­ established harmony, occasionalism, nor influxus physicus could be demonstrated. Therefore, one could choose one or even abstain.29 It simply was not helpful for a career in Prussia or the Empire, neither in church nor in state administration, to stick with a hardcore version of pre-­established harmony.30 But a quite different stand was taken by the younger Pietists who used Wolff only partially, such as Knutzen, Baumgarten, Meier, and—as I want to suggest—Kant. They fully agreed with Reinbeck and did not accept pre-­established harmony in any strict sense. They tried to avoid any independence of bodily motions from the succession of ideas in the souls as taught in pre-­established harmony (even if they kept the word as did Baumgarten). They all insisted in one way or another on the possibility of a mutual influence of soul and body, whereas the Wolffians saw this view in

27 For Lange’s extended network with the Prussian court see Goldenbaum (2004, 222–233, 270–279). 28  Watkins emphasized the rapidity of the exchange of polemic writings, their length, and the redundant character (1999, 149). This is easily explained by the methodus polemica, according to which one had to present the argument of the opponent before refuting it exhaustingly. Cf. Zedlers Universal-­Lexicon (1739–1764, vol. 20, cols. 13–37). 29  I do not see any support for the doctrine of influxus physicus in this statement but rather the attempt to escape the pressure to embrace it: “None of the three is completely explained or demonstrated; each of them still has its difficulties: Thus each person can maintain whichever one is most pleasing” (Gottsched 1983, 586. 30  Gottsched reported to Reinbeck that theology students did not dare to come to his lectures because they were afraid of not getting a position in church. Cf. Döring (1999, 72).

Ursula Goldenbaum  37 contradiction with the inertia principle even if they could not prove their pre-­ established harmony. It was somewhat tragic for the Wolffians that they could not benefit from the political change in 1740 that brought an end to the Pietists’ prevailing political influence in Prussia. Although Frederick II called back Christian Wolff and even offered him the co-­presidency of his renewed Academy in Berlin, the simultaneous appointment of the Newtonians and aggressive anti-­Wolffians, Maupertuis and Euler, as the Academy leaders replaced the former Pietist resistance against Wolff and his disciples with a new one, again backed by the strong support of the Prussian king. The opposition of Euler and Maupertuis to German Wolffianism is often presented as one between modern science and old-­fashioned metaphysics, and Leonhard Euler helped to shape this narrative in his publications (Goldenbaum  1999, 383–417). According to this again tendentious narrative it was the Newtonians who were productive in science, whereas Wolffians did not quite understand Newton’s significance. While Maupertuis’s hostility to Wolffianism was due to his jealousy of Wolff ’s amazing influence in Germany, the great mathematician Euler hated Wolffians and Leibnizians as atheists—for the very same reasons as Pietists and Orthodox theologians hated Wolffians and Leibnizians.31 Although historians of science often complain that the great Euler never became the Academy’s president because of the king’s preference for the French, Euler in fact held all the power in the Academy beginning at least in 1753. It was he who decided about finances, publications of the Academy, and to a significant extent about appointments and pensions of the members (Goldenbaum 2004, 558–569). Euler used his influence on Maupertuis to shape the Academy and even to intervene in the procedure of the Preisschriften to guarantee anti-­ Wolffian winners. He was not shy in enforcing his preferences, regardless of the topic or even the quality of the papers so long as they were written against Leibniz and Wolff.32 Thus while the Wolffians might have expected some relief from the new “enlightened” king in 1740, they faced again a tough opposition from Berlin, this time in clear alliance with “Newtonianism.” This opposition was even backed by the king himself who had turned from Wolff to Skepticism under Voltaire’s leadership before he took power in 1740. This new line of opposition came to light with the Preisfragen of the Academy and the elected Preisschriften in 1747, in 1752, and again in 1755. It is not without irony that it was the so-­ called enlightened king who forbade any public criticism of his Academy president Maupertuis in the public debate of 1752 and even made Voltaire’s satire against 31  Winter speaks of Euler’s uncompromising battle against Leibniz’s and Wolff ’s dangerous ideas and his intention to defend his Christian faith in revelation, cf. Winter (1957, 54). See also Spiess (1929, 129–131) and Goldenbaum (2004, 556–569). 32  Euler’s open partiality against Wolffians was widely known when Kant sent his book just to Leonhard Euler to ask him for judgment and even a review. See §4 of this chapter.

38  How Kant was Never a Wolffian the authoritarian president (and Euler) and in favor of the Wolffian Samuel König publicly burn (Goldenbaum 2004, 580–582). Samuel König, almost forgotten today, certainly in the history of philosophy, fits neither the Hegelian nor the Eulerian narrative. Having studied with the Bernoullis in Basel, he belonged to the small scientific elite (Waschkies  1987, 393–397) who could understand Leibniz’s and Newton’s mathematical work and scientific achievement. This Leibnizian even tried to reconcile Newton’s Mathematical Principles of Natural Philosophy with Leibnizian metaphysics (König 1748, reviewed in Nova Acta eruditorum 1749, 663–665). Unfortunately, he died soon after the end of his battle with Maupertuis and Euler in 1757. As with Leibniz and Wolff, König was not an opponent of Newton’s mechanics but of his metaphysics. He was twelve years older than Kant and caused the controversy with Maupertuis and Euler by his critical response to Maupertuis’s “discovery” of the principle of least action (König [1751]). Both papers belong in the context of the German discussion about the estimation of living forces. As Kant wrote his Estimation of Forces in 1746–1748 but could not publish it before 1749, Samuel König wrote his paper in 1748 but did not publish it before March 1751. Kant’s writing is often compared with d’Alembert’s Traité de dynamique (1743).33 But a comparison of Kant’s first writing with König’s essay shows that Kant cannot hold a candle to his Wolffian contemporary, neither in mathematics nor in mechanics. Of course, Kant could not yet have known König’s essay either. However, he obviously learned the most about Leibniz-­Wolffian living forces from one particular source—Der Frau Marquisinn von Chastellet Naturlehre an Ihren Sohn.34 That is the German translation of the Institutions de Physique of Gabrielle Emilie Le Tonnelier de Breteuil, the famous Marquise de Châtelet, published in French in 1740. The German translation was produced by the Wolffian and member of the Royal Academy at Berlin, Wolf Balthasar Adolph von Steinwehr, who regularly translated the transactions of the French Academy of Sciences.35 Madame de Châtelet wrote her book though—to put it mildly—under the direct influence of her teacher in the mathematics of Leibniz’s calculus—Samuel König.36 Thus the

33  Cf. B. Erdmann (1876, 8) and Adickes (1924a, 75). Cf. Kuehn (2003, 446 n. 128). 34  So also Waschkies (1987, 437). Concerning this German translation, Waschkies emphasizes that he found a German or Latin translation for every French work mentioned by Kant, cf. 514–515 n. 101. Waschkies also points to Hamann’s complain that Kant used him as a mere French translator (ibid.). 35  Steinwehr was a Wolffian teacher at the University of Leipzig, a contributor to the Nova Acta Eruditorum, and an active member of the Wolffian Deutsche Gesellschaft. Moreover, since 1736, he was the editor of the Neue Gelehrte Zeitungen from Leipzig before he became a professor, first in Göttingen, then in Frankfurt (Oder). As a reviewer of the Wertheim Bible he got involved in the public debate and had to appear for an interrogation at the court of Dresden. Cf. Goldenbaum (2004, 256–257). 36  König had taught higher mathematics to Madame de Châtelet in 1739 and 1740, discussing Leibniz-­Wolffian metaphysics with her as well. Her change from Newtonian positions (she had shared with her friends Voltaire and Maupertuis) to Wolff has been well known among her contemporaries as can be seen by Mairan’s sarcastic comments (included in the abovementioned German translation of her work). König’s relation to his noble student came to an end because she did not only owe him his

Ursula Goldenbaum  39 Institutions of Madame de Châtelet and even more König’s critique of Maupertuis’s principle of least action can indeed serve as a measure of what could be achieved by a young Wolffian philosopher who was out to challenge the leading scientists and philosophers of Europe and solve the problem of measuring living forces around 1750. As a result of such comparison, it would become obvious that Kant did not even grasp the scientific arguments of the controversy. Indeed, he was out to solve a completely different problem, that of the mutual influence of body and soul—the Pietist’s problem.

2  What Were the Issues Lutheran Theologians and Euler Had with Wolffianism? As Eric Watkins has pointed out clearly, the main problem that caused the long-­ standing battle between Lutheran theologians and Wolffians was the question whether the soul could have an impact on the body, and vice versa, whether the body could “convey” bodily sense perception to the soul. Three different metaphysical explanations for the relation between body and soul were under discussion: influxus physicus, pre-­established harmony, and, less so, occasionalism. The question arises, how could such an abstract philosophical question as the mind–body problem ever become the main subject of a theological-­political fight, last for more than three decades, and occupy the courts of the Empire? To be sure, this long-­ lasting debate was not stirred up or fueled by the Wolffians. Christian Wolff had settled this problem for himself at least in 1720— largely in agreement with Leibniz’s pre-­established harmony.37 The debate was rather caused by the anti-­Wolffians who not only disagreed with Leibniz and Wolff but held that pre-­established harmony would lead to atheism and fatalism and therefore destroy the Christian religion. Accordingly, they incessantly accused the Wolffians of being atheists and Spinozists and urged the political administration to take measures against Wolffianism as threatening not only the power of the church but the political state as well. Determinism, considered as fatalism and Spinozism, would take away moral responsibility and thus provide everyone an excuse to sin or commit crimes. As is well known, Joachim Lange successfully convinced the Prussian king by arguing that his deserting soldiers would be excused by Wolff because they had been determined to desert and thus could not be held responsible (Hinrichs 1971, 416–417). The deep gap between Leibniz, Wolff, and the Wolffians on the one hand and their opponents (in the first line Lutheran theologians) on the other, was their salary but even his expenses. In addition, we know of a private letter to a friend where he complained about her plagiarizing his lectures for her book. However, he never blamed her in public. 37  For the extended literature about Wolff ’s deviating from Leibniz, see Goldenbaum (2004, 27–28).

40  How Kant was Never a Wolffian radically different stance toward modern mechanics and mechanical philosophy. Whereas Leibniz and Wolff embraced them unconditionally, their opponents wanted to restrict mechanical thinking to mathematics, including applied mathematics (i.e. mechanics). Whereas the former fully accepted the principle of inertia as the basic principle of modern mechanics along with the law of conservation of force (still the foundation of modern science today), their opponents wanted to restrict the principle of inertia to the realm of mathematical bodies and did not care about a conservation law because it was up to God to conserve or destroy the world. However, when modern science began with Galileo, it was based precisely on the principle of inertia according to which no body could cease to move if not stopped by another body, nor cease to be at rest if not moved by another body. This was taken to be true for every body in the world. As a result though, no idea could ever move a body and no body could bring about an idea without violating the principle of inertia. This metaphysical consequence of the principle of inertia caused all the philosophical discussion about the relation of body and soul that arose in seventeenth and early eighteenth centuries. Descartes had made the conflict explicit and tried to solve it. Neither Descartes nor his partisans nor even Malebranche could provide convincing solutions though. But, if the question could not be answered sufficiently, Spinoza’s solution would be the only pos­ sible—albeit “dangerous”—result. This radical philosopher had already stated quite frankly that there is no influence between body and soul once the principle of inertia is accepted. Likewise, no liberum arbitrium could be accepted, i.e. no free will as an ability to freely choose what one wills without a reason. But—as Jacobi would express it in the later Spinozastreit—there was no good salvation in Spinoza (Jacobi’s letter to Moses Mendelssohn on November 4, 1783, in Jacobi (1987, 229). That is precisely what the anti-­ Wolffians repeated again and again— “mechanical absolutism” (Löscher 1735–1742, 1735, 239) would lead to Spinozism. What seems rather strange at first glance becomes quite obvious when we understand the close connectedness of the mind–body problem with the question of free will. Even today, the problem of free will stirs up controversies and emotional reactions. How can I act morally responsible if I cannot choose my action? It is also an issue for philosophy of law when it comes to justifying punishment. But free will was even more important for theologians in order to be able to blame and praise a member of the Christian community for her deeds as caused by nothing else than her free choice to act. If the deeds cannot be ascribed to the free will of an individual how could she possibly sin and therefore justifiably be punished? As soon as our will is determined by reasons or motives our will is no longer free to determine itself, without following a reason. Thus, the notion of a sinner seems to be impossible. Joachim Lange, in his aggressive writing against

Ursula Goldenbaum  41 the Wertheim Bible gets to the heart of his concerns with pre-­established harmony when he cries out: I only say this . . . that the author deduces the stubbornness [of Pharaoh in Exodus 7, 13 and following] from the nexus or the fatal connection of all things, and in this way ascribes it to God according to his pre-­established harmony. This nexus is the soul of the whole system of the mechanical philosophy.38

It is clear from this quote that Lange’s protest was not particularly directed against the Wertheim Bible but against the mechanical philosophy in general, especially against the principle of sufficient reason that demanded a reason or cause for any given event. The system of pre-­established harmony was seen as a hidden version of determinism and fatalism, in the same way as Friedrich Jacobi would later claim that all rationalism leads necessarily to Spinozism—that is, fatalism (Jacobi to Mendelssohn on November 4, 1783, in Jacobi 1987, 230–232). However, Leibniz, when he came up with his pre-­established harmony had indeed tried to “square the circle”; that is, to reconcile free will with the determinist foundation of modern science. He who had early recognized the enormous threat of this theoretical constellation developed his own metaphysical system of pre-­established harmony precisely to escape such a horrible “fatalism”—but without abandoning modern mechanical science.39 He wanted to save both, modern mechanics as well as Christian religion with its notion of free will. He succeeded by a strict distinction between the two realms, the phenomenal world of bodies in motion, subject to the principle of inertia, on the one hand, and the metaphysical level of substances or monads on the other. In this way, natural science would be restricted to the phenomena that could be exclusively explained by efficient causes whereas metaphysics would deal with substances and especially with those that were gifted with an intellect and thus free will. Because both realms would somehow be in correspondence though, Leibniz further distinguished between two levels of necessity within the realm of phenomena, one being absolutely necessary as in geometry but valid for abstract entities alone (i.e. for entia rationis). On the other hand, Leibniz insisted on the contingency of particular concrete things. Such things could be otherwise under different circumstances without including a contradiction. However, they would still be caused and nothing could happen by chance. Their complete concepts would still include all their properties virtually, just as in the case of geometrical

38  Lange (1735, 25, trans. UG); cf. also Goldenbaum (2004, 236). 39  See esp. Leibniz’s Confessio philosophi and his Letter to Magnus Wedderkopf as early documents of Leibniz’s struggle with determinism and even Spinozism, both in Leibniz (2005, 1–5 (= A II 1, 117–118) and 26–109 (= A VI 3, 116–149).

42  How Kant was Never a Wolffian concepts, and these would be known to God. However, their analysis would be infinite due to their infinite complexity. Freedom of the will, though, would not be the arbitrary choice, dismissed by Leibniz as an impossible “indifferentism.” It would belong to the intelligent substances being more active and more free the more they would understand themselves and their environment adequately. It is this view that puts Leibniz in the strongest opposition to the defenders of free will as arbitrary choice, not only to Lutheran theologians, but also to Locke, Newton, and finally Kant. Although Wolff slightly deviated from the Leibnizian system of pre-­established harmony, he clearly defended all of the abovementioned basic points of Leibniz’s theory (Corr  1975, 241–262). Although he gave more reality to bodies than Leibniz he clearly kept Leibniz’s strict distinction between the causality of bodies to be explained by efficient causes, and the succession of thoughts in their own causality, thus saving the uncompromised validity of the principle of inertia for the world of natural science, while defending the independence of substances (which would still obey the principle of sufficient reason though) (Goldenbaum 2011a, 29–41). Thus, both philosophers, Leibniz as well as Wolff, stuck with modern science and defended mechanical philosophy, explaining all phenomena by sufficient reasons, even if they could have been otherwise under different circumstances. Wolff ’s claim that the mathematical method is the philosophical method simply meant that philosophy is different from history because it explains things from their causes, in agreement with Hobbes’s and Leibniz’s understanding of philosophy (Wolff 1996, ch. IV, 126–163). The complete concept of a thing was its rule of construction or generation from which all properties of the thing could be deduced. In this way, knowledge of a complete concept or the essence of a thing was possible in principle, even beyond mathematics. Thus, the limits of human knowledge were not set firmly, but could be extended by working on better concepts or definitions. Therefore, the core of Wolff ’s (and other rationalists’) mathematical method is the doctrine of definitions. In his German Logic (1713), Wolff distinguishes definitions by words from definitions by causes, along the lines of the common Latin distinction of nominal and real definitions.40 Word definitions simply include a collection of properties known by experience or hearsay by which a thing can be distinguished from others. In contrast, real definitions indicate the possibility of a thing (i.e. whether its notion includes a contradiction). The special case of genetic or causal definitions was first emphasized by Thomas Hobbes. Such a definition even reveals the cause of the thing and thus its very essence. The model for this kind of definition was obviously

40  GW I 1, 141–151 (ch. 1, §§ 36–57). Cf. also Look in Ch. 1 of this volume.

Ursula Goldenbaum  43 geometry—where the construction of a figure shows at the same time its possibility.41 However, Wolff tried very hard to extend the realm of real and, if possible, even causal definitions beyond mathematics to close the gap between nominal and causal definitions. He succeeded by applying extensively Hobbes’s epistemological principle that we can cognize clearly and distinctly what we can generate. That is the epistemological background for Wolff ’s development of a philosophy of experiment and of technology or his ars inveniendi (Corr 1972, 323–334). He saw this theoretical approach as providing the possibility to advance our knowledge of real things in nature as well as society. By temporarily presupposing quasi-­genetic definitions for a demonstration, as a hypothesis, one could check them as well as their deductive consequences by experiment. The hypotheses could then be revised or rejected and the procedure could begin again. Is not this kind of procedure just what we still recognize under the label ‘hypothetico-­deductive method’? In this way Wolff transformed the generation of real things—their practical production—into a source of rational knowledge. Moreover, Wolff also used Hobbes’s suggestion that the essence of a thing could be expressed by any possible cause that could bring about the thing, not only the one that did. Thus, while consciously working empirically and with insufficient definitions— although within a rationally defined framework—Wolff pushed forward the limits of human understanding, particularly in engineering, agriculture, political science, and jurisprudence. Real particular things could be understood truly if they could be produced or at least changed by human beings in a controlled way; for instance, as the growing of grain (GW I 24). The strict gap between God’s and human knowledge was softened and was no longer a difference in kind. Like Galileo, and Leibniz, Wolff was convinced that humans could know strictly demonstrated things as well as God although they could rarely “intuit” them (i.e. grasp them instantaneously).42 Thus, it does not come as a surprise that it was the strategy of Rüdiger, Hoffmann, Crusius, Löscher, and Lange to emphasize the fundamental difference between mathematical knowledge and scientific knowledge of natural things whereby mechanical theory counted as applied mathematics. Whereas mathematics dealt with figures and numbers—produced by humans and thus arbitrarily—natural science as well as metaphysics and theology dealt with God’s creation and thus natural things (Löscher 1735–1742, 1735, 128–129; 1742, 78). Therefore, only mathematical concepts could be known by us in their very essence, whereas the essence of God’s creatures remained hidden to us. We could

41  Cf. Goldenbaum (2015). 42  One of the sentenced propositions in Galileo’s trial has been the following: “6. To ensure and to declare as false that a certain equality exists between the understanding of geometrical things by human and divine intellect” (quoted in Harig 1987, 282).

44  How Kant was Never a Wolffian know them only by observation, experience (equated with sense perception), induction, and abstraction. This was common sense among the anti-­Wolffians.43 Thus it does not come as a surprise that they already began their Locke studies at the end of the seventeenth century. Quite a few of Kant’s teachers were Aristotelians, Pietists, and also deeply interested in John Locke at the same time. Kant’s teacher and landlord Georg David Kypke translated Some thoughts on the Conduct of understanding in the search of truth together with A Discourse on Miracles (1755) and Knutzen is said to have worked on a translation of Locke’s Essay concerning human understanding until his death.44 It is one of the myths that Locke and English philosophy were received in German countries only in the second half of the eighteenth century. Many German journals such as the Freie Urtheile von gelehrten Sachen at Hamburg, the Hamburgische Magazin, the Göttingischen Zeitungen von gelehrten Sachen, and the even older Neuen Urtheile von gelehrten Sachen at Leipzig reported regularly in German about the newest European publications. As is well known, Kant came to know about the English author Thomas Wright and his “Newtonian” cosmology precisely by a review in the Freye Urtheile from Hamburg (Ak I 231:3; cf. Waschkies  1987, 513). These journals were all edited by Wolffians holding a wide and regular correspondence with German universities as well as with academic correspondents in Paris, London, the Netherlands, Russia, and Italy. 45 The extent to which Lutheran theologians embraced empiricism, while rejecting mechanical explanations of natural things as human haughtiness of reason, can already be seen in the well-­known debate between Stahl and Leibniz about mechanical explanation in biology (Duchesneau 2010, 135–158). Another example is Baumgarten’s invention of aesthetics as the science of sensuous, empirical knowledge, often misunderstood as re-­ evaluating the senses by overcoming one-­sided rationalism.46 In fact, he intended to found a science of sensuous knowledge that would restrict Wolff ’s philosophy to logic and math­em­ at­ics (including mechanics); that is, non-­real entities. According to Baumgarten, what we can know with certainty has no reality (mathematics) and what is real we cannot know in its essence but by sense perception alone. I should add that the identification of experience with sense perception and abstraction differs greatly

43  See Löscher’s suggestions to write a sound philosophy (Löscher 1735–1742, 1735, 139). 44  See Locke (1755). Kuehn sees Kypke vacillate between Aristotelianism and Pietism (Kuehn 2003, 74–75). For Knutzen see ibid. The Pietist school philosophy was Aristotelianism until the modern science forced it to modernize its teachings. Obviously, this school was ready to adopt Locke’s empiricism because of his agreement with the doctrine of influxus physicus. See n. 15 above. 45  On the Gelehrte Zeitungen see Kirchner (1928, 100–102); about those from Leipzig, Hamburg, Göttingen, and Berlin (in this succession) and their significance, see Goldenbaum (2004, 90–99, 522–529, and 710–714). 46  About the Pietist background of Baumgarten and Meier, see Goldenbaum (2011b).

Ursula Goldenbaum  45 from the rationalist understanding of experience as experiments led by the­or­et­ic­al­ly grounded hypotheses.47 But if Leibniz and Wolff were in favor of modern science, why did they reject Newton? The reason for the Leibnizian and Wolffian resistance to Newton’s theory of gravity was not the mechanical and mathematical character of the theory but the obscurity of the cause of gravity. For Newton, gravity served as one example of a variety of unknown and maybe unknowable “active principles” in the universe stemming directly from God. It was Newton’s reluctance or even his opposition to merely mechanical and thus natural explanations of bodily phenomena that caused Leibniz’s and Wolff ’s criticism. They rejected Newton’s mixing of theological arguments into mechanical theory. Leibniz and Wolff as well as the true Wolffians who understood modern science (i.e. Johann Bernoulli, Jacob Hermann, or Samuel König) tried to solve the tension between determinism in mechanics and free will in the world of human beings by distinguishing between the two realms—the phenomenal world of bodies and the world of substances—in order to allow science to search for efficient causes to explain phenomena and to allow metaphysics to deal with questions of thinking and its rules, with human and divine freedom, or with theological problems. In contrast, German Lutheran theologians, hardly capable of understanding Newton’s mathematics, were extremely comfortable with Newton’s “natural philosophy” and even more with his digs at Leibniz and pre-­established harmony. Newton clearly accepted free will as arbitrary choice, he allowed for final causes within science, and saw God’s will as a sufficient reason within natural science. He defended even the existence of ghosts. Newton did not see God as a great architect or geometrician as Leibniz but as the lord and master, just like the Lutheran theologians did. The Orthodox Lutheran theologian Löscher, in his polemics against “Absolutismus mechanicus” (Löscher 1735–1742, 1735, 239) (i.e. against the demand to explain everything in a mechanical and thus natural way) fully agrees with Newton that there is more in natural things, even in inanimate things, than only mass and mechanical motion (as could be seen in chemistry). Above all, it was especially the animated things in nature that could not be explained without reference to a vital force in nature that is of a higher order than mechanics. Last but not least, Löscher objected that mechanical philosophy would restrict human freedom as well as God’s power or would even make human freedom impossible. To be sure, Löscher did not even mention the scientific core of the Mathematical Principles, but he certainly knew Newton’s book and his overall presentation of “Newtonianism” is correct. In contrast, Leibniz, Wolff, and his partisans understood and appreciated Newton’s great book for those achievements we still hold in high regard, his 47  On the difference between rationalists and empiricists concerning the concept of experience, see Goldenbaum (2016).

46  How Kant was Never a Wolffian mathematical work and his experimental skills. But they insisted on mechanical (i.e. natural) explanations in science instead of obscure “active principles.” They also questioned Newton’s “empirical” credo and recommended the rationalist epistemology, well approving of experience and moreover experiment. As a result of our widespread contempt for Wolff and Wolffianism, it is often overlooked that Leibniz-­ Wolffian philosophy played the same role for the modernization of German philosophy as Descartes’s philosophy did for the French. What had been taught in German philosophy departments before Wolff was Aristotelian scholastic philosophy. Only with the spread of Wolffianism did modern science and especially mathematics get a hold at German universities. The first math­emat­ icians who deserved the title were exclusively Wolffians and this remained so until the arrival of Leonhard Euler in Berlin in 1741—Euler who had himself studied with the Wolffian Jean Bernoulli. It is symptomatic that the only textbook of Wolff that Kant ever used for his teaching was the Anfangsgründe aller mathematischen Wissenschaften, a beginner book.48

3  The Project of Kant’s Estimation of Forces What first catches the eye when studying Kant’s book49 is the amazing amount of literature the young philosopher mentions and discusses extensively—indeed the leading Cartesian and Wolffian scholars working in the field. At a second look, though, it becomes clear that he relied, above all, on the German translation of Madame de Châtelet’s Institutions de physique as mentioned earlier. Nevertheless, he did look up books and articles mentioned by her, especially Wolff ’s paper in the Acts of the Petersburg Academy.50 Thus he clearly studied the controversy he aimed to resolve. In spite of his impressive efforts, though, it soon becomes clear that he did not really grasp the point of the controversy. This is certainly due, to some extent, to the mediocre education in mathematics and mechanics that he obtained at the University of Königsberg. He simply did not understand mechanics and mathematics, in particular the mathematics of infinitesimals. However, it cannot be ignored that his missing the point of the controversy is above all a result of his burning wish to solve quite another problem—to show how the soul could indeed influence the body and likewise receive an influence from the body. 48  Cf. n. 6 above. About mathematical education at Königsberg in the first half of the eighteenth century in general and Kant’s in particular see Waschkies (1987, 117–171; see esp. 131 n. 57). 49  Kuehn sees this book as a defense of Leibniz’s living forces. Allegedly, Kant defended Leibniz’s pre-­established harmony (Kuehn 2003, 91–92). In fact, Kant, like all Pietists, tried to undermine the idea of pre-­established harmony. 50  Cf. Wolff (1728, 217–238). Interestingly, this volume also includes the two papers (mentioned by Kant) of Hermann (1728, 1–42) and of Bilfinger (1728, 43–120) and a further nine papers on higher mathematics by these and other authors (among them Daniel, Johann II, and Nicolaus Bernoulli) not mentioned by Kant.

Ursula Goldenbaum  47 This problem was the Pietist problem, and it had little to do with the controversy between Cartesians and Leibnizians about the measure of forces. These philo­ sophers and scientists discussed at the level of mechanics, even if their meta­phys­ ic­al views eventually backed their arguments. Having said this, I am ready to admit that Kant uses Leibniz-­ Wolffian terminology throughout his book. But so did Knutzen, Baumgarten, and Meier, who all tried to use pieces of modern philosophy to modernize Aristotelian school philosophy without ever becoming Wolffians. Thus, it does not come as a surprise that Kant praises Leibniz just for his agreement with Aristotle and their shared rejection of (mechanical) philosophers who would see a body’s force exclusively as acquired from outside; that is, from another body in motion (Ak 1:17, §1). Only the great Leibniz had ascribed an inner “essential force” to the body, even prior to extension.51 But when Kant then criticizes the Wolffians for using the term “vis motrix” in contrast to Leibniz’s “potentia agendi” he obviously misses the fact that Leibniz used both terms, the second on the metaphysical level, and the first in mechanics (i.e. on the level of phenomena—as did the Wolffians).52 Kant objects that vis motrix was an occult quality that could not serve as an explanation of the origin of motion—as if the name was supposed by the Wolffians to explain it.53 But what actually bothers Kant is clearly the identification of moving force with acting force or the connection of motion and action. Kant held that action should not be reduced to motion as it was done by modern mechanical philosophy (Ak 1:18, §2). In addition, it should not go unnoticed that Kant calls the force of a body or that of a substance not (with Leibniz) a vis or potentia agendi but rather (with Newton) a vis activa (Ak 1:18, §3; Newton 2004, 136–137). Of course, it could be said that Kant’s insistence on something beyond local motion is a reference to Leibniz anyway. That is certainly true but only as long as the laws of mechanics are acknowledged. Leibniz never tried to meddle his metaphysical notion of vis agendi et patiendi into mechanics, the science of phenomena. For Leibniz and Wolff and their partisans, bodies in motion had to be considered according to the laws of motion exclusively while substances had to be treated within metaphysics. Kant’s declared motive for insisting on active force instead of moving force is his own Pietist attempt to adapt modern science to the doctrine of influxus physicus. Already in §§4–6 he openly declares that the influxus physicus can be explained

51  The passage Kant had in mind might well have been the opening paragraph of Leibniz’s Specimen dynamicum. Cf. Leibniz (1969, 46). Although Leibniz’s wording indeed sounds similar to that of Kant, it has to be emphasized that Leibniz adds immediately that all corporal action is motion! Kant tried to attack this view. 52  The term can be found since Leibniz’s breakthrough with his conception of force in 1678, cf. A 1.10.236; 6.3.529; 6.4.1622:22; 6.4.2027:22; 6.4.2028:2; 6.4.2029:7; 6.4.2116:9 and 19; and 6.4.2423:5. 53  About the discussion of the Wolffian notion of “vis motrix” by Lutheran theologians see Löscher 1735–1742, 1737, 270–271.

48  How Kant was Never a Wolffian from the active force of substances.54 The key to Kant’s solution to the problem is to attribute an active force instead of local motion to matter—a force that could then act upon other substances (among them the soul) instead of only moving other bodies.55 Thus, calling the essential force of bodies active instead of moving is not a question of names but of dismissing the modern explanation of natural phenomena, according to mechanical laws, by bodies in motion. Kant then continues with a perfect circulus vitiosus: the soul can act externally because it is in a place, and it is in a place because a place indicates the mutual action of substances toward each other (§6). This is supposed to explain how a soul can receive effects from bodies, to be sure, not from their motion but from their active force. His substituting of vis motrix by active force seems to Kant indeed to solve the problem of influxus physicus that another “sharp philosopher” had not yet perfectly solved (§6). But there is no reason to assume that Kant did not see his own solution as an improvement on this ally rather than as a criticism of an opponent.56 In the following, I want to focus on this strong Pietist pattern in Kant’s earliest book: his strict distinction between mathematical/mechanical motion and active force. Before I turn to this discussion, I would like to emphasize that there are more Pietist patterns in Kant’s book that are not original at all but can be found again and again in Pietist literature, not only in the writings of philosophers but also of theologians. Moreover, since the 1730s, these arguments were increasingly used by Orthodox Lutheran theologians such as Löscher as well. One typical pattern of this kind is the always sarcastic acknowledgment that geometrical or mathematical demonstrations are unshakable. It can be found throughout Kant’s book. Another one is the sharp distinction between mathematics, as a science of arbitrary subjects created by men, and natural subjects, created by God—which is precisely the theological response to the unloved invincibility of mathematics. Whereas mathematics is ruled by necessity, thus allowing for strict demonstrations, nature is more complex and less necessary. Therefore, no strict demonstrations can be given for real and natural things. For Kant, this is a sharp qualitative distinction aiming for different rules in mathematics and mechanics on the one hand and physics (excluding mechanics) and other disciplines on the other hand. Kant also shares the sharp distinction between mathematical and natural bodies

54  “Wenn man die Kraft der Körper überhaupt nur eine wirkende Kraft nennt, so begreift man leicht, wie die Materie die Seele zu gewissen Vorstellungen bestimmen könne” (Ak 1:21, §6). 55  “Beide Schwierigkeiten verschwinden aber, und der physische Einfluß bekommt kein geringes Licht, wenn man die Kraft der Materie nicht auf die Rechnung der Bewegung, sondern der Wirkungen in andre Substanzen, die man nicht näher bestimmen darf, setzt. Denn die Frage, ob die Seele Bewegungen verursachen könne, das ist, ob sie eine bewegende Kraft habe, verwandelt sich in diese: ob ihre wesentliche Kraft zu einer Wirkung nach draußen könne bestimmt werden, das ist, ob sie außer sich in andere Wesen zu wirken und Veränderungen hervorzubringen fähig sei?” (Ak 1:20, §6). 56  It is generally accepted that Kant’s allusion to the “sharp philosopher” (Ak 1:21, §6) pointed to Knutzen. I do not see how though how this passage can be interpreted as a massive criticism of Knutzen as suggested by Kuehn (Kuehn 2003, 93).

Ursula Goldenbaum  49 (§114) that can be found not only in Crusius and Baumgarten57 but clearly is a common pattern in anti-­Wolffian literature since the 1720s, even in Lange or Löscher 1735–1742, 1736, 36–44). A third very typical Pietist pattern is the tendency to separate Wolff and Wolffianism from Leibniz (cf. §§2, 102–106 and Erläuterung des 105ten §)—a separation taken up by Hegel and his school of the history of philosophy. While the great Leibniz is praised, in spite of a few misleading formulations or even mistakes, Wolff and the Wolffians are attacked as fundamentally wrong, simplifying Leibniz and being reckless in their machine-­like drawing of conclusions. Here I want to focus though on Kant’s version of the separation of mathematics and mechanics from real or natural things; that is, his distinction of two qualitatively absolutely different kinds of motion (§15), by which he tries to secure the influxus physicus between body and soul. The one motion remains within the body to which it is communicated and lasts eternally if no obstacle comes its way. The other is a continuous effect of a continuously driving force where it does not even need a resistance to be annihilated. It is due solely to an external force and thus vanishes as soon as this force ceases to move it. While, at first glance, the first kind of motion seems to be local motion according to the principle of inertia, it is rather (for Kant) a motion that is due to an active force, and it will eventually violate this very basic principle of modern mechanics (as we will see below). However, a violation of the principle of inertia is already clear with the second kind of motion: in §16, Kant explicitly states that the second kind of motion can vanish as soon as the driving force ceases, without any resistance. The first kind of motion is not, contrary to first impressions, mechanical at all. What Kant calls “living force” deviates from the concept of both parties of the controversy, who both discussed mechanical forces. Kant subsumes both dead and Leibnizian living forces as they are discussed in the controversy as simply “dead forces.” In contrast, Kant’s “living force” is described as something very special (§16), ringing a bell of Lutheran metaphors.58 Not only is it alive, it can make a force living; it is eternal, somehow infinite, and free; it is rooted in an inner spring and able to nourish itself (§17). Of course, it has nothing to do with dead force, which is all mechanical force, related to mechanics (mathematics), and thus only momentary and—moreover—exclusively external. This mysterious living force of a body, moving freely through an infinitely subtle space can al­leged­ly be estimated by the sum of all its effects in eternity.

57 There is discussion whether Kant picked this up from Crusius or from Baumgarten, see Kuehn 2003, 447 n. 136. 58  Other Pietists as well bluntly connected Leibniz’s well-­defined term “living force” to the Lutheran pattern of “living faith” or of Thomasius’s “living knowledge.” See the intriguing essay by Alessandro Nannini about Alexander Baumgarten’s use of Leibniz’s term of “living forces” in his metaphysical project (Nannini [forthcoming]). Kuehn claims though that Kant would defend Leibniz’s living forces (Kuehn 2003, 91) although Leibniz did not talk about living forces other than such within mechanics.

50  How Kant was Never a Wolffian Already by his distinction of forces, Kant can easily be distinguished from Leibniz or Wolff although both do indeed also talk about forces on a metaphysical and a mechanical level. But they do not mix them. For Leibniz as well as Wolff, and their true partisans, living and dead forces belong to mechanics, as derivative forces, while the primitive forces—that is, potentiae agendi et patiendi—are treated within metaphysics and are not exposed to any measuring at all. Moreover, it is fundamental for Leibniz and Wolff that there is no way to give up the principle of inertia together with the law of conservation of force as Kant continuously does in his book. With a sigh, the Kant editor Adickes admits: “Wolff thinks much more in the spirit of authentic natural science than Kant. Nowhere . . . does he violate the law of inertia” (Adickes 1924a, 96, trans. UG). Having reduced the Leibniz-­Wolffian notion of living force to dead force, Kant uses the extended second part of the book with all its “mathematics” to show that the Cartesians are right to measure dead force—that is, all mechanical force (including Leibniz’s living forces)—by mv. It is here that Kant goes through all the literature mentioned above, but only in order to use the Cartesian arguments against the Leibnizians and to show that there are no living forces in mechanical motions. It is not my intention to discuss this further. Even Adickes resigned from going through all of Kant’s misunderstandings and mistakes (although he provides a pretty detailed description and explanation of Kant’s earliest book— certainly the best we have so far). I only want to point to Kant’s fundamental misunderstanding (to put it mildly) of Leibniz’s and Wolff ’s distinction of living and dead forces. They simply intended to distinguish a mere striving (conatus) to move from real local motion whereby they had the calculus available to calculate the infinity of infinitesimal solicitations that would bring about real motion. Kant instead abused Leibniz’s law of continuity by saying that if we think of a real motion as becoming smaller and even infinitely small it could be considered as rest and thus there would be no sufficient reason to distinguish between rest and mechanical local motion at all. Thus, real motion would be as much as pressure (a mere striving for motion) simply dead force, whereas living force would be something completely different in kind, as we already learned (§§ 24–26). As a result, Kant concludes, in nice agreement with Pietist patterns, that there could never be any consideration of living forces within mathematics (i.e. mechanics) (§28). The third part suggests Kant’s “solution” of the controversy which is simply to ascribe dead force to mathematical or mechanical bodies and living force to natural bodies that can act on each other. In order to ascribe the measure of mv2 to his living forces, Kant comes up with the “concept” of “intension” by which he understands something that occurs when a dead force becomes a living force, or gets “vivificated.” It has to be admitted that Kant’s term “vivification” is borrowed from Wolffians such as Madame de Châtelet. However (and again), they understand by that expression the integration of infinitesimal quantities of strivings to motion over time. Kant, lacking any understanding of the calculus, uses these terms to speculate about a new mysterious process driven by the inner,

Ursula Goldenbaum  51 infinite, free, and eternal forces of natural bodies. In §115 he underlines that a natural body differs from a mathematical one by having not only the force gained from an external body but one that it can increase in itself and by itself. In §117 he calls this force intension and attributes it to bodies with real force as distinguished from those external (mechanical) forces which cannot last. Thus, intension is related to a power of the body to persevere. In order to estimate the entire force of such a body with living force one needs to multiply the body’s velocity by the “intension,” assumed to be proportional to velocity. As a result of this magic argument, Kant then declares mv2 as the measure for living forces that can only be found in natural (but not in mathematical) bodies (§§122–124). §136 then states that living force can vanish without effect as well as originate from nothing, violating again the principle of inertia and the law of conservation (if this would still matter in such a wild metaphysical speculation). Kant even adds a completely inappropriate argument from experience(!) to drive his point home, thereby showing his poor understanding of experience in comparison to the Wolffian standards (§117).59 Adickes regrets that Kant, by violating the law of conservation, followed Newton instead of Leibniz thus being led to the false path (Adickes 1924a, 80). I would say, rather, that Kant chose to follow Newton because the latter’s view was a better fit to his Pietist understanding of the world. The lack of a real solution in Kant’s first book to the controversy concerning the estimation of forces is often excused by Kant scholars who point to the fact that Kant could not know yet that d’Alembert had solved the problem in 1743. Beside the fact that d’Alembert did make his point only after the public debate of Maupertuis and Samuel König in 1752,60 d’Alembert is not of any relevance for Kant’s argument, which does not discuss the same question at all. Kant did not even come close to meeting the standards of Wolffian or Cartesian discussion of the topic, and certainly not those of more recent mathematics. He simply does not speak about the same living force that d’Alembert, the Cartesians, or the Wolffians talked about, but about a mystical inner force of natural (i.e. non-­ mechanical) bodies that comes rather close to the active principles of Newton’s Queries of his Optics. I consider Adickes’s judgment as much more to the point when he declares the acknowledgment of the principle of inertia to be the key criterion when it comes to evaluating Kant’s achievement in his earliest book: It does not come as a surprise that the 22-­year-­old did not resolve a controversy whose participants included greater men than he. He cannot be blamed for that. 59  Adickes sighs, “daß Kant hier, wo er, soweit mir bekannt ist: zum ersten und einzigsten Mal in seinem Leben, eine Art Experiment anstellt, der Frucht seiner Bemühungen durch eigene Schuld verlustig geht, indem er nicht, wie es die Art methodischer Wissenschaft wäre, in den Dingen selbst nach tieferen gesetzmäßigen Zusammenhängen forscht, sondern die festgestellten Tatsachen in blinder Voreingenommenheit ohne weiteres in die Zwangsjacke seiner Lieblingstheorie einspannt” (Adickes 1924a, 102). 60  D’Alembert does so only in his later edition of the Traité from 1758, due to the public debate between Samuel König and Maupertuis. See Goldenbaum 2004, 509–651, here 635–637.

52  How Kant was Never a Wolffian But he can indeed be blamed for putting himself in sharp opposition to the principles of healthy natural science by a solution that includes a violation of the principle of inertia.  [Adickes 1924a, 69, trans. UG]

Despite its title and the extended discussion of scientific literature, Kant’s earliest book was far from being a solid scientific or mathematical book.61 Moreover, it did not fulfill what its author had promised in the preface: to solve the controversy of the Cartesians and Leibnizians about the true measure of forces. Instead, as becomes clear from the very first paragraphs, this book is written to solve the main problem of anti-­Wolffian philosophers and theologians, by showing the possibility of the influxus physicus.

4  Contemporary Judgment of Kant’s Book In order to conclude my argument that Kant was never a Wolffian, not even in his very first book, and rather that he set out from the Pietist discourse, I would like to point to the few judgments by Kant’s contemporaries that have come down to us. The best-­known contemporary statement, when it comes to a discussion of Kant’s Estimation of Forces—a statement mentioned by almost all Kant scholars— is certainly Lessing’s infamous verse (the epigraph quoted at the start of this chapter). But they quote it only to point out, with satisfaction, that Lessing “omitted” it in his later editions.62 This alleged omission is thus interpreted as if Lessing had changed his mind, although nobody gives any evidence for such an in­ter­pret­ ation. In addition, there lingers the silent objection whether the young Lessing had sufficient judgment regarding the challenging scientific and mathematical subject that Kant had taken up. But, in spite of the quite numerous mentions of Lessing’s verse in Kant scholarship, nobody seems to wonder what made Lessing write such an ironic dismissal, given his presumed lack of understanding? To be sure, although Lessing is notorious for ironic and sometimes sardonic criticism, he is equally famous for his precise and solid judgment. Moreover, he set up rules for criticism according to which beginners deserve a milder treatment. But what if Lessing was not the author—or not the only one? What if somebody else wrote the little verse, somebody who was sufficiently equipped with mathematics, mechanics, and metaphysics to be a solid judge of Kant’s writing? The little poem appeared indeed only once, in the weekend supplement of the 61 “Das tritt besonders klar hervor, wenn man die Schrift mit den Arbeiten wirklicher Naturwissenschaftler wie Euler, van Musschenbroek, Joh. Bernoulli, d’Alembert usw. vergleicht” (Adickes 1924a, 137). 62  Kant’s most recent biographer Kuehn states again: “Properly, Lessing suppressed this epigram in later editions of his work” (Kuehn 2003, 95). In fact, Lessing never included it in any of his editions, not even the one he published the same year. Cf. Lessing (1751a) .

Ursula Goldenbaum  53 Berlin newspaper that was edited, in alternating fashion, by either Lessing or--his best friend Christlob Mylius. Both had studied in Leipzig with Abraham Gotthelf Kästner. However, Mylius had focused on mathematics and science and, after winning the second place for his Preisschrift on winds in 1749, second to d’Alembert who won the price,63 he had moved to Berlin to work with Leonhard Euler.64 There he translated Maupertuis and Clairaut into German, made contact with the Wolffian mathematician Samuel König, built a European-­wide scientific correspondence, and regularly reviewed scientific literature for German journals.65 In 1749, Euler recommended him to the Academy of Petersburg66 but Mylius chose instead to lead a scientific expedition to East India in 1753, mentored by Albrecht von Haller. Unfortunately, he died on his way (in London) in 1754, at the age of 31. Christlob Mylius was clearly prepared to understand and make judgments about the topic of the estimation of forces and thus about Kant’s writing, in light of the then-­current scientific standards. It was due to the almost forgotten Mylius that the now famous Lessing had moved to Berlin in 1749 where the friends were roommates. They both lived from writing journals and published mutually in the journals one or both of them edited. There is quite some dispute among Lessing scholars about how to ascribe authorship of the anonymous reviews and contributions for the Berlin newspapers to Lessing or to Mylius—due to their similar witty style (Consentius 1902; see also Nisbet 2008, 136). Thus, the infamous Kant verse may well have been written by Mylius, whose judgment about Kant’s writing would be sound. But even if Lessing was the author of the verse, he certainly did not write it without solid advice from Mylius or even— from Samuel König, who visited Berlin in the fall of 1750 and met Lessing. Thus, there is no reason at all to assume that the verse was an accident due to a lack of insight on the part of the young Lessing who later omitted it after he came to acknowledge the great Kant. The verse was written for the day though, or rather for the weekend of the Berlin newspaper. It could not even be understood without its context, the then-­recent publication of Kant’s book. That’s it! There is no evidence (and no reason) that Lessing or Mylius should ever have recanted the verse. Moreover, their short dismissal is in clear agreement with the serious but also quite sarcastic review of the Nova Acta eruditorum whose author calls Kant’s book “soni sine mente” (i.e. sounds without meaning) (Nova Acta eruditorum 1752, 178).

63  For the Preisfrage see Harnack (1900, 305). For the fact that Mylius got the “Accessit”; that is, the second place, see Kästner to Maupertuis on May 16, 1752, in Le Sueur (1896, 290). 64  About Mylius see Trillmich (1914). 65  Cf. Lessing’s biographical preface to Mylius (1754, xxxvi). See ibid., 405. 66  Euler to Schumacher on December 18/29, 1749, in Die Berliner und die Petersburger Akademie (Berlin: Akademie Verlag, 1959), 185. Cf. also Euler’s statement from February 10/21, 1750, ibid., 189–190.

54  How Kant was Never a Wolffian The highly regarded and well-­ known Nova Acta eruditorum, publishing reviews and papers not only from and about German but also European scientists and scholars, was run by Friedrich Otto Mencke, the grandson of the founder of the Acta eruditorum, Otto Mencke. All three editors, grandfather, father, and son, were sympathetic to the Wolffians but strove for impartiality and high scientific standards.67 The author of the Kant review simply states that he does not understand what action of a substance beyond motion could be. He asks more questions, such as how a substantial force could be determined to act outside itself or to suffer the force of another substance. He then continues ironically: “Sed alia adhuc in eodem Capite habentur, sublimiora multo, quam quae capi a nobis possint” (Nova Acta eruditorum 1752, 178). To put it mildly, this review totally dismisses Kant’s book and does not take it as a serious contribution to the scientific controversy at all. The review in the Göttingische Gelehrte Anzeigen omits any evaluation of Kant’s book and limits itself to a mere report of Kant’s own argument—not unusual for reviews at that time. The author focuses on the metaphysical arguments, neglecting all the mathematical efforts of the second part. This not only suggests that the reviewer is not a Wolffian author, it also excludes any recognition of the controversy about the true measure of forces). What is particularly emphasized, however, is Kant’s strict distinction of “vis activa” and “vis motrix” in order to show that motion is not the only effect of force (Göttingische Gelehrte Anzeigen 1750, 291). The author then explicitly points to Kant’s actual problem: the alleged Wolffian confusion of “active force” and “vis motrix” that had allegedly caused the difficulties in understanding the mutual influence of bodies and souls. He also emphasizes the significant distinction between (1) motions that can persevere by themselves within the bodies to which they are communicated (continuing to eternity) and (2) those that perish as soon as the external force vanishes. Ultimately, the author of the review somehow confused the two sorts of motions— calling the first “dead” and the second “living” force (just the opposite of the way Kant had suggested). The author attaches importance to the statement that the book, although refuting (or rather determining more precisely) Leibniz’s measure of force, does so with appropriate respect toward the great Leibniz. An even more sympathetic review, though, appeared in the Frankfurtische Gelehrte Zeitungen, as early as on November 14, 1749 (reprint in Fischer [1985b, 214–218]). This very first review that we know of, was only noticed in Kant scholarship in 1985 by Harald-­Paul Fischer. Fischer showed, by agreements in wording 67  When Kästner, whose letter is often quoted to give evidence for the decline of the journal, wrote to Maupertuis that he should not bother about this old insignificant Latin journal, it should not be ignored that Kästner was trying to calm down an angry Academy president who had been denied his request to republish the Jugement against Samuel König in 1752 in this journal. See Goldenbaum 2004, 520. Notwithstanding this letter, Kästner was a regular contributor and supporter of the journal, beyond his time at the University Leipzig.

Ursula Goldenbaum  55 between the review and the very first letter of Kant that has come down to us, that Kant wrote this letter on August 23, 1749 to the Frankfurt newspaper (Fischer [1985a, 81]). We know of two letters Kant dated on that day. Both are cover letters he sent with his book in order to ask for comments and/or a review. One of them, a quite respectful personal letter, was found a few decades ago in Euler’s cor­re­ spond­ence in Petersburg (reprinted in a corrected version by Fischer [1985b, 214–218]). Leonhard Euler though, who could well have been considered a pos­ sible ally against the Wolffians,68 never answered the Kantian request for a review of his book. It seems that he did not even respond to the urging letter.69 The other letter has long been considered to have been directed to Albrecht von Haller, the editor of the Göttingische Gelehrte Zeitungen, thus allegedly giving rise to the known review in that journal on April 13, 1750 (290–294). Fischer’s argument that, instead, Kant wrote this letter to Frankfurt is however convincing, although it cannot be excluded that Kant wrote more letters of this kind and content, maybe to Göttingen as well (or even to Leipzig). However, Fischer is certainly right that Kant would have addressed the famous Albrecht von Haller rather in the same respectful way as he addressed Euler. The review from Frankfurt is not only written in a very sympathetic way but is also extraordinarily long.70 This is rather surprising because this journal did not belong to those journals that usually reviewed scientific or mathematical books as Fischer rightly underlines: “Reviews of philosophical-­scientific writings have been rarely published in the Franckfurtische Gelehrte Zeitungen. They rather focus on the announcement of juridical, exegetical, historical and economical works, i.e. writings that are concrete, informative, and of practical use. Theoretical essays such as that of Kant are exceptions.”71 Therefore he wonders how Kant’s writing would fit in such a journal. But in light of my argument in this chapter, there is no reason to be surprised at all. In fact, Kant’s project fits very well in the Franckfurtische Gelehrte Zeitungen. This journal had been initiated by the Pietist pastor Christian Münden in 1736, at the peak of the public debate of the Wertheim Bible and Wolffian philosophy. Münden had contributed a preface to Pastor Firnhaber’s Defense—a work directed at Firnhaber’s church administration explaining that he, being pastor at Wertheim, had indeed tried hard to hinder the  publication of the Wertheim Bible even if he did not succeed.72 Indeed, the Franckfurtische Gelehrte Zeitungen was from its foundation firmly under the 68  Laywine rightly points to this possible alliance (1993, 27–34). 69 Euler had published his own paper on the measure of force in the Memoires of the Berlin Academy (Euler (1746). It might well be that Kant’s emphasis on the fact that the printing of his book had already begun in 1746 (repeated by the authors of the reviews in Frankfurt and Göttingen) is due to his learning about Euler’s essay when it was too late to include consideration of it. 70  “Umso mehr verwundert, daß der Kantischen Schrift ein so breiter Raum zugestanden wird, denn in dem Zeitraum von über drei Monaten (3.Okt.-5. Dez.) ist ihre Besprechung die mit Abstand umfangreichste” (Fischer 1985a, 85). 71 Ibid. 72  Cf. Firnhaber (1737); about Firnhaber see Goldenbaum (2004, 302–303).

56  How Kant was Never a Wolffian control of the Frankfurt Pietist theologians until it were taken over by Goethe and his friends in 1772. The author of the review underlines precisely those points I mentioned above as being essentials of Kant’s book. He agrees with Kant that Leibnizian dead and living forces would in fact belong in the same species whereas truly living forces would belong to a totally different species of forces. In addition, the author is convinced that Kant’s “mathematical demonstrations” of the second part had sufficiently refuted all the famous Wolffians and shown that exclusively the Cartesian measure of force could be used for mathematical bodies. He then raises the big question whether there could still be a law in nature that had no place in a mathematical theory of motion and is glad to report that this has indeed been shown by Kant (Fischer 1985a, 88). He concludes his review with a summary of Kant’s deviations from Leibniz, underlining above all that living forces demand free motion that can preserve itself by means of its inner natural force (ibid., 89), thus being totally different in kind (toto genere) from any mechanical motion. It is obvious that this author had no problems at all understanding Kant’s genuine project, very much in contrast to the first review. However, he seems not to be trained in mechanics or mathematics at all. Thus, the sympathetic author of this review did recognize Kant’s book precisely as what it was—an attempt to refute the Leibnizian measure of force (by allegedly purely scientific methods) and to come up with a proof of the possibility of influxus physicus against the Wolffians, allegedly defeating them on their own territory. Moreover, the fact that Kant sent his book to the Frankfurt journal—a journal that was certainly not appropriate for evaluating scientific books (as Fischer rightly underlines)—indicates that Kant attached importance to a recognition of his book by the editors of this Pietist journal—recognition of it as a proof against the Wolffians and in favor of the influxus physicus. But no true Wolffian or Cartesian, and neither d’Alembert could ever have seen Kant’s book as a serious contribution to their discussion on the true measure of force.

3

Breaking with Rationalism Kant, Crusius, and the Priority of Existence Eric Watkins, University of California, San Diego*

Any adequate interpretation of Kant’s overall argument in the Critique of Pure Reason must be able to provide a compelling explanation of his most fundamental reason for rejecting the view he associated with “pure reason”; that is, with the kind of rationalist position developed by his immediate predecessors such as Leibniz, Wolff, and their followers. For Kant’s background and career are set entirely within the German tradition, and even though early Pietist strands of influence were later supplemented by a fascination with the British and Scottish empiricists from the late 1750s through the 1770s, this tradition was primarily rationalist in its philosophical orientation. Indeed, Kant’s audience and intended target in the first Critique, which he decided to write in German, not Latin, must have included precisely such German rationalists.1 Now it is completely natural to attempt to provide a philosophically rigorous and historically plausible explanation of Kant’s break with rationalism by focusing on his numerous explicit assertions about Leibniz, Wolff, and pure reason that occur in the first Critique, and it is to this end that various scholars (myself included) have undertaken interpretations of the Amphiboly, the various main sections of the Transcendental Dialectic (the Paralogisms, Antinomies, and Ideal of Pure Reason), the Doctrine of Method, and the second edition Preface. However, it is far from clear that any of these accounts has been truly successful in fully explaining Kant’s fundamental break with the rationalists. While many valuable interpretations of particular texts and incisive analyses of the relevant arguments are on offer, the attempts that are then made to move from these various specific points to a single larger idea or issue that is allegedly decisive with respect to his break with the rationalists, immediately face formidable problems. For example, it might naturally be thought that Kant’s fundamental break with the rationalists is due to his claim that the propositions of geometry consist of *  I thank John Grey, Tim Jankowiak, James Messina, Marius Stan, and Clinton Tolley, as well as audience members at the conference organized by Brandon Look at the University of Kentucky in September 2009 for helpful comments on earlier versions of this chapter. 1  For a selection and translation of texts from this context, see Watkins (2009). Eric Watkins, Breaking with Rationalism: Kant, Crusius, and the Priority of Existence. In: Leibniz and Kant. Edited by Brandon C. Look, Oxford University Press. © Oxford University Press 2021. DOI: 10.1093/oso/9780199606368.003.0003

58  Breaking with Rationalism synthetic a priori cognitions, a claim that is, in turn, supposed to entail Transcendental Idealism, but such a claim could easily seem to beg the question against Leibniz, who regards the truths of mathematics as, in effect, analytic through and through and would thus be completely unmoved by the examples Kant cites in support of his claim.2 Similarly, one might think that Kant’s most basic reason for rejecting Leibniz’s position is his epistemological requirement that to be cognized, objects must be not only thought, as Leibniz held, but also intuited. Yet insofar as Leibniz’s pre-­established harmony precludes the very possibility of causal interaction between finite substances, he has already rejected one of the basic presuppositions of Kant’s account of empirical intuition, namely that objects must ‘affect’ or act on us causally for sensible intuition to be possible. Like the view that emphasizes synthetic a priori cognition, this line of in­ter­pret­ ation can easily seem to beg the question against a Leibnizian rather than to justify a rejection of any rationalist position. Other attempts at explaining Kant’s break with the rationalists encounter difficulty by lacking explicit philosophical support in the text of the first Critique. Though Kant claims that cognition requires that we be able to establish the real rather than the merely logical possibility of objects of knowledge, the claim is not supported by any explicit or detailed justification in the text.3 Further, Kant’s crucial insistence that there be a distinction in kind rather than degree between our sensible and intellectual representations (intuitions and concepts) is similarly never given any sustained defense by Kant in the text. Kant’s argument in the Transcendental Aesthetic that our representation of space is an a priori intuition rather than an empirical concept does not establish that intuitions are different in kind from concepts,4 and in the Amphiboly (e.g. at A271/B327) the difference in kind is noted rather than explicitly argued for. Without having canvassed all possible attempts or, for that matter, discussed any of them in detail, it is tempting to suspect that Kant’s break with the rationalists was not based on a single simple move, or change of position, that is articulated and defended straightforwardly in the first Critique.5 In light of this suspicion, the possibility that Kant’s break with the rationalists might rest on a number of crucial ideas rather than one simple revolutionary thought, and further, that some of these ideas might lie relatively well hidden 2  For discussion see Anderson (2015). 3  See e.g. Chignell (2010a and 2010b). 4  It is true that the third and fourth arguments of the Metaphysical Exposition are relevant here insofar as they seem to rest on the assumption that concepts cannot be singular and cannot have in­fi­ nite intensions. However, he provides no arguments for these features of concepts, nor for characterizing the distinction between intuitions and concepts as a difference of kind. 5  Kant does construct an argument in the Prolegomena for the distinction between sensibility and understanding that is based on incongruent counterparts. However, in addition to the fact that Kant’s distinction between the faculties is in place well over a decade before he makes use of this argument to this end, it is not clear that it succeeds as an argument for the distinction being a distinction in kind rather than degree.

Eric Watkins  59 from view in his pre-­Critical period attains a certain degree of prima facie plausibility. Now the Critical Kant is typically dismissive of his own pre-­Critical position, suggesting, for example, that it is a “poorly understood” version of Leibniz’s position.6 Yet even if that were true, it could still provide some indication of the path that Kant actually took in arriving at his Critical position, and it is clearly possible that some of the insights that he arrived at early on are retained in the Critical period, even if in highly modified form due to the radical changes his position underwent in other respects. Further, the pre-­Critical Kant is by no means a dogmatic and unthinking imitator of either Leibniz’s or Wolff ’s views, a fact that is revealed by his decidedly independent thoughts about causality and Newtonian physics, to note only two examples of major innovations he undertook during this period.7 Kant’s pre-­Critical period thus offers fertile ground for ideas that, if taken together, could contribute to an explanation of his break with rationalism.8 To focus on those views from Kant’s pre-­Critical period that are most likely to be relevant, however, it is useful to consider the historical context in which Kant developed these views, and here the position developed by Christian August Crusius is particularly significant.9 Though Crusius’s position has deep roots in rationalism—the title of one of his main philosophical works is Sketch of the Necessary Truths of Reason—he is openly and deeply opposed to many of Leibniz’s and Wolff ’s fundamental tenets. In addition to his oft-­cited rejection of the principle of sufficient reason, he adopts a libertarian notion of free will and develops a comprehensive philosophy that draws inspiration from his Pietist precursors, who waged a public and at times vehement polemical battle with Wolff and his followers. For current purposes, however, what turns out to be most significant is not any of Crusius’s better-­known, explicitly articulated criticisms of 6  See e.g. Kant’s remark to this effect in the Metaphysical Foundations of Natural Sciences (4:507). 7  For discussion of Kant’s innovations on causality, see (Watkins 2005). For the pre-­Critical Kant’s relation to Newton’s physics, see both Kant’s Universal Natural History and Theory of the Heavens and Watkins (2013). Also relevant to this latter issue is Watkins (1997). 8  For detailed discussion of the pre-­Critical Kant’s reaction to Leibniz’s, Wolff ’s, and Baumgarten’s views on arguments for, and the different kinds of, simple beings, see (Watkins 2006). 9  Though Crusius is widely recognized by more scholarly and historically inclined interpreters as an important figure for appreciating the genesis of Kant’s views, more philosophically inclined readers have generally ignored his philosophy. The most important discussions of Crusius and of his influence on Kant are by Ernst Cassirer (Cassirer 1907); Heinz Heimsoeth (Heimsoeth 1956), esp. Part III, “Metaphysik und Kritik bei Chr. A. Crusius: Ein Beitrag zur ontologischen Vorgeschichte der Kritik der reinen Vernunft im 18. Jahrhundert”; Georgio Tonelli (Tonelli 1967); and Max Wundt (Wundt 1924 and 1945). With the exception of Tonelli’s relatively brief encyclopedia article, however, these discussions focus primarily on similarities between Crusius’s and the Critical Kant’s positions. For example, Heimsoeth shows that Crusius, despite his adherence to principles that Kant would ul­tim­ ate­ly view as dogmatic, anticipates Kant’s Critical position by holding a distinction between intuitive and symbolic cognition that bears some similarity to Kant’s distinction between knowledge and thinking, and by claiming that primitive forces are unintelligible to us, which is analogous to Kant’s claim that we cannot attain insight into the inner constitution of basic forces. Here, I propose to investigate Crusius’s influence on the pre-­Critical Kant such that it is then possible to ask what effects he might have had on the Critical Kant.

60  Breaking with Rationalism Leibniz and Wolff, but rather the basic philosophical structure that is set out at the beginning of the Sketch, where he distinguishes between existence and essence and attributes priority to the former over the latter in the course of providing detailed analyses of what belongs to the essence and existence of things. For it turns out that this structure is highly relevant to his own most fundamental reasons for rejecting rationalism. If one then turns to Kant’s pre-­Critical works with this context in mind, one has reason to attend to those aspects of his position that have not been thought to be particularly significant either in their own right or for their implications for the Critical turn and its rejection of the claims of pure reason. Though Kant adheres to elements of a broadly rationalist position in the New Elucidation by arguing for the principle of determining ground (a variant of the rationalists’ principle of sufficient reason), he restricts its application to contingently existing things. God’s existence, in particular, must be exempt from this principle, since there can be, he argues, no ground for God’s necessary existence. Then, in The Only Possible Argument, Kant argues in greater detail that God must be the ground of all possibility such that (God’s) existence has priority over all possibility, which constitutes the essence of things. He also argues there that existence is not a real predicate that would be intelligible to us (since real predicates would pertain to a thing’s essence). Instead, existence is presented as a simple absolute positing of a subject. So not only does existence lack a sufficient reason in the case of any necessary being, but at the most fundamental level existence is also, in some sense, an unintelligible brute positing that precedes all essence.10 What we thus find in two of Kant’s central works during the pre-­Critical period, the New Elucidation and The Only Possible Argument, is a position that resembles Crusius’s basic philosophical framework in several crucial respects. Specifically, he accepts the distinction between existence and essence as fundamental, posits the priority of existence over essence, and then uses these views to forge a deep and decisive break with rationalism. In §1 of this chapter, I lay out the basic structure of Crusius’s metaphysical position, explain the distinction he draws between essence and existence, and present his claim that existence has priority over possibility. In §2, I present Kant’s discussion of the principle of determining ground in the New Elucidation and his explicit statements regarding Crusius’s position before turning to the further development of his views concerning possibility and existence in The Only Possible Argument. In §3, I argue that the pre-­Critical Kant is influenced by Crusius in his fundamental break with Wolff ’s rationalism on the grounds that both Crusius and the pre-­ Critical Kant hold that possibility must be metaphysically grounded in something that exists without itself being grounded 10  This point is consistent with saying that the possibility of finite things precedes their existence, since the existence of God precedes the existence of the possibility of finite things.

Eric Watkins  61 in anything else. I also argue that Kant’s view at this time is best understood as an attempt to steer a middle course between the views of Wolff and Crusius. In the fourth and final §4, I present a brief sketch of how Crusius’s influence on the pre-­ Critical Kant’s break with rationalism might have implications for basic aspects of Kant’s Critical position, specifically the distinction he draws between sensibility and the understanding, which is fundamental to much of the Critical philosophy. Though the interpretation I present in these sections falls well short of providing a comprehensive explanation of Kant’s break with rationalism, one can see how it may be one indispensable element of a broader account.

1  Crusius on Existence and Essence In the first chapter of his Sketch of the Necessary Truths of Reason, published in 1745, Crusius provides an overview of the structure of his metaphysics, claiming that a comprehensive metaphysical system can emerge from a proper analysis of any object that is present to our senses: §9. Outline of Ontology. Thus, whoever is attentive and acute enough can abstract the entirety of ontology from any actually present object that comes before our senses, or at least repeat the entire science by observing what can be noticed and distinguished in that case. And precisely so that this can occur all the more easily, I want to provide a preview of the outline of ontology by means of which it will become easier to have insight into how one must be able to recognize the entire science from any example. First, I shall dissect the manifold that can be distinguished in the essence and existence of any actually present thing. Thereafter I have to explain those possible differences of things that can be surveyed a priori from the necessary essence of any actual thing due to which it must be simple or composite, necessary or contingent, finite or infinite.11

In this passage Crusius holds that the most basic distinction that must be made on encountering an actual object given to our senses is that between essence and existence. Only then can one consider those features found in the essence of a thing that can distinguish it a priori from other things. So, how should this distinction between essence and existence be understood? What belongs to the essence and what to the existence of a thing? Crusius’s initial characterization of essence occurs near the end of the second chapter of the ontology section of his Sketch:

11  The original German text is reprinted in Crusius (1964, 15–16). The English translation is found in Watkins (2009, 139).

62  Breaking with Rationalism §17. What the metaphysical essence is. For any thing whatsoever—one may take the word in either a broad or a narrow sense—one must think something through which one distinguishes it from other things. What one thinks for such a thing and what distinguishes it from others I shall call the metaphysical essence of a thing.  [Crusius 1964, 29–30; Watkins 2009, 141]12

Shortly thereafter, at the start of the next chapter, Crusius reiterates his basic characterization of essence and then draws an explicit contrast between it and existence: §19. Explanation of the enterprise. In an actual thing the existence and the essence are to be distinguished. And in this extended meaning, since one sets essence in opposition to existence, one considers as belonging to it [i.e., to the essence] everything that one thinks at all of a thing and through which one distinguishes it from other things. I have called this the metaphysical essence, §17. In the present chapter, we want to investigate it further and see what is to be distinguished in it.  [Crusius 1964, 31; Watkins 2009, 141]

Essence is explained in this passage in terms of what distinguishes one thing from another and this is distinct from the existence of that thing. In the remainder of the third chapter of the Sketch, Crusius distinguishes and clarifies a long list of notions that pertain to a thing’s essence: the metaphysical subject, the logical subject, accidents, substance, determination, properties, power (Kraft), causality, real and ideal grounds (a priori and a posteriori), active and existential grounds, fundamental essence, attributes, modes, as well as necessary and contingent essences. Though Crusius uses these notions to contrast his own position with Wolff ’s in several ways, he and Wolff are in fundamental agreement that one crucial function of an essence is that it allows one to distinguish one thing from all others. In the fourth chapter of his Sketch, Crusius then turns to provide a more detailed account of existence. §46. What existence is. If we represent something as existing, then the essence of our understanding requires us, apart from that through which we think it and distinguish it from others, to think also this in addition, that it exists somewhere and at some time, and thus we must also add in thought, beyond the meta­phys­ ic­al essence of the thing, a where and when (ubi & quando) that is attributed to it. For that reason existence is the predicate of a thing due to which it can also be found outside of thought somewhere and at some time. [Crusius 1964, 73; Watkins 2009, 145]

12  I have amended the translation slightly.

Eric Watkins  63 Crusius is making two basic points here. First, the existence of a thing is to be distinguished from the metaphysical essence of a thing. While the defining feature of a metaphysical essence is that it allows one to distinguish one thing from others, existence is defined as something positive that is not contained in the metaphysical essence. Second, Crusius clarifies the concept of existence by connecting it with the concepts of space and time. Specifically, the concept of existence necessarily contains the concepts of space and time such that the concepts of space and time can be abstracted from the concept of existence (as opposed to being abstracted from a thing’s essence, as was the case for various notions that were distinguished in the third chapter).13 As a result, Crusius accepts as a necessary truth that if something exists, it must do so somewhere and at some time (§48). Later in the Sketch, Crusius develops distinctive explanations of how this doctrine can be made consistent with the existence of God and the soul (which might seem not to be spatial). Thus, in §250 Crusius argues that given that God is an infinite and perfect substance, his existence in space entails his immeasurability (Unermesslichkeit) or immensity (immensitas)—God can be neither enclosed in nor excluded from space, but rather fills every space in the most perfect way— and in §§251–253 Crusius defends this position against several objections (such as whether this doctrine makes space into an eternal thing like God). In §§254–255, Crusius then maintains that God’s existence in time is tantamount to his being eternal, which is a perfect kind of duration, one without beginning or end. Similarly, the soul, in spite of its immateriality, can be said to fill a space by virtue of its capacity to act on matter, but without thereby being extended (§362). The passage that provides Crusius’s description of his concept of existence (§46) then continues as follows: That this is the true concept of existence can easily be comprehended thus. One posits what exists in opposition to what is merely thought or, which is the same, one distinguishes the actual being [das wirkliche Dasein] of a thing from its mere being [Sein] in thought. Mere being in thought constitutes possibility, as I will show further shortly. For that reason, some say that existence is complementum possibilitatis, which, however, can mean nothing other than that it consists in whatever is added that is required beyond possibility if an existence is supposed to arise. This is true, but it does not solve anything. Now when we pay attention to what the positive is that is added to the possible when it exists, it is impossible for us to think something other than this, [namely] that a where (ubi) and when

13  As Crusius notes: “Therefore, both of the concepts that lie in the concept of existence and which we now have to explain are that ubi & quando, or the concept of space and time, in the broadest sense” (Crusius 1964, §48; Watkins 2009, 146). See §§54 and 55 of the Sketch for clear statements that time and temporal concepts are abstractions from the existence of complete things.

64  Breaking with Rationalism (quando) can be affirmed outside of thought as well. As soon as we add this in thought in such fashion that we actually posit it, we think of something as existing. Accordingly, existence consists in a [thing that is thought] being somewhere and at some time.  [Crusius 1964, 74; Watkins 2009, 145–146]

Crusius is again making two central points here. First, he presents an argument for his distinction between essence and existence. The necessity of this distinction arises, on his account, from the contrast between what exists merely in thought (as a possibility) and what exists outside of thought. The argument seems to be that what is merely in thought corresponds to what is possible—that is, to the metaphysical essence of a thing—while what exists outside of thought must contain something in addition to what is possible. Now if something must be added to what is possible to make it an actual thing, then whatever it is that is added cannot be part of the essence of the thing, because it would not, in that case, be added to the essence and thus not allow for any contrast between what is merely thought and what exists outside of thought. What is added to the essence of a thing, which Crusius then explicates in a further step with space and time, is actual existence. Thus, while Crusius views existence as a predicate, it cannot be a predicate that pertains to the essence of a thing.14 Second, Crusius finds fault with Wolff ’s account of existence, which invokes the notion of the complement of possibility, arguing that it is vacuous. In his Latin ontology (Philosophia Prima sive Ontologia), Wolff clearly recognizes that existence is distinct from possibility (Wolff 1736). In §171 of that work he argues that just because something is possible does not entail that it exists, and in §172 he adds that the sufficient reason for existence is not contained in possibility, before noting in §173 that something more is required beyond the possibility of a being for that being to exist. But then in §174 he simply defines existence as the complement of possibility and claims that what takes something from the state of possibility to the state of actuality has been explained in its proper place, namely in theology, psych­ ology, and physics. Whatever one makes of his particular explanation in physics (e.g., of how the actuality of the world is brought about), in his ontology Wolff fails to provide a more than merely nominal account of what existence is beyond possibility. By distinguishing between the two as he has and associating space and time with existence, Crusius avoids providing a vacuous account of existence. Not only does Crusius draw a distinction between existence and essence, or possibility, and provide an argument for it, but, crucially, he also claims that existence has priority over possibility.15 In §57, he argues that: 14 Crusius rejects Descartes’s version of the ontological argument (in §235 of Sketch) on the grounds that the premises concern existence understood ideally, whereas the conclusion takes existence really. Crusius holds that God’s existence can be proven only on this basis of what God creates. 15  It is not at all uncommon to distinguish between possibility and existence, or actuality. It happens much less frequently that one explicitly prioritizes existence, or actuality, over possibility.

Eric Watkins  65 the concept of the actual is prior to the concept of the possible both according to nature and according to our cognition. First, I say that it is prior according to nature. For if nothing were actual, then nothing would be possible, because all possibility of a thing that does not yet exist is a causal connection between an existing thing and a thing that does not yet exist. Further, the concept of the actual is also prior to the concept of the possible according to our cognition. For our first concepts are [of] existing things, namely sensations, by which we can attain a concept of the possible only afterwards. In fact, even if one also wanted immediately to meditate a priori most precisely, then the concept of existence is certainly prior to the concept of possibility. For all I need for the concept of existence are the simple concepts of subsistence, coexistence, and succession. By contrast, for the concept of possibility I require the concept of causality, subsistence, and existence.  [Crusius 1964, 98; Watkins 2009, 150–151]

Crusius argues for the priority of actual existence over mere possibility in at least three distinct senses here. First, actuality is prior to possibility in an ontological sense (“according to nature”), because at least certain possibilities must be grounded in prior actualities. Second, it is prior in a broadly epistemological sense (“according to our cognition”), because we start with cognitions of actual things through sensations and then move on to cognition of what becomes possible through them. Finally, actuality is prior to possibility in a semantic sense, because the concept of possibility includes the concept of actual existence (along with the concepts of causality and substance).16 What is striking about this passage for current purposes, however, is Crusius’s position regarding the ontological sense of priority. Though Crusius seems, at times, to be interested in advancing the relatively modest view that a possibility cannot become an actuality unless some other actual thing makes that possible thing actual, there are also indications that he also wants to advance a more ambitious position. In §56 of the Sketch he distinguishes between ideal and real possibility and emphasizes that whereas an ideal possibility depends simply on nothing contradictory being asserted, real possibility depends on the actual existence of something that could cause that thing to exist. That is, Crusius seems to maintain that real possibility requires an actual cause. To be a real possibility, something that stands in a causal relation to this possibility must be actual, because otherwise it would be a merely ideal (or purely logical) possibility lacking any connection to actual existence.

16  Immediately preceding this quotation, Crusius adds the qualification, “although there is less in the concept of the possible than in the concept of the actual,” which complicates the semantic aspect of his claim.

66  Breaking with Rationalism One final feature of Crusius’s account of existence is of note here. Very early in the Sketch, prior to his most explicit descriptions of the concepts of essence and existence, Crusius draws a close connection between existence and sensation: §16. More precise determination of the distinguishing feature of actual things . . . It is ultimately always sensation. This makes clear that and why the distinguishing feature of actuality is ultimately always sensation in our understanding. For sensation is precisely that state of our understanding in which we are forced to think something immediately as existing, without first needing to cognize it through inferences and without it being the case that another state follows it from whose comparison with the former we perceive that we would have merely imagined something, which happens, e.g., in dreams.  [Crusius, 1964, 28; Watkins 2009, 141]

Crusius thus holds that sensation, understood as an involuntary, non-­inferential cognitive state, forms the ultimate justification for knowledge of existence; the justification for any claim about existence must ultimately be provided not through purely conceptual or inferential mediation, but by sensations. In the remainder of the Sketch, Crusius develops a comprehensive and systematic view of God, the world, and the place of human beings in it, one that attacks Wolff ’s position frequently and with considerable sophistication. For current purposes, however, what is important is the basic structure of his metaphysics, which contains a clear distinction between essence and existence, his claim that existence has priority over possibility, and a distinctive role for sensation in cognition.

2  The Pre-­Critical Kant Fast forward now a decade from 1745, when Crusius’s Sketch first appeared in print, to 1755, when Kant published his New Elucidation. The basic structure of this work is clear enough, with three sections clarifying the status and content of several purportedly basic logical-­metaphysical principles. In the first section Kant establishes that the principle of identity, which really consists of two distinct principles (one for affirmative, the other for negative truths), has priority over the principle of contradiction. In the second section he argues for the principle of determining ground, which entails, he argues in the third section, two novel metaphysical principles, the principles of succession and coexistence. What is of interest in the present context is Kant’s detailed treatment of the principle of determining ground in Propositions IV through IX in the second section. It is thus useful to consider the content of each of these Propositions and how the position they establish constitutes a fundamental break with rationalism before

Eric Watkins  67 considering how Kant develops his position further in The Only Possible Argument in Support of a Demonstration of the Existence of God in 1763.

2.1  Propositions IV through VIII of the New Elucidation In Proposition IV, Kant defines a ground as that which “converts things that are indeterminate into things that are determinate” (Ak. 1:392), and distinguishes between antecedently and consequently determining grounds; that is, between grounds of being or becoming (which make intelligible why something is the case) and grounds of knowing (which make intelligible that something is the case). In this context, he explicitly criticizes Wolff ’s official definition of ground because its use of the word “why” renders it circular. He also follows Crusius in replacing the phrase “sufficient ground” with “determining ground,” arguing that the notion “sufficient” is vague, while the notion of “determining” is clear insofar as a determining ground posits one predicate (determination) to the exclusion of its opposite. In Proposition V, Kant then provides a first formulation of the principle of determining ground: “Nothing is true without a determining ground” (Ak. 1:393). He presents an argument for it in two different versions, and, in a scholium, clarifies that the principle utilizes the notion of an antecedently determining ground and not simply that of a consequentially determining ground. It thus appears to be a strong principle insofar as it applies to the grounds of all truths and not just those that allow us to know certain truths. Kant concludes his discussion with the transitional sentence, “But let us proceed to the grounds which determine existence” (Ak. 1:394), which implies some distinction between grounds of truth and grounds of existence. In Proposition VI Kant makes a negative claim about grounds of existence, asserting: “To say that something has the ground of its existence within itself is absurd” (Ak. 1:394). That is, nothing can be the ground of its own existence. His reasoning is via a reductio. Since what grounds itself would also have to cause itself and “the concept of a cause is by nature prior to the concept of that which is caused” (Ak. 1:394), assuming something that has the ground of its existence within itself would have the implication that something could be prior to itself, which is absurd, regardless of whether the priority is taken in a temporal or a causal sense.17 In a corollary, Kant explicitly claims that according to this prop­os­ ition, anything that exists necessarily has no antecedently determining ground at all, either within or without, though at this point he has not yet argued that any 17 It is unclear whether Kant is taking causality here to be efficient causality alone or more inclusively.

68  Breaking with Rationalism such being exists. In the scholium Kant then uses Proposition VI to criticize those, such as Descartes, Wolff, and Baumgarten, who assert that God has the ground of his existence within himself. At the same time, Kant expresses some sympathy with the motivation that leads to the view about God criticized in the scholium. For if one were to accept that everything must have a ground (as Leibniz and Wolff do), then God’s existence would obviously have to have a ground. This ground would then necessarily be either internal or external to God. Since it is absurd to state that something external to God causes God’s existence, this ground would have to lie within God; that is, God would have to contain the ground of his existence within himself. Kant seems to think further that this line of thought can motivate one to accept (or at least be sympathetic to) the ontological argument. For if one pursues a chain of grounds back to its beginning, in this case one will reach the concept of God as the concept of a being that includes the ground of its own existence, which is precisely the kind of conceptual containment that the ontological argument attempts to exploit. Kant’s analysis of what goes wrong in the ontological argument is noteworthy. It can, however, easily be seen that this happens ideally, not really. Form for yourself the concept of some being or other in which there is a totality of reality. It must be conceded that, given this concept, existence also has to be attributed to this being. And, accordingly, the argument proceeds as follows: if all realities, without distinction of degree, are united together in a certain being, then that being exists. But if all those realities are only conceived as united together, then the existence of that being is also only an existence in ideas.  [Ak. 1:394]

Notice that this analysis, along with the objection based on it, does not rely on the idea expressed earlier that it is absurd to hold that any being could contain the ground of its existence within itself. Indeed, the concept of a being containing within itself all realities, including existence, seems to be granted, even if only for the sake of argument in this context. Instead, what Kant objects to is the idea that one could establish the existence of God outside of thought by means of the ontological argument, because it establishes only that God must exist merely in thought, as Crusius might put it. Thus, a trace of one element, but only of one element, of Kant’s slightly later more detailed rejection of the ontological argument seems already to be present here. Yet Kant’s primary interest here does not lie in detailing the inadequacies of the ontological argument, but rather in developing a principle that is closely related to his rejection of the idea that something could have the ground of its existence within itself. For in Proposition VII Kant states: “There is a being the existence of which is prior to the very possibility both of itself and of all things” (Ak. 1:395). That is, not only is it impossible that something could have the ground of its

Eric Watkins  69 existence within itself, but it is also the case that something must exist that has no ground at all, namely God. For such a being is required so as to be the ground of all possibility given that possibility requires the formal consistency of predicates that must be given by something already existing. Kant will articulate this line of reasoning in greater detail a few years later, in The Only Possible Argument, but what is most important for current purposes is Kant’s idea that possibilities require grounding along with his acceptance of the necessity of a being that is both ungrounded and prior to even its own possibility. That is, Kant’s most basic claim here is that in the case of God, an existence that is ungrounded has priority over all possibility.18 This constitutes one fundamental break with his rationalist predecessors because he is asserting the existence of something that has no ground (or reason) at all. In Proposition VIII, Kant returns to grounds of existence, asserting: “Nothing which exists contingently can be without a ground that determines its existence antecedently” (Ak. 1:396). That is, all contingently existing things must have antecedently determining grounds, even if God, as a necessary being, cannot. Though the details of Kant’s argument for this proposition are less than clear, the proposition itself rounds out his positive case for the principle of determining ground. Indeed, in the scholium to this proposition, Kant remarks: “Such is the demonstration of the principle of determining ground, which has now been fully illuminated by all the light of certainty” (Ak. 1:396). In addition to showing that Kant thinks that the principle can be proved, it makes explicit that he draws a distinction between grounds of truth and grounds of existence, maintains that the principle of determining ground has universal scope with respect to truth (just as Proposition V maintained), and holds that the principle of determining ground applies to all contingently existing things, but not to necessarily existing things, such as God (just as Propositions V through VIII had asserted).

2.2  Kant versus Wolff and Crusius Although Kant has completed his proof of the principle of determining ground at this point, he goes on to discuss how his treatment of the principle compares to the positions of Wolff and Crusius as well as the most significant objections that could be raised against it. In addition to his earlier rejection of Wolff ’s faulty definition of the notion of ground, Kant explicitly rejects Wolff ’s argument for the 18  In one instance Kant equivocates slightly on this point when he notes: “Of all beings, God is the only one in which existence is prior to, or, if you prefer, identical with, possibility” (Ak. 1:396). Unfortunately, Kant does not directly address how it is possible for an existence to precede its own essence or possibility. He does clearly think that if an existence has a ground, then its existence does not precede its essence (since presumably the ground would contain the possibility of the thing and hence its essence). But how it is possible that an existence could arise without there being a prior essence is not clarified.

70  Breaking with Rationalism principle of determining ground at the end of his discussion of Proposition VIII. Even without going into his rather murky analysis (Ak. 1:397–398), it is clear that he cannot accept Wolff ’s proof, since that proof, if it were sound, would establish the principle of determining ground for every existing thing, and not just for contingently existing things (if, that is, it is taken to apply to existence and not just to truth, which Wolff does not always distinguish clearly). The most central and important difference, however, concerns whether something could exist that has no ground. Wolff, as a rationalist, has to deny this possibility— everything, including God’s existence, must have a sufficient reason—whereas Kant breaks with the rationalists by arguing for one crucial restriction to the principle of determining ground, namely that even if everything else requires a determining ground, the existence of the necessary being that grounds all possibility precludes one. Kant’s most extended discussion of Crusius in the New Elucidation occurs in Proposition IX, where he considers the difficulties that “seem to beset the principle of determining ground” (Ak. 1:398). Kant’s rhetoric towards Crusius throughout this work is remarkably positive. At times, he is effusive with praise. Three times Kant refers to the “celebrated Crusius” and singles him out for “special honor” from “among the most penetrating philosophers of our age” (Ak. 1:396). In the very first sentence of Proposition IX, Kant notes that “among those who attack this principle [of determining ground], the most distinguished and penetrating Crusius is to be regarded, and rightly, as leading the assault. He alone of all those involved is able to bear the brunt of battle. I maintain that Crusius scarcely has an equal . . . in Germany” (Ak. 1:398). While there is certainly some truth in this kind of assessment—Crusius is one of the most incisive anti-­Wolffian German philosophers of his day—Kant is clearly also establishing a framework for judging his own accomplishments in this work. For if he can repel Crusius’s attacks on the principle of determining ground, then he can view himself “as having overcome every difficulty” (Ak. 1:398). According to Kant, Crusius’s attacks on the principle are twofold. First, Crusius complains that the principle, as formulated by Wolff, is ambiguous and vague, since the principle does not explicitly distinguish between ideal grounds, or grounds of knowing, and real grounds. Since Kant distinguishes between consequently and antecedently determining grounds and holds different versions of the principle for the different kinds of grounds, he is entitled to say: “Anyone who examines our various claims will find that I carefully distinguish the ground of truth from the ground of actuality” (Ak. 1:398). Crusius’s first objection thus does not apply to Kant’s position. Crusius’s second objection, by contrast, represents “a greater danger” (Ak. 1:398). It claims that Wolff ’s adherence to the principle of determining ground entails that he cannot avoid the immutable necessity of fatalism, which “impairs all freedom and morality” (Ak. 1:399). To this end, Crusius attacks the distinction

Eric Watkins  71 between absolute and hypothetical necessity, which Leibniz, Wolff, and their followers often used to try to blunt the force of this kind of objection. Kant goes to great lengths to respond to this objection, even constructing a fictitious dialogue between Caius and Titius to illustrate one of the distinctions that was, in his view, central to the dispute. He also considers a wide range of topics, including the nature of spontaneous action and, in the Supplements to Proposition IX, the possibility of divine foreknowledge and the indifference of equilibrium. In short, Kant defends a compatibilist account of freedom that is similar in fundamental respects to Leibniz’s and Wolff ’s. Kant concludes his discussion of the principle of determining ground in this section by showing, against Leibniz and Baumgarten, that certain other principles (such as the principle of the identity of indiscernibles) do not follow from the principle of determining ground. In this way, Kant has reason to think that he has defended (his version of) the principle of determining ground against Crusius’s objections.

2.3  Existence in The Only Possible Argument In The Only Possible Argument, Kant takes up for further discussion several issues that he had begun to articulate in the New Elucidation. In the course of preparing the setup for the novel theistic proof he develops in this work, Kant provides a more detailed analysis of existence. He begins by noting, negatively, that existence is not a predicate or a determination of a thing. To put the point in Crusian terms, existence is not part of the essence of a thing, for you can “draw up a list of all the predicates which might be thought to belong to Julius Caesar . . . You will quickly see that he can either exist with all these determinations or not exist at all” (Ak. 2:72).19 One of the arguments Kant employs to establish this negative claim proceeds on the basis of considerations in philosophical theology and metaphysics that Leibniz in particular drew attention to. When, according to Leibniz, God represents an individual as possible, God’s representation of that individual must be complete, since God, in virtue of his omniscience, cognizes every individual essence in its entirety as distinct from every other individual essence. However, Kant argues, if God’s representation of that possible individual is complete, then, if God decides to make it actual, then the predicate of existence cannot be added to that individual, because that representation was already complete. Therefore, existence cannot be a predicate like those that God thinks of as constituting the essence of any possible thing. Kant rightly notes, however, that one can still use “existence” as a predicate, such as in the proposition “God exists.” However, in that case, it is, Kant suggests, 19  It is relevant to note that Kant also uses this point (at Ak. 2:76–77) against Crusius’s claim that a possible individual would lack spatiotemporal location.

72  Breaking with Rationalism not a predicate “of the thing itself ” but rather “of the thought which one has of the thing” (ibid.). This idea leads Kant to claim positively that “existence is the absolute positing of a thing” (Ak. 2:73). Though Kant asserts that the notion of positing “is so simple that it is not possible to say anything further by way of elaboration” (ibid.), he does draw an instructive contrast between absolute and relative positing. Relative positing occurs when one posits a predicate concept of a subject concept. It is, Kant seems to think, a purely logical relation between the predicate concept and the subject concept whose truth conditions are internal to the judgment and that does not require that the subject exist (as e.g. in the judgment “Any square circle is round”). Thus, even an atheist can agree with the claim that God is omnipotent, since that claim asserts, Kant thinks, only that omnipotence is posited with respect to God, but not that God exists.20 Absolute positing, by contrast, “posits” that an actual object outside of thought corresponds to the subject concept. It is thus extra-­logical and involves what we would call the existential use of “is.” That is, unlike relative positing, which asserts a relation between two concepts in a judgment, absolute positing requires a different kind of relation, one that obtains between the judgment, its subject concept in particular, and something external to the judgment, namely the thing to which it refers; its truth conditions thus lie outside of the concepts that are involved in the judgment. Thus, despite grammatical appearances, “God is an existent thing” (when understood as a restatement of the claim that God exists) is not a case of relative positing (with “God” and “existence” being predicates posited relative to each other), but rather one that involves absolute positing, because the concept “God” is posited absolutely; its truth conditions require something in addition to any identity or containment relation that might obtain between concepts within a judgment, since they require that God actually exist. Now, given that the distinction between relative and absolute positing has been drawn, at least in part, in terms of whether a purely logical relation obtains between concepts within judgment or whether something extra-­ logical (or outside of judgment) is required, one might be tempted to associate this distinction with the distinction Kant draws, starting in the early 1760s, between logical and real grounds. Kant describes the law of contradiction as the logical ground of possibility, since it establishes the formal consistency of the concepts one is thinking of as possible, whereas something that exists must serve as the real ground of possibility. However, in the Negative Magnitudes, also published in 1763, Kant states very clearly that a logical ground pertains to what can be understood by appeal to the analysis of concepts according to the law of identity, whereas a real ground involves a connection that “does not take place in virtue of

20  Kant uses an interesting example to illustrate his point: “The God of Spinoza is subject to continuous change” (Ak. 2:74).

Eric Watkins  73 the law of contradiction” (Ak. 2:202).21 At this point, Kant has not identified what the principle of real grounds is, but the example that he has in mind is telling. For he notes that in cases where “because something is, something else is,” we have a grounding relation, but, as Hume has shown, it is not based on the principle of contradiction or identity. Thus, it is an instance of a real, not a logical ground. Later, in the Critical period, the relational categories will represent different kinds of real grounds, but since categories are pure concepts of the understanding, no appeal must be made beyond the concepts used in a judgment. As a result, Kant’s distinction between logical and real grounds falls within our understanding and cannot coincide with the distinction between relative and absolute positing, which goes beyond the understanding and its grasp of the relations between concepts in a judgment.

3  Crusius’s Influence on Kant’s Break with Rationalism Since we now have an account of the relevant philosophical views of both Crusius and the early pre-­Critical Kant, we are in a position to pose the question as to what might have led Kant to break with rationalism as he did.22 Though the situ­ ation is complicated, it is nonetheless plausible to think that Crusius was a de­cisive influence on the pre-­Critical Kant on precisely this point. Evidence for this derives from (i) the issue that is at the heart of their break with the rationalists, (ii) the arguments they offered in support of that break, and (iii) the fundamental motivations they had for doing so. Let us begin by considering the issue on which Crusius and Kant break with rationalism. Both break with Leibniz and Wolff by restricting the scope of the principle of determining ground. This contrasts with those, such as Hume, who break with rationalism either by denying that reason is a faculty that carries any normative weight or by disputing that reason is capable of forming the appropriate kind of representations.23 Further, Crusius and Kant both articulate this break by invoking a distinction between existence and possibility (or essence). It is true 21  Kant’s distinction between real and logical grounds thus represents an additional break with rationalism, and provides further evidence that one must speak, as suggested at the outset, of several smaller breaks rather than one all-­encompassing break. 22  I do not mean to suggest that Kant does not disagree with Leibniz, Wolff, and their followers about any other fundamental doctrines in the New Elucidation. Kant explicitly rejects a number of metaphysical principles that were associated at the time with the broadly Leibnizian position, including, importantly, the principle of the identity of indiscernibles (Ak. 1:407–410). I am inclined to think, however, that Kant’s rejection of these principles is either derivative on more fundamental points or at least no more fundamental than the break considered here. 23  In certain passages (according to which reason is and ought to be slave to the passions) Hume asserts that reason carries no normative force. In other passages (according to which there is no idea in the mind that is not a copy of a previous impression) Hume argues that there are limits to the relations of ideas that reason can form.

74  Breaking with Rationalism that, in The Only Possible Argument, Kant develops this idea further than Crusius does by arguing that existence is not a (real) predicate or determination of a thing (or of its essence) and that existence is an absolute positing of a thing (outside thought), rather than a relative positing of a relation between predicates (within thought). But these further developments emerge naturally from the issues raised by Crusius, since Crusius associates existence with sensation, where sensation is understood as something that does not require inferential mediation through judgments.24 Finally, and most importantly, they both use their account of this distinction to assert and make the case that existence is prior to possibility, which is a crucial element of their break with the rationalists insofar as both hold that the principle of determining ground cannot apply to those cases in which existence precedes possibility. Their priority claim is thus essential to their ability to state the nature and scope of their break with their rationalist predecessors. The argument that Kant takes to entail his break with the rationalists also seems to have been influenced by Crusius or at least displays remarkable structural similarities with Crusius’s argument. Crusius’s argument is based on the idea that real possibilities depend on actual things that could cause them to be actual and thus that give reason to view the possibilities of these things as being real rather than merely ideal possibilities. Kant’s line of reasoning is that possibility requires the consistency of predicates that must be given antecedently and can therefore be given only by a necessarily existing thing. This is not to deny that there are important differences between their positions, but the idea underlying both is that a possibility cannot be assumed, but rather must be explained, and the explanation of possibility must invoke some prior existing thing that actually exists (and is not merely possible). Finally, one can also discern a fundamental similarity in Crusius’s and Kant’s most basic motivations for restricting the scope of the principle of determining ground. On the face of it, their motivations appear to be different. Crusius is interested in restricting the principle because he thinks that only such a restriction can save the kind of libertarian freedom he attributes to both God and human beings. As we have seen, at this point in his career Kant rejects Crusius’s notion of libertarian freedom in favor of a compatibilist notion (both for God and for human beings).25 Kant’s motivation for limiting the principle of determining 24  This is not to deny that there are significant differences between Crusius’s and the early Kant’s positions, for Crusius does not seem to have Kant’s notion of absolute positing. What they share is an extra-­logical representation of existence (in the form of sensation and absolute positing). 25  It is significant that Crusius’s influence on the pre-­Critical Kant’s break with rationalism predates and thus must be distinct from Kant’s adoption of a libertarian conception of freedom. (Indeed, the case for Crusius being an important influence on the pre-­Critical Kant would have been much neater and tidier if Kant had adopted a libertarian conception of freedom in 1755. As it is, the historical real­ ities here are messier.) This observation raises the important question of why Kant changed his mind on such an important point if not on the basis of his first encounter with Crusius’s position some time prior to 1755.

Eric Watkins  75 ground lies in straightforward metaphysical considerations. Specifically, he thinks that whatever grounds, and is thus prior to, all possibility must exist and be ungrounded because the very idea of a self-­grounding being is incoherent. Despite these differences, however, Crusius and Kant are in fact motivated by a common objection to the rationalists and respond in similar ways to the problem that generates the objection. Specifically, both Crusius and Kant are dissatisfied with the rationalists’ strategy of providing an explanation in terms of a necessity that somehow grounds itself. Crusius thinks that such a strategy is incompatible with the contingency presupposed by libertarian freedom, while Kant thinks that the very idea of a self-­grounding necessity is incoherent, but both are focused on the modal consequences of a necessity that is supposed to ground itself; they are both dissatisfied with what they take to be the modal implications of the principle that is at the heart of the rationalists’ position. Further, they agree fundamentally about the best way to avoid the difficulty that the rationalists face. Both think that one ought to explain possibilities in terms of “brute” actualities. That’s the basic underlying point of agreement, but it is also what leads them to make their fundamental break with rationalism. If we step back from this important point of influence in the second section of the New Elucidation, we can see that it fits with Kant’s overall goal in this work: namely, to chart a course between the positions of Leibniz and Wolff, on the one hand, and Crusius, on the other. In the first section, Kant criticizes Leibniz and Wolff for not being clear about the nature of the principle of identity and the priority that obtains between that principle and the principle of contradiction. In the second section, he objects to Leibniz and Wolff further (drawing, as we have seen, on Crusius in the process) by restricting the scope of the principle of determining ground and by noting that several other principles (such as the principle of the identity of indiscernibles) do not follow from it, but he also takes issue with Crusius by defending the applicability of the principle of determining ground to the extent that he does and by rejecting Crusius’s incompatibilist conception of freedom. In the third section, Kant asserts that the principle of determining ground entails the principle of succession, which flatly contradicts Leibniz’s and Wolff ’s doctrine of pre-­established harmony, since it claims that finite substances that are causally isolated from each other cannot change. The principle of determining ground also entails, he argues, the Principle of Coexistence, which entails a rejection of Crusius’s notion of existential grounds, because it claims that finite substances do not stand in relation to each other in virtue of their existence alone.26

26  For a detailed discussion of Kant’s principles of succession and coexistence as well as of how, taken together, they chart a middle course between Leibniz and Crusius on the specific issue of causality, see (Watkins 2005, ch. 2).

76  Breaking with Rationalism In the course of trying to forge this kind of ambitious middle path, Kant is looking, in his pre-­Critical period, for the weaknesses of both of the schools of thought that dominated Germany at the time; and, while he is quite naturally influenced by points that both sides make against each other, one can see that he is already thinking in creative ways about the kind of position he ultimately wants to end up with. As a result, though Crusius influences Kant on several important points, his break with rationalism included, Kant does not simply adopt Crusius’s positive account as such, but rather attempts to rethink from his own perspective the specific points that Crusius raises and then embed them within a broader and deeper framework of his own that can, he hopes, capture the most fundamental motivations of both sides.

4  Kant’s Break with Rationalism in the Critical Period If we now return to the Critical Kant with this account of the early pre-­Critical Kant’s break with rationalism in mind, what insights might Crusius’s influence deliver? Specifically, what small steps could this appreciation of Crusius’s importance for the early Kant allow us to take for understanding the Critique of Pure Reason? Unfortunately, but also unsurprisingly, the interpretation provided above does not offer any immediate easy answers. The pre-­Critical Kant’s reasons for breaking with rationalism are not necessarily the same as the Critical Kant’s, especially if one keeps in mind the massive changes that take place when Kant undergoes his “Critical turn.” As a result, one cannot provide an adequate account of the Critical Kant’s break with rationalism without substantial further analysis and interpretive work that would extend beyond what is possible in the present context. However, I do want to point out, very briefly, one particularly intriguing possibility that is suggested by the pre-­Critical Kant’s reflections on the issue. As we saw above, several proposals as to what the Critical Kant’s reason might be for breaking with rationalism faced immediate objections from anyone who takes a Leibnizian position seriously. As we have seen, one distinctive feature of the pre-­ Critical Kant’s case for breaking with rationalism was his analysis of existence (in the course of distinguishing it from essence and asserting its priority). Existence is not, for Kant, a real predicate like those that are used to distinguish things from each other. And in fact, Leibniz’s own texts reveal some vacillation about precisely what position a rationalist ought to take on existence. On the one hand, in his discussion of the ontological argument Leibniz seems to treat existence as a perfection (i.e. a simple quality that is positive and absolute), which entails that it is what Kant would refer to as a real predicate that pertains to the essence of the perfect being. On the other hand, in other passages Leibniz seems to recognize that existence is not a quality, at least not in the same sense in which other

Eric Watkins  77 qualities are, since existence is predicated of qualities (rather than being itself a quality).27 Viewed in this context, the pre-­Critical Kant’s reflections on existence can be seen as picking up on a genuinely problematic or unresolved issue within Leibniz’s philosophy rather than begging the question against such a position. But how does this point bear on Kant’s Critical philosophy? Fundamental to Kant’s Critical turn is his distinction between sensibility and the understanding. This distinction is required for a number of Kant’s most basic doctrines. It is needed to define the distinction between appearances and things in themselves, because appearances are characterized as objects that are necessarily given to us through sensibility while things in themselves (or noumena, literally “things thought”) are understood in terms of the understanding (or reason). One must also appeal to it in explaining the analytic–synthetic distinction (and thus how synthetic a priori cognition is possible), since analytic judgments are explicable only in terms of purely logical principles that invoke the understanding or reason. Finally, it is required to describe the distinction between real and merely logical possibility, since logical possibility, like analytic truths, can be explicated only by means of the understanding alone. In short, Kant’s distinction between sensibility and understanding is presupposed by three of the most tempting accounts of the Critical Kant’s break with rationalism mentioned above. Now Leibniz wants to draw a distinction between sensibility and understanding too, but he does so in terms of degrees of clarity and distinctness. So, what is distinctive about Kant’s distinction between sensibility and understanding is that he conceives of the distinction as being one of kind rather than degree, a point that he makes clearly in the Amphiboly, though without much explicit argument. As a result, what Kant needs to be able to support a significant number of his Critical doctrines is a justification of this particular conception of the distinction between our faculties (along with a reason for assigning space and time to the faculty of sensibility).28 It is on precisely this point that the pre-­Critical Kant’s reflections on existence could prove to be relevant. The basic idea is that the nature of the distinction that 27  See the informative discussion of Leibniz’s position on existence in (Adams 1994, chs. 5 and 6). Specifically, Adam’s draws attention to the following undated memorandum by Leibniz: “Existence. It can be doubted very much whether existence is a perfection or degree of reality; for it can be doubted whether existence is one of those things that can be conceived—that is, one of the parts of essence; or whether it is only a certain imaginary concept, such as that of heat and cold, which is a denomination only of our perception, not of the nature of things. Yet if we consider more accurately, [we shall see] that we conceive something more when we think that a thing A exists, than when we think that it is possible. Therefore it seems to be true that existence is a certain degree of reality; or certainly that it is some relation to degrees of reality. Existence is not a degree of reality, however; for of every degree of reality it is possible to understand the existence as well as the possibility. Existence will therefore be the superiority of the degrees of reality of one thing over the degrees of reality of an opposed thing. That is, that which is more perfect than all things mutually incompatibles exists, and conversely what exists is more perfect than the non-­existent, but it is not true that existence itself is a perfection, since it is only a certain comparative relation [comparatio] of perfections among themselves” (Adams 1994, 165). 28  For fuller discussion see Watkins (2017).

78  Breaking with Rationalism the pre-­Critical Kant draws between essence and existence makes it plausible to view the faculties that represent essence and existence as distinct in kind.29 For if one thinks, as Crusius and the pre-­Critical Kant do, that existence is distinct in kind from essence, that is, that existence is something fundamentally distinct from the (real) predicates that make up an essence and would fall in the domain of judgment, then it is plausible to think that the faculties that one posits to represent essence and existence will be similarly distinct in kind. Specifically, as we saw earlier, Crusius argues that the predicates that constitute the essence of a thing are distinct from the concept of existence, which depends on space, time, and sensations, and the pre-­Critical Kant develops the thought further such that existence is to be understood in terms of absolute positing and thus in terms of something extra-­ logical, distinct from logical relations within judgment (or thought). As a result, there is, for both thinkers, clearly a difference in kind between existence and essence. The natural next step is to posit faculties that are distinct in kind to represent these different kinds of predicates.30 This is precisely what we find first in the Inaugural Dissertation, and then in more developed fashion in the Critique of Pure Reason. In fact, one of Kant’s most distinctive moves, that of assigning our a priori representations of space and time to sensibility, is clearly foreshadowed in certain respects by Crusius’s separation of space and time from the essence of things and their attribution to existence, which he connects with sensation. Crusius’s importance for Kant’s break with rationalism, first in the pre-­Critical and then in the Critical period, could thus prove to be quite extensive.31 Such a proposal would obviously need to be worked out further. Specifically, one would need to provide a much more detailed description of the position, a more precise formulation and defense of the argument for it, and responses to whatever objections might be raised against it. However, if that could be done, one would in fact have a fuller and more satisfying account of the Critical Kant’s break with Leibniz, Wolff, and their rationalist followers, one that was both firmly grounded in the historical context and the relevant texts and of considerable philosophical interest in its own right, without begging the question against the rationalists.

29 Paul Guyer (Guyer  2000) similarly suggests that Kant presents an argument in the New Elucidation for a dualistic position that is inconsistent with the positions of Leibniz and Wolff. 30  Significantly, Crusius does not take this step. He attributes sensations to the understanding. He may have thought this attribution appropriate due to his view that the understanding is capable of operating according to principles that are not purely logical (as Leibniz and Wolff would have understood the term), such as the principles of the uncombinable and the inseparable. See Crusius (1964, §15). 31  It is true that Kant will eventually want to stress the a priority of space and time, but it is clear that he first develops the idea that space and time require a distinct faculty, which then puts him in a position to note that this faculty, along with the basic representations it is responsible for, must be a priori.

4

Leibniz on the Ideality of Space Donald Rutherford, University of California, San Diego*

This division of the objects of our thoughts into substances, modes and relations is pretty much to my liking. I believe that qualities are just modifications of substances, and that the understanding adds relations. More follows from this than people think. New Essays, RB 1451 Leibniz’s theory of space has been closely linked to two philosophical doctrines. One is the doctrine of relationalism, according to which space is nothing more than the totality of actual or possible relations among bodies. The other is the doctrine of the ideality, or mind-­dependence, of space. Both of these doctrines are advanced in opposition to the view, defended by Newton and his follower Samuel Clarke, that space is something real and absolute—a container, as it were—within which bodies have absolute positions by virtue of being located in distinct regions of space.2 Leibniz is firm in rejecting this conception of space. Reiterating his position in his Fifth Letter to Clarke, he writes that in his view, “there need not be any real and absolute being answering to [the idea of space], distinct from the mind, and from all relations . . . Space is therefore something ideal” (LC V 104). The arguments of the Clarke correspondence reveal Leibniz’s reasons for rejecting the view that space is a “real and absolute being.”3 A central argument, *  I am grateful to the participants in the UCSD History of Philosophy Roundtable for discussion of an earlier draft of this chapter. Special thanks are owed to Michael Futch, Clinton Tolley, and two anonymous reviewers for subsequent probing of its claims. 1  Leibniz’s writings are cited according to the abbreviations given at the front of this volume. In some cases, I have taken the liberty of making small changes in the cited translation. Where no translation is cited, it is my own. 2  Leibniz’s exchange with Clarke has given rise to a large literature devoted both to the arguments of the correspondence (Broad 1981; Khamara 1993; Arthur 1994; Vailati 1997; Futch 2008) and to the analysis of Leibniz’s position as a species of spacetime theory (Earman 1989; Belot 2011). Neither of these topics is the focus of this chapter, which explores the metaphysical foundations of Leibniz’s theory of space based on a wide range of texts, including—but not limited to—the letters he exchanged with Clarke. 3  Pinning down precisely what this phrase means for the two sides of the controversy is not easy. See Earman (1989, ch. 1) on the multiple senses of “absoluteness” in play and the corresponding uncertainty about the content of relationalism. Newton’s definition in the Principia’s scholium on space and time does not clearly distinguish his position from Leibniz’s: “Absolute space, of its own

Donald Rutherford, Leibniz on the Ideality of Space. In: Leibniz and Kant. Edited by Brandon C. Look, Oxford University Press. © Oxford University Press 2021. DOI: 10.1093/oso/9780199606368.003.0004

80  Leibniz on the Ideality of Space premised on the principle of sufficient reason, challenges the consistency of the propositions (i) that space is “something absolutely uniform” and (ii) that there are distinguishable parts of space, in terms of which the positions of bodies can be designated.4 The uniformity of space, which Leibniz takes to be an assumption he holds in common with Newtonians, implies for him that bodies lack an absolute position and hence that some form of relationalism must be true. However, the exact form of this relationalism and the support it offers for the doctrine that space is ideal remain obscure. Some authors have questioned whether there is an inferential relation between the two doctrines. Although there is broad agreement that Leibniz accepts some version of relationalism, this view has been seen as separable from his idealism about space, which is motivated by different con­sid­er­ ations, such as his investigation of the “labyrinth of the continuum.”5 In what follows I argue for a close connection between Leibniz’s doctrine of the relational character of space and the doctrine that space is ideal, and I elaborate some of the consequences of the latter doctrine for his philosophy. The approach I take is anchored in basic principles of Leibniz’s metaphysics that are fixed long before his exchange with Clarke, as well as conclusions he reaches about space through work on the foundations of geometry. Throughout I assume that in his writings on metaphysics, physics, and mathematics, Leibniz operates with a single concept of space and a single account of the mode of cognition by which objects are apprehended as spatially related. However, I also claim that a metaphysical analysis of the concept of space—and not physical arguments based on considerations of motion or force—is primary in determining Leibniz’s theory of space.6 In §1 of this chapter, I locate the basis for the ideality doctrine in the general account Leibniz defends of the ontology of relations. In §§2 and 3, I extend this

nature, without reference to anything external, always remains homogeneous and immovable. Relative space is any movable measure or dimension of this absolute space; such a measure or dimension is determined by our senses from the situation of the space with respect to bodies and is popularly taken for immovable space” (Newton 1999, 408–409). This description of absolute space could be accepted by Leibniz; he certainly would not recognize his doctrine in the view that space is relative in Newton’s sense. 4  See LC III 5; and LC IV 18: “Space being uniform, there can be neither any external nor internal reason, by which to distinguish parts, and to make any choice between them.” 5  See Earman (1989, 15–16), citing Hartz and Cover  1988. Within the literature there has been disagreement about when, and whether, Leibniz adopts the view that space is ideal, as opposed to a “well-­ founded phenomenon.” Hartz and Cover locate the shift to this position in the 1690s. Crockett 2008 finds little support for this reading and argues that even in his early work Leibniz is plausibly read as thinking of space and time as ideal orders of existence. On Crockett’s account this is consistent with space and time also counting as “phenomena” in the broad sense Leibniz gives to that term, but not as “appearances.” For a passage contradicting this reading, see n. 32 below. 6  For a wide-­ranging account of Leibniz’s theory of space that differs from the approach pursued here, see Arthur 2013. Leibniz’s central arguments for the ideality of space apply analogously in the case of time (see LC V 49). Leibniz characterizes space and time as coordinate dimensions of existence (the orders of “coexistence” and “succession,” respectively); however, the metaphysics of time raises separate issues that I cannot take up here. For a comprehensive treatment, see Futch 2008.

Donald Rutherford  81 reading to include as a special case the ideality of space as an “order of coexistence,” and I distinguish as species of such an order discrete and continuous spaces. §4 examines the relationship between space and situs, the latter of which grounds our understanding of where something is relative to anything else. In §5, I sketch some of the consequences of my reading for Leibniz’s account of empirical knowledge and his rejection of indiscernible material objects. Here I also assess, briefly, the extent to which Leibniz anticipates Kant’s doctrine of the ideality of space and whether his position is fairly targeted by Kant in the first Critique’s “Amphiboly” section.

1  Relations as Ideal Leibniz’s basic ontological framework is expressed in the epigraph to this chapter: “This division of the objects of our thoughts into substances, modes and relations is pretty much to my liking. I believe that qualities are just modifications of substances, and that the understanding adds relations” (RB 145). Leibniz’s statement reflects his efforts to refine a common threefold division of the category of being into substances, qualities, and relations. His core conception of substance is that of an ultimate subject of predication: that of which other things are predicated which itself is not predicated of anything else.7 Leibniz offers reasons for limiting the class of substances to things that possess other essential properties (e.g. per se unity and activity); however, he begins from a conception of substance as a concrete thing of which other things are predicated.8 The things predicated of a substance are, broadly speaking, its properties. Here too, though, Leibniz imposes further restrictions. Most importantly, he insists that the qualities of a thing—properties predicated of it without relation to other things (e.g. the whiteness of Socrates)—are merely modifications of substance. By this he means that they are not things that have an existence independent of the substance they qualify. Qualities are not real abstract entities (i.e. universals) that are instantiated in individuals. Nor are qualities real accidents that can be separated from a substance, or that can migrate from one substance to another. In reality, qualities are only modes or states of a substance, or ways of being some particular concrete thing.9

7  “Furthermore, in every positive term, we conceive some subject, or thing [Rem], and an attribute or mode of the thing. An ultimate subject is called a substance. The remaining things are called accidents” (A VI 4, 388). 8  For a development of this point, see Rutherford (1995, chs. 4 and 5). Here I set aside the question of whether the category of substance includes only soul-­like entities (“monads”), or also ensouled bodies. On this, see the Introduction to LDB and Garber 2009. 9  See Leibniz’s letter to Des Bosses of September 20, 1712 (LDB 270–271); De realitate accidentium (ca.1688), A VI 4, 995–996.

82  Leibniz on the Ideality of Space In addition to any qualities a thing has by virtue of what it is, there are properties it has by virtue of its relations to other things. Concerning these relational properties, Leibniz defends two main theses. The first is a claim about the ideality of relations: [IR] Relations in themselves, i.e. relations taken independently of their relata, are not real or existing things. They are merely “beings of reason,” or “ideal” things.

Leibniz offers several arguments on behalf of [IR]. The most familiar, which I will call the “accident argument,” objects to the notion of a relation as an accident predicated of two or more relata. Whenever two individuals are related, there are accidents belonging to each of them in virtue of which they are related, but there is not a third accident that is the relation itself: It cannot be said that both of them are the subject of such an accident; for, if so, we should have an accident in two subjects, with one leg in one and the other in the other, which is contrary to the notion of accidents. Therefore we must say that this relation, in this third way of considering it, is indeed out of the subjects; but being neither a substance, nor an accident, it must be a mere ideal thing, the consideration of which is nevertheless useful. [LC V 47/GP VII 401; cf. LDB 326–327]

Leibniz’s argument draws its force from two assumptions: (i) an accident can be predicated of only a single subject (something that becomes more plausible if all accidents are modifications of their subjects); and (ii) the category of actual or existing things is limited to substances and their modifications. A relation being neither of these, it has no actual existence, but is merely an ideal thing.10 This line of thought might seem to beg the question against a defender of the reality of relations. If we are considering the relation as such, we are interested in more than just this or that instance of relatedness, as any putative accident would be. We have in mind instead a universal (e.g. . . . is next to . . . ; . . . is similar to . . . ) that can be instantiated in the case of any suitable actual or possible relata. With much of this objection Leibniz would agree. When considering the relation itself, we are considering something universal that can be applied in the case of any possible relata. Indeed, given this, we can locate the basis of a second argument (call it the “indeterminacy argument”) on which Leibniz relies in defending [IR]. 10  In at least one text, Leibniz offers a different argument with roots in Plato’s Parmenides. Suppose that relation R is a real accident by virtue of which it is true that Rab. If the relatedness of a and b requires a ground in some third existing thing, then so must the relatedness of R and a, and the re­lated­ness of R and b. Clearly, an infinite regress will result, hence the premise that relations are real accidents must be rejected. The argument appears in a manuscript from the Paris period (A VI 3, 399). For discussion of it, see Mugnai 2010 and 2012, 186–188.

Donald Rutherford  83 All actual or created things, he holds, are concrete and fully determinate individuals. Relations as such are neither: they are abstract and await completion by their relata. Thus, they are not actual things, but exist only as ideas: ways of representing how actual or possible things can be with respect to each other.11 As the contents of ideas by which God represents the unity of creation (and of anything he might create), relations possess a type of reality; however, this reality is limited to their being the eternal objects of divine thought and thus is distinct from the determinate existence of actual or created things.12 Leibniz combines his thesis about the ideality of relations with a second thesis about the grounding of relational facts. Here it is essential to distinguish two senses of “relation.” On the one hand, the term may refer to a relation abstracted from its relata (. . . R . . . ); in this sense relations are merely ideal, or beings of reason. But the term also is used by Leibniz to refer to instances of relations that hold of two or more existing things (aRb). With respect to this second sense of relation, he defends the thesis that all relations are grounded in, or result from, non-­relational properties of substances. I interpret this as a claim about the supervenience of the former on the latter: [SR] For any difference in the relatedness of substances a and b, there are differences in the modifications of a and b that fully determine how they are related; thus there can be no difference in how things are related without a difference in the relata from which their relation results.13

11  “I hold that when God produces something he produces it as an individual, and not as a universal of logic” (Theodicy, §390; GP VI 346/H 358). “[T]he nature of an individual must be complete and determinate” (A II 2, 48/AG 72). On the unreality of incomplete notions, see New Essays II.i.2 (RB 109–110); A VI 4, 991. Abstracta of all kinds, whether qualities or relations, fail to qualify as real because of their incompleteness; hence they are merely ideal. However, some complete beings also are ideal, namely, “purely possible substances,” which “have no other reality than that which they have in the divine understanding and in the active power of God” (A II 2, 78/LA 61). Determinacy or completeness is thus only a necessary, and not a sufficient, condition for being an actual or existing thing. 12  “The reality of relations is dependent on mind, as is that of truths; but they do not depend on the human mind, as there is a supreme intelligence which determines all of them from all time” (New Essays II.xxx.4; RB 265). See also RB 227. Although Leibniz is not always careful to distinguish them, he assigns two roles to God’s intellect in grounding relations: (i) relations as such, and eternal truths that depend on them, have reality as ideas in the divine mind; (ii) contingent relational truths are true in virtue of God’s representation of the relations among existing things. The second role is the basis of the doctrine, to be discussed presently, that relations are “results” of their relata. See Leibniz’s 1715 remarks on a book by Aloys Temmik: “Besides substances, or ultimate subjects, there are modifications of substances, which can be produced and destroyed in themselves; finally, there are relations, which are not produced in themselves, but result from other things that are produced, and which have a reality outside our intellect, for in truth they depend on no thinking thing. Nevertheless, they receive this [reality] from the divine intellect, without which nothing would be true. Therefore, two things are realized through the divine intellect alone: all eternal truths and, from among the contingent ones, relational truths” (LH IV 8 Bl. 60, quoted in Mugnai [1992, 155]). 13  Though the reading has fewer supporters than it once did, Leibniz’s thesis about the grounding of relations in non-­relational properties of substances has sometimes been construed as a stronger claim about the reducibility of relations to non-­relational properties. For debate on this point, see

84  Leibniz on the Ideality of Space Leibniz himself employs the language of supervenience in conjunction with the doctrine that relations are merely ideal: “A relation is an accident that is in several subjects, and it is only a result or supervenes [supervenit] with no change made in other things, if several things are thought at once; it is their co-­thinkability [concogitabilitas]” (A VI 4, 866).14 That substances a and b are related is determined by their non-­relational properties. There are no other facts about the world that are relevant to determining their relatedness. Still, the fact that a and b are related is explained by there being a relation, qua idea, through which a and b are thought together. For this reason, Leibniz says that a relation supervenes on the properties of its relata and that it itself is merely the “co-­thinkability” of those relata. Assuming that the scope of what is “co-­thinkable” is set by the content of the divine intellect, it follows that having fixed the non-­relational properties of substances, God has thereby also fixed their relational properties, and that there can be no difference in the latter without a difference in the former.15 The thesis [SR] is a close relative of Leibniz’s oft-­repeated claim that “there are no purely extrinsic denominations.”16 An extrinsic denomination in his parlance is a designation of a subject in a way that involves reference to some other subject. A purely extrinsic denomination of a would be a designation of a that comes to be true of it by virtue of its relation to some b, without any change occurring in a’s qualities. Standard examples are a’s being known by b or a’s being a certain distance from b. Throughout his writings, Leibniz rejects the possibility of purely extrinsic denominations. Whenever a substance acquires a new relational property (or a new extrinsic denomination can be said of it) because something to which it is related has changed, there must be a correlative change in the first substance as well. As he writes, “no one becomes a widower in India by the death of his wife in Europe unless a real change occurs in him, for every predicate is in fact contained in the nature of the subject” (A VI 4, 1503/L 365). This follows from Leibniz’s acceptance of a strong version of the principle of sufficient reason: for whatever is true of a substance, there must be a ground for that truth in the

Mates 1986; Cover 1989; Cover and O’Leary-­Hawthorne 1999. The interpretation advanced here is in broad agreement with that of Mugnai 1992 and 2012. 14  Cf. the texts cited in Mates (1986, 224–225); and Mugnai (2012, 182–183), who notes the difference between modern accounts of supervenience and Leibniz’s conception. 15 For a nuanced treatment that raises concerns about the grounding of relations within the ­physic­al world, see Jauernig 2010. Jauernig’s account throws into relief a question that I have left aside: what are the substances to which [SR] is to be applied? If they are limited to monads, as Leibniz’s late metaphysics holds, then what are we to say about the other concrete things acknowledged by him (e.g. bodies)? A quick answer is that there is no set of non-­relational facts about the latter; anything we say about bodies will be relational all the way down. Nevertheless, these facts too will supervene on non-­ relational facts about the modifications of substances. An extended version of [SR] would thus assert that for any difference in the relatedness of individuals a and b (whether or not they are substances), there are differences in the non-­relational properties of substances that fully determine how a and b are related. 16  New Essays II.i.2 (RB 110), and II.xxv.5 (RB 227); A VI 4, 1645/AG 32; C 8/MP 133.

Donald Rutherford  85 nature of the substance. If a is truly designated as related to some b, then this must reflect a fact about what a is as well as about what b is.17 The conjunction of [IR] and [SR] defines Leibniz’s theory of relations. It combines a thesis about the ideality of all relations (they are nothing more than ideas or truths about how things are or can be related) with the thesis that all instances of relatedness supervene on the non-­relational properties of substances. In what follows, I will be concerned primarily with the ramifications of [IR]. At this point, however, it is worth flagging the significance of [SR] for Leibniz’s doctrine of space and time. His rejection of ungrounded relations between individuals—in particular ungrounded spatiotemporal relations—is one of the main targets of Kant’s attack on his philosophy. Whereas for Kant two individuals can be differentiated (i.e. related as non-­ identical) on the basis of their spatiotemporal properties alone, Leibniz insists that any spatiotemporal difference by which they are distinguished as two must be grounded in some difference in the non-­relational properties of substances.18 To reiterate the main conclusion of this section: the principles of Leibniz’s metaphysics entail that any relation, considered in itself, is ideal. The accident argument shows that instances of relatedness are not existing entities in their own right, but facts that supervene on the modifications of individual substances. The indeterminacy argument adds to this the claim that a relation qua universal, applicable to actual and possible things, fails to satisfy a necessary condition on something’s being an actual or existing thing. It is not a determinate this-­or-­that, but a way that things in general can be with respect to each other. To this extent, relations are abstracta, whose reality is limited to that of objects of the divine intellect. This reflects Leibniz’s considered view of the nature of relations throughout his career. Thus, it is an essential basis for interpreting his defense of relationalism in his later discussions of space and time.

2  Order as a Species of Relation Although Leibniz denies to relations the ontological status of actual or existing things, a large part of his philosophy is devoted to explicating the relatedness of created things—or as he conceives it, “the unity in multiplicity.” Among the most important relational concepts on which he draws is that of order. In general, order is the relation that several things bear to each other insofar as they can be distinguished in accordance with some rule or principle.19 In contrast to the simple 17  A VI 4, 1645–1646/AG 32; C 9–10/MP 134–135. 18  I return to this topic in §5 of this chapter. 19  “Order is the relation of many things among themselves by which any one can be distinguished from any other” (A VI 4, 866; cf. 868). “Distinct thinkability [cogitabilitas] gives order to a thing . . . For order is simply the distinct relation of several things. And confusion is when several things are indeed

86  Leibniz on the Ideality of Space case of many things taken as one (e.g. the parts of a whole), in the case of order, one designates a specific relation by which elements can be distinguished from one another, as well as conditions under which primitive relations determine more complex relations. So, for example, if natural numbers are related to their nearest neighbors as immediate successors, they are similarly related to other numbers through the repetition of the same relation of succession; and this repetition determines the “position” of each natural number relative to that of every other number and rules by which they are “connected.”20 Like any relation, order, as a system of relations (or a relation of relations), is merely ideal. It is not an enumerable part of the world, not a substance or an accident, but an idea by means of which the distinctions among individuals are cognized.21 The relation of succession, which determines the order of natural numbers, is critical to Leibniz’s metaphysics. He himself draws the parallel between a series of numbers and the law of the series by which the succession of a substance’s states is determined.22 The same notion of an order of succession underwrites his analysis of time.23 The concept of space expresses a second fundamental type of order: not the order by which one thing succeeds another, but the order that a plurality of things, taken together, have with respect to each other. This is the order exemplified by the pieces on a chess board or the pixels on a video screen, which Leibniz describes as an “order of coexistence.” Leibniz’s recognition of this type of order can be traced to one of his earliest works, his 1666 dissertation De arte combinatoria: The whole itself (and thus number and totality) can be broken up into parts, smaller wholes as it were . . . And the disposition of the smallest parts, or of the parts assumed to be smallest (that is, the unities) in relation to each other and to the whole can itself also be varied. Such a disposition is called situs. [A VI 1, 171/L 77]

present but there is no principle [ratio] for distinguishing one from another” (C 535/MP 146; translation modified). For further texts and discussion, see Rutherford (1995, 26–35). Leibniz defines harmony, similarly, as “unity in variety” or “unity in multiplicity.” This supports a close connection between the concepts. As he writes to Wolff, “Order, regularity, and harmony come to the same thing” (GLW 172/AG 234). 20 See New Essays, II.xxv.6: “But there are instances of relations between several things at once, as occurs in an ordering [l’ordre] or in a genealogical true, which display the position and the connections of each of their terms or members. Even a figure such as a polygon involves the relation among all its sides” (RB 227). 21  Again, though “ideal,” order is not merely subjective, or dependent on human cognition, but possesses a type of reality as the content of a divine idea: “Relations and orderings [les ordres] are to some extent ‘beings of reason’ [l’être de raison], although they have their foundations in things; for one can say that their reality, like that of eternal truths and of possibilities, comes from the Supreme Reason” (RB 227). 22  Leibniz to De Volder, January 21, 1704 (GP II 263/L 534). 23  See Futch 2008.

Donald Rutherford  87 Leibniz claims that the notion of “disposition,” or arrangement, applies to any plurality of parts, “either real or conceptual” (A VI 1, 177/L 80). Given any plurality, we can take its elements simply as parts of a common whole, or we can consider how they are with respect to each other. In whatever way it is possible to do the latter, we presuppose some order according to which the elements are situated with respect to each other. It is this idea of an order of situs, or “coexistence,” that is the starting point for Leibniz’s analysis of space. As the passages from De arte combinatoria indicate, the idea appears very early in his writings. When Leibniz broaches the idea of space as an “order of situations” in the Clarke correspondence, we should see him as drawing on this background.24

3  Space as an Order of Coexistence In his Third Letter to Clarke, Leibniz advances the following account of space: I have said more than once, that I hold space to be something merely relative, as time is; that I hold it to be an order of coexistences, as time is an order of successions. For space denotes, in terms of possibility, an order of things which exist at the same time, considered as existing together, without entering into their manner of existing. And when many things are seen together, one perceives that order of things among themselves.  [LC III 4; cf. LC V 27–29]

We should resist the temptation to read too much into Leibniz’s statement that space is something “merely relative” (de purement relatif). In this general description of the nature of space, there is no suggestion of the relativity, or frame-­dependence, of spatial position. Rather, Leibniz’s statement is an expression of the idea that, fundamentally, space is just a certain order of things, or a system of relations. Specifically, he says, space is “an order of coexistences.” This def­in­ ition is repeated on many occasions.25 The notion of coexistence remains imperfectly analyzed by Leibniz. It is glossed as existing “at the same time” or existing “simultaneously,” and is contrasted with existing successively. This might make it seem that the idea of coexistence presupposes the temporality of existence, and that “coexisting” is limited to a qualification of the temporal existence of two or more things (i.e. they exist at the same time). This is not Leibniz’s intention. “Coexistence” is the basic mode by

24  Arthur 2013 finds an immediate source for Leibniz’s conception of situs in Hobbes’s De Corpore, II.xiv.20–21. This is consistent with the claim that the foundations of his position are laid early. The influence of Hobbes on Leibniz is most pronounced during the late 1660s and early 1670s. 25  GM VII 18/L 666; GP III 612/L 656; GP III 622; GP IV 568/L 583; LDB 140–141.

88  Leibniz on the Ideality of Space which things are taken as existing together, where the existence of one does not preclude the existence of the other.26 Such may be true of a plurality of things whose existence is taken as temporal, but equally it is a way of representing the relatedness of a plurality of timeless entities, such as the vertices of a triangle.27 Like any relation or order, space represents, “in terms of possibility,” a way things can exist with respect to each other. And, it does so “without entering into their manner of existing.” The basic thought underlying Leibniz’s analysis of space, then, is this: take any plurality of things; abstract from their distinguishing qualitative features, and ask: how do they coexist with respect to each other? If there is an answer to this question, because the supposition of their joint existence is consistent, we in effect are conceiving of a spatial order among them.28 Crucially, we must remember that space itself—the order of coexistence—is only ideal. It is not a concrete thing, or place, but a way of representing how things can coexist with respect to each other.29 As I emphasized in the case of Leibniz’s general theory of relations, we must be careful to distinguish space as an abstract order of coexistence and space as instantiated in the relations among a plurality of existing things. It is common to identify Leibniz’s “theory of space” with the latter conception, seeing space as a system of relations determined by the things located within it (e.g. physical space is determined by the relations among bodies that are, or can be, located with respect to each other). Yet this is not his fundamental notion of space. Space itself is an order of coexistence, which is something ideal: a representation of how things can coexist with respect to each other. Particular things come to be spatially related by being represented in terms of such an order. Consistent with the thesis [SR], the spatial relations thereby determined supervene on non-­ relational properties of their relata (or on non-­relational properties of substances on which the relata depend); but this assumes that the relata are related through a prior idea of space as an order of coexistence. Independent of such an order, individual substances and their non-­relational properties might exist, but they would not coexist with respect to each other. The latter presupposes that, at least with respect to ideas of the divine intellect, the ground of their “co-­thinkability,” things are represented as existing in relation to each other. Although these claims are far removed from common assumptions about space, they are basic to 26  “If a plurality of states of things is assumed to exist which involve no opposition to each other, they are said to exist simultaneously [simul]” (GM VII 17/L 666). 27 I omit for the moment Leibniz’s statement that representation of the coexistence of things involves perceiving the order that they have among themselves. I return to this point in the next section. 28  If the existence of one precludes the existence of another, then they can only be related tem­por­ al­ly, through an “order of succession.” 29 See New Essays II.xiii.17: “space is no more a substance than time is . . . It is a relationship: an order, not only among existents, but also among possibles as though they existed. But its truth and reality are grounded in God, like all eternal truths” (RB 149–150). Cf. RB 153–155.

Donald Rutherford  89 Leibniz’s approach. The challenge is to understand how from these starting points, he accounts for more familiar notions of space in mathematics, physics, and metaphysics. We may note first that, as an abstract order of coexistence, space can assume a variety of forms. For this reason, Leibniz’s definition is best understood as fixing a genus of which different species are determined through qualifications of the kind of order of coexistence they involve. One basic qualifier is whether the order is discrete or continuous.30 The possibility of the former kind of space is suggested by the example of a genealogical tree that Leibniz gives in his Fifth Letter to Clarke: [Space] can only be an ideal thing; containing a certain order, wherein the mind conceives the application of relations. In like manner, as the mind can fancy to itself an order made up of genealogical lines, whose bigness would consist only in the number of generations, wherein each person would have his place . . . And yet those genealogical places, lines, and spaces, though they should express real truth, would only be ideal things.  [LC V 47]

Leibniz does not claim that a genealogical tree is a species of space.31 Instead, he uses the example to motivate the thesis that space is an ideal order within which different individuals can occupy the same place (e.g. similarly related members of different families), and in which a place retains its relations to other places irrespective of whether or not they are occupied (there are places for my children whether or not I have any). Consequently, there are “real truths” about how I would be related to possible descendants who may or may not exist. The example illustrates nicely the generality of Leibniz’s conception of space as an order of coexistence. Although a genealogical tree is not an order of coexistence, the analogy shows that Leibniz frames his definition in a sufficiently broad way that it is not restricted to continuous, geometrical space but extends to spaces whose elements are discrete entities, and in which notions of place and distance can be applied in describing the relations among elements. Of course, by “space,” Leibniz almost always does mean three-­dimensional Euclidean space. His most general description of this space is that it is “pure absolute extension [Extensi puri absoluti].” Elaborating on this description, he continues, “by pure, I mean from all matter and change, and by absolute, I mean an unlimited space, containing all extension. All points are thus in the same space

30  Another qualifier is the mind on which the idea of space depends. To this point, I have assumed that this mind is God, the ground of all truth for Leibniz. But we will see in §5 of this chapter that finite minds, or monads, also can be the source of representations of order in terms of which things are perceived as spatially related. 31  A genealogical tree contains features of both an order of coexistence and an order of succession. The basic relation by which elements are related, in terms of which their “distance” is measured, is a successor or ancestor relation.

90  Leibniz on the Ideality of Space and can be related to each other.”32 Space taken absolutely is conceived as comprising all possible extension (along three dimensions33), and it is a property of such a space that within it every point has a definable relation to every other. The other feature of space mentioned by Leibniz is that it is “pure,” or uniform; that is, free of all matter and change. This feature is closely related to the continuity of space, a property on which Leibniz sometimes rests the case for space’s ideality. The relation between these two properties, however, is less straightforward than has sometimes been assumed. While continuity plays a role in Leibniz’s argument for the ideality of space, the deeper principle at stake is one deriving from his theory of relations, as argued in §1 of this chapter. The relation between the continuity of space and its ideality figures centrally in Leibniz’s resolution of the labyrinth of the continuum.34 His crucial insight is that it is a mistake to think of a continuum as resolvable into, or constructible from, determinate parts. No summation of points, or indivisible elements of space, can produce continuous extension. That way the labyrinth lies. Consequently, Leibniz believes, it is necessary to draw a sharp distinction between concrete matter, which is actually divided into parts and whose reality is grounded in indivisible simple substances, and space, which is not divided into parts and not composed of elements, though arbitrary divisions can be conceived in it. This formal distinction between the properties of matter and those of space (i.e. whether they are actually divided and grounded in unities) is the basis of the difference in their ontological status: whereas matter is determinate and real, because constituted from real things (simple substances), space is indeterminate and ideal: Composition is only in concretes . . . [I]n actual substantial things, the whole is a result or coming together of simple substances, or rather of a multitude of real unities. It is the confusion of the ideal with the actual which has muddled 32  Characteristica geometrica (hereafter CG), dated August 10, 1679, §9 (GM V 144). I rely on the text of Echeverría and Parmentier in Leibniz (1995, 150–151). In this relatively early work Leibniz immediately follows the quoted passage with this: “It is not important for the moment to know whether this space separated from matter is a substance or only a phenomenon, that is, a coherent appearance.” As Hartz and Cover (1988) argue, Leibniz’s mature view is that space is not a phe­nom­ enon or appearance, but “ideal.” Nevertheless, the conception of space conveyed in this definition is preserved in later writings: “Absolute space is the fullest place [locus plenissimus], or the place of all places . . . From four points not falling in the same plane there results absolute space [Spatium absolutum]. For every point whatsoever is uniquely determined by its situation in relation to four points not falling in the same plane” (ca.1714; GM VII 21/L 669). Absolutum is a technical term. An early study quotes the following definition from John Wilkins’s Essay towards a Real Character and a Philosophical Language (1668), which remains the basis of Leibniz’s understanding of it: “The absolute is that whose concept is unlimited, or that outside of which nothing of the same kind can be considered, that is, that which is capable of quantity and nonetheless involves no limits. Hence, absolute extension can be conceived, but not an absolute circle. God is an absolute being, for there is no other reality or perfection which is not in God. We may best say that the absolute is that which is purely positive in its kind” (A VI 4, 36). See also New Essays II.xiv.27 (RB 154), II.xvii.1–3 (RB 157–158) and II.xxv.10 (RB 228). 33  On the significance of this restriction, see n. 59 below. 34  See Arthur’s introduction to Leibniz 2001 and Levey 1998.

Donald Rutherford  91 everything and caused the labyrinth of the composition of the continuum. Those who make up a line from points have looked for the first elements in ideal things or relations, something completely contrary to what they should have done; and those who found that relations like number or space . . . cannot be formed by the coming together of points were wrong, for the most part, to deny that substantial realities have first elements, as if the substantial realities had no primitive unities, or as if there were no simple substances.  [GP IV 491/AG 146]35

Leibniz sometimes frames the difference between real aggregates and ideal continua as a difference in the direction of the part–whole relation. Whereas in the former, parts are prior to wholes, which are composed from them, in the latter the whole is prior to the parts, which are not actually present as determinate components, but merely conceivable through arbitrary divisions (points as boundaries of lines, lines as boundaries of planes, etc.).36 This lends support to the idea that it is continuity as such that explains the ideality of space, for it is characteristic of ideal things that they are not explicable in terms of prior parts or elements: The mass of bodies is actually divided in a determinate manner and there is nothing exactly continuous in it; but space or the perfect continuity that is in the idea marks only an undetermined possibility of dividing it as one will. In matter and in actual realities the whole is a result of the parts, but in ideas or possibles (which comprehend not only this universe but any other which can be conceived and which the divine understanding in fact represents to itself), the whole is prior to divisions.  [GP VII 562; cf. GP III 622]

The problem with drawing the distinction between the real and the ideal in terms of the distinction between the discrete and the continuous is that it risks ruling out as impossible discrete orders of coexistence or succession. Undoubtedly, the case of continuity is special for Leibniz, because he locates a conceptual impossibility in the construction of a continuum through a summation of point-­ like elements: either the elements themselves (infinitesimals) are incoherent, or there is no operation defined over them by which a continuous extension can be generated from them.37 This is in contrast to discrete orders, whose elements sum 35  See also LDB 141; GP VII 561–563. This explanation glosses over an important distinction for Leibniz between the parts into which a material thing is actually divided, and from which it is composed mereologically, and the simple substances that are the ground of its reality. For Leibniz, the latter are “constituents” of any material thing, but not parts of it (GP II 267–268). The distinction is extraneous to the point at issue here about the correct characterization of the ideal. For further discussion, see Rutherford (2008a). 36  “Strictly speaking, points and instants are not parts of time or space, and do not have parts either. They are only boundaries [extremités]” (RB 152). Cf. GP VI 627/AG 228. 37  Leibniz’s claim about the non-­additivity of infinitesimals to produce a continuum reflects limitations in his conception of mathematics. For a modern demonstration of what he argues is impossible, see Bell (2008).

92  Leibniz on the Ideality of Space additively. Nevertheless, although Leibniz believes that any continuum must be ideal because it cannot be resolved into simplest elements, it does not follow that there cannot also be discrete orders whose ideality is explained in a similar fashion. In this case, too, the order (e.g. of natural numbers) does not result from the summation of prior elements. On the contrary, the order as represented in thought is what makes possible the act of summing units, tracing a line of hereditary descent, or making a move in a game of chess. In the case of discrete orders as well, then, the whole must be prior to the parts, for it delineates a space of possible ways that things can be related to each other according to that order. If this is sufficient for ideality in the case of a continuous order, it is so also in the case of discrete orders.38 The other property that Leibniz associates with the ideality of a continuum is its uniformity. Continua are everywhere the same; each part or region is indistinguishable from every other. This, he maintains, is a property inconsistent with continua being actual or real things, because God misses no opportunity to create variety in the world, and even more basically, does not create things whose parts are indistinguishable.39 Significantly, though, this is a property that continua share with discrete orders of coexistence. Consider an infinite “pure absolute” space consisting of discrete nodes or cells. The elements of this space are as indistinguishable and lacking in variety as points determinable in a continuous space. The spaces have different geometrical structures, but if uniformity is an argument for the ideality of continuous space, then it is equally an argument for the ideality of a discrete space.40 Looking more deeply into Leibniz’s account, we see that the crucial factor that renders space ideal is that it is nothing more than an order or system of relations. Any such order has two features that support its description as “ideal,” according to the indeterminacy argument. First, any relation or order considered in itself is indeterminate, because it is incomplete; it awaits determination by suitably modified relata. Second, no relation or order can be defined in terms of a given set of relata, for it holds of any possible things that can be conceived as related in

38 As Crockett (2008, 56–57) argues, Leibniz does not give a satisfactory explanation of the discrete-­continuous distinction. Often the term “discrete” means for him “divided into actual parts” (or “grounded in true unities”), in which case anything discrete is a determinate existing thing. Here I aim to save room for ideal spaces with a variety of geometrical structures. On this, see Crockett (1999). 39 See New Essays II.i.2: “Things which are uniform, containing no variety, are always mere abstractions: for instance, time, space, and the other entities of pure mathematics. There is no body whose parts are at rest, and no substance which does not have something which distinguishes it from every other” (RB 110). Cf. GP VII 563. 40  Whether the same holds for discrete orders of succession is another question. The succession of natural numbers might be thought to lack the property of uniformity, because, e.g., the place of the number two is distinguishable from that of the number three. This, though, is an artifact of these numbers being defined relative to a fixed point (zero or one). As such, it is analogous to a relative space defined with respect to a body assumed to be at rest.

Donald Rutherford  93 the manner represented by the order. To this extent, the whole is prior to the parts in any ideal order.41 When Leibniz is being most careful in presenting his conception of space and time both of these points come out clearly. Space and time are ideal, because they are nothing more than orders of relations. And, insofar as they are ideal orders, they are the basis of necessary truths that apply equally to the actual and the possible: [Time] is only a principle of relations, a foundation of the order in things, insofar as one conceives their successive existence, or without them existing together. It must be the same with space. It is the foundation of the relation of the order of things, but insofar as they are conceived to exist together. Both of these foundations are true, although they are ideal . . . The imaginary possible participates as much as the actual in these foundations of order, and a novel could be as well regulated with respect to places and times as a true story.   [GP VII 564]42

The chief point to be emphasized is that Leibniz’s treatment of space is premised on a set of general considerations about the nature of relations and their ideality, and the assumption that space and time can be explicated in terms of two primitive systems of relations (the orders of coexistence and succession) that condition the existence of all actual or possible things. In brief, his theory incorporates the following claims: 1. An analysis of the concept of space reveals space to be ideal because it is not a substance or a property of substance, but only a certain order of things, namely, an order of coexistence. 2. The properties of an ideal order—that it is indeterminate and prior to its relata—hold for continuous and discrete spaces. While existing things instantiate a set of determinate spatial relations, the order of space is not exhausted by these relations. There are other ways things could have been and such things would be related according to the same spatial order.43 3. The “pure absolute extension” that is the object of Euclidean geometry is distinguished from other conceivable spatial orders by (among other things) 41  This is consistent with the account of Hartz and Cover, omitting the emphasis they place on the continuity of space and time: “Space and time . . . share a crucial feature with mathematical entities— roughly, that they are indifferent to what they apply to, or to what is placed in their jointly formed ‘grid.’ Together space and time constitute a continuous arbitrarily-­divisible conceptual ‘event-­space’ which, though usefully applied to bodies, is clearly not on an ontological par with them” (Hartz and Cover 1988, 499). I agree with all of this except the implication that the arbitrary divisibility of a continuum is definitive of its ontological status as an ideal entity. 42  See also GP IV 491/AG 146; GP IV 568/L 583; GP IV 568/L 583. 43  This agrees with the conclusion of Belot (2011, 173–185), who defends a reading of Leibniz as a “modal relationalist.”

94  Leibniz on the Ideality of Space the property of continuity, which entails that it cannot be resolved into simplest elements. This is the key insight of Leibniz’s analysis of the labyrinth of the continuum: points are not the simplest parts of a continuum but merely the boundaries of line segments, arbitrarily designated in space.

4 Space, Situs, and Geometry As an alternative to his definition of space as an “order of coexistences,” Leibniz also describes space as an “order of situations”: I don’t say . . . that space is an order or situation, but an order of situations; or (an order) according to which situations are disposed; and that abstract space is that order of situations conceived as possible. Space is therefore something ideal. [LC V 104]

Again, we may note the inference from space conceived as an abstract order embracing actual and possible things to the conclusion that space is merely ideal. In this text, however, the order is described not as an order of coexistence but an “order of situations,” or (in Latin) situs. The notion of situs is foundational to Leibniz’s understanding of geometry and the ontology of space.44 The core idea is presented in his Fifth Letter to Clarke. Whenever we consider that many things exist at once, and “observe in them a certain order of co-­existence, according to which the relation of one thing to another is more or less simple,” we conceive “their situation or distance” (LC V 47). Other comments expand on this statement. Situation is not equivalent to the idea of distance, but it involves it. “Situs is a mode of coexistence. Therefore, it involves not only quantity but also quality” (GM VII 18/L 667). “Situs is a certain relation of coexistence among a plurality of entities; it is known through other coexisting things which serve as intermediates, that is, which have a simpler relation of coexistence to the original entities” (GM VII 25/L 671). Broadly speaking, situs connotes the way in which anything is arranged, or disposed, with respect to other things within a single order of coexistence. It answers the question where the thing is in relation to other things.45 All things that coexist have a situs with respect to each other. Thus, any order of coexistence must also be an order of situations. So understood, situs incorporates notions of 44  See De Risi (2007) for the most comprehensive treatment of the topic. Many of the points made in this section receive more extensive discussion, backed by a detailed survey of the textual evidence, in De Risi’s book. The significance of the analysis situs for Leibniz’s metaphysics of space has been emphasized by Arthur in papers dating back to the 1980s. Citations can be found in Arthur (2013). 45 From manuscript studies: “Situs is the relation of one thing to another according to place [locum]” (Leibniz 1995, 304); “Situs is nothing but that state of a thing by which it can be understood to exist in a certain way with extended things; in other words, a mode of coexisting” (Leibniz 1995, 276).

Donald Rutherford  95 both distance and orientation. The former is highlighted in the preceding quotations. The relative situation of two things is more or less simple, based on the number of things separating them. This reflects how near to, or far from, each other they are. But the situs of any two objects equally, and more basically, expresses how things are oriented with respect to each other, or the shape or configuration determined by them. Thus, Leibniz emphasizes, the order of situ­ations involves notions of both quantity and quality.46 Both of these features of situs, distance and orientation, imply that, though conceived as the property of an object, situs is an inherently relational property—one that an object possesses only in relation to other things co-­locatable with it. This explains some of the ambiguity in Leibniz’s descriptions of situs. Sometimes it is represented as a mode of a thing (“a mode of coexistence”, GM VII 18); at other times, as a relation among things (“a certain relation of coexisting among several things”, GM VII 25).47 This ambiguity can be resolved if we focus on the principles of Leibniz’s ontology. If we are thinking of the situs of a concrete thing, a body or a substance, he holds that for any such thing there is a unique relational accident (or set of relational accidents) that reflects where it is relative to the other things that coexist with it. As he explains to Clarke, any body that nominally occupies the same place as another body will nonetheless have a distinct situs relative to its neighbors, for the situs is an individual accident of the body.48 At the same time, Leibniz maintains that relational accidents, designated by what he calls “extrinsic de­nom­in­ ations,” are not ontologically basic. For any such accidents truly predicated of related things, there must be modifications of substances in virtue of which they bear the relevant relation to each other. Given this, we can infer that situs, as a relational accident, or “mode of coexistence,” of concrete things (i.e. bodies), must be grounded in modifications of monads. Considerable evidence supports this as Leibniz’s position, but it is not a topic that I can pursue further here.49 Instead, I will concentrate on Leibniz’s treatment of situs as a relational property of geometrical objects. Here the ontological considerations are quite different. Geometrical objects (points, lines, circles, planes, spheres, etc.) are not concrete res. They are inherently indeterminate abstract entities, whose ontological status

46  Although Leibniz usually takes situs to include properties of distance and orientation, in at least one study he relates it more closely to the latter: “The states in which extended things exist are their magnitude . . . and their shape . . . In magnitude, there is a relation to the number of parts; in shape, [there is a relation] to situs. Situs, such as to be parallel, inclined, straight, oblique, concave, assymptotic, secant or tangent with respect to something else” (A VI 4, 600). 47  De Risi (2007, 99 n. 113) points out that although these passages are presented as being from the same essay in Gerhardt’s edition, they appear in two separate manuscripts. 48  LC V 47 (GP VII 400–401). 49  Thus, it is not situs, but monadic perceptual states that are the ultimate bases of relations of coexistence: “things that differ in place must express their place, that is, they must express the things surrounding, and thus they must be distinguished not only by place, that is, not by an extrinsic denomination alone, as is commonly thought” (GP II 250/AG 175). See also C 9/MP 133; RB 110; and the discussion of Mugnai (2012).

96  Leibniz on the Ideality of Space is the same as that of space itself; they are entia rationis, whose reality is limited to that of divine ideas. For this reason, there is no question of giving a deeper ana­ lysis of the relational properties of such objects in terms of modifications of substances. In the final analysis, geometrical objects are nothing more than certain sets of relations that can be designated, or discerned, within a given space, or order of coexistence. In support of this conclusion, I will briefly consider Leibniz’s elaboration of the technical notion of situs within geometry and the connection he establishes between geometrical knowledge and the perceivability of spatial relations. Leibniz proposes the project of an analysis situs in response to what he sees as the inadequacies of existing treatments of geometry, especially accounts in which the properties of geometrical objects are limited to those that can be represented by numbers or ratios. Such accounts, he believes, are incapable of expressing certain kinds of geometrical knowledge that we in fact possess. A prime example is knowledge of the properties of similar figures—figures having the same form, or “quality,” but different magnitudes (e.g. differently sized circles or equilateral triangles). Based solely on their internal relational properties (e.g. the ratio of radius to area in a circle), such figures are indistinguishable; yet there is a difference that can be discerned between them when they are presented together, as well as a systematic relation between the property of similarity and other geometrical properties.50 Leibniz regards situs as the appropriate concept for analyzing such properties. Without attempting to convey the full scope of his project, I will highlight several features of the account that bear on his conception of geometrical space. Situs is explicated in terms of the dual notions of space and point: Space is pure absolute extension, or what possesses all magnitude and no situs. A point is pure absolute situs, or what has situs absolutely and through itself but no magnitude.  [Leibniz 1995, 138]

Space, by definition, contains all extension and for that reason has no situation with respect to anything else. By contrast, a point is identified with a primitive situation, or that which is locatable relative to any other point or figure in space.51 50  The project of the analysis situs is announced in writings from the late 1670s (see the well-­known letter to Huygens of September 8, 1679, printed at GM II 17–20/L 248–249), and is pursued in many works, including those devoted to a reconstruction of the foundations of Euclidean geometry. On the latter, see the texts collected in Leibniz 1995, and in De Risi 2007, which includes a provisional chron­ ology of the works relating to the project (117–126). 51  “We understand a point to be, of all things having situs, that whose situs is simplest and hence lacks extension, or in which there is nothing else having extension except itself . . . We understand space to be that which is most absolute in extension, and so lacks situs, and thus there is in it everything else having situs or extension, and consequently many things existing at the same time and having situs, and also having situs among themselves” (A VI 4, 606).

Donald Rutherford  97 To avoid misconstruing Leibniz’s position, we must be clear on two things. First, Leibniz does not argue that Euclidean space is to be understood as a construction from points, or situations, conceived as geometrical simples. To suggest this, would be to fall back into the paradoxes associated with the labyrinth of the continuum. Although a point is defined as having the simplest situation, it is a situation in space, the cognition of which is a condition for discerning the relations of geometrical objects. (I will return to this claim shortly.) Second, although points are assigned a primitive situation relative to other points, “in itself,” the situs of any point is indistinguishable from that of every other. This reinforces the conclusion that in itself no point is distinct from any other point and no point has an intrinsic situation in space. The situs of points can be distinguished only if two or more points are related in space.52 Consistent with this account, Leibniz maintains that from a single point nothing else is determined, but that from any two points there follows a unique geometrical figure: With any one point assumed, nothing further is determined, but with two points (or even more) assumed, by that fact something else in addition is determined, which follows uniquely from those givens. But that which is determined from the assumption of two points is the simplest extension that passes through them, which we call a straight line [rectam]. Certainly, when the mind designates two points in space, by that fact it conceives an indeterminate straight line passing through them. For it is one thing to regard single things singly, and another to regard them both as existing at the same time and so the situation they have with respect to each other.  [Leibniz 1995, 278]

We are thus to conceive of a straight line as a construction in space, when the mind distinguishes two points and their relation with respect to each other. The simplest such relation gives a new entity, the straight line, whose properties are determined by the relative situations of the points. Leibniz regards it as significant that a straight line (or any other geometrical figure) can be specified without reference to the distance between its endpoints. Take any two points in space and 52  “But we assume before all things that there is a certain absolute extension, in which nothing is considered in itself except extension, and in which are many things having situs. This is commonly called universal space. It is in itself everywhere similar to itself, and consequently proceeds to infinity, and that which is located in it can continually change situs in it” (A VI 4, 605–606). Here I dis­ agree with Arthur’s interpretation. He argues that Leibniz construes “space as constituted by points, but not composed of them. Since each point has its situation to all the others, this is what an in­stant­ an­eous space consists in: it is a system of relations of situation, an order of situations” (Arthur 2014, 155). Although this may be true of physical space in which relations of situation are grounded in accidents of their relata, it is not true of points in a purely geometrical space. In this case, it makes no sense to speak of space as constituted by a plurality of object-­like points. In itself, every point has exactly the same situation; they are only given as distinct via relations designated within a given order of situations.

98  Leibniz on the Ideality of Space a line is thereby determined in thought. Beyond this, we can draw further distinctions among such figures in terms of the distance between the points, but the straight line is a figure uniquely determined by the situs of any two points. Likewise, a plane is uniquely determined by the situs of any three points. And more complicated figures are determined by relations of situation (involving both orientation and distance) among these prior figures. So, for example, the surface of a sphere is defined as the locus of all those points having the same situs with respect to a given point (Leibniz 1995, 242). The geometrical property that any such class of figures (straight lines, circles, equilateral triangles, etc.) has in common is that they are similar to each other. The formal properties determined by the relative situation of given points or lines are the same for each member of the class. For this reason, if we consider just the determination of any figure by simpler figures, we as yet have no basis for distinguishing larger or small examples of it. Taken by itself, every straight line determined by two points is indistinguishable from every other.53 We move beyond this only when we represent the lines in relation to each other (or to some third thing) by perceiving them together, or “coperceiving” them: Similars are those things that cannot be distinguished as single things considered in themselves, such as two equilateral triangles. For we can find no attribute, no property in one that we cannot also find in the other . . . Yet if they are perceived together [simul], there immediately appears a difference; one is larger than the other. The same can happen even if they are not perceived together, provided that something like a means or measure is assumed, which is first applied to one, or to each in itself, and it is noted how the first, or a part of it, agrees with the measure or a part of it, and then the same measure is also applied to the other. And so I am accustomed to say that similars cannot be distinguished except by being coperceived [per comperceptionem].  [CG §31; Leibniz 1995, 182]54

Leibniz emphasizes the importance of similarity as a relation that holds among geometrical figures when we abstract from considerations of magnitude. The form of such figures, which we represent in terms of relations of situation among their determinants, is insufficient to distinguish between larger or smaller examples of them. Represented in itself, each is just a straight line, a sphere, etc. To distinguish geometrical figures as larger, smaller, or of the same magnitude

53  “When two points are perceived at the same time, by that fact there is perceived the situs of each with respect to the other. But any situs between two points is similar, and so they are distinguished by coperception or magnitude alone” (CG §105; Leibniz 1995, 228). 54  See also A VI 4, 564–565 (“Similars are those things that cannot be distinguished by concept, but by perception”); A VI 4, 74; A VI 4, 382; A VI 4, 514; A VI 4, 626–627; GM VII 30; CG §27 (Leibniz 1995, 178–180).

Donald Rutherford  99 (thereby establishing them as “congruent”55), instances of them must be perceived together, or in relation to a third thing, which serves as a common measure. Although Leibniz appeals to perception specifically in connection with the discernability of similar figures, his treatment of Euclidean geometry as a whole hinges on the relation between geometrical space and the perceptual capacities of human beings. For anything to have a situs in space is for it to be perceived to have situs. Two points have a situs relative to each other only insofar as they perceived to be so related. In general, any judgment about the form of a given figure, about how many such figures there are, or about whether they are similar or congruent to each other, rests on our ability to represent perceptually the form of Euclidean space.56 Leibniz does not, to be sure, think of geometrical knowledge as empirical knowledge that is established inductively through a survey of cases. Nevertheless, his account of geometry also does not identify it with purely conceptual knowledge, determined through necessary relations among the contents of intellectual ideas. Geometry involves an understanding of necessary relations among the properties of spatial figures, but the figures themselves (determined by relations of situs) are given to us through perception. Leibniz’s most general characterization of mathematics is as the science of the imagination.57 The imagination has as its object two types of properties: quantity or magnitude, over which is defined relations of equality and inequality, and quality or form, over which is defined relations of similarity and dissimilarity (A VI 4, 514). Geometry is that branch of mathematics in which “magnitude and similarity are applied to situs” (A VI 4, 362); that is, the branch of mathematics whose subject matter is form and magnitude insofar as they apply to situations in space and relations among situations. Geometry consists of knowledge of necessary relations among the form and magnitude of spatial figures, which 55  “[T]hose things that are both similar and equal are congruent” (CG §32; Leibniz 1995, 184). Cf. A VI 4, 168, 382, 514. 56  Leibniz assumes that if two figures are congruent in his sense—similar and of equal magnitudes—then they have the same situation and hence can be superimposed on one another or made to occupy the same place (GM II 22; GM V 172). Kant famously observed that this assumption fails in the case of “incongruent counterparts,” figures that meet the conditions of similarity and equality but that cannot be superimposed because of an internal difference of direction, e.g. otherwise indistinguishable right- and left-­handed spirals. (See “Concerning the Ultimate Ground of the Differentiation of Directions in Space,” 1768, in Kant (1992b), and the discussion in Storrie [2013].) The existence of incongruent counterparts demonstrates a limitation of Leibniz’s formulation of the analysis situs based on his definition of congruence, but it does not invalidate the general approach to the geometry of space ascribed to him here. Given the connection he establishes between situs and perception, Leibniz could in principle extend his account to include sameness of direction as a necessary condition for congruence understood as sameness of situs. 57  In his 1702 letter to Sophie Charlotte “On What is Independent of Sense and Matter,” he stresses the role of the imagination in the representation of mathematical concepts: “And these clear and distinct ideas, subject to imagination, are the objects of the mathematical sciences, namely arithmetic and geometry, which are pure mathematical sciences” (GP VI 501/AG 187–188). Cf. GM III 243; A VI 4 511, 513–514; RB 128.

100  Leibniz on the Ideality of Space implies our ability to represent those properties conceptually, in a way that supports demonstrative reasoning. What Leibniz’s writings on the analysis situs reveal, however, is that this knowledge presupposes a prior perceptual (or imaginative) representation of three-­dimensional Euclidean space. Without such a representation, we would be unable to grasp the idea of situs as it applies to geometry and the distinction between similar and congruent figures. We are able to comprehend geometrical truths only by having the objects of those truths immediately present to us in perception—and this in a way that does not depend on our experience of this or that particular figure. Although Leibniz does not explicitly formulate the idea of space as an a priori form of perception, his treatment of Euclidean geometry requires something like this as the basis of our capacity for geometrical knowledge. With respect to situs, we are only able to make assertions about the properties of extended figures if they are not just thought but perceived: [A] diversity of things, if the particular nature of them is not considered and they are regarded only as thought, produces a multitude of things or number, but if they are supposed not only to be thought but also to be perceived together, from this arises situation and extension.  [A VI 4, 381]58

Unless we perceive a plurality of things together, we do not represent them as extended, or as having the property of being situated with respect to each other. Consequently, geometry presupposes a perceptual, and not merely conceptual, representation of space. And, this representation must be “pure,” or prior to the determination of any figures in it, for space itself, according to Leibniz, is “pure absolute extension.” We are in a position to draw the following further conclusions about Leibniz’s theory of space. By virtue of what it is, an order of coexistence, any space is something ideal: an indeterminate order of relations. Three-­ dimensional Euclidean space is unique among conceivable species of space, because it is given to us perceptually as “pure absolute extension,” which is prior to the de­ter­min­ ation in it of geometrical figures, or the perception of spatially related bodies.59 58  See also CG §105: “When two points are perceived together, by that fact there is perceived the situs of each with respect to the other” (Leibniz  1995, 228); and Demonstratio axiomatum Euclidis: “That two points have situs I express thus: AC, that is, I conceive that both are perceived at the same time . . . For to perceive situs is to perceive two points and nothing besides, except this one thing, that nothing besides can be known, and this is the very concept of a straight line” (A VI 4, 173). “Situs contains two things: points are given by position, or are perceived, and they are perceived together, or their distance is perceived” (A VI 4, 174). 59 Leibniz argues that the three-­dimensionality of space follows “by a geometrical necessity, because the geometers have been able to prove that there are only three straight lines perpendicular to one another that can intersect at the same point” (Theodicy, §351; GP VI 323/H 335). If such a proof presupposes a perceptual representation of space, the necessity in question is weaker than logical necessity. The proof would not show that no other dimensionality of space is possible, but only that we

Donald Rutherford  101 No plausible reconstruction of Leibniz’s view can identify space with an order abstracted from a plurality of existing bodies.60 The representation of space as a whole must come first, making it in this sense as “absolute” as the rival Newtonian account of space. The difference is that, for Leibniz, absolute space is not a concrete thing, or substance, consisting of distinguishable parts or regions, but the representation of a uniform order, within which different figures can be designated and coexisting objects related to one another.

5  Empirical Knowledge and the Ideality of Space: A Kantian Coda In this essay I have aimed to reconstruct Leibniz’s theory of space in the broadest possible terms, beginning with his definition of space as an “order of coexistence.” From this definition alone it follows, given the principles of his metaphysics, that any space is ideal or mind-­dependent. Within Leibniz’s philosophy, however, space appears in a variety of guises that must be carefully distinguished. In mathematics, he is concerned with a purely abstract conception of space as an “order of situations.” Here he is mainly interested in giving a deeper account of the foundations of Euclidean geometry. Despite the limitations of his own analysis situs, his approach anticipates later developments in geometry that investigate the structures of a much wider range of abstract spaces. Elsewhere in his philosophy, Leibniz targets a different notion of space as an order of coexisting things. In this case, too, space itself is merely ideal, but the focus now is on how an order of coexistence is instantiated in the relations among a plurality of individuals. Leibniz confronts this question under two separate headings. First, space is one of the two coordinate dimensions under which God conceives of the “capacity” of a world. All and only those things that stand in well-­defined relations of coexistence and succession can belong to the same cannot—given Leibniz’s understanding of geometry—demonstrate the possibility of more or fewer dimensions. The claim that space is Euclidean in every possible world depends upon this assumption. See the discussion of Belot 2011, 182–184. 60 Contrast Arthur: “unlike later thinkers, Leibniz still conceives space as an abstraction from existing extended things, rather than as a self-­standing entity. Thus the distance involved in congruence is in the first instance that between situated bodies, bodies in extenso, and only derivatively in the space abstracted from them” (Arthur 2014, 156; cf. 159). Leibniz’s comments in LC V 47 (and LC V 104, citing the same passage) have been read as supporting such an interpretation. I believe he frames his position poorly in these sections. In LC IV 41, he expresses a view like the one I have attributed to him: “[Space] does not depend upon such or such a situation of bodies; but it is that order, which renders bodies capable of being situated, and by which they have a situation among themselves when they exist together; as time is that order, with respect to their successive position. But if there were no creatures, space and time would only be in the ideas of God.” A similar explanation of the order of time is found in the New Essays: “A train of perceptions arouses the idea of duration in us, but it does not create it. Our perceptions never provide a sufficiently constant and regular train to correspond to the passage of time, which is a simple and uniform continuum like a straight line” (RB 152).

102  Leibniz on the Ideality of Space world. This is a fundamental commitment of his metaphysics, which holds even if the things God creates (monads) do not stand in real external relations to one another.61 Second, space is the order that holds among extended bodies, each of which is assumed to be locatable relative to any other body. This is the space in which we perceive bodies to exist and whose precise representation forms a background to physical theories. One of the most difficult problems in the interpretation of Leibniz’s philosophy is how to frame the relationship between the order of coexistence relating mind-­ like monads and that relating extended bodies. This is not an issue that I can address here.62 What I want to do instead is to argue for a closer connection than is usually acknowledged between Leibniz’s geometrical account of space as an order of situations and the physical space of bodies. The way in which I propose to do this is by highlighting the common ground of these two spaces and their ideality in the perceptual capacity of finite minds. If, as Leibniz argues, our knowledge of Euclidean geometry presupposes a perceptual representation of space as “pure absolute extension,” then such a representation is equally a condition for the empirical cognition of bodies. The same capacity that accounts for our ability to represent relations of situs among points of space explains why we represent existing things as spatially extended and as related in ways that conform to the principles of Euclidean geometry. We represent them in this way because Euclidean space is, in effect, the form of human (and, more generally, monadic) perception.63 Leibniz does not draw this conclusion as explicitly as I have done, but it is supported by many things he says. He defines perception as the representation of a multitude in a unity.64 In any act of empirical perception—perception directed at the world—many things are represented as extended and coexisting.65 For Leibniz, to represent things as extended and coexisting is to represent them as 61 See GP VII 303–304/AG 150–151; GP IV 568/L 583; and the discussion in Messina and Rutherford (2009). 62  For attempts to do so, see Adams (1994); Rutherford (1995). 63  As perceived, space is not just an abstract idea represented in the divine mind: it is a representation that orders each monad’s perceptual field. Space must be the form of monadic perception and not its content (an appearance or the representation of an existing thing), for only in this way can Leibniz uphold the doctrine that space is ideal and not, like matter, a well-­founded phenomenon. 64  “The passing state which involves and represents a multitude in the unity or in the simple substance is nothing other than what one calls perception” (“Monadology,” §14; GP VI 608/AG 214). 65  In a note associated with the Des Bosses correspondence, Leibniz writes: “Extension . . . is only a multitude of coordinated co-­perceptions or phenomena, insofar as these have a common order of coexisting. I simultaneously perceive A, B, and C, and the co-­perception of A and B is something different from that of A and C or that of B and C, though without considering what is different in each of them or how A, B, and C differ intrinsically. And, in observing this, I say that I perceive space and extension” (LDB 312–313). Similar ideas are expressed in remarks to De Volder: “I believe that perception is involved in extension and even in motion” (LDV 104–105); “Extension . . . is resolved into plurality, continuity, and coexistence, i.e., the existence of parts at one and the same time. Plurality also belongs to number and continuity, also to time and motion, while coexistence is only added in that which is extended” (LDV 72–73).

Donald Rutherford  103 having a situation with respect to each other. We have seen, however, that things having a situation presupposes an order of situations by which they are related. The representation of this order—Euclidean space—is a condition for the representation of things as situated with respect to each other. For this reason, space can be taken as the form of empirical perception. Thus, whatever we perceive as coexisting, we perceive as spatially related. Consistent with this, Leibniz further asserts that it is a property of perception that we represent things as external to us.66 This too is plausibly understood in terms of his positing space as the form of perception: in representing things perceptually, we represent them as extended and as having different but locatable positions relative to us.67 On this account, Euclidean space is both the object of geometrical knowledge and a condition of the possibility of empirical knowledge, because it is the order according to which monads (including human minds) perceive particulars as actually or possibly coexisting. Such an interpretation points to a decidedly Kantian reading of Leibniz’s philosophy. Just as for Kant, space for Leibniz is in effect the a priori form of outer sensibility.68 Because our knowledge of existing things is conditioned in this way, what we are given in perception are only phenomena: things as they are represented according to our mode of perception, from which we are entitled to draw no conclusions about things in themselves. Indeed, on Leibniz’s account, we have positive grounds for denying that things in themselves (monads) can be given in perception. On the basis of intellectual ideas, we can understand the essential properties of monads. However, given their simplicity, we cannot perceive monads as particulars. Whatever we perceive as existing must be a thing consisting of partes extra partes; that is, an extended thing spatially related to other extended things.69

66  “The soul knows things, because God has put into it a principle representative of things without” (LC IV 30). Cf. PNG 2, 4; and his letter to Wagner of June 4, 1710: “this correspondence of internal and external, or the representation of the external in the internal, of the composite in the simple, of a multitude in a unity, in fact constitutes perception” (GP VII 529). 67  “From this it also follows that every simple substance represents an aggregate of external things, and that in those external things, represented in diverse ways, there consists both the diversity and the harmony of souls. Each soul will represent proximately the phenomena of its own organic body, but remotely those of others which act on its own body” (C 14/MP 176). 68 See Critique, A 22–31/B 37–45. This conclusion is affirmed by De Risi 2007, 575, who embeds it in a broader theoretical framework than I do here. The conclusion must be qualified in several ways. I do not claim that Leibniz possesses Kant’s notions of the a priori or intuition; nor do I attribute to Leibniz Kant’s arguments for his doctrine of space. My suggestion is limited to the proposal that Leibniz defends, however dogmatically, an analogous conception of space in his treatment of Euclidean geometry and his account of monadic perception. A further point to be emphasized is that Leibniz has, as Kant recognizes, an underdeveloped theory of human cognition. I am focusing on a monad’s capacity for empirical perception, which involves the representation of particulars as coexisting and external (with respect to the perceiver and each other). However, Leibniz ascribes all cognitive acts to the same power of perception, or the representation of the many in a unity. This raises add­ ition­al problems that I set aside here. 69 “Spirits, souls, and in general, simple substances or monads cannot be apprehended by the senses or the imagination because they lack parts” (GP VII 501). Leibniz assumes that one can know

104  Leibniz on the Ideality of Space This is not how Kant depicts Leibniz’s philosophy in the Critique of Pure Reason. In the section of the Critique entitled “On the Amphiboly of Concepts of Reflection,” Kant criticizes Leibniz for failing to appreciate the special status of space as the a priori form of outer intuition. Broadly, his charge is that Leibniz lacks an adequate distinction between understanding and sensibility, and as a result construes all knowledge as a confused intellectual representation of things in themselves.70 On Kant’s reading, space for Leibniz is a system of relations that applies in the first place to things in themselves, which are confusedly represented as spatially related bodies.71 Kant finds in this the rationale for Leibniz’s claim that two existing things cannot differ merely in their spatial positions: whenever two things are taken as different because of their different locations, there must be some further difference in their qualities by which they can be distinguished. This is the import of Leibniz’s principle of the identity of indiscernibles, which Kant accepts as valid for noumena; that is, things as represented by the understanding alone. Kant’s objection is that the objects given in perception are not noumena, but phenomena, which are not subject to the principle of the identity of indiscernibles. In this case, non-­identity requires nothing more than discernability through spatial location: “since they are objects of sensibility, and the understanding with regard to them is not of pure but of empirical use, multi­pli­city and numerical difference are already given by space itself as the condition of outer appearances” (A 264/B 320). In his famous description of the search for two perfectly similar leaves, Leibniz indeed argues that if the leaves are two, they cannot differ merely in their spatial locations, but that there must be in addition some qualitative difference in terms of which they can be distinguished: There is no such thing as two individuals indiscernible from each other. An ingenious gentleman of my acquaintance, discoursing with me in the presence

one’s own mind directly through reflexive intellectual knowledge; however, he does not extend this to knowledge of other monads. 70  See A 270/B 327. Kant expresses this in the claim that Leibniz lacks a “transcendental topic”: “a doctrine that would thoroughly protect against false pretenses of the pure understanding and illusions arising therefrom by always distinguishing to which cognitive power the concepts properly belong” (A 268/B 324). For an analysis of Kant’s arguments in the “Amphiboly,” see Pereboom (1991). As I have suggested (n. 68 above), Leibniz can be faulted for failing to give an adequate explanation of the distinct powers of sensibility and understanding, which he combines under the heading of perception. Less clear is whether he confounds the sensible and the intelligible in precisely the ways Kant alleges. What follows is a sketch of how I think Leibniz’s position should be understood, which may serve as the basis for a fuller evaluation of the charge of amphiboly. 71  “Thus space and time became the intelligible form of this connection of the things (substances and their states) in themselves. The things, however, were intelligible substances (substantiae noumena). Nevertheless [Leibniz] wanted to make these concepts valid for appearances, since he conceded to sensibility no kind of intuition of its own, but rather sought everything in the understanding, even the empirical representation of objects, and left nothing for the senses but the contemptible occupation of confusing and upsetting the representations of the former” (A 275/B 331).

Donald Rutherford  105 of Her Electoral Highness, the Princess Sophia, in the garden of Herrenhausen, thought he could find two leaves perfectly alike. The Princess defied him to do it, and he ran all over the garden a long time to look for some; but it was to no purpose. Two drops of water or milk, viewed with a microscope, will appear distinguishable from each other. This is an argument against atoms; which are confuted, as well as the vacuum, by the principles of true metaphysics.  [LC IV 4]72

The “principles of true metaphysics” that rule against the existence of in­dis­cern­ ible objects are the “great principles of a sufficient reason [PSR], and of the identity of indiscernibles [PII]” (LC IV 5). The import of these principles cannot be that they contradict the very idea of congruent spatial figures (i.e. figures having the same form and magnitude) for such figures are a presupposition of Leibniz’s treatment of Euclidean geometry. With respect to the conditions of perception, numerically different congruent figures are possible. We are capable of apprehending them in imagination and of reasoning about their properties. Any restrictions imposed by the PSR and PII, therefore, must come through the implications they hold for things regarded not just as geometrical objects but as determinate things existing in nature. Leibniz ties questions of which things can or cannot exist in nature to considerations of what God would or would not create. He assumes that God does not act with an arbitrary or indifferent will, but in a way that is guided by perfect wisdom. On these grounds he infers that God would not create numerically non-­identical things that cannot be distinguished on the basis of their qualities. To suppose that God had acted in this way would be to suppose that he had acted without a sufficient reason, for in this case God would be blocked from having specific knowledge of the difference between this individual and that one. Such individuals would be indiscernible to God and could exist only through a brute act of will, which God would not exercise (LC IV 1–2; LC V 21–25). To this point it may seem that Leibniz must be guilty of the charge brought by Kant: he has imposed on the discernability of phenomena (leaves, drops of water) a condition that applies only to things as created by God (i.e. things in themselves). Even if we allow that God would not will the creation of two indiscernible substances, since he would have no way of distinguishing them in thought, there is no reason to suppose that the same holds for indiscernible phenomena. In attempting to block this possibility through the restriction he imposes on the

72 See also New Essays, II.xxvii.1–3 (RB 230–231); and for the contrasting view, Kant in the “Amphiboly”: “Thus, in the case of two drops of water one can completely abstract from all inner differences (of quality and quantity), and it is enough that they be intuited in different places at the same time in order for them to be held to be numerically different” (A 263–264/B 319–320).

106  Leibniz on the Ideality of Space matter of phenomenal objects, Leibniz has in effect treated those phenomena as noumena. In the remainder of this section I will try to show that Leibniz has a defense to make against this charge—one that reinforces the connection between his doctrine of the ideality of space and his account of empirical knowledge. As applied to the case of two phenomenal objects, two leaves or two drops of water, the PII has the consequence that there must be some discernible difference between them. Leibniz’s emphasis on the discernability of the difference indicates that he is thinking of a difference in the arrangement of the material parts of the objects and not some deeper, potentially undetectable difference in their nou­ menal grounds. His description of the two leaves in the garden of Herrenhausen implies that it must be possible to find a difference between them—one detectable by normal means, given sufficient powers of magnification. Obviously, Leibniz does not question that the two leaves are given as two on the basis of their different spatial positions. This difference, and the relation of situation between them, are given immediately in perception, as all properties of situs are. To this extent, Leibniz agrees with Kant on the role of space as an a priori condition of empirical knowledge. The PII does not negate this view of space, but imposes a further condition on the phenomena that exist in nature, namely, that no two phenomena—material things given in perception—are perfectly similar or have the same geometrical form.73 Recall that similar figures for Leibniz are objects that are qualitatively indiscernible and hence cannot be distinguished in thought; for instance, spheres of different sizes or cubes of the same size located at different places in space. When conceived individually on the basis of their form or quality, any two such figures are indiscernible, in effect the same figure. They can be distinguished as two only if they are “coperceived”; that is, perceived together in space, along with a common measure. Although perfectly similar figures are imaginable given the form of our perception, such figures cannot exist in nature. According to Leibniz, we conceive them by means of “incomplete and abstract notions, when things are considered only in a certain respect, but not in every way, as, for example, when we consider shapes alone and neglect the matter that has shape” (A VI 4, 1645/AG 32). For any two existing things, there must be some difference in their qualities on the basis of which they can be cognized as two. Leibniz’s confidence on this point relies on his understanding of what God would and would not do consistent with his wisdom. Specifically, he argues that the order God gives to nature precludes 73  “From these considerations it also follows that, in nature, there cannot be two individual things [res singulares] that differ in number alone . . . [F]or never do we find two eggs or two leaves or two blades of grass in a garden that are perfectly similar” (A VI 4, 1645/AG 32). Cf. A II 2, 288–289; and LC V 23: “I claimed that in sensible things one never finds two that are indiscernible, and that (for example) one will not find two leaves in a garden or two drops of water that are perfectly similar” (GP VII 394, emphasis added).

Donald Rutherford  107 the existence of qualitatively indiscernible objects. Although the supposition of two pieces of matter perfectly alike, “seems to be possible in abstract terms,” he writes to Clarke, “it is not consistent with the order of things, nor with the divine wisdom by which nothing is admitted without reason” (LC V 21; GP VII 394). Order is as much a feature of the phenomenal world as it is of the intelligible world of substances.74 As we have seen, order in general “is simply the distinct relation of several things” (C 535/MP 146). Rational or intelligible order—order expressive of divine wisdom—adds to this the requirement that the relations among things be thinkable in accordance with “fitting” principles, such as the principle of sufficient reason, the principle of continuity, and the principle of equipollence.75 For this requirement to be met, Leibniz holds, the ordered things themselves must be thinkable as different on the basis of their qualities. That is, it must be possible to conceive (and not merely perceive) how they differ from each other and how they are related.76 This is the basis of the order God gives to the world, which excludes the existence of perfectly similar, or qualitatively in­dis­cern­ ible, figures. Within this scheme, the order of the world is balanced by the variety of things God chooses to create. Because God aims to produce the world of greatest harmony—the greatest variety unified by the most fitting order—he neglects no opportunity for creating different things; that is, things thinkable as different on the basis of their form. At the level of phenomena, this is accomplished by creating matter that is infinitely divided through the motion of its parts.77 Leibniz takes it to follow from this that no material thing has a determinate shape representable by a finitely complex concept (A VI 4, 1622/MP 81). Furthermore, he 74  “Whenever we come to know something of the works of God, we find order in it” (to Hartsoeker, 1711; GP III 529). “The whole of nature is connected [liée] by the bonds of order” (to Jaquelot, 1704; GP VI 570/WF 199). 75  To Bayle, Leibniz writes that the soul “is expressive of phenomena according to metaphysico-­ mathematical laws of nature, that is, according to the order that best conforms to intelligence and reason” (GP III 72). Cf. GP II 168/L 515–516; GP III 383/WF 135; GP IV 568/WF 123; GP VI 602/AG 210–211. The order of situs shows that there are distinct relations that are not intelligible in Leibniz’s sense. We immediately perceive that similar things are located in different places, or are of different sizes, even though we do not comprehend their difference “absolutely,” through qualities of the things themselves. 76  Leibniz defines the relevant distinction in a study from the 1680s: “A quality is that according to which things can be distinguished conceptually. A quantity is that which is in things even when they have the same quality or are similar; it may be discerned by the experience of two subjects perceived at the same time” (A VI 4, 993). See also “Metaphysical Foundations of Mathematics”: “Quality . . . is that which can be known in things when they are observed singly, without requiring compresence. Such are the attributes that are explained by a definition or through various modifications which they involve” (GM VII 19/L 667). 77  “According to my demonstrations, every part of matter is actually subdivided into parts differently moved, and no one of them is perfectly like another” (LC V 22; GP VII 394/AG 333). See also GP II 306; GP VII 562–563. Leibniz offers several arguments on behalf of the infinite division of matter beyond that discussed here. They include its role in underwriting the doctrine of universe expression (GP III 383/WF 135; GP VI 618/AG 221; GP VII 561) and the inconsistency of the rival atomist doctrine with the principle of sufficient reason (GP III 507, 519, 527, 532–533).

108  Leibniz on the Ideality of Space infers that no two things have the same shape or form. Although two material things may appear similar, on closer examination they will be found to have different internal configurations and different external shapes, both reflecting the infinite division of matter. When we reason about material things using classical geometrical concepts, we routinely ignore the microscopic differences between them. Nevertheless, the relevant differences exist within the structure of matter and they must become evident at some level of analysis.78 The preceding argument relies on theological reasoning that Kant would eschew, but it shows how Leibniz can insist on the validity of the PII for phenomena, while denying the charge of amphiboly. In doing so, he need not be seen as trading on a conception of phenomena as confused representations of noumena. He need only claim that insofar as phenomena are taken as real or existing, they must be distinguishable conceptually, and not merely perceptually, on the basis of their form. This is to make conceptual discernability a necessary condition for real or existing phenomena, but it is not to reduce those phenomena to objects of pure understanding, since all phenomena are given in perception, in conformity with its a priori conditions. We can deepen this account by drawing out its connection to the no purely extrinsic denominations thesis introduced in §1 of this chapter. It follows from this thesis, Leibniz writes, that “there cannot be two atoms which are at the same time similar in shape and equal in magnitude to each other; for example, two equal cubes. Such notions are mathematical, that is, they are abstract and not real” (C 8/ MP 133). Mere differences in size or location are purely extrinsic denominations: objects can be imagined to change in these respects without the changes being reflected in the qualities of the relata. Consequently, such differences cannot be the basis on which existing things are distinguished from each other. Leibniz’s reasons for blocking such scenarios again can be traced to con­sid­er­ ations of divine wisdom—in particular, what is required by God’s representation of the order of nature. In real things, change (or difference) must be reflected in discernible differences in the qualities of objects. Leibniz accommodates this demand by maintaining that, within any world, objects must be related by more than spatial relations. Whenever objects change their spatial relations, they also change their causal relations and these changes are reflected in differences in their matter. Leibniz’s particular theory of causation, pre-­established harmony, does not alter this conclusion. Within any possible world, he writes, “all is connected.” 78  Leibniz offers one of his fullest statements on this topic in a letter to the Electress Sophie of October 31, 1705 (GP VII 563–564). See also A VI 4, 1648/AG 34; GP IV 568/WF 123. Note the consistency of the claims that no material thing has a shape that can be perfectly represented by a finitely complex concept and that for any two prima facie similar material things there is some degree of magnification at which the two could be distinguished by a finite mind. Pace Leibniz in the garden, there is no guarantee that we will succeed in identifying this difference, but there is some finite power of magnification that would allow us to recognize that the two have different shapes. On Leibniz’s treatment of the property of shape, see Levey (1998 and 2005); Garber (2009, 158–162).

Donald Rutherford  109 This means that everything is causally conditioned by the active and passive states of everything else.79 Consequently, although spatial relations by themselves fail the test of the no purely extrinsic denominations thesis, causal relations, which are tied to differences of situation, make up for the deficit. Whenever two things have a different situation with respect to each other, they also have different causal relations, and these relations involve each altering its active and passive states in lawful ways relative to those of the things to which it is spatiotemporally related. Once again, then, Leibniz has the means to avoid Kant’s charge that he confounds phenomena with noumena. If the no purely extrinsic denominations thesis entails only that any change in relational properties must be reflected in changes in the discernible qualities of the relata, this can be construed entirely at the level of phenomena: bodies change their physical properties as a function of their spatiotemporal locations. Since no two bodies have the same spatiotemporal location and each is responsive to the state of everything else in the universe, no two bodies will be indiscernible with respect to their physical properties (e.g. the arrangement and internal motion of their parts). This is all that Leibniz’s example of the two leaves requires. Yet we have seen that Leibniz also makes a stronger claim about the grounding of relational facts—one that goes beyond the requirement of discernible qualitative differences. At a fundamental level, he claims, all facts about the actual world are grounded in facts about substances and their modifications. More specifically, according to the thesis [SR], any difference in relational properties must supervene on the non-­relational properties of substances. He expresses the same demand in a stronger version of his no purely extrinsic denominations thesis, according to which any difference in extrinsic denominations must be grounded in differences in the intrinsic denominations of substances.80 Leibniz’s commitment to the grounding of the relational properties of phenomena in the modifications of monads suggests another way in which he might be found guilty of Kant’s charge of amphiboly. Since his metaphysics entails that all relational truths, including all truths about the physical world, supervene on truths about the modifications of substances, the ultimate differences by which phenomenal things are distinguished must be located in what Leibniz calls “the innermost [intimam] nature of body”; that is, substantial powers of acting and resisting (GM VI 235/AG 130). Without such a ground in substance, there would no basis for positing a real change in things or in the relations of things.81 79  Principia Logico-­Metaphysica (ca.1689) (A VI 4, 1646–1647/AG 33); Theodicy, §9; New Essays, II.xxv.5 (RB 227). 80  I draw here on what I referred to above (n. 15) as the extended version of [SR]. See n. 49 above for texts supporting the stronger reading of the no purely extrinsic denominations thesis. 81  See Part II of the Specimen dynamicum: “We must realize, above all, that force is something real in substances, even in created substances, while space, time, and motion are, to a certain extent, beings of reason, and are true or real, not per se, but only to the extent that they involve either the divine attributes . . . or the force in created substances . . . It follows that motion taken apart from force, that is,

110  Leibniz on the Ideality of Space It is important to recognize, however, that this is a metaphysical claim that has no direct connection to the application Leibniz makes of the PII to phenomenal objects. His vaunted claim about the possibility of always finding some difference between such objects does not presuppose epistemic access to the underlying difference makers, but is satisfied through differences discernible on the basis of geometrical properties alone (i.e. differences in the number, arrangement, and motion of parts). Such a position is consistent with the idea that the qualitative differences by which material things are discerned remain relational all the way down. In distinguishing one bit of matter from another on the basis of the configuration of its parts, we never reach bedrock in a representation of the modifications of monads. All the thesis [SR] entails is that wherever there is a difference characterizable in relational terms, there is some difference in the intrinsic modifications of monads on which the former supervenes. It provides no basis for thinking that we could in principle cognize that monadic ground.82 For Leibniz, no characterization of a body’s states in purely geometrical terms can capture the “innermost reality” of its substantial force. Yet he also denies that this reality is ever given to us distinctly in perception or that its apprehension is a precondition for empirical knowledge. In external phenomena, all we ever distinctly apprehend are changes in the size, shape, and motion of bodies, from which we infer changes in their dynamical properties. Consequently, the underlying differences in bodies by which they are distinguished must be tracked through differences in properties that are geometrically represented: differences in their size, shape, and motion. In short, Leibniz’s position is that while the principles of his metaphysics require that any difference between things ultimately be grounded in differences in the modifications of substance, our access to those differences is exclusively through differences in properties characterizable in geometrical terms.83 By drawing out the importance for Leibniz of space as a condition of empirical knowledge, we can be more precise about where the fissures between his and Kant’s philosophies lie. One of the most significant differences is the epistemic authority Leibniz assigns to reason. Leibniz claims, and Kant denies, that on the motion insofar as it is taken to contain only geometrical notions (size, shape, and their change), is really nothing but the change of situation, and furthermore, that as far as the phenomena are concerned, motion is a pure relation” (GM VI 247/AG 130). Cf. De ipsa natura, §13 (GP IV 514/AG 164–165). 82  As I interpret Leibniz, material things have an “inner reality” in monads but this reality cannot be cognized via the representation of them qua extended. Contrast Kant: “The inner determinations of a substantia phenomenon in space . . . are nothing but relations, and it is itself entirely a sum total of mere relations” (A 265/B 321). 83  On this, see Rutherford (2008b). When I say “characterizable in geometrical terms,” I do not mean to deny Leibniz’s claim against Descartes that a body must be understood to involve distinct properties of force over and above modes of extension. In physics, however, properties of force and matter are specified in terms of quantities of time and distance, which themselves are inherently relational.

Donald Rutherford  111 basis of reason alone we are able not simply to think, but to know, reality as it is in itself. If the preceding interpretation is correct, Leibniz does not believe that we know monads directly, as things in themselves. We are limited to formulating ideas of them, which allow us to understand their essential properties. When we represent the existence of things directly (insofar as this is possible in Leibniz’s philosophy), we necessarily do so under the forms of our perception as spatiotemporally related bodies. Although cursory, these remarks suggest how Kant in the “Amphiboly” misconstrues the thrust of Leibniz’s philosophy. Much as in Kant’s own system, for Leibniz, we only know external existing things as phenomena. These phenomena are enduring, three-­dimensional objects, among which we locate ourselves and on the basis of which we navigate the world and carry out the inquiries of science. Over and above this empirical reality, we can raise “transcendental” questions about the conditions of the possibility of the world of existing things. The questions the two philosophers pose in this regard are different: for Leibniz, they are questions about the being or essence of the things we take as existing; for Kant they are questions about the possibility of our experience of the things we take as existing. Nevertheless, the context that allows these questions to be posed is similar for the two philosophers, resting on their shared assumption of space as ideal and the a priori form of outer perception.

5

Leibniz and the Transcendental Deduction Alison Laywine, McGill University

The purpose of this chapter is to explore the significance of Kant’s engagement with Leibniz for the Transcendental Deduction in the Critique of Pure Reason. I  will argue in the first section of the chapter that the goal of the Deduction is cosmological: it will succeed in showing that the pure concepts of the understanding relate a priori to objects if it succeeds in showing that human understanding uses these concepts to construct a world out of the appearances that are sensibly given to us in space and time. Kant more usually prefers to speak of “nature” in this context. But I will argue that what he calls nature has a world-­ like character and that he himself was aware of this. I will try to confirm my suggestion not only by looking at the Deduction itself and parallel passages in the Prolegomena, but also at the presentation in the first Critique of the “System of Cosmological Ideas” (A 408–420/B 435–448).1 The hope of §1 of the chapter is to pull out what is distinctive in the “cosmology of experience” of the Transcendental Deduction and also to understand, at least in a programmatic way, how it guides the argument. That will set things up for the second section of the chapter. For it is natural to ask what led Kant to conceive of cosmology as he does in the Deduction. In §2, I will address this question from two angles. First, I will show that we can find an ancestor of this cosmology in Kant’s early metaphysics. Second, I will try to make it plausible that the ancestor cosmology was the result of efforts on Kant’s part to work out the cosmological implications of a challenge about perception raised by Leibniz for Newton in his correspondence with Clarke. If indeed the ancestor cosmology helped shape Kant’s conception of the cosmology of experience at work in the Deduction, as I will argue, then it makes sense to interpret the Deduction, at least with respect to the characterization of its goal, as a peculiar and remarkable elaboration of the cosmological lessons Kant learned from Leibniz as early as 1755. That will set things up for the third and

1  Throughout this chapter I refer to the Academy Edition of Kant’s writings. I more usually cite volume number, page number, and line number. The exception is for the Critique of Pure Reason. Here I refer to the page numbers of the A and B versions, but as printed in the Academy Edition (volumes IV and III respectively). All translations are my own.

Alison Laywine, Leibniz and the Transcendental Deduction In: Leibniz and Kant. Edited by: Brandon C. Look, Oxford University Press. © Oxford University Press 2021. DOI: 10.1093/oso/9780199606368.003.0005

Alison Laywine  113 final section of the chapter. It is intrinsically interesting to note Leibniz’s significance for the Transcendental Deduction, but it is just as interesting to note the distinctive ways in which the Deduction adapts and repackages some of what Leibniz had to offer—in ways that Leibniz himself had little or nothing to do with. When we discover that one imaginative philosopher has taken up an idea from another, we almost always discover that the idea is developed in ways and to ends that its author would never have imagined: that too is intrinsically interesting. In §3, I will call attention to just two of Kant’s adaptations of the lessons he learned from Leibniz: the first is his conception of human understanding; the second is the way he conceives the relationship between the idea that the understanding legislates laws to nature and his account in the Deduction of the way knowledge relates to objects. I will not develop all of my suggestions in full detail, because I have worked them out in other articles (and in my 2020 book), and I will give references along the way. Here I will step back from the details, developed in these other places, and present for discussion a program for reading the Deduction.

1  Cosmology and the Goal of the Transcendental Deduction Kant states the goal of the Transcendental Deduction in a couple of different ways. On the one hand, it is supposed to explain how the pure concepts of the understanding relate to objects a priori (B 117). On the other hand, it is supposed to show that these concepts make experience possible (B 126). The one statement is meant to complement the other. For Kant thinks that if his Deduction can show that the pure concepts make experience possible, it will have succeeded in showing that they relate to objects a priori—not to just any objects, of course, but only to objects of experience. By the end of the Deduction, however, Kant claims to have shown how nature is possible—a claim that Kant himself finds startling (at least, that is what he says), because it goes hand in hand with a further claim that the laws of nature are somehow legislated by the understanding: no nature without these laws; no laws without the understanding; hence the possibility of nature rests on the legislation of the understanding. Both versions of the Deduction end on this note. In the 1781 A-­edition, Kant remarks: “As overblown and ridiculous as it even thus sounds to say: the understanding is itself the source of the laws of nature and hence of the formal unity of nature, such a claim is nevertheless just right and suited to the object, namely experience” (A 128). He puts it this way in the 1787 B-­edition: “There is nothing more disconcerting than the way the laws of appearances in nature must agree a priori with the understanding and its a priori form, i.e., with its faculty for combining the manifold as such . . . But as mere representations, [appearances] stand under no law of connection at all save that which the connecting faculty [sc. the understanding—AL] prescribes” (B 164). There may be many differences between

114  Leibniz and the Transcendental Deduction the two versions of the Deduction. But it is striking that, even after having revised the first version so thoroughly, Kant keeps the “overblown” and “ridiculous” sounding claim just mentioned and reserves for it the same dramatic flourish and the same place of honor—namely the end—in the second version. This is disconcerting for a reason Kant seems to have overlooked: it suggests that there is a fundamental mismatch—in both versions—between what he thinks he has achieved by the end of the Deduction and the two goals he announced at the beginning. This is how it will seem, at least, depending on what one takes to be the meaning of “nature” and “experience.”2 If we assume that experience is just whatever we know through the senses and that nature is whatever we mean when we speak of “nature red in tooth and claw,” then obviously the result of the Deduction and its stated goals will be at odds with each other. But that’s not the case, as it turns out, because of Kant’s idiosyncratic use of the two terms. As implied at least in the passage I just quoted from A 128, they mean the same thing, though not what they ordinarily mean even in philosophical discussion. If, then, Kant can show that the pure concepts of the understanding make experience possible (in the relevant sense), he will have shown that the understanding prescribes laws to nature (in the relevant sense). Having shown all this, he will have shown that the pure concepts of the understanding relate to objects a priori. But to get a better fix on what Kant thinks 2  In what follows, I will try to characterize the goal of the Deduction in the light of what Kant means by “nature” in §36 of the Prolegomena. That means I will disregard certain familiar issues associated with this problem, notably the relevance of the remarks Kant makes in the preface to the A-­Edition at A xvi–xvii about the “two sides” of the Deduction: the one having to do with the “objects of the understanding” and whose task is thus to “prove the objective validity of its concepts a priori” (which he says is essential to his task); the other having to do with “the understanding itself according to its possibility and the cognitive powers upon which it rests” (which he says is not essential to his task). I am also not going to consider the difference between the so-­called “objective deduction” and the “subjective deduction.” For a sensitive discussion of these important issues, one may read with profit Wolfgang Carl’s commentary on the A-­Deduction. Let me just add that I am deliberately following his example in that commentary in the following respect. Carl points out that, at the end of the day, in order to make sense of what Kant takes to be the goal of the Deduction, one cannot simply consider his programmatic remarks about the Deduction. One must also examine what he takes to be its result (Carl 1992, 54ff.). I am struck by the fact that Kant can be taken to say explicitly at the end of both versions of the Deduction that the result is to have shown that the categories make nature possible in the formal sense (as characterized in Prolegomena §36). That is my point of departure. I take Kant to be making a very robust claim: Kant believes he has shown that all appearances are necessarily subject to the categories and, to that extent, are connected under laws in such a way as to form a single, universal, law-­governed experience (so he is not taking experience for granted). That is what qualifies the categories to be the “principles of nature in the formal sense,” as Kant puts it in Prolegomena §36. For a quite different way of reading the Deduction in light of Prolegomena §36 (and the Second Analogy), see Thöle (1991). I will be taking the talk about nature and its laws at the end of the Deduction to stress the dynamic unity of a single, universal experience. Thöle seems to take it to stress the necessity of connections among appearances, as when we take them to stand in the relation of cause and effect. It is natural for Thöle to place the stress here, because his interpretative strategy is to read the Deduction in light of the Second Analogy. But perhaps because that is his strategy, he seems tone deaf to an idea that is surely in play already in the Deduction and that will be important for me here: in the dynamic unity that is experience, causal sequences are related to one another in a common world—the causal sequences are not disconnected from one another.

Alison Laywine  115 he is doing in the Transcendental Deduction, we need to understand more precisely what he means by “nature.” §36 of the Prolegomena can shed some light on this. Kant distinguishes two senses of the word “nature” there. He says first of all that “nature in the material sense” is to be understood as “the totality of appearances” (4.318.8). The use of the word “totality” [Inbegriff] in this passage is connected with talk about principles. It is in the nature of a principle to apply to the whole class of things to which it applies—that is, the “totality” of them—because otherwise none of the relevant things would be possible as the kind of thing that they are all supposed to be. In Prolegomena §36, Kant is asking, first of all, what principle or principles make appearances possible as such. The principles in question cannot fail to apply to the sum total of all possible appearances, because otherwise nothing at all could appear to us. We know from the Transcendental Aesthetic of the Critique of Pure Reason that the sought-­after principles are space and time taken to be the pure forms of inner and outer sense respectively. Space and time are thus the principles of nature in the “material sense.” The word “material” comes in to play for no better reason than that they—space and time—make appearances possible as the cognitive stuff or matter out of which nature as such comes to be. A complete account of nature in the material sense and its underlying principles is given in the Transcendental Aesthetic. So nature in this sense is not the immediate concern of the Transcendental Deduction. The Transcendental Deduction is concerned with nature in the other sense of the word. Kant says in Prolegomena §36 that nature in the formal sense is to be understood as “the totality of rules under which all appearances must stand if they are to be thought of as connected in one experience” (4.318.16–18). An explicit association of nature and experience is found here. Experience depends on all those rules3 by virtue of which we can think of appearances as universally connected with one another. The principle that governs all those rules and makes them possible as such is what makes experience possible as such. But it is also the principle of nature in the formal sense. Since experience and nature in the formal 3  Here “rule” (Regel) seems to be synonymous with “law” (Gesetz). But at A 126, there seems to be a slight contrast. Laws seem to be a special class of rules, namely just those rules that are “objective.” Some rules are just regularities that the understanding has detected: Kant has just finished saying that the understanding is “always busy scouting out appearances in the hope of detecting any rule among them.” Empirically detected rules may well include things that are nothing more than regularities, or what the understanding takes to be regularities. But what the understanding takes to be a regularity need not be a law, since it need not be objective—if, say, the understanding is either mistaken in construing an empirically detected pattern as a regularity, or if it cannot find a way to rigorously express what the regularity is, or to confirm that it does indeed obtain. It is possible that something like this distinction between rule and law is in the back of Kant’s mind in Prolegomena §36. But if it is, the rules that are at issue in that passage will turn out not only to be laws—rules that are objective— but also those higher laws of which Kant speaks briefly at A 126: the laws that he there unambiguously represents the understanding as sometimes establishing empirically (with perhaps the aid of mathematics, as in the case of Newton’s law of gravitation) are “particular determinations” of these higher laws.

116  Leibniz and the Transcendental Deduction sense ultimately have the same principle, they must be the same thing, or very close to the same thing. If a qualification is in order, perhaps it is this: while experience is the system of universally interconnected appearances that arises by the application of the relevant rules, nature in the formal sense is just the totality of all those rules whose application gives rise to experience. Perhaps another way to put it is this: experience is nature taken as a composite of matter or form. Either way, though, the principle of the one is the principle of the other. If, then, the Transcendental Deduction must explain how experience is possible as such, it must explain how nature in the formal sense is possible. But this thought needs a bit more unpacking. We learn from Prolegomena §36 that the principle of nature in the “formal sense” makes possible certain rules to which all possible appearances are subject. The concern for principles dictates that the sum total of these rules stands under the relevant principle, since no such rule would be possible without it. But, as we have seen, the class of rules under consideration here is restricted to just those that make it possible to “think appearances as connected in one experience.” It is conceivable, given the high level of generality in Prolegomena §36 and the Deduction itself, that not every rule that applies to appearances will satisfy this condition. For all we know at this stage of the game, it might be a rule that appearances as such be countable.4 But it is not obvious that this rule makes possible the thinking of appearances as connected in one experience. For perhaps it will turn out that our countability rule cannot apply to appearances unless the rules in fact at issue in Prolegomena §36 are already in play. It seems plausible that the only appearances I could possibly count are those that I can already conceive as being connected with me and with one another in one and the same experience. I might volunteer to count the swans in Abbotsbury, England for the next swan census in the summer of 2017. Swans are possible appearances; they are surely subject to our countability rule. But if the Abbotsbury swanherd instructs me and

4  As we make our way through the Transcendental Analytic, we learn which rules constitute nature in the formal sense. As I argue in the conclusion of my book (Laywine 2020), they are, in the first instance, the “Analogies of Experience.” The rule I am imagining here is not one of the three “Analogies.” The purpose of the exercise here is to press Kant to clarify what he takes to be the upshot of Prolegomena §36 with regard to “nature” in the formal sense. One simple way to do that is to imagine he is confronted by a reader completely innocent of what is to come. Such a reader might wonder, reasonably enough, which rules are at issue when Kant talks about this “totality of rules” that govern appearances. It would seem natural to me, if I try to remember what it was like to read the Prolegomena for the first time, to wonder whether one such rule is that appearances be countable: surely experience would not be possible if appearances did not at a minimum conform to this rule? When we imagine Kant confronted by such a reader and that reader’s question, how do we imagine he would respond? Might we learn something from what we imagine his response would be? I think we do learn something. I imagine that he would concede, as he must, that, yes of course, appearances must be countable and measurable in all sorts of ways. But I also imagine that he would deny that counting, measuring, and so forth would be possible unless appearances have first been connected with one another and with the people doing the counting and measuring. If that is right, then the exercise already points to cosmological overtones in Prolegomena §36 and related passages. See below.

Alison Laywine  117 the other volunteers to include in the count all those swans that are connected to one another in an alternate, parallel experience cut off from the one in which we are rounding up swans and counting, we will have to tell him politely that the parallel swans are “as good as nothing for us.” The rules at issue in Prolegomena §36, whatever they may be, are such as to connect with one another all possible appearances in our thought of them. Whatever makes these rules possible as such is the principle of nature in the “formal sense.” The word “formal” comes into play here, because the relevant principle ultimately imposes a certain form or structure on the matter of nature. The “formal unity of nature” mentioned in passing in the passage from the Deduction at A 128 I quoted in the first paragraph of this section is a term Kant uses to designate that structure. It is well chosen, though highly abstract: by virtue of the relevant rules and hence by virtue of the principle that makes these rules possible as such, appearances are somehow unified in our thought of them. Their unity depends in no way on the specific content of our thought: it will not matter whether we are thinking of them as swans or geese or other aquatic birds. That is what makes the unity of appearances “formal.” But it is also universal in a special sense. There is some kind of universal, formal unity in all our thinking, just because all our thinking uses concepts. Concepts have formal unity, and they are universal in scope, by virtue of their “logical form.” As Kant puts it at A 107, “All knowledge requires a concept, however incomplete and obscure it may be. Still this concept is always something universal—as far as its form is concerned—that serves as a rule.”5 It will serve as a rule insofar as we can apply it in our thinking to a whole class of things. That is what gives our thinking unity and universality in one fell swoop: the concept Swan applies not to just one bird of a certain description, nor only to a pair, but to all of them; in so doing, it gives us a way of taking together— uniting in thought—a whole bunch of disparate things, namely all those things that we take to be swans. We can do this, because our concept gives us the criteria or marks of whatever we may regard in thought as belonging to the relevant class. Kant’s own example is the concept Body, which, by virtue of its logical form, is applicable to all those things that exhibit Extension, Shape, and Impenetrability. But it is not obvious, on reflection, that the universal, formal unity characteristic of our thinking by virtue of our use of concepts is universal in the same sense as the formal unity of nature. By means of concepts, we can think all sorts of things: for example, that swans are aquatic birds and that some swans are mute, while others are very vocal. But, again, it seems plausible that we could not begin to

5  Here again we find talk of rules. Concepts are rules for thinking. The rules at issue before were for connecting appearances. Both kinds of rule deserve to be called rules, because they exhibit universality of some kind. The important thing to note here is that they exhibit different kinds of universality and that is what will make them different kinds of rules.

118  Leibniz and the Transcendental Deduction apply concepts like Body or Swan, much less have an opportunity to form them in the first place, unless we could already think of appearances as universally connected in one experience. It does not even make sense to speak of parallel swans in a parallel experience cut off from the one that cognitively engages us. If they are indeed “as good as nothing for us,” we cannot really apply the concept Swan to them, or any other concept for that matter. If we are going to apply the concept Swan to a whole class of things, we need in principle to have available to our thinking, from among all possible appearances, what we take to be the plausible swan candidates. This requires at a minimum that all plausible swan candidates can appear to us under the conditions of space and time taken to be the principles of nature in the material sense. But that will not be enough all by itself, because our plausible swan candidates and all other sensibly given things might appear in pockets of space and slices of time so totally unconnected from one another that they would not be much better than nothing for us.6 At best, they would be a chaotic jumble of things that we could not make any sense of. It must be possible at least to think of all possible appearances—including the plausible swan candidates and ourselves—as parts of a single, connected experience; that is, as universally related to one another not just in space and not just in time, but in space and time mixed together in such a way that we can think of them and ourselves as potentially having a shared history directed towards a shared future, no matter how far and wide we and they may be scattered throughout the universe. We might characterize the formal unity of our concepts as logical. But the formal unity of nature or experience might well be characterized as “cosmic”—a thought that I will return to later. If these considerations faithfully capture the thought underlying Prolegomena §36, they imply that the formal unity imparted to our thinking by the logical form of our concepts itself depends on the universal, formal unity of nature and hence that the latter is indeed more fundamental than the former. This implies that all our thinking, however mundane or recondite, depends on the principle of nature in the formal sense, which in turn suggests a very general game plan for the Transcendental Deduction. The task of the Deduction is to show that the categories of the understanding relate a priori to objects. The way to do that is to show that they make possible all our use of concepts and hence all our thinking as such. For given that we think about objects, the categories cannot fail to relate to objects a priori, if indeed there 6  We might think that this is not a worry, so long as appearances are given to us in space (as well as time) and given that there is supposed to be a single, universal space, as Kant claims in the Transcendental Aesthetic. But that Kant does not himself take this to be enough to put the worry to rest is clear from this remark: “For appearances could indeed be so constituted that the understanding did not find them to conform to the conditions of its unity and everything lay in such confusion that, e.g., in the succession of appearances nothing presented itself that gave to hand a rule of synthesis and thus corresponded to the concept of cause and effect so that this concept were thus completely empty, vain and without meaning” (A 90/B 123).

Alison Laywine  119 can be no thinking without them. As it turns out, we find such an argument running from §17 to §20 of the B-­Deduction. §17 tries to show that all our thinking and hence all use of our understanding depends on subjecting the manifold of an intuition as such to the original unity of apperception. But then §19 tries to show that subjecting the manifold of an intuition as such to the original unity of apperception is the logical function of judgment.7 But, then, if it is true that the categories are nothing other than logical functions of judgment, it will turn out that all our thinking and all use of our understanding depends on the categories, with the caveat, of course, that the categories can apply only to objects of experience (§22). As Kant says in §21, this is a “start” on the Transcendental Deduction (B 144). But it is indeed only a start, because the argument that the categories are logical functions of judgment does not show all by itself that they are the principles of nature in the formal sense. Without a further argument to that effect, we will be missing something crucial: the guarantee that we can in principle think of all possible appearances as connected with one another under laws in one, single experience.8 The rest of the Deduction in the B-­edition tries to secure that guarantee—as too the argument in the A-­edition. That is why the Deduction terminates, in both editions, with the “overblown and ridiculous” claim that the understanding “prescribes laws to nature.” If these laws depend on the categories, so does nature in the formal sense. If our ability to think of all possible appearances as connected in a single experience is necessary for our ability to think at all, then the categories make possible all thinking and all use of our understanding. Hence, they relate a priori

7  Of course, the game plan imagined here is tailor fitted to the B-­Deduction. It is the merit of Béatrice Longuenesse to have pointed out the central role of judgment in the B-­Deduction. See her (1998). I take the interest of the B-­Deduction to be something like this. I take Kant to be saying, first of all, that all thinking depends on the power to judge (itself made possible by the categories) and that the power to judge would itself be impossible unless we can orient ourselves in the world, that is build a world out of all possible appearances. If Kant can show that we build the world out of appearances by means of the categories, he will have shown that they make possible the power of judging. In other words, the theory of judgment does not, by itself, get Kant a cosmology of experience; it will be the cosmology of experience that lies at the basis of the theory of judgment. This move may well raise further complications for him. But if so, they fall out of the scope of this chapter. Since it is true, as Longuenesse points out, that judgment plays no role in the A-­Deduction, one may wonder how the argument was supposed to go in the earlier version of the text. Here I will just hand wave: the cosmology of experience is derived somehow from pure apperception in the so-­called “argument from above” in Part Three of the A-­Deduction; it is used as the point of departure of a regressive argument that is supposed to take us back to pure apperception in the “argument from below.” Let me just add that I develop these difficult ideas at greater length and with greater care in chs. 3 and 4 of my book (Laywine 2020). 8 Is this not the best way to understand why the B-­Deduction has two steps in light of the interpretative strategy raised by Henrich 1973? My strategy for understanding the proof-­structure of the B-­Deduction implies that we must take very seriously the rhetoric at the end of §26 about the understanding prescribing laws to “nature” (B 163–165)—indeed that we stop thinking of it as overblown rhetoric and treat it as what it purports to be: a statement of the conclusion of the Transcendental Deduction.

120  Leibniz and the Transcendental Deduction to objects, though only to things as they appear and not to things as they are in themselves. These considerations lead me to the observation that will serve as the point of departure for this paper and that I alluded to earlier: nature as such (as a composite of form and matter) or experience, as understood by Kant, has a world-­ like character. My observation will initially seem off the mark, because one might worry that the talk of worlds and cosmology will immediately raise the specter of Antinomies. But it is striking that Kant himself explicitly calls attention to the world-­like character of nature. At A 418–419/B 447 of the Critique of Pure Reason, he says: We have two expressions, world and nature, that sometimes run into each other [in einander laufen]. The first means the mathematical whole of all appearances and the totality of their synthesis, in large as well as small, that is both in the advance of the same through composition and through division. But the very same world is called nature insofar as it is regarded as a dynamic whole and one looks not to the aggregation in space or time, so as to bring it about as a magnitude, but rather to the unity in the existence of appearances.

Kant is saying that “world” and “nature” are used synonymously, but that, for the sake of precision, a certain distinction should be made, namely the one now familiar to us from Prolegomena §36 between nature in the formal sense and nature in the material sense. We should reserve “world,” properly speaking, for the latter and “nature” for the former. But the interest of the passage is precisely that it acknowledges a certain natural inclination Kant expects us to have: when we reflect on “unity in the existence of appearances,” we will tend to speak indiscriminately of “world” and “nature.” For nature in the formal sense does indeed have a world-­like character. To be sure, context matters: and, indeed, in such a way, one might object, as to undermine the claim I just made. Our passage from A 418–419/B 447 appears in the presentation of the “system of cosmological ideas.” Kant is moved to reflect on the meaning of “world” there, precisely because he is about to take us through the Antinomies—the inconsistent set of claims and counter-­claims generated by any attempt of pure reason to deduce what follows from its concept of a world, considered from different angles, when it makes its characteristic demand for the unconditioned. But while all of this is true, it does nothing in fact to undermine my fundamental claim. Nor does it cast into doubt the significance of this passage, as I understand it, for Prolegomena §36. On the contrary, the one passage is a nice complement to the other. First of all, Kant says right at the beginning of the “System of Cosmological Ideas,” a few pages before our passage from A 418–419/B 447, that “the understanding alone is the source from which pure and transcendental concepts

Alison Laywine  121 can proceed.” He goes on directly to say that “reason, in fact, produces no concept at all; at best it frees the concept of the understanding from the inevitable restrictions of a possible experience and thus tries to extend it beyond the limits of the empirical, while nevertheless staying in connection with it [sc. the empirical—AL]” (A 408–409/B 435). Pure reason has no special concept of a world that it cooks up for itself. Its point of departure must be some concept formed by the understanding that properly applies only within, or with respect to, the bounds of a possible experience. Reason takes over this concept and tries to extend its use, but it does not try to sever the connection of this concept with the empirical altogether. All of the special cosmological ideas reviewed in the Antinomy arise in the same way. They begin as pure concepts of the understanding, which is why the table of cosmological ideas formally mirrors the table of categories (A 415/B 443). Reason happily grants that these concepts apply to things as they appear, but then it tries to apply them to other things as well. In order to see the implication for our passage from A 418–419/B 447 and its relation to Prolegomena §36, we will have to flesh out this thought in a bit more detail. Each of the four cosmological ideas is the result of reason’s demand: whenever the conditioned is given, so too the whole sum of conditions and therewith the absolutely unconditioned. Moreover, each idea is the result of making this demand in a special way: reason fastens on to just those categories in conformity with which its demand for the unconditioned can play out as a linear sequence of ascending conditions, the one subordinate to the next, as it rises through the conditions in the direction of the absolutely unconditioned. Thus, for example, the category of cause and effect yields the idea of the “absolute completeness of the coming to be of an appearance as such.” For any given appearance, taken as effect, reason can ask, what is its condition? When the cause of that appearance is given, by way of an answer, it will ask for the condition of that condition, and so on. This will be the pattern for all the cosmological ideas. But in addition to the four cosmological ideas, pure reason surely must have the idea of a world as such, even though this idea does not formally correspond to any given category on the relevant table. For one thing, without such an idea, it would be unable to understand what makes cosmological ideas cosmological. Reason itself will say, presumably without any help from Kant, that the cosmological ideas concern a world as such, and not the rational soul or the divine being, because it claims to have knowledge of all these special things and hence also knowledge of the differences among them. But it could not claim to have any such knowledge without having the relevant concepts. Since reason cannot produce any pure or transcendental concepts on its own, it must borrow its concept of a world from the understanding. I take Kant to imply in our passage from A 418–419/B 447, given the surrounding context of the passage, that the concept of a world, precisely insofar as it ori­gin­ ates in the understanding, is distinct from the four cosmological ideas. It does not

122  Leibniz and the Transcendental Deduction follow their pattern of a linear sequence of ascending conditions; rather it is the idea of some kind of cosmological whole, considered either mathematically as the sum total of all that is or all that can be, or dynamically as unity in the existence of the cosmic whole.9 Thus it should be understood as reason’s appropriation of the understanding’s concept of nature, which Kant will no doubt claim is best and most precisely characterized in Prolegomena §36: either as nature in the material sense, if we attend to all possible appearances as such, or as nature in the formal sense, if we attend to the laws that produce “unity in the existence of appearances.” Before pure reason has submitted to Kant’s critique, it will never lose sight of the empirical significance of the concept of nature, in either sense, even without precisely understanding what its principles are and what they imply for its own peculiar interests as pure reason (that is even without having read the Prolegomena). However, it will deny systematically that this is its only significance inasmuch as it regards the world or nature and any sensibly given part of it as requiring ultimate explanation—explanation that will take it upward towards the unconditioned. A second, but complementary consideration comes into play here. Unless reason appropriates for its own use the understanding’s concept of nature and— more specifically—the concept of nature in the formal sense, it will never be able to orient itself in the different linear sequences of conditions it may perhaps come to fix on. If it seeks the whole causal sequence of conditions of a certain appearance A, the whole sequence of conditions of some other appearance B, and likewise that for yet another appearance C, it will have to make intelligible to itself somehow that these different linear sequences are inscribed in a single, united, dynamic whole and so are somehow related to one another: as reaching back jointly to a shared history in a shared cosmic arena. If pure reason were to fix on each linear sequence, or even just on the general idea of a linear sequence as such, with no care for cosmic context, we could doubt whether its ideas are properly cosmological: a linear sequence of conditions is just a string of beaded conditions and not a world, unless reason has a way in principle of conceiving of it as part of a universal, dynamically united whole. But, in view of our passage from A 418–419/B 447, it is plain that reason never loses sight of the cosmology in its linear sequences. That is surely because it has elaborated for itself the idea of a world as such out of the concept of nature in the formal sense produced or presupposed by the understanding. Thus, the presentation of the cosmological ideas in the introduction to the Antinomy section of the first Critique is indeed a complement to Prolegomena §36. Moreover, it gives us the best evidence that experience or nature as a composite of matter and form is taken by Kant himself to be world-­like. For it indicates that he takes pure reason itself to recognize its world-­like character and 9  This thought is borne out by the last paragraph of the section as a whole at A 419–420/B 447–448.

Alison Laywine  123 to be tempted by this recognition to form ideas that are genuinely cosmological and that therefore generate the Antinomies. This has an important implication for the Transcendental Deduction. At the end of the day, the Transcendental Deduction will achieve its goal just in case it can show how nature in the formal sense is possible (namely by means of the categories). Hence, its goal is fundamentally cosmological. It culminates with what we might call a “cosmology of experience” (that is an explanation of how experience comes to have its world-­like character).10 This does not mean that the Transcendental Deduction is threatened by Antinomies,11 because there is no essential connection between Antinomies and cosmology as such: they are a threat only to the characteristic treatment of cosmology by pure reason before it is summoned by Kant to account for itself. Prolegomena §36 indicates that the conception of world, nature, or experience that guides the Transcendental Deduction towards its goal is a very distinctive one. We may, and should, ask what philosophical motivations led Kant to articulate this conception. If we neglect to do this, we will have overlooked something important; namely, what it is that the Transcendental Deduction is ultimately striving towards. The way to answer this question, at least to start out with, is to go back briefly to Kant’s early metaphysics, because it includes a cosmology that can easily be seen to advance the distinctive conception of a world driving the Transcendental Deduction. Precisely because it motivates and defends this conception without the complications of the later argument, it can help us see more clearly than the Deduction itself where the Deduction is headed. To be sure, this metaphysics is vulnerable to the Antinomies from Kant’s later point of view. But all that means is that Kant had some more thinking to do. In particular, he needed to find a way to immunize his conception of a world so that it would not be vulnerable to the Antinomies. By the 1780s, Kant thought he had succeeded in doing that: it consists in reminding pure reason that it cannot ask the understanding to apply its categories to anything other than things as they appear to us. To be sure, the Transcendental Deduction is concerned ultimately with a question that did not occur to Kant until years after his early metaphysics. It is supposed to put to rest the worry that the pure concepts of the understanding are “empty, null and devoid of all meaning” (A 91/B 123). But this just means that the conception of a world available from the early metaphysics had to be adapted once Kant recognized that this is indeed a worry that he needed to settle: hence the world at issue in the Deduction will be one formed out of things as they appear, not things as they are. But it remains a world nevertheless.

10  This claim forms the program of my book (Laywine 2020). 11  Any more than the “Analogies of Experience” are threatened by Antinomies. Note the explicit talk of worlds in the Third Analogy in the footnote to A 218/B 265. Of course, there are clear parallels with Prolegomena §36 at A 216/B 263.

124  Leibniz and the Transcendental Deduction

2  Kant’s Early Metaphysics: Leibniz and the Right and Wrong Ways of Doing Cosmology As early as the mid-­1750s, Kant was asking, in effect, the question relevant for our purposes. Given that God has created all the finite things he will ever create, what will it take to form a world out of them? One might think that the question is pointless, because as long as God has finished creating, we already have everything we need: the world is just the sum total of all God’s creatures. This will be the response of anybody who thinks it is enough to characterize a world as such as a whole that is not itself a part. But Kant was never satisfied with this characterization for the following reason. There can be no world unless the things that God created somehow externally relate to one another. If they do not, we might as well say that there are as many worlds as there are creatures, which is to say that there is no world at all. In order to get a world, God not only has to create all the things that will have a place and part in it, he will also have to provide the condition under which external relations among these things will be possible. Hence, he will have to subject his creatures to laws of some sort. These will be the laws of “community” by virtue of which God’s creatures relate to one another in a single, unified whole: the world. The train of reflection I just sketched can be found in Section Three of Kant’s Nova dilucidatio of 1755. It is the ancestor of the distinction in Prolegomena §36 between nature in the material sense and nature in the formal sense. We might say on behalf of the early Kant, but using the language of his later self, that the sum total of all the finite things God created is a world, but only in the material sense,12 while the laws he laid down to allow for the external relations among these things, as in a single, unified whole, is a world in the formal sense: a world as such is a composite of matter and form. We might also use the talk of principles characteristic of Prolegomena §36 in order to capture Kant’s thinking in the mid-­1750s. As principle of a world in the material sense, the early Kant will invoke not just God’s willingness to bring creatures into existence at all, but whatever makes a created substance possible as such.13 It is not important for us to linger on what this might be taken to mean. 12  There is a slight modal difference here. When Kant speaks of nature in the material sense in Prolegomena §36 and asks after its principle, he means the sum total of all possible appearances. I am suggesting, for now, that we can imagine the early Kant speaking of a world in the material sense and that what he means (at least for now) is the sum total of all creatures actually created by God. There are no doubt issues that the early Kant would have had to discuss concerning creatures insofar as they are possible and insofar as possible creatures are made actual when God created them. But that discussion, such as it played out or might have played out in Kant’s early metaphysics, is not relevant for my purposes. What matters here is just that there seems to be a rough distinction between a world in the material sense and a world in the formal sense in Kant’s early metaphysics that clearly parallels his distinction in the senses of nature in Prolegomena §36. It is that distinction in Kant’s early metaphysics and the details of its account of a world in the formal sense that are the focus of what follows. 13  It would be in the spelling out of this talk of principles that the modal difference mentioned in the previous footnote between Kant’s early treatment of a world in the material sense and his discussion of nature in the material sense in Prolegomena §36 would be overcome. But, again, this is not the place to do that. That discussion would also have to involve a discussion about the relation between

Alison Laywine  125 The more important thing for our purposes is what the early Kant invokes as the principle of the form of the world: in Section Three of the Nova dilucidatio, this is the thing Kant calls the “schema of the divine intellect”; that is, the supremely wise providential plan for the world that God uses to guide his choice of the laws of community that will govern his creatures and make a world out of them. The surest sign that it is natural and right to characterize Kant’s early cosmology in the language of Prolegomena §36 is found in the inaugural dissertation of 1770: on the one hand, Section Four of the dissertation reprises the cosmology of the Nova dilucidatio without revision; but on the other hand, it explicitly uses the talk of a “principle of the form of a world”—both in the title and throughout the body of the text. The new twist in the dissertation is a fundamental distinction between the sensible world and the intelligible world. It is natural to think that, while the talk of a “principle of the form of the intelligible world” in Section Four of the dissertation harkens back to the Nova dilucidatio, its talk of a “principle of the form of the sensible world” in Section Three anticipates, and is completed by, the considerations that will ultimately be elaborated in Prolegomena §36. Be that as it may, our concern for now is the early cosmology of the Nova dilucidatio and its distinctive way of characterizing a world and its relation to what we might call the “principle of the form” of a world as such. Perhaps the thing to say about it is this: had the later Kant been willing to reflect on its significance, he would no doubt have taken this cosmology to be a representative (and philosophically the most astute14) sample of pure reason’s appropriation of the concept of nature elaborated by the understanding and properly used only with respect to the conditions of possibility of experience. If the early Kant went wrong, by his later lights, it would be in failing to see that pure reason requires discipline and that the concept of a world as such must be understood in the terms spelled out in Prolegomena §36. This is where Leibniz comes in. For, as I shall argue, it was he who gave Kant an important tip on how to spell out his early cosmology in the distinctive way we have just seen. The text of reference is his correspondence with Samuel Clarke. I will return to Kant’s early cosmology at the end of this section and the significance of its Leibnizean overtones for the Transcendental Deduction at the end of this section and in the section that follows. Leibniz initiated his correspondence with Clarke in 1715 out of concern for the state of natural religion in England, weakened—he said—by the rise of materialism: “some take souls to be corporeal; others, God himself ” (352).15 The the early metaphysics (more specifically Kant’s account of real possibility and its relation to the highest reality contained in God as necessary existence in the Beweisgrund of 1763 and related passages) and the Transcendental Aesthetic in the Critique of Pure Reason. 14  I am emboldened to say this on Kant’s behalf in light of what I take to be the way he took up a challenge Leibniz presented to Newton in his correspondence with Clarke. See below. 15  All references to the Leibniz–Clarke correspondence are by page number to vol. VII of GP. I have cited Clarke’s English as it appears in GP, but I have myself translated into English Leibniz’s French. This has the unfortunate effect of making the English version of Leibniz’s interventions look more modern than Clarke’s. So it goes.

126  Leibniz and the Transcendental Deduction chief culprit, in the first case, was Locke; in the second case, it was Newton. The evidence against Newton is his claim in the appendix to the Opticks that, as Leibniz puts it, “space is the organ used by God for sensing things” (352). It is easy enough to see what motivates Leibniz’s accusation: any God that used organs of sense, as we do, would have a material body, as we do. But Clarke objects to Leibniz’s construal of Newton’s words: “Sir Isaac Newton doth not say, that Space is the Organ which God makes use of to perceive things by; nor that he has need of any Medium at all, whereby to perceive things” (353). Clarke goes on to present what he takes to be the right characterization of Newton’s view. He thereby sets himself up for the Leibnizian rejoinder that will ultimately prove to be of interest to us. For Clarke says, on Newton’s behalf, “on the contrary, . . . he [sc. God—AL], being Omnipresent, perceives all Things by his immediate Presence to them, in all Space wherever they are, without the Intervention or Assistance of any Organ or Medium whatsoever” (353). God perceives all things other than himself not by any bodily organ like an eye or an ear, but by his immediate presence to them. Space is where these things happen to be; God, by his existence, is present to them. This is all that Newton is supposed to have meant when he characterized space as God’s “sensorium.” Clarke says, moreover, that the word “sensorium” is more usually used with respect to perception in created animals. But he denies that Newton’s extended use of the word commits him to anything like the materialism suspected by Leibniz. For, even as applied to living creatures, the word “sensorium” does not mean physical organ of sensation, as Leibniz assumes in his first letter (352) and then explicitly claims in his second (356). Rather, it means “Place of Sensation” (360). It is, as Clarke explains, that place where, for example, “the Mind of Man, by its immediate Presence to the Pictures or Images of Things, form’d in the Brain by means of the Organs of Sensation, sees those Pictures as if they were the things themselves” (353). Unlike God, the human mind is not immediately present to any of the things themselves, but rather only to images of them produced in the brain, presumably as the result of physical contact between these things—assuming they are material—and the associated human body. Just as God perceives all things other than himself by his immediate presence to them, so the human mind perceives those things in physical contact with the associated human body by its immediate presence to the images of these things left on the brain. We imagine, of course, that the immediate objects of our perception are external bodies, rather than brain states. In this we are mistaken. But it is no mistake to think that we are perceiving something. We do so by the mind’s presence to this something. To be present to anything is to be somewhere. Clarke says that this place of the mind’s presence to the physical images sketched out on the brain is just what Newton calls the mind’s “sensorium.” Since space is that place where God is present to all things and, to that extent, perceives them, we might just as well follow Newton’s example and call it God’s “sensorium.” But

Alison Laywine  127 Clarke is confident that this use of the word is free from material connotations, since it in no way imputes to God the use of bodily organs. Leibniz’s reply to all this is—I think—both interesting in its own right and the point of reference for the aspects of Kant’s early cosmology that are relevant for us here. The most important passage for our purposes is found in Leibniz’s Second Letter to Clarke: It is assumed [sc. by Clarke and Newton—AL] that the presence of the soul is enough for it, the soul, to apprehend [pour qu’elle s’apperçoive] what takes place in the brain. But this is precisely what Father Malebranche16 and the whole Cartesian school deny, and rightly so. Something quite else is required besides mere presence for one thing to represent [représente] what takes place in another. For that to be explicable, some kind of communication is needed, some sort of influence either of the things amongst themselves [entre elles] or of a common cause. According to Mr. Newton, space is intimately present to the body that it contains and that is commensurate with it [et qui est commensuré avec luy]. Does it follow, as a result, that space apprehends what takes place in the body and that it remembers what takes place in the body even after the body has been removed therefrom? Besides the fact that the soul is indivisible, such that we could only imagine its immediate presence in the body as concentrated in one point, how then could it apprehend what takes place beyond this point?  [356–357]

Leibniz’s challenge to Clarke and Newton is spot on. The soul’s immediate presence to the body, or even just some part of the body, cannot explain perception all by itself. If it could, Newton and his followers would have to embrace the queer conclusion that space itself has perception or—anyway— perception of the bodies that occupy it, since it is immediately present to them. The human mind has perception not by immediate presence alone, whether to external bodies or their corresponding brain states, but rather by virtue of some kind of “communication.” Leibniz himself is committed to pre-­ established harmony. He will naturally deny—at the end of the day—that this communication takes the form of a real influence of mind and body on each other. But he will also deny that it takes the form of a real influence of their common cause, if that is taken to imply anything like the occasionalism of Malebranche. Creatures really do act, according to Leibniz; hence, the communication requisite for perception in them must involve the influence of the mind on itself and the influence of the body on itself, but coordinated by God’s infinite wisdom and assisted by his divine concursus.

16  See e.g. Book Three, Part Two, ch. 1, §1 of Malebranche’s Recherche de la vérité in Œuvres I, ed. Geneviève Rodis-­Lewis and Germain Malbreil in Bibliothèque de la Pléiade (Paris: Gallimard, 1979).

128  Leibniz and the Transcendental Deduction Of course, the same challenge can be raised for Newton’s account (if that is what it really is) of perception in God. God’s immediate presence to all things other than himself cannot possibly explain his perception of them. Even in this case, some kind of communication is required. Leibniz puts it as follows: “The reason that God apprehends everything [s’apperçoit de tout] is not his mere presence, but rather his operation: it is because he conserves things by an action that continually produces whatever share of goodness and perfection they may have” (357). God perceives his creatures by his continual influence on them. He would cease to perceive them at all, if he ceased to exercise this influence. Without this influence, they would, of course, cease to be. But even if we could somehow imagine per impossible that they persisted in their existence, with the same share of goodness and perfection bestowed on them in the original act of creation, God would cease to perceive them, though it might be imagined—with enough stretching—that he would remain immediately present to them. Leibniz’s challenge to Clarke and Newton starts out as a challenge to defend natural religion from a certain kind of materialism. It quickly converts itself into a challenge to explain perception in human (and other animal) souls.17 This challenge in turn has a natural extension into general cosmology—an extension that Leibniz himself does not explore in the Clarke correspondence. But it is easy enough to articulate this extension for him. As we do so, we will find ourselves approaching the cosmological position adopted by the early Kant. There seems to have been a natural temptation in the early modern period to think that whatever resources will be needed to explain the mind–body union might well be enlisted to account for that vaster union of all created substances that we call the world. If, say, we were going to pump for pre-­established harmony in the one case, we would do so in the other.18 But if we think, as Newton and Clarke do, that the mind–body union is just the effect of the mind’s immediate presence to the body, then it will be natural for us, at least in the historical context, to think that the world as a whole is just the global effect of any given creature’s presence to others. Perhaps we will stop short of the idea that any given creature

17  I should point out that Clarke does reply—or tries to reply—to this challenge in section four of his second letter. He says only this: “It was never supposed, that the Presence of the Soul was sufficient, but only that it is necessary in order to Perception. Without being present to the images of the Things perceived, it could not possibly perceive them: But being present is not sufficient, without being also a Living Substance” (360). This holds both for God and living, sentient creatures. This is clearly the start of a reply to Leibniz, but only a start. The thought that life is required for perception in addition to presence never gets developed beyond its enunciation. The reason for this, I think, is that, in his next letter, Clarke becomes much more interested in engaging with Leibniz about the nature of space and time as such and what it means to say that God always has a sufficient reason for what he does. 18  In the case of pre-­established harmony, our hand would be forced anyway, because we would want to commit ourselves to a cosmology that as little violates the law of conservation of living forces as our account of the mind–body union. In the case of occasionalism, our hand would be forced too, because we would want to commit ourselves to a cosmology that treats creatures as true causes no more than our account of the mind–body union does.

Alison Laywine  129 is immediately present to all others. But we will have no reason to resist the idea that it is immediately present to some of them, and, by the mediation of others, indirectly present to the rest of them. A commitment to anything like that which Clarke represents as the Newtonian account of the mind–body union will make such an account of what it is to be a world natural, if not ineluctable. But, then, any such account will leave itself open to Leibniz’s challenge to the related account of the mind–body union. A union of any sort will require some kind of communication and hence some kind of influence, either that of things themselves or that of their common cause. Now I realize that I have stated the general cosmological version of Leibniz’s challenge so quickly and at such a high level of abstraction that it may now be harder to feel its punch than it was in the case of the mind–body union. After all, it might be objected that, as long as creatures have come into being, they will, by their mere presence, collectively add up to a world. To get a world as such may well require a union of some kind. But perhaps this union is much less complicated than that of mind and body; certainly, it seems to involve much less intimacy: whatever relationship there may be between the earth and the moon, it is not such that the moon seems to perceive what’s going on in the earth or that the earth seems to carry out the moon’s volitions. Before returning to Kant, there is one all too quick thing I can do, in the face of this objection, to restore the punch to Leibniz’s challenge extended now to general cosmology. However lacking in intimacy a union may be, it must nevertheless unite its members to count as a union at all. This means that its members must be appropriately adapted or suited so as to seem, at least outwardly, to take note of each other somehow. If God had decided to create a less perfect world than this, he could have brought into being all sorts of creatures none of which were in any way adapted in its powers and activities to any of the others. Such creatures would be absolutely cut off from one another, as we might imagine physical bodies that could pass through each other without resistance and thus occupy the same space at the same time. We would be willing to characterize such creatures as present, and even perhaps as present to one another. But the fact of their simply “being there” will only get us a list of creatures. A list of creatures will not give us a union that will convert them into a world. We cannot specify the nature of this union without positively committing ourselves to some conception of God’s wisdom and hence to that system of causes most compatible with it. The challenge so far is thus neutral with respect to metaphysics. That makes it indeterminate. We know where Leibniz himself will ultimately come down. He will say that the union by virtue of which all the monads collectively constitute this world, and hence the most perfect of all possible worlds, is universal pre-­established harmony. But the challenge, as I have stated it here, like the one Leibniz himself puts to Clarke and Newton, is polemical in nature (as challenges always are) and addressed to any philosophical adversary—innocent of, or allergic to, Leibniz’s metaphysics—who

130  Leibniz and the Transcendental Deduction imagines that the sum total of all creatures is a world just by including all of them. Can such an adversary account for the being of the world as a world, on his or her own terms, without reducing it to a mere catalog of creatures? That is the challenge; it is a tough one. It seems to me, then, that, if we felt the punch of the original challenge, we should still feel the punch of its extension to the world as a whole. But even if we do not feel its punch, I believe that Kant did. He almost certainly read the Leibniz–Clarke correspondence.19 Though I cannot prove it conclusively, I think it is likely that he worked out for himself the cosmological implications of Leibniz’s challenge to Clarke and Newton to explain perception, along the lines that I just spelled out, or something like them. The evidence for my claim is first of all that these implications are philosophically natural and second of all that they yield something close to the distinction in Prolegomena §36 between “nature in the material sense” and “nature in the formal sense”—and something even closer still to the ancestor of this distinction in Kant’s early cosmology. For Leibniz’s challenge implies, in the general cosmological case, that the catalog conception of the world as just the sum total of all creatures is what Kant might have called a world “in the material sense,” whereas the union that unites all creatures is what gets us a genuine world-­whole and its principle is therefore what Kant might have called the “principle of the world in the formal sense.” The resonances here with the passage about world and nature from A 418–419/B 447 in the presentation of the system of cosmological ideas in the Critique of Pure Reason that I quoted and discussed in the previous section are obvious. I do not mean to suggest that Kant was ever merely a cheerleader on Leibniz’s side of this issue. In the 1750s, Kant was as determined to reject pre-­established harmony as a system of general cosmology as he was to reject it as a system of

19  I do not know of a passage in Kant’s writing where Kant explicitly mentions Leibniz, Clarke, and their correspondence. But given his long-­standing interest in all of the issues discussed in this correspondence and the fact that it seems to have been published numerous times almost immediately after Leibniz’s death in languages we know Kant could read, and was widely discussed in philosophical circles, I do not think that I risk very much in making this assertion. Indeed, Lambert explicitly refers to the correspondence in his letter to Kant of 1770; the nature of his reference is such as to indicate that he is taking for granted that Kant knew the correspondence too. I would like to think that Kant had read the correspondence already by the 1750s. Perhaps that thought is a bit riskier. But I doubt it. The Leibniz–Clarke correspondence had already established itself by then as a classic. For its early publication history, see Ed Dellian’s introduction to his German translation of Clarke’s 1717 edition, A Collection of Papers which passed between the late Learned Mr. Leibnitz and Dr. Clarke in the Years 1715 and 1716, relating to the Principles of Natural Philosophy and Religion. This was published by Felix Meiner, Hamburg in the Philosophische Bibliothek series in 1990. It may be too that Leibniz expresses similar ideas with similar implications elsewhere in a works accessible to Kant (perhaps in Les Nouveaux essais). But I have not yet managed to track down anything quite like it, with quite the same punch. The key thing, though, for my purposes is that Kant will be seen to develop in his early cosmology an idea that is Leibnizian and that he even knew to be Leibnizean. For that claim to be at all plausible, the Leibnizean idea at issue had to be in a work by Leibniz already published in Kant’s early days and widely circulated enough for it to be believable that Kant actually read it in his early days. The Leibniz–Clarke correspondence fits the bill very nicely.

Alison Laywine  131 psychology. He wanted to make the case for real interaction as the alternative in both cases. He hoped to show, in other words, that no change can take place in any created thing unless some other created thing externally acts on it. This is, in effect, the statement of his Principle of Succession in Section Three of the Nova dilucidatio (1.410.17–20). But I believe that the project of the Nova dilucidatio shows an unmistakable sensitivity to what I have been calling Leibniz’s challenge and its cosmological implications. Perhaps the thing to say is that the early Kant takes himself to be the philosophical competitor of Leibniz who can successfully meet this challenge, namely by conceding the fundamental insight that animates it, without buying into pre-­ established harmony. This emerges from his elaboration of the other principle in Section Three of the Nova dilucidatio, the so-­called Principle of Coexistence. The Principle of Coexistence states that “finite substances do not relate to one another by any relation, through their existence alone, nor are they contained by any community, except insofar as they are sustained in accordance with their mutual relations by the common cause of their existence, namely the divine intellect” (1.412.36–413.2). I take this principle to state very succinctly, and in general terms, the fundamental point I represented Leibniz as making against Clarke: the union we find among creatures cannot be accounted for from their simply “being there”—or, as Leibniz himself puts it, using Clarke’s language— from their simply being “present to one another”; that union requires some kind of influence, ultimately that of their common cause. Kant’s proof of the principle simply restates the point a bit more fully.20 It rests on the uncontroversial idea that created substances exist separately from one another; the one is not the cause of existence to the other. So, unless the author of their otherwise separate existence provided for some kind of union that would embrace them all, they would remain cut off from one another, just insofar as they exist. There would be a collection of creatures, a world in the material sense, but not a genuine world-­ whole in the formal sense. The Principle of Coexistence is important in Kant’s early cosmology precisely because of the way it can be used to radicalize this distinction. It can do this by teaching us to recognize how many possibilities for world-­making were open to God even after he created everything he wished to create. For the Principle of Coexistence teaches us that, having brought into existence everything he wished to create, God was then free to form the world as we know it or to abstain from world-­making altogether. He could have elected not to lay down any laws of community. But he was also free to form more than one world or to form a world from some of his creatures while excluding others from it. The world, as we know it, arose because God subjected his creatures to certain laws of 20  Hence it is not very good as a proof, but it is very useful for understanding what Kant took the thrust of his principle to be.

132  Leibniz and the Transcendental Deduction community by virtue of which every creature externally relates to every other. If God had decided to subject some of his creatures to laws of community other than those that govern this world, there would be two different worlds such that nothing in the one world could externally relate to anything in this world. It would be like in the movies: two parallel universes. If God decided to suspend the law of community governing the second world, the creatures of that world would cease to belong to any world at all. They could not become a part of this world, unless God decided to subject them to the laws that govern us. The Principle of Coexistence thus radicalizes the cosmological implications of Leibniz’s challenge to Clarke and Newton—indeed, so much so that it calls into question one crucial assumption that Leibniz had been willing to share with his two English interlocutors (at least for the sake of argument); namely, that to be is to be somewhere and hence that things can exist in space without being embraced in any kind of union or community. The Principle of Coexistence says, to the contrary, that there must be some kind of divinely legislated community for creatures even for them to have spatial relations with one another. Without such a community, creatures might well be, and yet be nowhere at all (1.414.10–20). It is clear that Kant never gave up the lessons from the Principle of Coexistence as I just spelled them out. He uses them—unsuccessfully, I think—in the inaugural dissertation of 1770.21 But, more to the point, he uses them again in the Transcendental Deduction. We find a very concise, but explicit statement of these lessons in a passage from the A-­Deduction, which reads as follows: “There is only one experience in which all perceptions are represented as in thoroughgoing and law-­bound connection . . . If we speak of experience in the plural, then what we really mean is just so many perceptions, insofar as such perceptions belong to one and the same universal experience” (A 110). It is obvious that this talk of experience in the singular evokes the notion of nature in the formal sense from Prolegomena §36. But it should be just as obvious now that it harkens back to Kant’s idea in the 1750s that a world is a unified whole in which all the parts mutually relate to one another under divine laws. The parts here are no longer created substances that owe their existence to God. They are rather parts that we represent by perceptions; that is, they are things insofar as they appear to us and

21  For the argument in defense of this claim, see Laywine (2003). The short of the argument is this. When Kant, in effect, asks after the principle of the form of the sensible world in Section Three of the dissertation, he answers with the resources of his new account of space and time as pure intuitions. We readers recognize this answer as an anticipation of the Transcendental Aesthetic of the Critique of Pure Reason. But we also know from Prolegomena §36 that the Transcendental Aesthetic provides us with the principles of nature in the material sense, but not with the principle of nature in the formal sense. Thus, from the point of view of Kant’s later development, we can see that Kant tries to answer the question of the principle of the form of the sensible world, in Section Three of the dissertation, by adducing the principles of the matter of the sensible world. This will not do, because it ultimately conflates a contribution of sensibility with that of the understanding. I suggest that Kant came to see this by the mid-­1770s and that the Duisburg Nachlaß is his first try at correcting the problem.

Alison Laywine  133 insofar as we are conscious of them as appearing to us. The whole of which they are part is not, as such, the world that God providentially manages by laws of universal community; rather, they are as much of that world as we can rationally reconstruct by interpreting our perceptions as connected by laws according to certain expectations of the understanding. It is precisely here that the cosmological lessons of the 1750s become relevant. For we know from the Transcendental Aesthetic that all appearances, and hence all our perceptions as the special class of appearances actually accompanied by consciousness, are possible by virtue of space and time taken to be the pure forms of sensibility. As such, they are spatially and temporally determined. But this leaves undetermined how—or indeed whether—the understanding can interpret them as being connected under laws. To put it in the language of Prolegomena §36, the pure forms of sensibility guarantee us nature in the material sense; that is, the sum total of possible appearances, but they do nothing to guarantee us nature in the formal sense. As Kant himself puts it in the prefatory remarks to both A and B versions of the Deduction: “appearances might perhaps be so constituted that the understanding could find in them nothing conformable to the conditions of its unity” (A 90/B 123). By implication, the lessons that held true in the 1750s for God’s own world-­ making have been applied to our efforts to make models of the world using the resources available to our finite cognitive powers. For we cannot expect to produce experience in the singular—nature in the formal sense—just on the strength of having a whole bunch of appearances. A whole bunch of appearances is no more sufficient for having experience in the singular than the existence of creatures was sufficient for having a world. If it is true that the Transcendental Deduction cannot achieve its aim without elaborating in principle a cosmology of experience, and if it is also true that this cosmology of experience follows the pattern of thinking already in evidence in the Nova dilucidatio, then it must be acknowledged that Leibniz helped Kant figure out how to characterize the goal of the Transcendental Deduction. His challenge to Clarke and Newton to give a philosophically adequate account of perception has a natural extension into general cosmology. We find this extension clearly elaborated in Section Three of the Nova dilucidatio. The cosmological lessons of the Nova dilucidatio are just as clearly on display (with the appropriate adaptations) in the Transcendental Deduction and Prolegomena §36. Kant reprised in the Transcendental Deduction the cosmological lessons he learned from Leibniz for one simple reason: he had to show how thinking is possible for human intellects, and he understood more clearly than anybody else before him that human intellects cannot think at all, unless they think in world-­like terms. We cannot entertain any thoughts about Napoleon unless we can situate him in our thinking in the parts and times of the world in which we believe (perhaps mistakenly) he lived and died and unless we can situate those parts and times of the world with respect to our own. If we cannot intellectually frame a setting for

134  Leibniz and the Transcendental Deduction him in the world (itself taken as an object of our thought and knowledge), Napoleon will be “as good as nothing for us.”

3  Some of the Ways Kant Adapted and Repackaged the Lessons He Learned from Leibniz Now, by way of conclusion, I would like to explore some of the ways Kant applied and elaborated the cosmological lessons he learned from Leibniz in the mid-­1750s for the purposes of the Transcendental Deduction. The interest in doing this is to see how these lessons may have guided Kant’s understanding of what he was doing in the Deduction and what sorts of special adaptations he had to make in order to take them on board. I do not have room enough to say everything I could on this subject, so I will present and discuss briefly just two ways the Transcendental Deduction develops the cosmological lessons of the 1750s. The first one is very simple, but important in its way; the second is more difficult. The Transcendental Deduction is—as we all know—an argument about the understanding as a faculty of thought. But because of the cosmological overtones, and because of the nature of the cosmology whose overtones we hear in it, it naturally invites us to compare and contrast our understanding with that of God. The issue is no longer what God can do to make a world out of things that he himself has created; it is rather what the understanding can do to construe things that have been given to our sensibility as belonging to experience. So, we will insist on a fundamental, and familiar, contrast. God’s intellect is not only world-­ designing; it is creative in the literal sense of the word; that is, it brings the parts of the world into existence out of nothing. A finite understanding such as our own is not creative: it is discursive (that is. it knows things only insofar as it forms judgments about them and then only insofar as they have been given to the faculty of sensibility to which it is associated). Thus, it is natural to say that the Transcendental Deduction is a reflection on human cognitive finitude. Our finitude is distinctive in that we have two, radically different faculties of knowledge: the understanding and sensibility—the one a faculty of thinking, the other a faculty of being sensibly affected by objects. That we can have knowledge of objects is possible only to the extent that these two faculties cooperate. Kant himself draws a related contrast early in the B-­Deduction, namely that between a finite discursive understanding like our own and an intuitive understanding capable of generating its own manifold through itself or through the mere thought or consciousness of itself (B 135, 138–139). Now there is nothing wrong with saying these things. They happen to be true. But they are not everything there is to say about the matter. For it is just as important to notice how much God and the human understanding have in common. Even for Kant, there is a sense in which we were made in God’s image.

Alison Laywine  135 To start with, the scopos of our understanding22 and that of God are the same in an important respect; namely, they are directed at the same thing: either the same thing under two different aspects, or two different things the one of which may be regarded as an analogue of the other. The scopos of God’s creative activity is the world as a whole. But so too for human understanding. Experience in the singular is not some one mouse in the pantry, or even a pair of mice; it is as much of the whole—of which mouse and pantry are part—as we can reconstruct in a finite time interval, however short/however long. I do not explicitly spell out to myself where we are in the galaxy when I confront the evidence of a mouse infestation. But it is Kant’s view that I should, in principle, be able to do so, in order to assess the evidence competently and then figure out what to do about its implications. To be sure, God grasps all of the whole all at once, whereas we have to strive for it, and it always gets away from us. Nevertheless, the whole is as much our scopos as it is his. But there is something else we have in common: we both relate to the scopos as a legislator. God converts the sum total of all the things he created into a world by subjecting them to universal laws of community. The cosmological lessons of the 1750s are built into our cognitive duality in such a way that only our understanding has a chance of converting appearances into experience in the singular. The principles of our sensibility—space and time as pure intuitions—give us all possible appearances or nature in the material sense, as we learn from Prolegomena §36. But they can do nothing all by themselves to give us nature in the formal sense. This falls to the understanding, but only on the condition that it can somehow legislate to nature; that is, provide for the universal laws under which all possible appearances may be regarded as belonging to the same unified whole. Just as God had to legislate universal laws of community to form a world out of his creatures, so the understanding has to legislate universal laws to appearances in order to give us nature in the formal sense. I cannot believe that Kant was unaware of this parallel. It can be documented that the cosmological lessons of the 1750s continued to be present in his mind into the mid-­1770s. I think it is in this connection that we should understand his self-­conscious claim at the end of both versions of the Deduction that the understanding legislates laws to nature. It is clear—perhaps more so in the B-­version—that he expects his reader to find this claim astonishing. But, so far as it goes, it is not at all astonishing, at least not in light of the cosmological lessons of the 1750s and Kant’s willingness to treat God’s relation to the world as a paradigm for the relation of our understanding to nature (in the relevant sense). It is perhaps a sign of how significant Kant took the paradigm to be for evaluating our power of judgment that it is in evidence not only in both versions of the Transcendental Deduction, but also in the 22  By “scopos,” I mean target: that towards which our understanding or God’s understanding is directed.

136  Leibniz and the Transcendental Deduction Groundwork for the Metaphysics of Morals where the test of our maxims for contradiction in the will depends on seeing whether, in acting on the relevant maxim, we can imagine ourselves as legislating laws to a world that is practically viable as a world. That was the first of the two points I promised. But it brings me to the second, which in turn brings together two different, but related themes in the Transcendental Deduction. The first concerns what Kant may mean by the legislating of laws to nature by the understanding and the significance he assigns it; the second concerns his understanding of the way knowledge relates to an object. It is obvious that these two themes go together. For the Transcendental Deduction claims to show how all knowledge and thinking more generally relates to objects, namely by means of the categories; and, it claims to have succeeded in showing this, if it has established that the understanding is the source of laws for nature. I spelled out in broad brush strokes what this means for the argument strategy of the Deduction in the first section of this chapter. But what I would like to explore now is the way the two considerations just mentioned fit together and are jointly animated by the cosmology at work in the Deduction. I will start with the second consideration. Kant characterizes knowledge in §17 of the B-­ Deduction as “given representations” (that is sensible representations) that relate to an object. He seems to have a very peculiar, but very interesting understanding of how such a relationship is possible. As I understand §17 of the B-­Deduction and what I take to be related passages in the A-­Deduction (A 104ff.), Kant wants first of all to exclude what he takes to be a problematic understanding of this relationship. This understanding begins with the recognition that a claim to have knowledge of something is a claim to truth. If this claim to truth is justified, we are under some kind of constraint imposed upon us by the object itself: we cannot say anything we please about it; we can say only what it allows us to say. This means that the relationship between the object and our knowledge is necessary. None of this is unreasonable so far. But the problematic account of this relationship can think of only one way to account for the relevant necessity. It says that the object is something separate or distinct from us and hence our knowledge. If it were not at some kind of remove from us, and fully independent of us as knowers, the constraint we are under would be lifted; and, we might well try to pass off our subjective expectations for the truth itself. I think Kant recognizes that this view has a certain plausibility. But it also seems clear that he thinks it is aporetic. To see why, just try to imagine how this view imagines that we justify our claims to have knowledge of an object. It will have to say that it is a matter of confrontation between the object and what we say about it to see whether there is some kind of match, as though we were trying to coordinate—say—towels and color samples. But this is not what we typically do. Suppose I claim to have knowledge of conic sections and you challenge me on the truth of my claim. In order to satisfy you

Alison Laywine  137 that I am right, I will summon the evidence. But this typically involves testing my current claim not against something that stands outside and independent of our knowledge—“a something = X,” as Kant would say. On the contrary, it will involve testing it against other items of knowledge, namely the ones that I can convince you are both relevant and already secure from any doubts you may have. This means that the relationship between knowledge and object is indeed necessary: I am constrained in what I say about the object by what I already know about it. But at the same time the relationship is much more intimate than could be imagined on the problematic view under discussion. The object—as an object of knowledge—is itself an inevitable byproduct of knowledge. It is what we build up out of what we already know. The best illustrations of the idea come from classical geometry. How do I establish that I really do know something about conic sections? It will not be by bouncing a hyperbola off your forehead. It will be by mustering the elements of Apollonius’s treatise on the subject—the relevant items of prior knowledge—and successfully using them first of all to construct a hyperbola and then to characterize it in terms of its “principal property,” as Apollonius does in Book One, Proposition 12 of the Conica. I cannot successfully construct this object (and then proceed with its characterization) unless I really know it: I put my knowledge on display in the act of constructing and characterizing it. Having successfully constructed it, there can be no further doubt about my claim to know it—or to know things about it, provided that I can show you how to use its construction and certain auxiliary constructions to establish the conditions a locus of points must satisfy in order to count as a hyperbola. There is no reason to think that Leibniz played a direct role in Kant’s account of knowledge and the relation of knowledge to its object. I think that Kant was driven to this account, in part by recognizing that ideas about empirical knowledge he had tried to develop in the so-­called Duisburg Nachlaß in the mid-­1770s may have presupposed the aporetic conception of knowledge and its object,23 and in part by reflecting in a creative way on the idea Lambert had been 23  I elaborate these ideas more fully in the third section of ch. 2 of my book (Laywine 2020). The thought, in brief, is this. One of the distinctive things about the Duisburg Nachlaß is that, on the one hand, it assumes the account of space and time as pure intuitions from the inaugural dissertation of 1770 and, on the other hand, it also seems to assume that the intellect has a direct intellectual grasp of its own nature as a thinking substance. It apparently uses this intellectual grasp to make sense of what is given to it under the conditions of space and time. But the making sense involved here seems to be a kind of projection of what it knows about itself on to what is sensibly given. To be just a bit more specific (and the Duisburg Nachlaß does not get much more specific than this), the intellect construes elements of what is sensibly given to it as standing in certain relations with one another (e.g. the relation of cause and effect and that of part to composite whole) on the model of, or following the pattern, of the relations in which it knows by this immediate intellectual grasp that it stands (as a thinking substance) to its own ideas. But it seems reasonable to think that, however the details of this idea might be developed, the idea itself naturally goes hand in hand with the aporetic conception of know­ ledge and its object. For if we ask Kant of the mid-­1770s how he can assure us that this projection of our intellect on to what is sensibly given is really knowledge of something out there in the world

138  Leibniz and the Transcendental Deduction keen to share with him in the late 1760s of using the notion of a Euclidean postulate to reform metaphysics.24 But all of that is another story. What I propose to do now is to make plausible the idea that the new and improved non-­aporetic conception of knowledge and its object, as I just sketched it, gave Kant a way to adapt the cosmological lessons of the 1750s for the purposes of the Transcendental Deduction. To be more specific, I think it gave him a way of conceiving what the understanding does when it legislates the laws of nature. I cannot make a full case for this idea here and now. But here are some considerations that I hope will make it plausible. First of all, it should be uncontroversial that Kant’s new and improved understanding of the relation between knowledge and its object only makes sense for finite knowers like us and that it stands in sharp contrast with what we may well want to say about God as an infinite knower. Nevertheless, as knowers, he and we have this much in common: the relation between knowledge and the object of knowledge is—for us, as for him—a necessary one and an intimate one such that the object is produced by knowledge. But the difference, of course, is that, for God, the act of knowing and the act of creating are one and the same: he creates the object of his knowledge exactly as he knows it, and he knows it exactly as he creates it. If we now take seriously the idea that the world is the scopos of God’s intellect, it follows by Leibniz’s insight that God creates and knows the world by virtue of knowing all his creatures as subject to the relevant laws of community. For the world comes to be a world, rather than the ultimate heap of all finite things, only if such laws are imposed on it. Thus, it is by legislating the laws of community that God knows the world as a world in the strict sense. But if we now take seriously the idea that the scopos of God’s intellect and that of our own are one the same, we have to say that the object of our knowledge will ideally be the whole world (that is nature as a composite of matter and form or universal experience in the singular). But given that we are finite knowers who cannot generate the object of knowledge by creation, we must construct it in something like the way we construct the objects of geometry: we will build it up out of what is sensibly given to us under the formal conditions of our sensibility as a kind of model in thought by confronting our expectations of it against what we take to be the relevant items of knowledge already secured from doubt. But our construction will depend on imagining this model as subject to the laws of community, just as geometrical constructions and the knowledge they embody ultimately depend on Euclid’s constructive postulates: we do in thought what God does in deed. If we rather than just a subjective “projection” of its subjective expectations of what is out there, I am not sure he could say much beyond appealing to the thought he will clearly reject in §17 of the B-­Deduction and the parallel passages in the A-­Deduction: “Oh well,” we might imagine him to say, “to do that, we have to confront what we claim to know with whatever is out there, and whatever is out there will serve as a corrective.” That is the thought in brief. 24  For the argument, see Laywine (2010).

Alison Laywine  139 ask what this could mean more concretely, it might do to consider Kant’s work of cosmogony from 1755, the Universal Theory and Natural History of the Heavens. I  grant that, from the point of view of the Critique, this work is vulnerable to Antinomies, because it rests on dogmatic claims about the world’s infinite extension in space. But, in retrospect, it nicely illustrates the idea I have been discussing, precisely because it is uncluttered by the complications of Kant’s critical philosophy. The problem of the Universal Theory and Natural History is to explain why the planets in our solar system orbit around the sun on the same plane and in the same direction, and why our solar system orbits with others around the center of the galaxy on the same plane and in the same direction and why the Milky Way is a greater system still of the same structure formed by our galaxy and others like it. The scopos of this question is the physical universe as a whole. The strategy for answering the question is first of all to imagine hypothetically that, in the act of creation, God subjected matter to Newton’s laws of motion and bestowed upon it a force of repulsion and a universal force of attraction by reason of the inverse square of the distance. But then, on the further assumption that God randomly scattered particles of matter throughout all of space, Kant tries to show that matter would have organized itself into a vast system of systems of the iterated structure we actually observe as a consequence of the original forces of repulsion and attraction operating under the relevant laws. In short, Kant’s explanation uses the laws of motion to build up a model in thought of the physical universe. The test is whether the model is a world that we can recognize as our own.25 I think that some such ideas as the ones I just sketched lie behind the  “­astonishing” conclusion announced at the end of the Deduction that the understanding legislates laws to nature. To be sure, the cosmogony of the Universal Theory and Natural History of the Heavens invokes specific laws of nature: they are Newton’s laws of motion. The “astonishing” conclusion of the Transcendental Deduction concerns an act of legislation prior to the application of any specific natural laws to any specific physical system or system of systems of however many iterations. This act of legislation is the one that makes all others possible. As Kant himself puts it at B 165 at the very end of the B-­Deduction: “All

25  One might perhaps worry that this talk of testing the model will reintroduce the aporetic conception of knowledge and the way it relates to its object. For how else will the test be carried out if not by confronting the model Kant has constructed with the physical universe? Is the physical universe not something that lies outside Kant’s mind and that therefore acts as a brake on, or corrective to, his conjectures? The answer is both “yes” and “no.” Of course, the physical universe lies outside the mind. But, as an object of knowledge, it just is the product of our conjectures. Testing these conjectures means putting the object of knowledge itself to the test. In the case at hand, that will involve making astronomical observations. But no such observations can be made without a lot of theoretical commitments to how we think measurements of time, distance, and position are to be made. In short, the observations themselves are theoretical constructions and not the effect of any kind of confrontation between the world out there and our pronouncements about it.

140  Leibniz and the Transcendental Deduction possible perceptions, hence also all that which can ever reach empirical consciousness, that is all appearances of nature, must, with respect to their combination [ihrer Verbindung nach], stand under the categories; nature (considered merely as nature as such) depends on [sc. the categories—AL] as the original ground of its necessary lawfulness (as natura formaliter spectata).” If we follow the early Kant’s cue in the Universal Theory and Natural History and try to build up an account of how the physical universe took shape, by assuming certain initial conditions, we will find ourselves engaged in repeated acts of legislation: we will keep invoking Newton’s laws of motion. These acts of legislation will ultimately rest on the categories,26 for they are the formal conditions that allow our understanding to engage in any such legislation at all. This does not mean that the particular laws of nature that form the content of our legislation can be derived deductively from the categories. Kant says explicitly a few lines after the passage that I just quoted that we learn the particular laws of nature through experience. But our recognition that these laws are laws and more especially our appeal to them in building up what we know about an object of knowledge, be it the physical universe or some smaller system, depends on the categories. It is in this sense that the cat­ egor­ies prescribe laws to nature and thus make possible all knowledge and hence all objects inasmuch as they are objects of knowledge.

4 Conclusion If Kant can successfully defend his astonishing conclusion that the categories prescribe laws to nature, he will have succeeded in carrying out the Transcendental Deduction. For he will have accounted for all possible intellectual world-­building, whether on a small scale or large. He will have thereby accounted for all thinking and thus all use of the understanding, if it is true, as he believes, that some kind of intellectual world-­building is required by all use of the understanding. If indeed the categories lie at the basis of an original act of legislation, they cannot fail to relate a priori to objects. For the effect of legislating the laws of nature is to make possible empirical knowledge of the world. But such knowledge, as knowledge, relates to an object; so the categories must relate to this object a priori, just to the extent that they make this relation possible. Were it not for the categories, there would only be nature in the material sense. As Kant puts it in A-­Deduction at A 111, “it would be possible for our soul to be filled with a swarm of appearances without experience being able to arise therefrom. As a result, all relation of knowledge to objects would fall away, because the connection according to universal and necessary laws would be missing for it; hence to be sure they would 26  To be a little bit more specific, they depend on the schematized relational categories, as presented in the “Analogies of Experience.” I argue for this claim in the conclusion to Laywine (2020).

Alison Laywine  141 be thoughtless intuition, but never knowledge, and thus they would be as good as nothing for us.” I have had to leave out of account many important details. But if I am right about what Kant is trying to do, then it seems to me that the strategy of the Transcendental Deduction is to adapt the cosmological lessons he learned from Leibniz to the new and improved conception of knowledge and its object that he came up with in response to lessons he learned from Lambert.

6

Bodies, Matter, Monads, and Things in Themselves Nicholas F. Stang, University of Toronto*

1 Introduction There is a well-­known tension in Leibniz’s later philosophical writings, from approximately 1704 onwards, concerning the ontological status of bodies and matter. On the one hand, some texts suggest that in this period Leibniz held a phenomenalist view, according to which bodies are merely the “mutual dream” of the monads; on the other hand, some texts seem to contain a more “realist” view of bodies according to which they are “aggregates” [aggregata] of monads, or are “composed” [composés] of monads. In the past few decades a number of scholars have attempted to reconcile these two strands in Leibniz, or argued that they cannot be reconciled, that they represent an unresolved tension in his metaphysics.1 It is less commonly appreciated, however, that a structurally very similar tension appears in Kant’s Critical philosophy: some passages suggest that Kant was a phe­ nom­en­al­ist about empirical objects in space (bodies), while others suggest a more realist view on which bodies just are things in themselves considered under a certain guise or description (“as they appear to us”). This tension has given rise to one of the oldest, and most intractable, debates about the nature of his transcendental idealism: are bodies (appearances) and things in themselves distinct kinds of object, or are they two different ways of considering one and the same domain of objects? Or, more dramatically, is Kant’s transcendental idealism merely a more complex form of Berkeley’s phenomenalism, or something quite different? While this tension in Kant’s idealism has been the object of philosophical and scholarly scrutiny far longer than the corresponding discussion about Leibniz,

*  I would like to thank the other participants in the Kentucky Kant-­Leibniz conference for their comments on a much earlier (and quite different) version of this paper. Clinton Tolley, James Messina and Robert Pippin also gave me very helpful comments on this version. Beau Madison Mount read and extensively commented on the antepenultimate draft; I owe him an enormous debt of gratitude for his painstaking eye for detail and unerring philosophical insights. I would also like to thank Brandon Look and an anonymous referee for very helpful comments on the penultimate version. 1  Jolley (1986), Wilson (1989), Rutherford (1994; 2008), Hoffmann (1996), Arthur (1998), Lodge (2001), and Garber (2009). For a survey of older literature on the subject, see Adams (1994, 217 n. 2). Nicholas F. Stang, Bodies, Matter, Monads, and Things in Themselves. In: Leibniz and Kant. Edited by Brandon C. Look, Oxford University Press. © Oxford University Press 2021. DOI: 10.1093/oso/9780199606368.003.0006

Nicholas F. Stang  143 the discussion of Kant has, to some extent, calcified into a debate between “one object” readings and “two object” readings—or, as I prefer to call them, “identity” readings and “non-­identity” readings.2 Interpretations of Kant’s idealism often proceed by giving some general considerations in favor of either the identity or non-­identity view, selectively citing some passages that support it, acknowledging that there are texts that support the other interpretation, and admitting that there are other ways of reading Kant. By contrast, the discussion of Leibniz’s ontology of bodies has been guided by the attempt to balance both strands in Leibniz. My aim in this paper is to examine the tension between phenomenalism and realism in Kant through the lens of the structurally similar tension in the later Leibniz. I do this for two reasons. First of all, I think that recent Leibniz scholarship has achieved a greater level of philosophical sophistication on these issues than corresponding literature on Kant.3 This is partly because Leibniz has a richer set of technical notions for characterizing the relation between monads and bodies than Kant has for characterizing the relation between things in themselves and appearances: aggregation, immediate requisition, resulting, and “being in.” The other reason I approach Kant through a Leibnizian lens is the obvious one of influence. While I do not explicitly address the influence of Leibniz on Kant, it is plausible that Kant is picking up Leibniz’s complex ontology of body and

2  “One object or two?” is a bad way of characterizing this dispute because “two object” readers, despite the name, are not committed to thinking that there is in general one and only one thing in itself that appears as a given appearance. The dispute really turns on whether appearances and things in themselves are numerically identical or not. “Two object” or “non-­identity” readers hold that things in themselves are not in general numerically identical to appearances; appearances and things in themselves, on this view, are distinct kinds of objects. “One object” or “identity” readings hold that the distinction between appearances and things in themselves is not a distinction between kinds of objects but between two ways of considering one and the same set of objects. Prominent “identity” readers include Allais (2004) and Allison (1983/2004) (although Allison’s view is very different from Allais). Aquila (1979) offers a full-­throated defense of the “non-­identity” view; Adams (1997b) and Ameriks (1982b) come down on the side of non-­identity, although not as strongly as Aquila. The debate is complicated somewhat by the fact that many “non-­identity” readers hold an identity view about the self: the “empirical self ” and the “noumenal” self are one and the same entity considered under two guises. Aquila (1979b) holds the non-­identity view even about the self. On this issue, see Adams (1997b), Aquila (1979) and Ameriks’s (1982b) discussion of Aquila. The interpretation offered by Langton (1998) does not fit easily into either the “identity” or the “non-­identity” camp because her official view is that appearances are not “objects” at all but relational properties of things in themselves, substances with intrinsic properties. This could be developed either in an “identity” direction (talk of appearances is just talk of substances qua their relational properties) or in a “non-­identity” direction (appearances are properties of substances, and thus are numerically distinct from the substances in which they inhere). The now standard distinction between “one object” readings and “two object” readings is originally found in Ameriks (1982b). Several important works on Kant’s idealism (e.g. Lucy Allais’s Manifest Reality) were published in the years between my writing of this chapter and the publication of this volume, so I cannot discuss them here. 3  The situation has improved somewhat in the years since I wrote this paper. Recent work by Lucy Allais, Tobias Rosefeldt, and Colin Marshall have brought a new level of textual and philosophical sophistication to the debate.

144  Bodies, Matter, Monads, and Things in Themselves adapting it to his own ends, albeit with significant modifications.4 To answer adequately the question of historical influence, one would need to reconstruct Kant’s knowledge of Leibniz’s metaphysics of bodies on the basis of Leibnizian texts available in 1781, as well the way in which philosophers like Wolff and Baumgarten constructed their own monadological theories of matter; that project lies outside the scope of the present chapter.5 Before continuing, I want to flesh out slightly my claim that there is a structural similarity between Kant and Leibniz’s ontology of bodies. The principal element in this common structure is a distinction between two classes of entities. The entities in the first class exist “in themselves”; that is, they exist independently of whether they are perceived or otherwise represented by any finite mind. According to Leibniz, at least in his later writings, the entities in the first class are non-­extended non-­composite mind-­like substances, monads; Kant’s term for these entities is “things in themselves,” and he is necessarily more cautious than Leibniz in offering a positive characterization of them. Both Leibniz and Kant call entities in the second class “phenomena” and both of them include bodies and matter among the phenomena. Phenomena exist at least partly in virtue of the contents of the representational mental states of minds or mind-­like entities. Both thinkers hold that human minds, at some level of description,6 are among the “in itself ” entities of the first class, and both think that human minds are among those whose representational states ground (at least partly) the existence of phenomena. Both thinkers also hold that phenomena possess certain of their properties (e.g. spatiotemporal properties) in virtue of the contents of the representational states of such minds, but my focus in this chapter will be the ways in which the existence of phenomena, rather than their properties, is or is not grounded in representations of them. It is clear that Leibniz and Kant disagree on important points about the nature and relation of the entities in these two classes, although it is less clear what their precise views are. For instance, Leibniz has highly determinate views about the 4  Kant’s acceptance of intersubstantial causation and a distinction in kind, rather than in degree, between sensibility and understanding means that, no matter the parallels, there will be significant differences in their views about bodies. 5  See Radner (1998) and Watkins (2006) for more on the Leibnizian and Wolffian background to Kant’s own theory of matter. See Brandon Look (Ch. 1 of this volume) for a discussion of the history of Leibniz reception in Germany philosophy before, and by, Kant. I should also note that, while I am restricting my attention to the “later” Leibniz in this chapter, there is now a lively scholarly debate as to whether the views of the later Leibniz are continuous with the views expressed in the 1680s in such texts as the Discourse on Metaphysics and the Arnauld correspondence (neither of which were avail­ able to Kant); I am going to ignore that debate. Cf. Adams (1994) and Garber (2009) opposing views on the “continuity” of Leibniz’s theory of bodies. 6  For the later Leibniz, as a thinking substance, a monad, I am an entity in the first class, but my body belongs in the second class. Considered as an organism, I am a body (phenomenon) united to a dominant monad (my soul, a substance); organisms for Leibniz are less substantial than monads but more substantial than mere bodies. Kant’s views on the ontology of the “self ” are harder to discern, but it is relatively clear that considered as I am in myself, I belong in the first class of entities.

Nicholas F. Stang  145 nature of “in itself ” entities: they are non-­extended, each of them perceives the entire world, they do not causally influence one another, God creates them with harmonized perceptions, etc. Kant’s official position is that we can have no theoretical knowledge whatsoever about entities in the first class, things in themselves, but he does venture some limited claims about them, and in some places even claims to know that they exist and are non-­ spatiotemporal.7 Determining Kant’s exact views on the extent and nature of our knowledge of things in themselves is a complicated task, but one I will not pursue in this chapter. Leibniz thinks that entities in the first class are substances in a non-­ derivative sense, while entities in the second class are substances only in a derivative sense.8 Kant thinks that some entities in the second class are substances; specifically, bodies in space (composed of matter). His official view is that categories like “substance” and “causation” lack what he calls “objective validity” when applied to entities in the first class, but he nonetheless frequently uses these very categories when discussing them.9 These are real points of disagreement between Kant and Leibniz, and I do not want to downplay them. But in this chapter I am going to focus on the broad structure of agreement between the two thinkers, not their differences. Leibniz and Kant not only make a similar distinction between entities that exist independently of being represented (monads, things in themselves) and those that depend upon being represented (bodies, phenomena); they also express these views in ways that raise very similar hurdles for commentators trying to interpret their texts. In the case of both philosophers, some texts support a narrowly phenomenalist reading on which bodies are nothing over and above 7  Regarding the existence of things in themselves, Kant writes in Prolegomena §32: “the understanding, just by the fact that it accepts appearances, also admits to the existence of things in themselves [gesteht auch das Daseyn von Dingen an sich selbst zu]” (Ak. 4:315; cf. 4:354). This is significant because Kant here attributes the category of “existence” [Dasein] to things in themselves, and not merely problematically. The prima facie meaning of the passage is that in claiming that empirical objects are appearances, one is thereby committed to claiming that there exist non-­empirical things in themselves that appear as those objects; if the former is a possible item of knowledge—and, given that it is one of the central tents of the KrV, it is hard to see how it might not be—then the latter is as well. Regarding their non-­spatiotemporal character, Kant claims at A 48 that it is “ungezweifelt gewiß und nicht blos möglich oder auch wahrscheinlich” that space and time are merely subjective, hence, that things in themselves are non-­spatiotemporal. 8  GP VI 590/AG 265; GP II 275–276/AG 181–182; GP III 606/L 655. 9  He repeatedly uses causal concepts to describe things in themselves, writing that they “affect” us (A 190/B 235; Ak. 4:289, 314, 318, 451), that they possess forces (Ak. 8:153–154), that they are causes (A 387, A 494/B 522, Ak. 8:215), etc. In some texts he also applies the modal category of possibility to things in themselves (R5184, 5723, 5177). Kant asserts the substantiality of God in his lectures on rational theology (Ak. 28:1037; cf. 28:600, 800, 805, 1163, 1261); I take him to mean that it is sub­ject­ ive­ly necessary for us to hypothesize a substantial God, but we cannot assert the existence of such a being with apodictic certainty (a priori cognition). The Paralogisms section of the first Critique is usually interpreted to mean that I cannot know that my soul is a substance. However, in his metaphysics lectures Kant repeatedly claims that I can know through apperception that I am a substance, rather than a modification of some other substance (as Spinoza held); Ameriks (1982a) draws on these and other texts to challenge the standard interpretation of the Paralogisms.

146  Bodies, Matter, Monads, and Things in Themselves the intentional contents of some class of representations (infinite monadic perceptions, experience). For instance: Matter and motion are not so much substances or things as the phenomena of perceivers, whose reality is located in the harmony of perceivers with themselves (at different times) and with other perceivers.  [GP II, 270/L 537]10 For the appearances, as mere representations, are in themselves real only in perception. To call an appearance a real thing prior to perception means either that in the continuation of experience we must encounter such a perception, or it has no meaning at all.  [A 493/B 521–522]11

But other texts suggest a realist view, on which an individual body is either identical to an individual thing in itself (Kant) or to an aggregate of monads (Leibniz), or is in some other way grounded in “in itself ” entities (things in themselves, monads). For instance: Everything is full in nature. There are simple substances everywhere, actually separated from one another by their own actions, which continually change their relations; and each distinct simple substance or monad, which makes up the center of a composite substance (an animal, for example) and is the principle of its unity, is surrounded by a mass composed [composée] of an infinity of other monads, which constitute [constitituent] the body belonging to this central monad, through whose properties the monad represents the things outside it, similarly to the way a center does. [“Principles of Nature and Grace,” GP VI, 599/AG 207]12

10  See also AG 307; L 363–365; GP VI 590/AG 265. In the De Volder correspondence see Leibniz’s letter of 1705 (GP II 275/AG 182) and Leibniz’s letter to De Volder of January 19, 1707 (GP II 283/AG 186). In his letter of June 30, 1704, De Volder himself seems to be interpreting Leibniz as a phe­nom­en­ al­ist (GP II 272). 11  See esp. the entire section “Transcendental idealism as the key to solving the cosmological dialectic” (A 491–497/B 519–525) as well as the Fourth Paralogism in the A edition (A 367–380, esp. A 376f. and the note on A 374–375). Cf. A 59/B 42; A 383; A 506/B 534; Ak. 4:354, 4:506. Allais (2004) objects to any phenomenalist reading of Kant by pointing out, correctly, that on Kant’s view there are empirical objects we can never perceive (A 226/B 273, Ak. 8:205). However, if we understood phe­ nom­en­al­ism as the view that empirical objects exist partly in virtue of facts about the contents of subjects’ perceptual states, this is compatible with holding that there exist empirical objects we never directly perceive. Van Cleve (1999) makes a similar point. I realize that these phenomenalist-­sounding texts can be read otherwise. While, ultimately, I think that Kant ultimately holds a form of phe­nom­en­ al­ism, I’m not assuming that here. My point is simply that there is a strand in Kant’s texts that supports the attribution of phenomenalism, just as there is in Leibniz. 12  See also GP II 281/AG 185; GP IV 498/AG 146–147; GM III, 537/AG 167; and GP VI 607/L 643. However, in his letter to De Volder of June 30, 1704, Leibniz explicitly denies that matter is composed of monads: “properly, speaking, matter isn’t composed [componitur] of constitutive unities, but results from them, since matter, that is, extended mass is only a phenomenon grounded in things, like a rainbow or a parhelion, and all reality belongs only to unities” (GP II 268/AG 179).

Nicholas F. Stang  147 [T]he same objects can be considered from two different sides, on the one side as objects of the senses and the understanding for experience, and on the other side as objects that are merely thought at most for isolated reason striving beyond the bounds of experience.  [Note to B xix, Kant’s emphasis]13

Interpreters of Kant and Leibniz, therefore, face similar problems when confronted with these texts: how do we square the apparent commitment to phenomenalism with the apparent commitment to a more realist view about bodies? How can bodies simultaneously be (aggregates of) substances/things in themselves, and be the mere phenomena of perception/experience? I begin by examining Leibniz’s views on matter and bodies from the perspective of an influential essay by Donald Rutherford that squarely confronts the question: how do we do justice to Leibniz’s idealism without reducing it to mere phenomenalism? Rutherford argues that Leibniz is an idealist, in that he accepts that everything that exists in the finite concrete world14 is either a monad or exists in virtue of the contents of monadic perceptions, but he is not a phenomenalist, because, although bodies exist in virtue of the contents of monadic perceptions, they have a reality “over and above” these perceptual contents. Precisely characterizing the sense in which bodies can be something “over and above” the contents of monadic perceptions, yet exist in virtue of them, will be one of the main tasks of §§2 and 3 of this chapter.15 In §4 of the chapter I turn to Kant and compare his views on the ontological status of empirical objects in space (bodies) with what my early discussion has revealed about Leibniz’s theory. Having distinguished several different idealist views about bodies in the context of Leibniz’s philosophy, I consider which of these idealist theses Kant is committed to. I conclude that, their substantial differences aside, Kant and Leibniz have similar views about the ontological status of bodies. In particular, both philosophers agree that bodies exist in virtue of 13  See also A 3/B 5, B xx, B xxvi–xxvii, and B 69. 14  The qualification “in the finite concrete world” is meant to exclude God, who is infinite, and “abstract” items like concepts, propositions, numbers, etc. that, according to Leibniz, exist in virtue of God’s awareness of them. For more on Leibniz’s ontology of abstract objects, see Mates (1989). 15  A note on how I am using the terms “ground” and “in virtue of.” By “ground” I refer to an asymmetric relation of metaphysical determination between facts, which can (though need not) hold between mutually necessarily entailing facts. “In virtue of ” refers to the converse of the grounding relation; that fact that p is said to obtain in virtue of the fact that q just in case the fact that q grounds the fact that p. There has been much recent work on the relation of grounding in contemporary metaphysics; see esp. the essays collected in Correia and Schnieder (2012). In an eighteenth-­century context, though, it is important to distinguish (metaphysical) grounding from efficient causation; the precise details of how to do so depend upon our background theory of grounding and of causation, but for now this rough gloss will suffice: a grounded fact is nothing “over and above” what grounds it, while a cause is either a distinct substance (Kant) or an earlier state of one and the same substance (Leibniz). There is more metaphysical “distance” between a cause and its effect than there is between a ground and its consequences. The distinction between metaphysical ground and efficient causes cor­ res­ponds to the distinction between ratio essendi and ratio fiendi in German rationalism, a distinction Kant inherits. See Smit (2009) and ch. 7 of Stang (2016) for more on this distinction.

148  Bodies, Matter, Monads, and Things in Themselves things in themselves (monads) appearing to cognitive subjects (monads), and that to be a body just is to be the appearance of things in themselves (monads) to cognitive subjects (monads). I conclude by suggesting that the notion of force is the key to combining the realist and phenomenalist strands in each thinker. Both thinkers hold that bodies are more than mere illusions, in part because the (primitive) forces in things in themselves (monads) appear as the (derivative) forces in bodies.

2  Preliminaries: Idealism, Essence, and Existence Rutherford begins his essay by very helpfully distinguishing two kinds of idealism that Leibniz might accept: substance idealism, “the idea that the only things that meet the strictest conditions on being a substance are unextended, mind-­like entities”;16 and matter idealism, “the idea that material things exist only as appearances, ideas, or the contents of mental representations.”17 Rutherford argues that, while Leibniz is a substance idealist in his later period, he is not a matter idealist. Because much will hinge on exactly what is, and is not, compatible with these forms of idealism, it is worth being very precise about what they mean. The locus classicus for the later Leibniz’s substance idealism is his remark in a letter to De Volder that: Indeed, considering 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 phenomena of perceivers, the reality of which is situated in the harmony of the perceivers with themselves (at different times) and with other perceivers.  [GP II 270/AG 181]

In this passage, and many others, Leibniz claims that everything there is, is either a monad, a perception or an appetition of a monad, or is a “phenomenon” of these. (Henceforth, I am going to just refer to monadic “perceptions,” by which I mean both perceptions and appetitions). Assuming that a phenomenon of monads exists in virtue of facts about monads, this means that Leibniz is committed at the very least to the following principle: Substance Idealism:  Everything that exists in the concrete realm is either a monad, a perception of a monad, or exists in virtue of facts about monads and their perceptions.18 16  Rutherford 2008a, 142. 17 Ibid. 18  I am restricting attention to finite monads; in some texts, Leibniz refers to God’s mind as the supreme monad. Consequently, I am ignoring, for the purposes of this discussion, the strand in

Nicholas F. Stang  149 I think Rutherford would accept this as a characterization of substance ideal­ ism because it retains the core idea of substance idealism from his paper: monads are substances in the most genuine sense because everything else that exists, exists in virtue of monads and their perceptions. Note, though, that, as formulated, Substance Idealism is compatible with the claim that bodies exist in virtue of facts about monads and their perceptions, but do not exist in virtue of being themselves perceived by monads. Strictly speaking, Substance Idealism is compatible with bodies never being perceived by monads, as long as the existence of those bodies is grounded in some other way by facts about monads’ perceptions. For instance, if bodies were complex wholes composed of monads, which wholes exist in virtue of harmonious relations among monadic perceptions, Substance Idealism would be satisfied even if ­monads never perceive the bodies they compose.19 Whatever view about the relation between bodies and monads Leibniz is asserting in the De Volder passage, it is at least as strong as Substance Idealism. Rutherford distinguishes Substance Idealism from what he calls “Matter Idealism,” “the idea that material things exist only as appearances, ideas, or the contents of mental representations.”20 He argues that Leibniz is not a Matter Idealist, in this sense, because bodies have a “reality” over and above being represented by monadic perceptions; they do not exist only “as” the objects of monadic perceptions. He packs a lot into the idea of an object existing only as the content of a mental representation, so before continuing I want to separate a few different claims that might constitute “matter idealism,” first: Weak Matter Idealism:  For all bodies B, if B exists, B exists partly in virtue of facts about monadic perceptions of B.21 Weak Matter Idealism adds an important element missing from Substance Idealism: that the existence of bodies is not merely grounded in facts about monads, but are grounded (at least partly) in facts about those monads’

Leibniz’s thought, noted by both Adams (1994) and Rutherford (2008a) that takes bodies to be the phenomena of God’s perceptions. 19  However, Leibniz does not think that monads are parts of bodies. See the letter to De Volder of June 30, 1704 (quoted in n. 11 above). 20  Rutherford (2008a, 142). Cf. Adams (1994, 146), Lodge (2001, 472). 21  Some readers will object that it is incoherent to suppose that something can exist in virtue of its standing in a relation to something else, for its existence is a partial ground of the fact that it stands in that relation. For instance, the relational fact that a monad perceives a body is partially grounded in the fact that this body exists; it cannot therefore be a ground of the existence of that body (given the irreflexivity of grounding). While I agree with this line of reasoning as a piece of metaphysics, I do not think it is correct to import this assumption into Leibniz (or Kant for that matter). I think both philo­ sophers accept that for some values of x, x exists in virtue of a fact about x. For instance, Leibniz holds that God exists in virtue of a fact about God: his essence contains existence. Kant, I think, holds that a phenomenon exists in virtue of a fact about it: it is experienced.

150  Bodies, Matter, Monads, and Things in Themselves perceptions of bodies. That Leibniz is a Weak Idealist about Matter is relatively uncontroversial; he repeatedly describes bodies as aggregates, and writes that aggregates are “semimental” [semimentalis] because their unity, hence their existence, depends upon a perceiving mind.22 Since the conjunction of Substance Idealism and Weak Matter Idealism is not going to be further in question in this chapter, henceforth I will use the term “Substance Idealism” to refer to that conjunction. Intuitively, the idea behind the renaming is that that we have added to Substance Idealism the requirement that the existence of bodies must be grounded, at least partly, in facts about monadic perceptions of them. It excludes the purely compositional relation between bodies and monads I mentioned earlier and hence gets better at Leibniz’s intentions in the De Volder passage. But Rutherford means something stronger by “matter idealism,” since he does not deny that Leibniz held Weak Matter Idealism. Remember, he defines “matter idealism” as the view that bodies exist only “as” the objects of monadic perceptions. There are at least two different ways of understanding this. First: Idealism about the Existence of B odies:  For all bodies B, if B exists, B exists wholly in virtue of facts about coherent monadic perceptions of B. The idea behind this principle (“Existence Idealism,” for short) is that all there is to a given body B existing is there being monads that have coherent perceptions of B. One might well ask what “coherent perceptions” are; intuitively, this is supposed to capture the requirement that B’s existence is grounded in a plurality of monads having internally consistent perceptions of B that agree in representational content with their perceptions of other bodies.23 But the details of the coherence relation among the perceptions that ground the existence of bodies is not relevant here; it suffices to notice the structure of the view: there is some relation of coherence among monadic perceptions, and monadic perceptions that stand in that relation ground the existence of bodies. But Rutherford might have something else in mind by “matter idealism.” If bodies exist only “as” the objects of monadic perceptions then, intuitively, they are nothing “over and above” the objects of monadic perceptions. All there is to being a body is being the object of monadic perceptions. These sound like claims about the essence of bodies, not about what grounds their existence (what it is in virtue of which they exist). So, we might understand “matter idealism” as follows: Idealism about the Essence of B odies: (I) The essence of being a body (= what it is to be a body) is to be the object of coherent monadic perceptions, 22  RB 146; Des Bosses correspondence (GP II 304). 23  Cf. “On the Method of Distinguishing Real from Imaginary Phenomena” (GP VII, 319–222/L 363–365).

Nicholas F. Stang  151 and (II) For any body B, the essence of B (= what it is to be B) is to be the object of a certain set of coherent monadic perceptions.24 I will refer to this view as “Essence Idealism.” This gives a quite strong sense in which bodies might be “nothing over and above” monadic perceptions: their essences are exhausted by being the objects of those perceptions. Clause (II) states that for any body B there is a coherent set of monadic perceptions such that the essence of B is to be the object of those coherent perceptions, but it does not say which set of perceptions that is. The natural answer is to combine Essence Idealism with Existence Idealism, and further specify those monadic perceptions as the monadic perceptions that enter into the facts that ground the existence of B.25 So, on this package of views, for any body B, B exists in virtue of facts about monadic perceptions of B, and those very monadic perceptions determine the essence of B: to be B is to be the object of those very monadic perceptions. This makes bodies, in a very strong sense, dependent upon monadic perceptions. On this view, all there is to being a given body is being the object of some particular coherent monadic perceptions. Perhaps, in denying that Leibniz is a matter idealist, Rutherford merely means to deny that Leibniz held that conjunction of views.26 Before continuing, though, I want to address a point that will shortly become important. One might think that Essence Idealism entails Existence Idealism, so, by the contrapositive, the negation of Existence Idealism entails the negation of Essence Idealism. If this is correct, if Leibniz denies idealism about the existence of bodies, he would thereby be committed to denying idealism about their essences. But it is incorrect; Essence Idealism does not entail Existence Idealism. To see why, note that that Essence Idealism is a claim of the form: (1)  The essence of X is to bear relation R to the Ys. where X is a given body, the Ys are some monads, and R is the relation of being coherently perceived by the Ys. Existence idealism is a claim of the form: 24  The first clause (idealism about the essence of being a body) does not entail the second clause (idealism about the essence of individual bodies). For instance, assume that the essence of being a father is having at least one child and assume that every father is essentially a father. It does not follow that for any father x, there is a child such that x is essentially the father of that child. This example is due to Beau Mount. Paul Hoffmann anticipates my distinction between idealism about the essence and idealism about the existence of bodies by distinguishing between whether, for Leibniz, the being (essence) or the unity (existence) of bodies depends upon perception (Hoffmann 1996, 118). 25  This means that the view I go on to describe in the rest of this paragraph is stronger than the conjunction of Essence Idealism and Existence Idealism: the quantifier over monadic perceptions has been moved out to have scope over the grounding relations, so for each body there are some monadic perceptions that body that grounds its existence and constitute its essence. Thanks to Beau Mount for pressing me on this point. 26  Lodge (2001, 473) cites a passage from the Arnauld correspondence in which Leibniz denies what I have called “idealism about the essence of bodies” (GP II 96/LA 121). However, this passage is from the mid-­1680s, and thus falls outside of the purview of this chapter.

152  Bodies, Matter, Monads, and Things in Themselves (2)  The fact that X exists is grounded in the fact that X bears relation R to Y. But claims of the form of (1) do not in general entail claims of the form (2). For instance, consider the following claim about the essentiality of origin: (1*)  The essence of GWL is to be the son of Friedrich Leibniz and Catharina Schmuck. does not entail the following claim about ontological dependence: (2*)  The fact that GWL exists is grounded in the fact that GWL is the child born to Friedrich Leibniz and Catharina Schmuck. Regardless of whether one thinks (1*) or the essentiality of origin is true, and of whether one thinks (2*) is true, (1*) does not entail (2*). There is nothing incoherent about claiming that to be Leibniz is to be the child of his parents, but he does not exist in virtue of being their child. Surely, one could accept (1*) and hold that, when Leibniz exists, he exists in virtue of, say, his organic parts being appropriately unified and functioning. This shows that an entity need not, in general, exist in virtue of facts about the objects that are mentioned in its essence (i.e. the objects such that the entity’s essence is to be related to those objects in certain ways). So, there is room for Leibniz to be an idealist about the existence of bodies, but not about their essences. In much of the essay, Rutherford takes Robert Adams’s “qualified realist” interpretation of Leibniz as his stalking-­horse and argues that Adams mistakenly assimilates Leibniz’s view to “matter idealism.” However, the distinctions I have made among various things “matter idealism” might be (idealism about the existence or the essence of bodies) allow us to see that Rutherford may not be as successful as he thinks in finding an interpretive alternative to Adams.27 Adams agrees with Rutherford that Leibniz rejects what I have called “Existence Idealism” about bodies. According to Adams, the existence of a body is grounded in monadic perceptions of that body as well as there being monads “in” that body; Adams expresses this point by writing: “since all bodies have substances [monads] ‘in’ them, they can be regarded as appearances of substances as well as appearances to substances [monads].”28 So, in this sense, Adams’s Leibniz is not an Existence Idealist, as Rutherford himself acknowledges (Rutherford 2008a, 160). This means that Rutherford succeeds in distinguishing his interpretation from 27  The same ambiguity affects Adams’s claim that “Leibniz does not believe that phenomena have any being except in the existence or occurrence of qualities or modifications of perceiving substances” (Adams  1994, 223). If by “being” Adams means existence, then this is false on his own view. If by “being” he means essence, it is unclear why he assumes this is true of Leibniz. 28  Adams (1994, 240f.). Cf. Jolley (1986, 48) and Hoffmann (1996, 115).

Nicholas F. Stang  153 Leibniz only if he can show that his reconstruction of Leibniz’s theory rejects Essence Idealism; if Rutherford cannot do this, then there is no clear sense in which his Leibniz is any more “Matter Realist” than that of Adams. This is why it is important that, as I argued in the previous paragraph, Essence Idealism does not entail Existence Idealism, and, equivalently, the negation of Existence Idealism does not entail the negation of Essence Idealism. By rejecting Existence Idealism, Rutherford’s Leibniz is not thereby committed to rejecting Essence Idealism. In §3, I will argue that, for all Rutherford shows, Leibniz might nonetheless be an Essence Idealist about bodies. However, I conclude by sketching my own argument that Leibniz in fact rejects Idealism about the essence of bodies.

3  Rutherford and Adams on Leibniz The guiding idea of Rutherford’s interpretation is that while Leibniz is a Substance Idealist (and hence holds that bodies exist at least partly in virtue of being perceived by monads), bodies have a “reality” above and beyond that which they have as the intentional objects of monadic perceptions, a reality they inherit from the monads from which they result (to use Leibniz’s technical terminology). The crucial issue about Rutherford’s interpretation, and how it differs from Adams, is how to understand the idea that bodies get their “reality” from the monads “in” them. Rutherford offers the following model for reconciling Substance Idealism and Leibniz’s alleged Matter Realism: bodies exist partly in virtue of monadic perceptions, because monadic perceptions give them the unity without which they would not exist, but the unity conferred by monadic perception does not exhaust the reality of the monads that make up the body, the monads that are unified by perception. In this section I will try to unpack this model and determine whether it succeeds in reconciling Substance Idealism with Matter Realism, and whether it constitutes a real interpretive alternative to Adams. Rutherford’s model would show that Leibniz is a Matter Realist if Leibniz were to hold that monads are parts of bodies: a group of monads only compose a body if other monads agree in perceiving those monads as parts of the body. On such a view, the existence of a body is partly grounded in monadic perceptions of the body, but the essence of the body is not merely to be the object of those monadic perceptions; to be a given body is to be a complex whole composed of a given set of monads, whose principle of unity is perception by other monads. On such a view, Leibniz would not be an Essence Idealist in my sense; in Rutherford’s terms, he would be a Substance Idealist and a Matter Realist. The reality of bodies would be something “above and beyond” that of monadic perceptions because bodies would literally be composed of monads, substances that exist (and thus are real) independently of being perceived. But, as Rutherford himself points out, Leibniz

154  Bodies, Matter, Monads, and Things in Themselves denies that monads are literally parts of bodies. Leibniz writes in a letter to De Volder: properly, speaking, matter isn’t composed [componitur] of constitutive unities, but results from them, since matter, that is, extended mass is only a phe­nom­ enon grounded in things, like a rainbow or a parhelion, and all reality belongs only to unities . . . substantial unities aren’t really parts, but the foundations of phenomena.  [June 30, 1704; GP II 268/AG 179]29

So, the compositional model of how bodies have a reality “over and above” being the objects of monadic perception, whatever its merits, is not Leibniz’s. But Rutherford suggests another way for understanding how Leibniz can be a Substance Idealist (bodies exist at least partly in virtue of monadic perceptions of them) without being a Matter Idealist (without thinking that “all there is” to a body is its being the object of monadic perceptions). According to Rutherford, Leibniz conceives of matter as an “inherently plural mass.” But it is not entirely clear what Rutherford means by this intriguing suggestion. One thing he might mean is that “matter” (materia, matière), as Leibniz uses it, is a “mass noun”: it refers to a kind of stuff, not to individual objects. Syntactically, the marker of a mass noun “x” is that “counting expressions” like “how many x’s?” are not well-­ formed. For instance, “how many matters?” is ill-­formed (at least, in the relevant sense of “matter”), just like “how many waters?” or “how many nitrogens?” (likewise). Where x is a mass noun, the expression “some x” can be used to refer to a quantity or portion of x. By contrast, “body” (corpus, un corps) is a count noun. (It can also be used generically to refer to the kind body, but I’m going to ignore the generic usage.) Syntactically, the mark of a count noun is that the question “how many x’s?” is well-­formed. For instance, the questions “how many pencils?,” “how many kittens?,” and “how many bodies?” are all well-­formed, which means that “body” is a count noun, just like “pencil” and “kitten.” This means bodies are individual objects. The relation between matter and bodies, on this interpretation, is that every body is made of some matter (a quantity of stuff). What is the matter that makes up bodies? Rutherford’s idea that matter is a kind of stuff, rather than a kind of object or thing, only helps us understand the relation between bodies and monads, and thus determine the sense in which Leibniz is or is not a “matter idealist,” if we have a grasp on the relation between monads and the stuff matter. And here, I think, Rutherford’s suggestion faces a significant problem. Rutherford might claim, that matter, for Leibniz, is monads, or, more precisely, the matter of a body is some monads. But this is incoherent, if we are assuming that matter is a stuff, because monads are countable individual

29  Cf. GP II 436/AG 199; A VI.4.1670/AG 105; GM III 542/AG 168.

Nicholas F. Stang  155 objects, not stuff, so a particular quantity of the stuff matter cannot be literally identical to the monads “in” that matter, or from which that matter results. The identity statement “the matter of body B = the monads in B” is ill-­formed because the term on the left-­hand refers to a quantity of stuff, while the term on the right-­ hand side refers to some countable individuals.30 Observe, we can ask intelligibly “how many monads?” but we cannot ask “how many matters?” But this means that if matter is a kind of stuff rather than a kind of thing, then the matter in a body is not identical to the monads from which it results. Rutherford’s interesting suggestion that “matter” is a kind of stuff, therefore, gets us no closer to understanding the relation between matter and monads. What, then, is the relation between the matter of a body, and the monads from which that body results? Rutherford claims that it is identity: bodies are made of monads. This only makes sense, as I argued in the last paragraph, if both halves of the identity claim are count nouns. This requires giving up Rutherford’s suggestion that matter is an “inherently plural mass.” But the monads “in” a body cannot be literally identical to the matter of that body, because, in passage after passage, Leibniz contrasts matter with substances, writing that matter is a “phenomenon” rather than a substance.31 These passages become hard to understand under the supposition that matter itself is just some substances. Even if matter is monads, this gets us no closer to understanding how Leibniz is a “Matter Realist,” or how Rutherford’s Leibniz is any more realist than Adams’s, without understanding what bodies are, and whether this commits Leibniz to Idealism about the Essence of Bodies. What, then, on Rutherford’s view, is the essence of a Leibnizian body? One might think that his answer should be: (3)  For any body B, the essence of B is to be some monads coherently perceived by a (perhaps over-­lapping) set of monads as composing B. But this cannot be right, for reasons given earlier: to be a body is not to be the monads that compose the body, because the body is not identical to those monads, even if the body is perceived as the aggregate of those monads (i.e. even if all the conditions for the existence of the body are satisfied). The monads are the matter of the body, a plural collection of substances, while the body is an individual. Compare a Leibnizian body to a set, such as {a,b}. The essence of {a,b} is not to be a and b; although the existence of {a,b} is grounded by the fact that a exists and b exists, that is not all there is to the essence of {a,b}. To be the set {a,b}

30  See, however, Sider (2007) for a partial defense of the view that masses (e.g. some matter) can be identical to individual objects (e.g. monads). 31  E.g. GP II 270/L 537; GP II 276/AG 182; GP III 636/L 659; and GP IV 356/L 384.

156  Bodies, Matter, Monads, and Things in Themselves is to be the set whose members are a and b (and nothing else). So, we might instead think that Leibniz’s view is that: (4)  For any body B, the essence of B is to be a whole composed of some monads, such that those monads are perceived by monads in certain determinate ways. If this were Leibniz’s view, he could hold onto Weak Existence Idealism while rejecting both Existence Idealism and Essence Idealism. However, this is not Leibniz’s position because, as we have seen, Leibniz denies that monads are parts of bodies. This means that bodies are not literally composed of monads. There may be some other sense of “composition” in which bodies are composed of monads, but Rutherford—and Leibniz as well—owes us an account of what that is; the literal, ordinary sense of composition is that a thing is composed of its parts. Rutherford has given an account of what grounds the existence of Leibnizian bodies; but, on this score, he has not distinguished his interpretation from Adams’s interpretation. Rutherford needs to explain what the relation is between monads and bodies in such a way that avoids the attribution of Essence Idealism; otherwise, his view is no less “matter idealist” than Adams’s. In the later sections of his essay, Rutherford distinguishes several different technical Leibnizian notions that might characterize the relation between monads and bodies: aggregation, resulting-­from, being-­in, and resulting. However, I will argue, none of these is sufficient to support Rutherford’s interpretation. Each of them, when interpreted correctly, is merely a consequence of his Substance Idealism. Consequently, none of them succeeds in explaining why the essence of a body is not merely to be the intentional object of monadic perceptions. None of them succeeds in making Rutherford’s Leibniz any more “realist” about bodies than Adams’s. 1. Aggregation. In numerous texts, Leibniz claims that bodies are aggregates of monads. As noted earlier, though, we cannot understand this in the natural way, because monads are not parts of bodies. However, the alternative to the straightforward reading is to understand Leibniz’s claims that bodies are aggregates of monads as meaning that bodies bear some other metaphysical grounding relation to aggregates of monads: they result from aggregates of monads, they are “in” aggregates of monads, or aggregates of monads are immediate requisites of them. 2. Resulting. Rutherford cites a text in which Leibniz defines the resulting relation as follows: “I understand that to result [resultare], which is immediately understood to be posited, when those things from which it results have been posited” (A VI 4, 310). The most natural way of reading this def­ in­ition is: the X-­facts result from the Y-­facts just in case the X-­facts are immediately grounded in the Y-­facts (i.e. the Y-­facts ground the X-­facts) and if the Z-­facts ground the X-­facts, either they are identical to the Y-­facts

Nicholas F. Stang  157 or they ground the Y-­facts. Clearly, for Leibniz, facts about bodies result from facts about monads, but this is just a consequence of his Substance Idealism. So, by pointing out that bodies result from monads, Rutherford fails to distinguish his interpretation from Adams’s, and thus fails to show that his Leibniz is any more “realist” about matter than Adams’s. Nor does the fact that bodies “result” from monads entail that Essence Idealism is false; for all this shows, Leibniz might still be an Idealist about the essence of bodies. 3. Being-­in/immediate requisition. Rutherford sees Leibniz as identifying “being in” with immediate requisition. Leibniz frequently talks about monads being everywhere “in” matter, and of monads being “immediate requisites” of bodies. Rutherford quotes the following Leibnizian definition of immediate requisition: “if A is an immediate requisite of B, A is said to be in B, that is, A must not be posterior in nature to B, and with A supposed to exist, it must follow that B also does not exist, and this consequence must be immediate, independent of any change, action or passion” (A VI 4, 650).32 Clearly, for Leibniz, this relation holds between monads and bodies: it would not be possible for there to be bodies unless there were monads. However, this is just a consequence of Substance Idealism. That monads are immediate requisites of bodies does not entail that Leibniz is a Matter Realist, in either of the two senses distinguished earlier. Specifically, it does not entail that Leibniz rejects Essence Idealism, and constitutes no difference between Rutherford’s and Adams’s interpretations. Since, when properly interpreted, all of these relations hold between bodies and monads whether or not Leibniz is an Idealist about the essence of bodies, I conclude that Rutherford has not given us grounds to deny that Leibniz is an Idealist about the essence of bodies. Since Rutherford does not succeed in distinguishing Leibniz’s view from Essence Idealism, nor from Adams’s qualified realist view, I conclude that he does not undermine Adams’s interpretation. Rutherford seems to think that Adams’s view attributes a kind of “matter idealism” to Leibniz; if he is correct about this, then Rutherford’s Leibniz is just as much a “matter idealist” Leibniz as Adams’s. I will make only the weaker claim that neither Adams nor Rutherford’s interpretation is incompatible with Essence Idealism; neither of them show that Leibniz would deny that the essence of a body is to be the object of coherent monadic perceptions. I conclude that, for all they show, bodies may indeed exist only “as” objects of monadic perceptions (to use Rutherford’s phrase). Finally, Rutherford considers another way in which Leibnizian bodies are something “over and above” the objects of monadic perceptions: they possess active forces, which are the manifestation or modification of the primitive active forces in monads. One of the most consistent features of Leibniz’s metaphysics, from at least the 1680s on, is his anti-­Cartesian insistence that it is part of the 32  Rutherford (2008a, 167).

158  Bodies, Matter, Monads, and Things in Themselves essence of bodies that they have forces, that they are not inert. Whether this Leibnizian doctrine is inconsistent with Essence Idealism hinges on whether bodies can have forces in virtue of being perceived to have forces. If so, then it would be consistent for Leibniz to hold that what it is to be a body is to be the object of coherent monadic perceptions which represent the body as having certain dynamic properties. But I am not going to delve into that issue here. I will consider the relation between monadic and corporeal forces in §4 of this chapter, when comparing Leibniz’s views on force to Kant’s. In the remainder of this section I want to sketch a “Matter Realist” reading of Leibniz in Rutherford’s sense; that is, an interpretation on which there is more to the essence of a given body than being the object of a coherent set of monadic perception and which thus vindicates Rutherford’s claim that Leibniz is a “Matter Realist” in both of the two senses distinguished in §1 of this chapter: he denies both Existence and Essence Idealism about bodies. However, as far as I can tell, my reading is just as much available to Adams as it is to Rutherford; if successful, therefore, my reading would not vindicate Rutherford’s claim to have found a more “realist” Leibniz than Adams. What follows is a rational reconstruction of Leibniz’s views. It is not explicitly stated (or denied) in any texts of which I am aware. Consequently, I offer no direct textual support in its favor; it constitutes merely one view that Leibniz could have held, consistent with the texts cited above by me, and those discussed by Adams and Rutherford. Earlier, I distinguished Idealism about the essence of the kind body and Idealism about the essence of individual bodies. I give a Leibnizian argument against the first idealist thesis—idealism about what it is to be a body—and then I use that to motivate a Leibnizian argument against the second idealist thesis— idealism about what it is to be B, where B is a body. Earlier, I argued that Leibniz is committed to: (1)  For any body B, if B exists, B exists partly in virtue of facts about monadic perceptions of B and partly in virtue of facts about the monads appearing as B. This is the consequence of Leibniz’s acceptance of Weak Matter Idealism—the claim that a body exists partly in virtue of facts about monadic perceptions of it—and rejection of Existence Idealism—the claim that a body exists wholly in virtue of facts about monadic perceptions of it. The beginning of my argument is the following plausible principle to which Leibniz can appeal: (2)  If the X-­facts obtain in virtue of the Y-­facts this is explained by the essences of the things involved in the X-­facts and the Y-­facts.33

33  Cf. Rosen (2010).

Nicholas F. Stang  159 For instance, to use a Leibnizian example, if the facts about what is possible obtain in virtue of facts about God’s intellect (as Leibniz claims in Mon 43), then this is explained by the essences of the possible things and the essence of God and his intellect. But note that claim (1) is a claim about all bodies as such. It is plausible, then, that (1) is explained not merely by the individual essences of each body but by the essence of the kind body. Specifically, it is plausible that (1) is explained by the fact that it is part of the essence of body (what it is to be a body) to be the object of coherent monadic perceptions and it is part of that essence to be the appearance of other monads (perceived as the organic parts). There may be more to the essence of body than that, so we can say schematically: (3)  Being a body = being the object of coherent monadic perceptions and being the appearance of monads perceived as organic parts and F1 where this is understood as specifying the essence of being a body, what it is to be a body, and F1 is just a placeholder for whatever other information is included in the essence of body. However, we do not need to complete the schema—we do not need to know what F1 is—to observe that this entails that Idealism about the Essence of Body is false: to be a body is not merely to be the object of monadic perceptions. It is also to be the appearance of monads to other monads. But this allows us to mount a further argument, against Idealism about the Essence of Bodies, since it is plausible that, for any body B, it is part of the essence of B (what it is to be B) to be a body (i.e. every body is essentially a body). But this means that the essence of any body can be given schematically as: (4)  For any X, for X to be B = X is a body and X is the unique body such that F2 where this is understood as specifying what it is to be B and F2 is a placeholder for whatever additional information is included in the essence of B. But we do not need to know how to fill in F2 because, substituting in (1), we can see that (2) is equivalent to: (2*) Being B = [being the object of coherent monadic perceptions and being the appearance of monads perceived as organic parts and F1] and X is the unique body such that F2 without knowing how to fill in F2 we can see that this means that there is more to being B than being the object of monadic perceptions, because being B involves at least being the appearance of some monads to other monads. So even without being able to fully specify what the essence of an individual body is (or what the essence of body is) we can see Idealism about the essence of individual bodies is false, on Leibnizian grounds. To be an individual body B is to be the object of

160  Bodies, Matter, Monads, and Things in Themselves certain coherent monadic perceptions, but also to be the appearance of various monads whose organic bodies are perceived as the infinitely enfolded organic parts of B and perhaps more. So, there are very general reasons why Leibniz would deny that to be B is merely to be the object of a set of coherent monadic perceptions. This is merely a sketch of an argument, but, if correct, it supplies what, I have argued, is missing from Rutherford’s interpretation: an account of why there is more to the essence of an individual Leibnizian body than merely being the object of monadic perceptions. But this strategy is also available to Adams. So, if Rutherford is successful in showing that Leibniz is a “Matter Realist” (in both senses distinguished in §1), then so is Adams.

4 Kant So far, we have distinguished several different kinds of idealist positions that Leibniz might hold, examined the logical relations among them, and considered which is the most defensible reading. In this section I distinguish corresponding positions within Kant’s philosophy. This comparison with Kant requires a change in formulation. Rather than monads, it is “things in themselves” that are said to be ground phenomena and consequently the representations that ground phenomena are not monadic perceptions, but subjects’ experiences. Human cognitive subjects, considered as they are “in themselves” rather than as they appear to themselves, are things in themselves, entities that do not exist in virtue of being experienced by other things in themselves—in Leibnizian terms, they are genuine substances. Translating the Leibnizian notion of body and matter into Kantian terms is slightly more complicated, though. “Matter” is an empirical concept, which Kant variously analyzes as “movable in space” (Metaphysical Foundations of Natural Science) or as “impenetrable, lifeless and extended” (Critique of Pure Reason). In Metaphysical Foundations Kant gives this definition of “body’: “a body, in the physical sense, is a matter between determinate boundaries” (Ak. 4:524). I take this to mean that “body” is a count noun; it refers to a determinate quantity of matter within specified boundaries. Consequently, it is an empirical concept, just like “matter” itself. That “matter” is an empirical concept means that, although all outer objects in space are material (“matter” is the highest empirical concept of nature), this is not determined by the forms of experience. So, the correct Kantian analogue to the Leibnizian concept “body” is “outer appearance” or “empirical object in space,” the objects that empirically happen to be material. But in order to avoid these clunky expressions, I am going to simply use the term “body.” In doing so, I am using the expression differently than Kant uses it.34 34  I do not mean to give the impression at this point that these Leibnizian notions can be translated “without remainder” in to Kantian ones. In Kantian terms, all monads are things in themselves but we

Nicholas F. Stang  161 We can now consider the Kantian analogues of the various idealist theses we considered in the context of Leibniz’s philosophy, for instance: Substance Idealism:  Every finite thing that exists is either a thing in itself, or exists in virtue of facts about the representational states of things in themselves.35 This claim partitions every finite thing into two classes: things in themselves, and things that exist in virtue of facts about things in themselves and the contents of experiences. Determining whether Kant is committed to Substance Idealism, so formulated, is difficult. On the other hand, some of Kant’s statements in his lectures on rational theology suggest that he might in fact have been a Substance Idealist; assuming that God is a thing in itself in the relevant sense, there are passages in his lectures on rational theology that suggest that Kant thinks that everything that exists, exists in virtue of God’s intellectual intuition of it.36 However, it is unclear whether these statements express Kant’s own considered theoretical commitments, because it is unclear what their epistemic status is. Within the Critical system what could warrant Kant in claiming that God is the ground of absolutely everything there is? While Kant does deploy such Critical doctrines as the ideality of space and time, the casual community of all phenomenal substances and the unknowability of things in themselves in the lectures on rational theology, he never squarely addresses the question of how, or whether, they are compatible with his Critical restrictions on theoretical cognition. He does claim that, while we cannot prove with apodictic certainty (i.e. a priori) that there is a God, we are subjectively required by our rational faculty to assume that there is a God and to conceive of him in certain determinate ways (the content of which he outlines).37 He claims further that while we cannot have positive knowledge of God’s inner nature, we can have negative knowledge that some conceptions of God are incorrect (e.g. Spinozism, polytheism, deism), and that apparent positive doctrines

cannot know whether all (indeed, any) things in themselves are monads or not. My point about “translation” should be read in the context of my remarks in the Introduction about the parallel structure of Leibniz’s and Kant’s ontology of matter. All I am claiming is that things in themselves and monads are playing a structurally similar role. 35  This does not presuppose that all things in themselves have representational states or are minds (something we could never know on theoretical grounds, according to Kant). It states that the existence of bodies depends upon those things in themselves that do have representational states (according to Leibniz, all of them). 36 E.g. Religionslehre Pölitz, Ak. 28:1054 (esp. within the larger context of 1033–1056). I think Kant’s considered view is that if there is a God and God is the ens originarium—the being whose existence grounds the possibility of all other beings—God grounds actually existing beings through his will not through his understanding (God does not have senses, so it is misleading to talk about his perceptions), which is the traditional position (see Mon 48). 37  Ak. 28:1036, 1046.

162  Bodies, Matter, Monads, and Things in Themselves about God (e.g. that he is a substance) are to be understood analogically.38 Fully unpacking this, and determining whether it constitutes a violation of Critical doctrines (either because Kant had misgivings about those doctrines, or because he was inclined to present a more positive “spin” to his critique of rational the­ ology when publicly exercising his duties as a servant of the Prussian state) or not, is a significant undertaking and I will not attempt it here. I bring it up merely to point out that determining whether Kant is really committed to Substance Idealism requires dealing with some of the most vexed questions in the Critical system (e.g. exactly what it means that we cannot have theoretical cognition of things in themselves). The next Idealist thesis we considered in Leibniz was: Weak Matter Idealism:  For all outer phenomena B, if B exists, B exists partly in virtue of facts about subjects’ experience of B. It seems fairly clear that Kant is committed to this. There is abundant textual evidence that bodies exist at least partly in virtue of subjects’ experiences of them.39 In fact, some of those texts appear to support attributing to him something stronger, namely: Idealism about the Existence of B odies:  For all phenomena B, if B exists, B exists wholly in virtue of facts about subjects’ experiences of B. However, there are reasons to think that Kant would reject this particular Idealist thesis. In the Critique of Pure Reason, he writes: “representation in itself does not produce its objects in so far as existence is concerned” (A 92/B 125). By “produce” [hervorbringen] I think Kant means “ground,” for surely the efficient causal production of objects by representation is not what is in question here (in a parenthesis he makes clear that he is here bracketing the relation of objects to the will, which can causally produce its object). In the larger context of the passage Kant is contrasting two ways in which representation and object can “meet and relate to one another”: either the object makes the representation possible, or the representation makes the object possible. He goes on to argue that our a priori concepts make their objects possible, but not in respect of their existence, merely their form. I take this to mean that the existence of the objects of our representations is not grounded in facts about the contents of our representations. One might assume that by claiming that representations do not “produce” the existence of their objects, Kant is making the anodyne point that I can have a

38  Ibid., 1048–1049.

39  See n. 10 above.

Nicholas F. Stang  163 non-­veridical perceptual episode (e.g. a hallucination) that represents an object outside of me without there being such an object. But can there be an ideally coherent and connected series of representations (let us assume, even across different subjects) of bodies, without those bodies existing? In other words, is it just that individual representations do not ground the existence of their object, but they do ground their existence at the limit of coherence and intersubjective agreement? Or, to put it in Leibnizian terms, does Kant hold that for bodies “existence consists in this, in sensation keeping certain laws . . . moreover sleep need not differ from waking by any intrinsic reality, but only by the form or order of sensations; wherefore there is no reason to ask whether there are any bodies outside us [because their existence is guaranteed by the coherence of our perceptions]” (A VI 3, 511)?40 One might think that, for Kant, the ideal coherence of a set of representations does guarantee the existence of its objects. Kant repeatedly makes claims about empirical objects that are most naturally read as meaning that for an empirical object to exist is just for us to experience it. For instance: Accordingly, the objects of experience are never given in themselves, but only in experience, and they do not exist at all outside it. That there could be inhabitants of the moon [daß es Einwohner im Monde geben könne], even though no human being has perceived them, must of course be admitted; but this means only that in the possible progress of experience we could encounter them [treffen könnten]; for everything is actual that stands in one context with a perception in accordance with the laws of the empirical progression. Thus they are real when they stand in empirical connection with my real consciousness, although they are not therefore real in themselves, i.e. outside this progress of experience . . . For the appearances, as mere representations, are in themselves real only in perception. To call an appearance a real thing prior to perception means either that in the continuation of experience we must encounter such a perception, or it has no meaning at all.  [A 493/B 521] One must note well this paradoxical but correct proposition, that nothing is in space except what is represented in it. For space itself is nothing other than representation; consequently, what is in it must be contained in representation, and nothing at all is in space except as it is really represented in it. A prop­os­ ition which must of course sound peculiar is that a thing exists only in the representation of it; but it loses its offensive character here, because the things with which we have to do are not things in themselves but only appearances, i.e., representations.  [A 374 n.]

40  Quoted in Adams (1994, 238).

164  Bodies, Matter, Monads, and Things in Themselves First, a note on how I am interpreting these texts. One might be tempted to read the second sentence of the first passage as meaning: if it is possible for there to be men on the moon, this just means it is possible for us to experience them. But this reading—on which Kant is merely asserting a connection between possibilities— is contradicted by the final sentence: “to call an appearance a real thing prior to perception means either that in the continuation of experience we must encounter [treffen müssen] such a perception, or it has no meaning at all.” If Kant were merely equating the possibility of there being moon-­men with the possibility of experiencing moon-­men, why would he write that to say that an appearance exists before we have perceived it is to claim that we must eventually experience it? Now, it might be objected that Kant here is making only a claim about a sufficient condition on actual existence: anything connected with our perceptions by empirical causal laws is actual, although there may be actual objects that we neither experience either by directly perceiving or indirectly perceiving their traces. But then his claim about moon-­men would be undermined: if it is possible for there to be actual objects that are never in any sense experienced, then the possibility of there being moon-­men does not reduce to the possibility of our experiencing moon-­men. In the second passage Kant makes as clear as one could hope that, for an object in space, to exist is to be represented. Now, the second passage leaves open what kind of representation is sufficient to ground the existence of an object in space; surely Kant does not mean to deny that I can have a hallucination as of a pink elephant, even if no pink elephants exist in space. The natural answer to this question, in light of the first passage, is that it is experience of objects in space (bodies) that is sufficient to ground their existence. The idea that experience of bodies is sufficient to ground their existence may strike some readers as absurd. Surely, I can have a hallucinatory experience of an object when in fact no such object exists? The key point here is that Kant’s technical notion “experience” [Erfahrung] does not refer to just any perceptual episode with objective purport but consists in at least a unified and coherent set of perceptions that exhibit exception-­less empirical laws and whose content obeys the transcendental principles of experience (it represents an absolutely permanent set of substances that underlie all apparent generation and corruption). As Kant says at the outset of the A Deduction: there is only one experience, in which all perceptions are represented as in thoroughgoing and lawlike connection, just as there is only one space and time, in which all forms of appearances and all relation of being or non-­being takes place. If one speaks of different experiences, they are only so many perceptions insofar as they belong to one and the same universal experience.  [A 110]

So, in this sense, hallucinations and non-­veridical perceptions would not count as experiences because they do not cohere with the “universal experience,” which, I have been arguing, grounds the existence of phenomena.

Nicholas F. Stang  165 Determining exactly what it takes for a connected series of perceptions to constitute experience lies far outside the scope of this essay; indeed, it would take little less than a commentary on the entire Transcendental Analytic. But even without answering that monumental question we are in a position to see an apparent tension in Kant’s view. On the one hand, he denies that “representation in itself ” grounds the existence of its objects. On the other hand, he holds that experience grounds the existence of its objects. The answer to this question lies in what Kant means by the phrase “representation in itself.” The phrase suggests that Kant means to be talking of representations considered solely in themselves, or from the side of the subject or subjects having them. In this sense, “representation in itself ” does not require an external object, although it may represent there being one. “Representation in itself ” is related to what philosophers now call “narrow content.” Any state subjectively indistinguishable from my present state is the same state of “representation in itself ”; it has the same narrow content, assuming the common view that narrow content supervenes on phenomenal states.41 However, two subjectively indistinguishable states that are the same considered as “representations in themselves” may differ in important respects: one may veridically record the existence of an external object, while the other does not. In order to avoid running afoul of Kant’s doctrine that only a connected series of representations is an experience, let us focus not on episodic representations but on the entire representational sequence of a single mind over its lifetime. If I am correct about what Kant means by “representation in itself ” at A 92/B 125, then his claim that representation in itself does not produce the existence of its object is compatible with his view that the existence of empirical objects is grounded in the contents of experience only if experience is not merely “representation in itself ”; that is, if whether a representational sequence is an experience does not depend only on “the subjective side,” its narrow content. How could it be that experience is not merely representation in itself (i.e. that whether I am experiencing or merely seeming to experience does not supervene on the qualitative character of my representational sequence)? One of the central tenets of the Critique of Pure Reason is that experience requires sensibility, which he defines as “the capacity to acquire representations through the way in which we are affected by objects” (A 19/B 33); the Introduction to the B edition begins: “how else could the cognitive faculty be awakened into experience if not through objects that stimulate our senses and in part themselves produce representations . . . ?” (B 1). This naturally gives rise to the question: which are the objects that stimulate our senses, produce representations and thus “awaken” the cognitive faculty into experience? Given that Kant ends the first 41  “Narrow content” in contemporary philosophy typically refers to mental (or semantic) content (if there is any) that supervenes on the intrinsic states of a subject. “Wide content” is content that does not so supervene. Obviously, I am using these terms in a slightly broader sense in Kant and Leibniz. The classic defenses of wide content (and inspirations for the distinction) are Putnam (1975) and Burge (1979).

166  Bodies, Matter, Monads, and Things in Themselves paragraph of the B edition by saying “as far as time is concerned, then, no cognition in us precedes experience,” one might think that he has in mind exclusively a process of causal affection that occurs in time and, thus, the affecting objects can only be spatiotemporal empirical objects. However, there are compelling reasons, both textual and philosophical, for thinking that Kant also admits a noumenal affection of our sensibility by non-­empirical objects, things in themselves. First of all, he straightforwardly says as much in On a Discovery, and numerous other texts: Having raised the question “Who (what) gives sensible sensibility its matter, namely sensations?” [Eberhard] believes himself to have pronounced against the Critique when he says “We may choose what we will—we will never arrive at things in themselves.” Now that, of course, is the constant contention of the Critique; save that it posits this ground of the matter of sensory representation not once again in things, as objects of the senses, but in something super-­ sensible, which grounds the latter, and of which we can have no cognition. It says that the objects as things-­in-­themselves give the matter to empirical intuition (they contain the ground by which to determine the faculty of representation in accordance with its sensibility), but they are not the matter thereof. [“On a Discovery,” Ak. 8:215]42

Aside from the historical question of whether Kant accepted non-­empirical affection (to which the answer must be a resounding “Yes!”) there is the philosophical question of whether he should. Consider, however, the following argument: (1) Empirical objects exist partly in virtue of the sensory content of experience. (2) If the fact that p is among the grounds of the fact that q then the fact that q does not cause it to be the case that p. (3) If empirical objects affect the subject to produce the sensory contents of experience, then the fact that they exist causes it to be the case that ex­peri­ ence has the sensory content it does. (C1) ∴ Empirical objects do not affect the subject to produce the sensory contents of experience. (C2) ∴ If the sensory content of experience has a causal ground, it is not in empirical objects. (4) Receptivity is the faculty of being affected by objects distinct from the subject and thereby acquiring sensory representations. So, the sensory content of experience has a causal ground in objects distinct from the subject. (C3) ∴The sensory content of experience has a causal ground in non-­ empirical objects distinct from the subject. 42  Cf. A 190/B 235; A 387; A 494/B 522; Ak. 4:289, 4:314, 4:318, 4:451.

Nicholas F. Stang  167 This is an argument that the objects that affect our sensibility and produce the sensory representations which the faculty of understanding goes about synthesizing, combining, etc. are not empirical objects. Intuitively, the idea of the argument is that, since empirical objects are appearances, they exist in virtue of the contents of experience, and thus by (2) cannot be among the causes of that sensory content; appearances cannot “reach” back to cause the very experiences in virtue of which they exist. F. H. Jacobi famously quipped that “without the presupposition [of the thing in itself] I was unable to enter into [Kant's] system, but with it I was unable to stay within it.”43 I take it that the first half of this remark expresses the argument up to (C2)—that the thing in itself and noumenal affection are indispensable to Kant’s theory of experience—and the second half of the remark continues the objection to (C3) and then raises the objection: Kant’s restriction of the categories to empirical objects does not allow us to draw (C3). Kant’s theory of experience both requires, and does not allow him to posit, noumenal affection by things in themselves. Note that (C1) and (C2) are not subject to the objection that categories like “cause-­effect” can only be applied to empirical objects. Even if (4) inadmissibly applies “cause” effect to non-­ empirical objects, (C2) stands. However, I think this objection misunderstands Kant’s “restriction” thesis. Kant holds that the categories are cognitions only with respect to empirical objects; we can only know that empirical objects fall under the categories, and we can only know principles involving the categories (e.g. the persistence of substance) if those principles are restricted to empirical objects. Not only can we think about any object as falling under the categories; we must, because we cannot think about any object whatsoever without applying the categories to it!44 The respect in which the Jacobi-­ inspired objection is correct is that Kant’s Critical epistemology seems to entail that we cannot know through theoretical reason premise (4) or (C3). Thus, while it is clear that Kant is committed to the noumenal affection of sensibility by non-­empirical objects the Jacobi objection points out a serious problem about how he could be rationally warranted in adopting that commitment, but I will not further pursue the point here. If I am indeed correct that Kant thinks that because the receptivity of sensibility requires affection by a non-­empirical object, this offers a natural explanation of why experience is not “representation in itself ” and thus why Kant’s claim at A 92/B 125 that “representation in itself ” does not “produce” the existence of its object is compatible with claiming that the existence of empirical objects is grounded in experience of them: to be experiencing is not merely to enjoy a representational sequence with certain highly unified contents, but for that representational sequence to be the product of causal affection by things in 43  From Jacobi (1787), vol. II, 109. 44  See B xxvi; A 88/B 120; B 167 n.; A254/B 309; Ak. 5:43 and 5:55.

168  Bodies, Matter, Monads, and Things in Themselves themselves. In other words, experience necessarily requires affection by something “transcendentally external” to the subject (or subjects) of experience. This proposal—that experience for Kant is by definition the product of causal affection by things in themselves—is confirmed by Kant’s repeated insistence throughout the Critical philosophy that the idea of an appearance that is not the appearance of something that is not itself an appearance (i.e. the appearance of a thing in itself) is absurd. Kant repeatedly maintains that since empirical objects are appearances, there must be objects that are non-­empirical, and hence which are not appearances, which appear as those empirical objects.45 This is not a causal inference to the existence of things in themselves, but a conceptual requirement on what it is to be an appearance: if x is an appearance, then there is a y such that y is not an appearance and y appears as x. Kant also seems to conclude that if this y is not an appearance it must be a thing in itself. This fits well with the current proposal—that experience is essentially the product of noumenal affection—because appearances are essentially the appearances of thing in themselves, and experience is essentially causally related to things in themselves, then it stands to reason that experience is essentially the experience of appearances; that is, objects that exist in virtue of things in themselves appearing to us by causally affecting us. What does this have to do with whether Kant is committed to Idealism about the Existence of Bodies? First of all, it shows that it matters how we translate the Leibnizian notion of “perception” into Kantian terms. For consider the following views Kant might hold: R-­I dealism about the Existence of B odies:  For any body B, if B exists, B exists wholly in virtue of facts about the contents of subjects’ representations of B. E-Idealism about the Existence of B odies:  For all bodies B, if B exists, B exists wholly in virtue of facts about the contents of subjects’ ex­peri­ ences of B. In Kant’s technical terminology, R-­Idealism is the claim that representations in themselves (i.e. merely in virtue of their narrow content) ground the existence of bodies; as we have seen, he rejects this view. Whether he holds E-­Idealism depends upon whether I am right that experience essentially involves noumenal affection by things in themselves; that is, whether it is possible to experience bodies without that experience constituting the appearance of things in 45 The locus classicus for this view is Kant’s following remark in the B preface: “the reservation must also be well noted that even if we cognize these same objects as things in themselves, we at least must be able to think of them as things in themselves. For otherwise there would follow the absurd prop­os­ ition that there is an appearance without anything that appears” (B xxvi). Cf. Ak. 4:315, 354.

Nicholas F. Stang  169 themselves as those bodies. If I am right that, for Kant, experience necessarily involves affection by things in themselves then Kant is committed to E-­Idealism. If a body exists, then its existence is grounded in the fact that noumenal affection produces in subjects representations (with the appropriate content) which ground an experience of that body such that those things in themselves appear as that body. But notice further that Kant’s commitment to E-­Idealism does not constitute a difference from Leibniz, who, I argued in §1, rejects Idealism about the Existence of Bodies. The reason this does not constitute a difference with Leibniz is that Leibnizian “perception” is a notion of “representation in itself ”; whether a mental state is a Leibnizian perception supervenes on its narrow content. That a state is a perception depends only upon the “subjective side”; it does not require that there be an external object answering to that state (causally or otherwise). So, Leibniz and Kant agree that the mere fact that subjects enjoy representational mental states as of bodies does not entail that there are such bodies; for each thinker more must be added to ground the existence of bodies. For Leibniz, there must be monads “in” each part of the represented bodies. Ultimately, this means there must be an infinitely enfolded series of organic bodies, each dominated by a monad perceiving the entire world from the point of view of that organic boy. For Kant, it means that the subjects’ experience of bodies must constitute the appearance of things in themselves to those subjects as those bodies, which requires at least that there be things in themselves causally affecting those subjects, giving rise to the sensory contents that are synthesized into experience of those bodies. The remaining idealist doctrine we need to consider is: Idealism about the Essence of B odies: (I) The essence of being a body is being the object of subjects’ representations (with appropriate content), and (II) For any body B, the essence of B is being the object of a certain set of subjects’ representations (with appropriate content).46 46  The following objection may occur to some readers. There is also a strand in Kant’s thinking that is extremely pessimistic about our ability to know essences; if this pessimism about essence represents Kant’s considered view, then he must be agnostic vis-­à-­vis both conjuncts of Idealism about the Essences of Bodies. In various texts Kant distinguishes between the logical essence of a concept and the real essence of the object or objects of the concept. The logical essence of the concept is the analytic marks we think in the concept—for instance the logical essence of is , and —while Kant defines real essence as “the first inner ground of all that belongs to the possibility of a thing” (Mrognovius, Ak. 29:820). I take this to mean that for Kant the real essence of a thing is a complexes of properties that constrain what is possible for those objects: anything that is a consequence of the essence is a necessary property of the object, while anything that is compatible with the essence but is not entailed by it is a contingent property of the object. The properties that make up the essence are the essential properties of the object (essentialia). Kant repeatedly claims that we cannot know real essences, notably in his letter to Reinhold of May 12, 1789 (see Ak. 11:37). That text might seem especially damning for my project, since it is the real essence of matter/body itself that Kant there claims we cannot know. However, when put in the context of other texts, this text is in fact fully compatible with my interpretation. Kant, following Baumgarten, distinguishes between the essence of

170  Bodies, Matter, Monads, and Things in Themselves Just as with Leibniz, whether Kant accepts the first conjunct of Idealism about the Essence of Bodies—which I will call Idealism about Being a Body—depends upon whether the representations involved are mere representations or experience. For, as we have seen, there is more to being a body than being the intentional content of representation; to be a body is to be the intentional object of representations with a certain content in which things in themselves are appearing as that body. So, if what Kant calls “representation in itself ” is the relevant notion, Kant rejects Idealism about Being a Body. Does he accept it as formulated with experience, where experience is understood as involving appearance of things in themselves to subjects? E-­I dealism about Being a B ody:  To be a body is to be the intentional object of a certain set of subjects’ experiences (with appropriate content). On this view, to be a body is to be represented by experiences with a certain content—for instance, to be represented as an impenetrable, lifeless extended object (to adopt the analysis of in the Critique of Pure Reason rather than in the Metaphysical Foundations)—where that experience is the product of noumenal affection by things in themselves that appear in the content of those experiences as those bodies. Determining Kant’s views on these matters is always somewhat difficult because he never states his views on the ontology of phenomena very clearly, and he thinks our knowledge of the noumenal basis of phenomena is very limited, but this seems to express Kant’s view about what it is to be a body. Now let us turn to the second half of Idealism about the Essence of Bodies: for any body B, the essence of B is to be the object of certain experiences. Call this “Idealism about the Essences of Individual Bodies.” In §3 I sketched a short Leibnizian argument against Idealism about the Essences of Individual Bodies. I think that a similar argument is available to Kant. The crucial premise here is that every body is essentially a body. I know of only one text, from the Duisberg Nachlaß, in which Kant explicitly addresses this issue. In this unpublished note Kant is discussing the comparison of concepts in judgments, and distinguishes three cases, of which the third is: an object, the properties that make up the essence (essentialia), the properties that follow necessarily from those properties (attributa), and the contingent properties of the object that are compatible with, but sufficiently grounded in, its essence (affectiones). Even if we cannot know the essence of we may be able to know some part of its essence, specifically, we might know that it is part of the essence of matter to be an appearance of things in themselves, and we may be able to know some of the properties that follow from the essential properties of matter. In other logic lectures, a context in which he frequently claims that we cannot know the real essence of objects, Kant explicitly claims that we can know some of the essential properties of matter; see Metaphysik Pölitz, Ak. 28:553. I conclude that this strand of skepticism about our knowledge of essences does not preclude Kant from adopting an Idealism about the essence of body, or its negation. Kant might hold that we know part of the essence of body, and that it includes being the appearance of things in themselves, or that we know that, whatever the essence of matter is, it does not include things in themselves, without claiming that we can know the complete essence of matter.

Nicholas F. Stang  171 But if a cannot be separated from b in x, i.e. no x which is a body is indivisible, then one must see that the x which is thought through a can never be thought through non-­a, that no being which has the nature of a body can become incorporeal, and that the a in itself is not a predicate in respect of x, but a reciprocal concept [sondern mit ihm Wechselbegriff sei] and thus holds of a substance.  [R4767; Ak. 18:654]47

Kant here asserts two different necessary connections: an object x which cannot fail to fall under the concept a and the concept b is contained in a. Consequently, x cannot fail to fall under b. The second necessary connection is the familiar Kantian one of analytic containment. The first necessary connection is more problematic. Kant’s characterization of this necessary connection in the last sentence suggests that he has in mind the idea that a substance and its essence are only “rationally distinct”; the essence of x is not a predicate of x but a merely a way of expressing x’s nature. The object x and its essence a are “reciprocal concepts” because it is impossible for x to exist without a being its essence; a can be substituted for x in all contexts, even modal ones, salva veritate. Kant’s example of the relation between an object, an essence and an essential property contained in that essence is: a body x, being a body, and being impenetrable. If every body is essentially a body, and being a body necessarily involves being the appearance of things in themselves to subjects, as I argued earlier, then the essence of any individual body entails being the appearance of things in themselves (this is an attribute of any individual body). So, being an individual body is more than being the intentional object of certain representations “in themselves.” But once again, whether Kant is an Idealist about the Essences of Individual Bodies depends upon whether the representations involved are “mere representations” or experience, which necessarily involves the appearance of things in themselves to subjects: R-­I dealism about the Essences of Individual B odies: For any phenomenon B, the essence of B (= what it is to be B) is to be the object of a certain set of representations “in themselves.” E-­I dealism about the Essences of Individual B odies: For any phe­ nom­enon B, the essence of B (= what it is to be B) is to be the object of a certain set of experiences.

47 It has long been appreciated that the Duisberg Nachlaß can shed light on the Critique; Longuenesse (1998) in particular offers an extensive argument that Kant’s Critical conception of the role of the logical function of categorical judgment and its relation to the category of substance should be understood in light of this passage. Although it is a pre-­Critical text, I think that nothing in Kant’s mature doctrine in the Critique is incompatible with.

172  Bodies, Matter, Monads, and Things in Themselves But the argument I just sketched is only a reason for Kant to reject R-­Idealism about the Essences of Individual Bodies: it does not entail that the essence of an individual body is not to be the object of a certain experience because being an object of experience necessarily involves being the appearance of things in themselves to subjects, and all I have argued is that individual bodies are essentially the appearances of things in themselves. Let us now focus on E-­Idealism about the Essences of Individual Bodies, the idea that all there is to being a given body is being the object of a certain set of experiences. Let us assume for the moment that experiences are individuated by their representational content and perhaps by the subjects whose experiences they are; they are not individuated by the things in themselves that appear in those experiences, although experience is essentially the appearance of some things in themselves. (In other words, experiences are essentially appearances of things in themselves to subjects, but not essentially the appearance of any particular things in themselves). Perhaps it is easiest to get a handle on the content of E-­Idealism by understanding what it denies. Take a particular body B, say, the boat sailing downstream from the Second Analogy of Experience. If Idealism about the Essences of Individual Bodies is true, then while this boat is essentially an appearance of things in themselves, it is not essentially the appearance of any particular things in themselves. No particular things in themselves enter into the essence of B; to be B is to be an object represented as having certain properties in an experience that is the appearance of things in themselves to subjects, but not any particular things in themselves. To get a more intuitive grasp on this issue, notice that on so-­called “One Object” readings of Kant’s idealism an individual appearance is essentially the appearance of an individual thing in itself; namely, the thing in itself to which it is numerically identical. Let “B” refer to the appearance, in this case, the boat. Let “T” refer to the thing in itself that appears as B, if there is only one. On “One Object” readings, B = T. Since B is numerically identical to T, in fact, it is simply T qua appearing to us, this very appearance B could not exist without being the appearance of T (i.e. without being T qua appearing to us). So insofar as one is tempted by a One Object reading of Kant’s idealism, there is strong reason to think that Kant would reject Idealism about the Essence of Individual Bodies, and hold that it is part of the essence of a given body to be the appearance of a particular thing in itself in itself, namely, the thing in itself it, considered independently of how it appear to us, is. Independently of whether one is a “One” or “Two” Object reader of Kant’s ideal­ism, though, there are very general reasons to expect that Kant would not accept Idealism about Essences of Individual Bodies. If we are truly ignorant of the noumenal “side” or “aspect” of appearances—that is, we are ignorant of the intrinsic nature of the things in themselves that affect us and appear to us as these objects—it would stand to reason that, for all we know, individual bodies are

Nicholas F. Stang  173 essentially appearances of a particular thing in itself or group of things in themselves. Within Kant’s epistemology, what would rationally warrant us in excluding the possibility that for a body B there is some group G of things in themselves such that B is essentially the appearance of the members of G to us? According to this line of reasoning, Kant should neither accept nor reject Idealism about the Essences of Individual Bodies; we cannot know whether it is part of the essence of individual bodies to be appearances of particular things in themselves. I am not sure this exhausts the reasons for accepting or rejecting this Idealist thesis in the context of Kant’s system; there may be reasons originating from this theory of the will, practical agency, or the immortality of the soul for making more determinate claims about the involvement of particular things in themselves in the essences of individual appearances. But for reasons of space, I will have to leave the discussion there. Our examination of Leibniz’s and Kant’s views on the ontological status of bodies and matter has revealed that in addition to the structural similar tension in both philosophers—there are texts that support a phenomenalist reading and texts that suggest a more realist reading of bodies—they agree to a surprising extent, once we abstract from differences in their background theory (e.g. Leibniz’s monadology, Kant’s doctrine of noumenal ignorance). Specifically, we have seen that each of them denies a version of the following thesis: Idealism about the Existence of B odies:  For any body B, the existence of B is grounded solely in facts about the representational contents of ­monads/subjects’ perceptions/representations of B. Substituting the appropriate Leibnizian or Kantian terms, this is a view each philosopher would reject. Similarly, both would accept the weaker claim that: Weak Idealism about B odies:  For any body B, the fact that B exists is grounded partly in facts about the representational contents of monads/subjects’ perceptions/representations of B. And their reasons for accepting Weak Idealism but rejecting Strong Idealism are the same: the facts that ground the existence of bodies are not merely facts about the “narrow” representational content of monads’/subjects’ representations but require, in addition, facts about the grounds of those representations in monads/ things in themselves. In Leibniz this is the requirement that there be monads in each organic part of the body that perceive the whole universe of bodies “as if ” from that spatial location. In Kant it is the requirement that there be things in themselves appearing as the body in virtue of noumenal affecting subjects and thereby producing the experience of that body. What this brings out is the main difference between Leibniz and Kant over the “extra factor” that must be required

174  Bodies, Matter, Monads, and Things in Themselves to ground the existence of bodies (beyond the narrow representational contents of subjects’/monads’ perceptions) follows from differences in their background theory: Leibniz’s denial of intersubstantial causation and Kant’s doctrine of noumenal ignorance. Because Leibniz denies the possibility of intersubstantial causation, the “appearance” relation by which monads appear to one another as a world of infinitely enveloped organic bodies is necessarily non-­causal. Leibniz cashes out this appearance relation in terms of an isomorphism between the world of monads and the phenomenal world of bodies. Likewise, Kant is much more agnostic about the involvement of things in themselves in bodies than Leibniz is because he puts much more radical limits on our knowledge of the extra-­phenomenal reality of things in themselves. Where Leibniz is willing to posit an infinity of monads, one dominating each organic part of each body, Kant thinks we are irredeemably ignorant of the intrinsic natures of things in themselves and the specific details of their involvement in bodies. This puts a limit on how similar their theories can ultimately be. I would like to conclude by discussing one final similarity in their theories of the ontological status of bodies and matter. Leibniz and Kant share an anti-­ Cartesian conception of matter, according to which bodies possess dynamic properties and real moving forces. And both philosophers claim that the forces possessed by bodies are the appearance or manifestation of the forces of monads (Leibniz) or things in themselves (Kant). Leibniz is clearer on this point, for he holds that primitive forces are perceptual capacities in monads, while derivative forces, the moving forces possessed by bodies studied by the science of dynamics, are manifestations or appearances of those primitive forces. Consider these two texts from the De Volder correspondence: Derivative forces I relegate to the phenomena, but I think it is clear that primitive forces cannot be anything but internal tendencies of simple substances, because of which, by a certain law of their nature, they pass from perception to perception.  [GP II 275/AG 181] The forces that arise from mass and velocity are derivative and belong to aggregates or phenomena [bodies]. When I speak of the primitive force remaining, I do not mean the conservation of the total power to move, which was discussed between us earlier, but the entelechy which always expresses that total force as well as other things. And certainly derivative forces are nothing but modifications and results of the primitive [forces].  [GP II 251/L 530]

What does it mean for a force that is essentially a force of motion in space and time to be the appearance of a non-­spatial force of perception and appetition in a nonextended simple substance? Ordinarily, if we say that x appears as y we are assuming at least that (i) x is among the causes of our perception of y, and (ii) x’s

Nicholas F. Stang  175 perceptible features are responsible for y’s perceptible features in some relatively systematic way. This model is significantly complicated when x is something that cannot be directly perceived; that is, when x is something that we can only perceive indirectly by perceiving an object y where x appears to us as y. If we drop the requirement that the features that manifest themselves in the appearance y are perceptible features of x, then we get some traction on Leibniz’s view. There is a systematic correlation between the features of monads and the features of bodies as they appear to us: monads perceive the entire corporeal universe with varying degrees of clarity and these differences in clarity correspond to their spatial positions and their relations of domination. If monad m perceives monad n more distinctly than monad n perceives monad m then the organic body of m contains the organic body of n. So, there is a systematic correlation between the force of perception in monads and its degrees of clarity and distinctness and relations of containment in organic bodies. It is somewhat harder to see how there can be systematic correlation between the derivative forces in bodies and the perceptive and appetitive forces in monads. This is in part because of Leibniz’s unclarity surrounding the derivative forces of bodies. But if we assume that bodies have derivative force in virtue of being perceived by monads as having derivative force, then we can say the following: the derivative forces of bodies supervene globally on the degrees of clarity and distinctness present in the perceptual forces of monads. So, there is a function from the total perceptual state of all the monads to the complete set of derivative forces and motions possessed by bodies. Some sense can be made of the idea that primitive forces appear as derivative forces. Kant also describes the forces of things in themselves as appearing in empirical objects, although (appropriately, given his epistemology) he is more reticent about this than Leibniz is. Kant claims that our empirical character is an appearance of our intelligible character (A 538–558/B 566–586); in the Groundwork he claims that “the world of understanding contains the ground of the world of sense and of its laws” (Ak. 4:453); and in the Prolegomena he writes that “reason is the cause of these natural laws and is therefore free” (Ak. 4:346).48 However, he is in a worse position than Leibniz to account for this. There is a problem, we saw earl­ ier, in understanding how the features (forces or otherwise) of a non-­spatial 48  This view is more clearly articulated in the Inaugural Dissertation of 1770, in which Kant is more optimistic about our knowledge of noumena: “the human mind is only affected by external things, and the world is only exposed to its view, lying open before it to infinity, in so far as the mind, itself, together with all other things, is sustained by the same infinite force of one being. Hence, the mind only senses external things in virtue of the presence of the same common sustaining cause. Accordingly, space, which is the sensitively cognized universal and necessary condition of the co-­ presence of all things, can be called phenomenal omnipresence” (Ak. 2:409–410); see also Ak. 2:357, 2:391, and 2: 407. If space is the appearance of divine omnipresence (the way in which divine omnipresence manifests to us, given our form of intuition) then the forces encountered in space, it stands to reason, are the appearances of the forces sustained by the infinite force of one being, God. Whether this remains Kant’s Critical view would require further substantiation.

176  Bodies, Matter, Monads, and Things in Themselves object can appear as the features (forces or otherwise) of spatiotemporal object. Bracketing problems about the unschematized use of the cat­egor­ies, Kant can claim that forces in things in themselves appear as the forces in empirical objects in the following sense: things in themselves affect us, which produces in us ex­peri­ence of empirical objects and this experience represents those empirical objects as having forces that correspond in some way to the forces by things in themselves affect us. Whether this is a genuine sense in which the forces in things in themselves appear as the forces in empirical objects depends upon what “cor­ res­ponds in some way” means. The mere fact of causal dependence is not enough to make it the case that empirical forces are appearances of noumenal forces; causal dependence is compatible with a weaker relation, on which noumenal forces are merely the external cause of empirical forces. What more is required for noumenal forces to appear as empirical forces? What more do we need to pack into “corresponds in some way”? Leibniz had a natural answer at his disposal: due to the pre-­established harmony, there is an isomorphism between the (degrees of clarity and distinctness possessed by) perceptual forces in monads and the motive forces in bodies. It is not clear there is anything satisfactory for Kant to say here; it is not clear that Kant can vindicate the idea that noumenal forces appear as empirical forces without violating the strictures of his own epistemology. I began this chapter by discussing the tension between phenomenalism and a more realist view of matter in Leibniz and Kant. Leibniz maintains that it is only because the primitive forces of monads manifest as the motive forces of bodies that bodies have more reality than the objects of coherent dreams, or phantasms.49 The main systematic purpose to which Kant puts the claim that noumenal objects appear as, rather than merely cause, empirical objects, is in the context of his theory of the will: because our noumenal character appears as our empirical character, we have some rational warrant to hope that a continuously improving empirical character (the best Kant thinks is possible for radically evil creatures like us) is a sign of a noumenal character that has not completely subordinated the moral law to self-­interest. This suggests that for both philosophers the main sense in which empirical objects are not mere phenomena is that the causal ­powers of non-­empirical objects, substances in their own right, appear in those empirical objects. Perhaps the best place to look to reconcile the phenomenalist and realist strands in Leibniz’s and Kant’s philosophy is their theory of non-­ empirical forces and causes and how these manifest in the empirical world. But that will have to wait for another occasion.

49 “On the Method of Distinguishing Real from Imaginary Phenomena” (A VI 4, 1500ff./L 363–365)

7 Kant and the (Alleged) Leibnizian Misconception of the Difference between Sensible and Intellectual Representations Anja Jauernig, New York University*

The Leibnizian philosophy was a preferred target for Kant for most of his career. He attacks the Leibnizians on various fronts but the objection that occurs most frequently in his writings is that their conception of the difference between sensible and intellectual representations is untenable. Kant spells out this complaint in different ways but the central objection is that the Leibnizians regard intellectual representations as more fundamental and understand the difference between them and sensible representations as a merely “logical” difference, concerning their form—namely, their degree of distinctness and confusedness— while in truth it is a difference in kind, concerning their nature, origin, and con­ tent.1 For ease of communication, I shall call this criticism the “misconception objection,” and use the expression “the misconception” as short for “the Leibnizian misconception of the difference between sensible and intellectual representations.” The misconception objection has received its share of attention in the litera­ ture. Several Leibniz scholars have undertaken to defend Leibniz against it,2 and

*  I presented earlier versions of parts of this chapter at the joint meeting of the North American Kant Society and the Leibniz Society of North America in Lexington, Kentucky, and at the Oxford Seminar in Early Modern Philosophy in Oxford, UK, both in 2009. I want to thank both audiences for helpful questions. 1  See FM, Ak. 20:278: “Regarding the Leibnizian principle of the logical difference of the indistinct­ ness and distinctness of representations, when he claims that the former kind of representation, which we called mere intuition, is basically only the confused concept of its object, and, thus, intuition is distinguished from concepts of things merely with respect to the degree of consciousness, not spe­cif­ ic­al­ly . . . ” See FM, Ak. 20:285: “a kind of enchanted world, to the assumption of which the famous man [Leibniz] could only have been misled by taking sensible representations as appearances, and not, as it should be, for a kind of representation that is completely distinct from all concepts, namely, intuition, but for a cognition, albeit confused, through concepts, which have their seat in the under­ standing, not sensibility.” See Anth, Ak. 7:140–141 n.: “To posit sensibility [to consist] merely in the indistinctness of representations, but intellectuality in distinctness, and, thereby, to posit a merely formal (logical) difference of consciousness instead of a real (psychological) [difference] that concerns not only the form but also the content of thinking, was a big mistake of the Leibniz-­Wolffian school . . . ” See ÜE, Ak. 8:219–220. 2  See McRae (1976, esp. ch. 5); Brandom (1981); Parkinson (1981, 1982); Wilson (2005). Anja Jauernig, Kant and the (Alleged) Leibnizian Misconception of the Difference between Sensible and Intellectual Representations. In: Leibniz and Kant. Edited by Brandon C. Look, Oxford University Press. © Oxford University Press 2021. DOI: 10.1093/oso/9780199606368.003.0007

178  Kant and the (Alleged) Leibnizian Misconception most Kant scholars who discuss Kant’s own conception of the relation between sensible and intellectual representations at least mention the misconception by way of contrast. It is more difficult, however, to find detailed reconstructions of what exactly the misconception objection amounts to,3 or more explicit discus­ sions of the precise relation of the misconception to other problems that Kant identifies with the Leibnizian philosophy, or of Kant’s underlying concerns that would explain why he regards the misconception as so significant that he keeps coming back to it over and over again.4 These underresearched aspects of the misconception objection are the focus of the first part of this chapter. The ques­ tion of whether the underlying worries that fuel Kant’s interest in the misconcep­ tion apply to Leibniz’s philosophy has also been largely neglected so far. This question will be taken up in the second part of this chapter. For the purposes of the following discussion, I will set aside the thorny problem of whether Kant’s objections are aimed at Leibniz himself, or whether Wolff, or some other Leibniz-­Wolffians are the primarily intended addressees. I shall use the label “the Leibnizians” as an intentionally vague cover term to signify Kant’s target, with the understanding that more could and should be said at a future occasion about whose doctrines Kant takes himself to be criticizing. Another limitation of our discussion is that we will only talk about the misconception objection as it plays out in Kant’s critique of Leibnizian epistemology, philosophy of mind, and metaphysics. There is another version of the objection that targets the Leibnizian classification of the concept of beauty as a confused concept, namely, a confused concept of perfection.5 This version of the misconception is part of an attack of Leibnizian aesthetics and will not be considered here.

2  The Misconception Objection Explained What exactly are the representations whose difference the Leibnizians allegedly mischaracterize, according to Kant? The relevant contrast between sensible and intellectual representations at the heart of the misconception objection is closely related to the contrast between intuitions and concepts, that is, between representations produced by sensibility, which are singular and immediately relate to their objects, on the one hand, and representations of the understanding, which are general and only mediately relate to their objects, on the other hand.6 More specifically, the contrast that Kant has predominantly in mind is between intuitions, on the one hand, and non-­sensible concepts, on the other hand, that is, 3  Some useful remarks can be found in Martin (1966). 4  Some relevant considerations appear in Finster (1986); Langton (1998, esp. 73–74, 89–93); and Schneider (2004). 5  See KU, Ak. 5:228–229; EEKU, Ak. 20:226–227. 6  See B 33/A 19; A 320/B 376–377; Prol, Ak. 4:282; Log, Ak. 9:91; FM, Ak. 20:325; Anth, Ak. 7:196.

Anja Jauernig  179 concepts that do not contain any sensible elements and have their origin exclusively in our intellectual faculty. All non-­sensible concepts are a priori, but the converse does not hold. In addition to empirical concepts, Kant also classifies the a priori concepts of mathematics as sensible, on the grounds that they contain their sensible application conditions, that is, their schemata, which originate in the imagination, the faculty that mediates between sensibility and the under­ standing.7 Also note that Kant’s critique does not only extend to the Leibnizian conception of sensations and empirical intuitions, that is, sense perceptions, but also to their conception of our representations of space and time, which Kant identifies as a priori intuitions. In sum, Kant’s objection concerns the Leibnizian conception of the difference between (what Kant calls) a priori and a posteriori intuitions, on the one hand, and non-­sensible (and, thus, a priori) concepts, on the other hand. What does it mean for a representation to be distinct or indistinct, on Kant’s view? The difference of the form of cognition is based on a condition that accompanies all cognition, namely, consciousness. If I am conscious of a representation, it is clear; if I am not conscious of it, it is obscure . . . All of our clear representa­ tions . . . can be distinguished with respect to distinctness and indistinctness. If we are conscious of the whole representation, but not of the manifold that is contained in it, it is indistinct.  [Log, Ak. 9:33–34]8

Passages like this one suggest the following definitions. A representation is clear for a cognizer if, and only if, the cognizer is conscious of the whole representation; it is obscure otherwise. A clear representation is distinct for a cognizer if, and only if, the cognizer is individually conscious of some of its parts; it is indistinct otherwise. These definitions raise several questions, in particular, what exactly it means to be conscious of a “whole” representation, and how one could be conscious of a whole representation without being individually conscious of any of its parts. It is also not clear if Kant’s ultimately intended conception of indistinctness is really supposed to be as strong as his explicit characterizations suggest. It is possible that he had a weaker conception in mind, according to which a representation is indistinct if, and only if, most of its parts, or a significant number of its parts, are obscure. There is no need to pursue these questions here; whatever the details of Kant’s conception of indistinctness, the central operative notion is the obscurity of (all/most/many of) the parts of the representation, which, as we will see, is also at the core of Leibniz’s understanding of what it 7  See B 177–181/A 138–142; A 124; B 151–152. 8  See V-­Lo/Wiener, Ak. 24:805; V-­Lo/Pölitz, Ak. 24:510–511, 536; V-­Lo/Busolt, Ak. 24:617; V-­Lo/ Dohna, Ak. 24:729.

180  Kant and the (Alleged) Leibnizian Misconception means for a perception to be confused. Also note that Kant repeatedly chides the Leibnizians for using the label “confused” for representations that, according to him, should properly be called “indistinct.” To Kant’s mind, the opposite of “con­ fused” is not “distinct” but “ordered.”9 In order not to complicate matters too much, I shall use “confused” and “indistinct” interchangeably, namely, in the sense of “indistinct” as defined by Kant, which corresponds to the Leibnizian “confused.” How does Kant understand the allegedly problematic Leibnizian conception of the relation between sensible and intellectual representations? At some places, he describes the Leibnizians as claiming that (1) sensible representations are indistinct concepts; at other places he describes them as claiming that (2) sensible representations are indistinct, while intellectual representations are distinct. These two claims amount to slightly different views about the difference between sensible and intellectual representations. A Leibnizian who endorses (1) can legitimately be accused of not regarding this difference as a difference in kind, but a Leibnizian who endorses (2) could insist that, on his view, sensible and intellectual representations do belong to different kinds of representations: the indistinct kind and the distinct kind. It is worth noting that (1) is not only stronger than (2) but also considerably more implausible. How could the confusion of a concept, that is, the becoming unconscious of its parts, possibly ever lead to, say, a red-­sensation? In addition to (1) and (2), there is yet another way in which one might characterize Kant’s understanding of the Leibnizian conception of the relation between sensible and intellectual representations, which occupies a middle ground between (1) and (2) in terms of strength and plausibility. Kant appears to read the Leibnizians as holding that (3) any content that is represented confusedly in a sensible representation can also be represented distinctly in a (possibly complex) intellectual representation. In contrast to (1), (3) does not implausibly identify sensible representations with indistinct concepts, but it preserves the reductionist spirit of (1) and goes beyond (2) in implying that sensible representations can be fully translated into intellectual ones. Kant appears to regard at least some Leibnizians as committed to (1) even though several of his related, subsequent objections, as we will see, only require the weaker (3). To my mind, when Kant uses (2) to describe the Leibnizian position, he is concerned to make the more general point that even if distinctness and indistinctness can be used to mark the specific difference between certain kinds of representations, these features do not constitute the specific difference between intellectual and sensible representations. It is important to appreciate that Kant is not only objecting to the Leibnizian understanding of the difference between sensible and intellectual representations 9  See Anth, Ak. 7:138; V-­Lo/Pölitz, Ak. 24:511; V-­Lo/Wiener, Ak. 24:805; V-­Lo/Busolt, Ak. 24:617; R178, Ak. 15:67; R4641, Ak. 17:622.

Anja Jauernig  181 as a merely “logical” difference, but also to their privileging of the intellectual over the sensible. As just indicated, intellectual representations are more fundamental in that all sensible representations are reducible to, or translatable into, them, but they are also privileged in that they afford us the best kind of cognition that we are capable of, that is, distinct cognition. Indeed, as we will see, there is another thesis that expresses the superiority of the intellectual compared to the sensible that goes beyond the misconception proper but that Kant often associates with it and also ascribes to the Leibnizians, namely, that by means of purely intellectual representations, and purely intellectual representations alone, we can represent what is ontologically basic, that is, things in themselves. Even more strongly, Kant also seems to regard it as an integral part of the Leibnizian view that the most fundamental level of reality is exactly as our pure concepts present it to be, on which more below. On what grounds does Kant deem the Leibnizian conception of the difference between sensible and intellectual representations a misconception? The first reason for this assessment is that, in Kant’s eyes, the Leibnizian conception, at least as captured in (1) and (2), is in conflict with the facts on the ground, so to speak. Kant protests that there are sensible representations that are distinct without thereby ceasing to be sensible, just as there are intellectual representations that are indistinct without thereby ceasing to be intellectual. For example, if I see a house and can clearly make out its roof, windows, and door, my perception of the house is (at least partly) distinct, yet it remains a perception all the same. Cognitions that are distinct or indistinct are distinguished only depending on the degree to which I am conscious of the representation. Sensible representa­ tions can be very distinct; but still they remain sensible; on the other hand, representations of the understanding can be very indistinct, but they still remain intellectual.  [V-­Met-­L2/Pölitz, Ak. 28:584]10

Since both sensible and intellectual representations can be distinct and indistinct to various degrees without losing their sensible or intellectual nature, their specific difference cannot consist in their different degrees of distinctness/ indistinctness, nor can sensible representations be identified with indistinct concepts. The second reason for Kant’s classification of the Leibnizian conception of the difference between sensible and intellectual representations as a misconception is rather obvious: he has a different story to tell about how these two kinds of representations are distinguished, and he takes his own story to be correct. Kant identifies various differences between sensible and intellectual representations, 10 See MSI, Ak. 2:394; R 207, Ak. 15:79; V-­ Lo/Pölitz, Ak. 24:511–512; V-­ Th/Baumbach, Ak. 28:1266.

182  Kant and the (Alleged) Leibnizian Misconception but two deserve to be singled out. First, in the context of criticizing the Leibnizian philosophy, he tends to emphasize that sensible and intellectual representations originate in distinct, heterogeneous cognitive faculties. The former have their ori­ gin in sensibility, the latter in the understanding. Each one of these cognitions carries the sign of its origin, so that the former, even if they are very distinct, are called sensible according to their origin, and the latter, even if they are very confused, are called concepts of the understanding. [MSI, Ak. 2:395]

Sensibility and the understanding are distinguished in that the former is a recep­ tive, passive faculty that needs to be affected in order to produce representations, while the understanding is a spontaneous, active faculty, which generates repre­ sentations from its own depth, as it were.11 Sensible cognitions are not sensible because they are confused, but because they arise in the mind insofar it is affected by objects. Intellectual representations, on the other hand, are not intellectual because they are distinct, but because they originate in ourselves. For distinctness or obscurity are merely forms, which pertain to both sensible and intellectual representations. But they [the representations] are sensible and intellectual according to their origin; may they be distinct or confused.  [V-­Met-­L1/Pölitz, Ak. 28:229–230]12

So, another way of formulating the misconception objection that is implicit in Kant’s emphasis of the different origins of sensible and intellectual representations would be to say that the Leibnizians fail to recognize that our faculty of cognition comprises two distinct stems, an active and a passive one. The Leibnizians know only one basic, active faculty of cognition, from which, ultimately, all representa­ tions are supposed to originate. The second difference between sensible and intellectual representations that is centrally relevant for Kant’s critique of the Leibnizians is that, on Kant’s view, intuitions and concepts have very different representational formats, as one might put it, which, in turn, is responsible for further differences concerning their structure and their representational capabilities. Intuitions are (primarily) pictorial or image-­ like with a built-­ in point of view, which grounds their characteristic structural features such as the priority of the whole compared to the parts and their infinite complexity, and makes them capable of representing individuals and spatial orientations, for example. Concepts, on the other hand, do 11  See B 93/A 68; B 74/A 50; B 33/A 19; Log, Ak. 9:36; Anth, Ak. 7:140–141. 12  See Log, Ak. 9:35–36. Pure intuitions also originate in ourselves, but they still depend on affec­ tions, namely, self-­affections. See B 67–68.

Anja Jauernig  183 not have any picture-­like aspects and are defined in terms of their containment relations to other concepts, which grounds their characteristic structural features such as their finite complexity, and makes them capable of representing classes of objects.13 The relevance of these differences for Kant’s critique of the Leibnizians is that they support the rejection of (3) and its implication that any content that can be represented in sensible terms can also be represented in intellectual terms. The Leibniz-­Wolffian philosophy . . . considers the difference between sensibility and the intellectual merely as logical, whereas it is obviously transcendental and does not only concern the form of distinctness and indistinctness but their ori­ gin and content.  [B 62/A 44, my emphasis]

On Kant’s view, certain contents can only be represented in sensible terms, which, as we will see, he takes to show that the referents of our sensible representations must be distinct from the referents of our purely intellectual representations.

2  The Significance of the Misconception To my mind, there are two closely related general reasons why Kant thinks that the misconception is an especially detrimental error. First, it is the most basic mistake of the Leibnizians in that virtually all of their other major mistakes and all other major problematic aspects of their philosophy depend on it. Second, Kant’s own conception of the specific differences between sensible and intellectual representations occupies an especially central place in his theoretical philosophy and is directly involved in many of his most innovative doctrines. That is, the misconception is not only the most fundamental mistake of the Leibnizians, from the perspective of Kant’s philosophy, it is one of the worst mistakes they could have made. In order to better understand why Kant regards the misconception as so significant, we shall look at five major mistakes or problematic features of the Leibnizian philosophy that Kant takes to be closely tied to the misconception, and contrast them with the corresponding doctrines that Kant advocates instead. In most cases, the connection of the problem or mistake and the misconception is not a matter of implication. The misconception functions more like a prejudice that prevents the Leibnizians from seeing (what Kant takes to be) the correct view that would otherwise have been within their reach. In addition to shedding light on Kant’s relation to the Leibnizian philosophy, the main benefit of analyzing why Kant regards the misconception as so crucial is that it tells us much about Kant’s

13  See B 40; B 47–48/A 32.

184  Kant and the (Alleged) Leibnizian Misconception own philosophical development. All of the insights that the Leibnizians failed to have due to the misconception are insights that mark important steps in Kant’s journey that culminates in the transcendental philosophy. In the following, when describing the Leibnizian position I will no longer explicitly add the qualification that the description reflects Kant’s interpretation, but this is how I should be understood. Whether Kant’s reading does justice to the Leibnizians (and to which ones) is a question for another day. First problem. On the Leibnizian view, sensible representations are not essential to cognition; more specifically, they are believed not to play any role in our a priori knowledge in metaphysics and the exact sciences. Kant thinks that, due to the misconception, the Leibnizians hold that sensible representations are dispensable and not necessary for the cognition of objects. By contrast, it is one of the main tenets of Kant’s theoretical philosophy that cognition (at least in the strict sense) requires both concepts and intuitions; mere concepts without intuitions remain empty. [It] was a big mistake of the Leibniz-­Wolffian school . . . to regard sensibility as merely a lack (of clarity of the partial representations), and hence as indistinct­ ness, but the quality of the representations of the understanding as distinctness; while in truth the former is something very positive and an essential addition to the latter in order to bring about a cognition.  [Anth, Ak. 7:140–141 n.]14

It is not obvious how Kant moves from diagnosing the misconception to the ver­ dict that the Leibnizians fail to see that sensible representations are essential to cognition, nor is it evident on first glance how exactly this verdict should be understood. The latter depends on the meaning of “cognition” (Erkenntnis) in this context, which Kant uses in several senses. Among the things that Kant might have in mind are that the Leibnizians are committed to saying that (a) sensible representations are not essential to representing objects (in the “weighty” sense, as one might say), or that they are committed to saying that (b) sensible representations are not essential to knowledge. Note that neither one of these claims follows from the misconception. The assertion that sensible representations can be reduced to, or translated into, intellectual representations is a thesis about the nature of sensible representations, which, by itself, has no implications for whether we can actually perform these transformations, how we acquire various representations, or how we can ascertain or demonstrate the truth of substantive claims about the world. One could hold that, even though sensible representations are in principle reducible to intellectual ones, they still are essential to our representations of objects (in the weighty sense)—which, clearly, is also what

14  See B 76/A 51–52.

Anja Jauernig  185 Kant is talking about—because, due to our limited cognitive abilities, we are incapable of resolving the confusion that is present in our sensible representations or of entertaining the very complex distinct concepts that would be required to represent objects (in the weighty sense) in purely intellectual terms. In the same vein, one could insist that even if our purely intellectual representations to which all sensible representations can be reduced are innate, it could still be the case that we could not acquire or learn these innate ideas, in the sense of becoming ex­pli­ cit­ly aware of them, without sense perception.15 Finally, even if we took all true propositions to be analytic—as the Leibnizians do—one could still insist that due to our cognitive limitations we cannot perform the very complex analyses that would be needed to demonstrate the truth of contingent propositions, for which, accordingly, we have to rely on sense perception.16 To my mind, while Kant presumably would give his blessing to the accusation that the Leibnizians mistakenly hold (a) and (b), his main concerns in connection with the misconception and its implications for the Leibnizian account of cognition are directed elsewhere, namely, at a more specific version of (b) and the Leibnizian account of our a priori knowledge in metaphysics and the exact sciences. It is fairly clear that Kant takes the Leibnizians to believe, on account of the misconception, that we can acquire knowledge of substantive necessary truths by pure thinking. Since they leave for sensibility only the “despicable” job of confusing our representations in perception, the Leibnizians fail to see that the understanding is not our only source for a priori knowledge, but that sensibility is such a source as well, and, indeed, that without sensibility there would be no a priori synthetic knowledge at all.17 This failure, in turn, is responsible for two fur­ ther grave mistakes. First, it prevents the Leibnizians from appreciating the futil­ ity of engaging in speculative metaphysics, a discipline that purports to deliver a priori synthetic knowledge on the basis of pure concepts alone.18 Second, it has the consequence that the Leibnizians miss the fact that we do not only possess a priori concepts but also a priori intuitions, namely, the intuitions of space and time. They regard our representations of space and time as a posteriori concepts, and use their alleged confusedness to explain their seeming independence from

15  If the claim that innate ideas have to be learned sounds strange, recall that this is what Leibniz himself advocates. See RB book I, ch. i. 16  Parkinson suggests a similar response on behalf of Leibniz after identifying as one of the central upshots of Kant’s misconception objection the accusation that Leibniz advocates the prescription that “the person who wants to understand nature ought not to observe or experiment, but should confine himself to a priori thought” (Parkinson 1982, 17; see Parkinson 1981, 308). He replies that this is a mere caricature of Leibniz’s actual views, who insists that we can know contingent truths only by means of our sense experience. I agree with Parkinson’s reply on behalf of Leibniz, but I disagree with his assessment that said accusation is what Kant is ultimately after with his misconception objection. 17  See B 332/A 276. 18  See ÜE, Ak. 8:250–251.

186  Kant and the (Alleged) Leibnizian Misconception the actual spatiotemporal relations of bodies. As a result, they cannot account for the a priori synthetic status of geometry and of the pure theory of motion.19 Second problem. The Leibnizians cannot refute “fanciful” idealism or solipsism. Since the Leibnizians think that there is only one basic, active faculty of cognition, they have no firm epistemological grounds for denying that there is only one monad that exists in the world—in addition to God, who exists outside of the world. I take it that this problem is Kant’s underlying concern when, in discussing the misconception, he insists over and over again that the proper way to characterize the difference between sensible and intellectual representations is to highlight their origin in specifically distinct faculties.20 Kant’s claim that our experience of objects depends just as much on our sensibility, which is essentially passive, as it depends on our understanding, which is active, is what allows him to defend his unfaltering commitment to the existence of things in themselves, the Kantian equivalent of Leibnizian monads. Spelling out this defense requires considerable work, but the basic idea is that since sensibility is passive and produces “outer” sensations, that is, sensations that are spatially ordered, only in response to an affection by external things, the presence of outer sensations in our consciousness indicates that there must be something outside us that affects our sensibility, and (for somewhat complicated reasons that we cannot get into here) this something must be a thing in itself. After . . . asking “Who (what) gives sensibility its matter, i.e., the sensations?” he [Eberhard] believes himself to have spoken against the Critique in saying: “We can choose what we want—we end up with things in themselves.” Now, that is exactly the constant assertion of the Critique; only that it locates this ground of the matter of sensible representation not itself in things, as objects of the senses, but in something supersensible, which is the underlying ground of the former and of which we can have no cognition.  [ÜE, Ak. 8:315]21 19  See R 4851, Ak. 18:9: “Plato regarded all a priori cognition as intellectual. Leibniz as well; thus, they did not recognize the sensible [nature] of space and time. Leibniz declared it also to be intellec­ tual but confused.” See R 28, Ak. 23:24: “Space and time are not merely logical forms of our sensibility, i.e., they do not consist in that we represent to ourselves actual relations in a confused way; for how could we a priori deduce from this synthetic and true principles?” See B 56–57/A 39–40. 20  See GMS, Ak. 4:451: “Once this distinction [between sensible and intellectual representations] is made (if only through the indicated difference between representations that are given to us from somewhere else and while we are passive, and the representation that we just bring forth from our­ selves and while proving our activity) it follows automatically that one has to admit behind the appearances something else that is not appearance, namely, the things in themselves, although we humbly acknowledge directly on our own accord that we cannot get closer to them and that we cannot know what they are in themselves on the grounds that we can never be acquainted with them but [can know them] merely as they affect us.” 21  See Prol, Ak. 4:315: “Indeed, if we regard the objects of the senses as mere appearances, as is proper, we admit through this at the same time that a thing itself is their ground, although we do not know the same as it is in itself but merely its appearance, i.e., the way in which our senses are affected by this unknown something. Thus, the understanding, precisely in assuming appearances, admits the existence of things in themselves.” See B 306–307; B 61/A 44; B 72; B 235/A 190; A 251–252; B 334/A

Anja Jauernig  187 It is a crucial feature of Kant’s philosophy that it includes a justification for the claim that things in themselves that ground appearances exist, not least because this claim marks one central respect in which Kant’s idealism differs from the “fanciful” idealism of Berkeley, which several of Kant’s early readers perceived to be virtually identical to Kant’s position.22 Since the Leibnizians do not admit sens­ibil­ity as a self-­standing, passive cognitive faculty that is essential to our ex­peri­ence, the indicated refutation of fanciful idealism is not available to them. Third problem. The Leibnizians confuse noumena with phenomena; in particular, they think that we perceive things in themselves, albeit only confusedly. The Leibnizians fail to see that the objects that can be fully represented in purely conceptual terms (noumena) are distinct from the objects that our sensible representations present to us (phenomena).23 They mistakenly regard the differ­ ence between noumena and phenomena as merely reflecting a difference in the way in which we consider the same entities, namely, distinctly, in terms of pure concepts, on the one hand, and confusedly, in terms of sensible representations, on the other hand. Pure concepts represent these entities as they really are; sens­ ible representations represent them as they appear to us. This mistake, the “confu­ sion of the object of the pure understanding with the appearance” (B 326/A 270), is the focus of attention in the chapter on the so-­called “amphiboly of the con­ cepts of reflection,” which contains Kant’s most sustained critical engagement with the Leibnizian philosophy in the Critique of Pure Reason. Note that there are different ways in which one can confuse noumena and phenomena. For example, Locke is guilty of a version of this confusion as well, but in contrast to the Leibnizians who “intellectualized appearances,” that is, who took phenomena to be confusedly represented noumena, he “sensitized (sensifizierte) the concepts of the understanding” (B 327/A 271), that is, he took the concepts of the understanding to be worked-­over, glorified ideas of sensation (and reflection). The Leibnizians regard intellectual representations—and, with them, noumena—as fundamental, whereas in Locke’s eyes sensible representations are more basic. As indicated above, Kant takes it to be part and parcel of this privileging of the intellectual that the Leibnizians believe that our purely intellectual representations capture the most fundamental aspects of reality, and, accordingly, that noumena, the objects of our purely intellectual representations, are (or exactly correspond to) things in themselves. This means that concomitant 278; A 358; A 379; A 387; B 522/A 494; B 566/A 538; B 723–724/A 695–696; Prol, Ak. 4:288–289, 290, 318, 337, 354, 360; GMS, Ak. 4:451, 459; ÜE, Ak. 8: 203, 205, 207, 208, 219; KpV, Ak. 5:6. For a detailed reconstruction of Kant’s argument for the existence of things in themselves, see Jauernig 2021, esp. ch. 5. 22  See Prol, Ak. 4:293–294. For a detailed account of the difference between Kant’s and Berkeley’s idealism, see Jauernig 2021, esp. chs 3 and 5. 23  See MSI, Ak. 2:395. The term “noumenon” is a moving target in Kant’s work. In the following, I shall use it in its traditional sense, that is, as synonymous with “object of the pure understanding” or “intelligible object.”

188  Kant and the (Alleged) Leibnizian Misconception with their confusion of noumena with phenomena, the Leibnizians also confuse things in themselves with appearances. Most notably, they are committed to the view that we perceive things in themselves, albeit only confusedly.24 Kant clearly thinks that the Leibnizian confusion of noumena with phenomena is closely related to their misconception. It is not immediately obvious what Kant takes this relation to be, but there are at least two connections between the misconception and the relevant confusion that can plausibly be assumed to have played some role in Kant’s thinking. First, the misconception effectively amounts to taking away what Kant seems to regard as good reasons to assume that the objects of sensible representations and the objects of purely intellectual representations are distinct entities. One of these reasons is that there are certain properties that can only be represented in intuition, as for instance specific spatial orientations, to pick a prominent example. Since some empirical objects, or phenomena, are essentially handed, for example, human hands, it follows that some empirical objects cannot be fully represented in purely conceptual term. And since (by assumption) noumena can be fully represented in conceptual terms, we can conclude that some empirical objects are not noumena. Generalizing this result, it follows that phenomena and noumena are distinct entities. (And on the assumption that noumena are, or correspond to, things in themselves, that is, that our pure concepts adequately represent the world at the ontologically most fundamental level, it also follows that empirical objects are not things in themselves.)25 A related, more general reason for classifying noumena and phenomena as distinct entities rests on Kant’s claims that our sensibility produces representations only as a result of being affected, that it has its own a priori forms, space and time, and that space and time are nothing but forms of our sensibility. It follows from these claims that the objects that our sensible representations represent are not things in themselves but appearances, that is, objects that depend (at least in part) on the subjective conditions of our cognitive apparatus.26 Anybody who thinks that noumena are things in themselves, as the Leibnizians do, should conclude from this that phenomena and noumena are distinct entities. The Leibnizians are prevented from discovering either one of these arguments—which Kant endorses, up to a point, namely, the point at which it is assumed that our purely conceptual representations represent things in themselves—because their misconception commits them to the theses that anything that can be represented in intuition can also be represented in purely conceptual terms, and that sensibility is not a self-­standing mental faculty that makes significant contributions to our cognitions of objects.

24  See the references in n. 28 below. 25  See MSI, Ak. 2:403; Prol, Ak. 4:285–286; MAN, Ak. 4:483–484.

26  See B 42–59/A 26–42.

Anja Jauernig  189 The second connection between the misconception and the confusion of phe­ nomena with noumena is that Kant seems to regard it as self-­evident that neither the becoming unconscious of the parts of a representation nor the analysis and making clear of the parts of a representation can change its referent. For ease of communication, I shall call these two processes the “confusing” and “disfusing” of a representation, respectively. The disfusing or confusing of a representation only makes its content more or less explicit but does not alter what it represents. So, if sensible representations and intellectual representations were merely distin­ guished by their different degrees of distinctness, as the misconception stipulates, and if we assume that intellectual representations and, with them, noumena are fundamental, it would follow that the objects of sensible representations are nou­ mena, albeit confusedly represented ones. Kant himself, of course, rejects this line of reasoning, on account of rejecting the misconception. No matter how distinct we manage to make a sensible representation, it will always remain sensible, and, accordingly, its referent will always remain a phenomenon. This means, in particular, that it is not the case that we perceive things in themselves, albeit confusedly; we do not perceive them at all. Even if we could bring our intuition to the highest degree of distinctness, we would not get any closer to the nature of things in themselves . . . The Leibniz-­ Wolffian philosophy, thus, gave a completely illegitimate perspective to all inves­ tigations concerning the nature and origin of our cognitions, in that it considered the difference between sensibility and the intellectual as merely logical, even though it is obviously transcendental and does not merely concern the form of distinctness or indistinctness [of representations] but their origin and content, such that by means of the former we do not merely cognize the nature of things in themselves indistinctly but rather not at all, and as soon as we take away our subjective nature, the represented object with the properties that sensible in­tu­ ition gave to it, is nowhere to be found, nor can be found, insofar as this sub­ject­ ive nature precisely determines it as appearance.  [B 60–62/A 43–44]27

One of the reasons why the Leibnizian confusion of noumena with phenomena is so significant is that it lies at the heart of the aforementioned amphiboly of the concepts of reflection, which, in turn, is directly responsible for a number of other grave errors. Fourth problem. The Leibnizians fail to recognize that the concepts of reflection are “amphibolous,” that is, that each one of them can be understood in two

27  See Prol, Ak. 4:290; B 60/A 43; GMS, Ak. 4:451; FM, Ak. 20:278.

190  Kant and the (Alleged) Leibnizian Misconception different ways, which leads them to mistakenly regard certain metaphysical prin­ ciples as valid for phenomena, while in truth they are only valid for noumena. It is important to emphasize—since it is often overlooked—that Kant’s primary complaint against the Leibnizians in the amphiboly chapter is not that they overstep the limits of knowledge by enunciating a priori synthetic principles about things in themselves.28 To be sure, the Leibnizians are, in fact, guilty of this transgression, and Kant does complain about it. But he does so elsewhere, for example, in the transcendental dialectic. His main focus in the amphiboly chapter is on a different kind of mistake, namely, the indicated unwarranted extension of certain a priori synthetic principles to the sensible realm. Even if our pure concepts did give us reliable cognitive access to things in themselves, there would still be a problem, namely, that, contrary to the view of the Leibnizians, their intellectual principles are not valid for sensible objects.29 This is the problem that Kant addresses in the amphiboly chapter. Concepts of reflection are concepts that govern the comparison between representations (“logical comparison”) or between objects or properties (“objective comparison”) by specifying certain respects with reference to which the comparison is supposed to be performed. These respects are identity and diversity, harmony and opposition, inner and outer, and matter and form. That is, in the relevant reflections of comparison we ask whether two representations or objects/properties are identical or distinct, whether they are opposed to each other or harmonize with each other, whether their relation is an “inner” or an “outer” one—which, in this context, seems to come down to the question of whether one of them is essential to the other or not—and, finally, whether they are related as “matter” and “form,” and, if they are, which one is logically/ ontologically more fundamental. As will be explained in more detail below, by spelling out what these concepts of reflection mean one can formulate general synthetic principles that capture important features of concepts as well as of objects and their properties. The potential problem posed by the concepts of reflection is that they mean different things, depending on whether the representations that are supposed to be compared are intellectual or sensible, or, similarly, depending on whether the objects/properties that are supposed to be compared are noumenal or phenomenal. That is, the conditions that define what it means for concepts or noumena to be identical or distinct, or opposed or 28  Paton reads Kant in this way. See Paton (1969, 74). Also see Schneider (2004, 71). In the same vein, several commentators have expressed the (misguided) criticism that Kant’s critique of Leibniz in the amphiboly chapter rests on special teachings of Kant’s transcendental idealism, namely, in particu­ lar, the claim that our knowledge is restricted to appearances, which means that the Leibnizians, who do not accept transcendental idealism, need not worry about Kant’s critique. See Willaschek (1998, 347). 29  See B 332/A 276: “But even if we could say something synthetically about things in themselves through the pure understanding (which, however, is impossible), it could not be extended to appear­ ances, which do not represent things in themselves.”

Anja Jauernig  191 harmonious etc., differ from the conditions that define what it means for sensible representations or phenomena to be identical or distinct, or opposed or harmonious etc. Since the Leibnizians view sensible representations as not specifically different from intellectual representations and, thus, confuse noumena with phenomena, they are oblivious to the ambiguity or amphiboly of the concepts of reflection. And since they take intellectual representations to be more fundamental, they understand the concepts of reflections only in the sense appropriate for a comparison of intellectual representation or noumena. All of this leads them to think that the general principles that can be formulated on the basis of the concepts of reflection that are appropriate for intellectual representations or noumena are valid for things in general, that is, that they are universally valid, and, thus, include phenomena under their scope as well. Deceived by the amphiboly of the concepts of reflection, the famous Leibniz created an intellectual system of the world, or rather thought that he cognized the inner quality of things, in comparing all objects only through the understanding and the abstract formal concepts of its thinking . . . He compared all things with each other merely through concepts, and discovered, naturally, no other differences than the ones through which the understanding distin­ guishes its pure concepts from each other. The conditions of sensible intuition, which have their own differences, he did not regard as original; for sensibility was for him only a confused kind of representing and no special source for representations; appearance was for him the representation of a thing in itself, although distinguished from the cognition through the understanding accord­ ing to its logical form . . . Thus, Leibniz compared the objects of sense with each other only in the understanding as things in general.  [B 326/A 270]

Before looking in more detail at these comparisons and their problematic results, it is important to highlight that the root cause of the deception of the Leibnizians by the amphiboly of the concepts of reflection is the misconception, as Kant clearly acknowledges in the previous quotation, and not, as he misleadingly claims in framing the amphiboly chapter, the Leibnizians’ neglect to engage in “transcendental reflection.” Transcendental reflection is supposed to precede the comparison of our representations and is intended to determine whether the objects of these representations are noumena or phenomena, so as to ensure that the concepts of reflection are used in the appropriate sense.30 To be sure, Kant is right to say that without transcendentally reflecting in this way one runs the risk of ending up with “seemingly synthetic principles that critical reason cannot accept, which are grounded merely in a transcendental amphiboly, i.e., a

30  See B 325/A 260; B 316–317/A 260–261; B 318–319/A 262–263.

192  Kant and the (Alleged) Leibnizian Misconception confusion of the object of the pure understanding with the appearance” (B 326/A 270). But at the same time, transcendental reflection will not be of any help, and will not make much sense, to anybody who thinks that there is no specific difference between intellectual and sensible representations, or between noumena and phenomena to begin with. Transcendental reflection presupposes a conception of the difference between sensible and intellectual representations as a difference in kind, which is precisely what the Leibnizians (allegedly) lack. This shows that the ultimate culprit for their deception by the amphiboly is the misconception. Only after correcting the misconception could it even have occurred to them to engage in transcendental reflection. Corresponding to the abovementioned four respects in which concepts or objects and their properties can be compared, there are four amphibolies and, accordingly, four synthetic principles about noumena that the Leibnizians mistakenly regard as valid for phenomena too. Kant’s text does not make it entirely clear how he conceives of the relation between the comparison of representations and the comparison of objects, and how exactly he thinks the Leibnizians arrived at the problematic synthetic principles. Sometimes he apparently wants to say that the relevant properties of noumena can be directly read off the corresponding properties of pure concepts.31 At other times, the cru­ cial assumption on which the derivations rely seems to be that whatever cannot be represented by means of pure concepts cannot be true of noumena.32 In some cases, the explicit or presumable derivation of the relevant features of noumena from the comparison of concepts or their representational capacities also appears to involve certain additional implicit assumptions. For instance, it seems to be taken for granted at several places that no concept has a relational predicate as an essential constituent, which, then, might be seen as giving license to the claim that none of the essential properties of an object of the pure understanding are extrinsic. For our present purposes, there is no need to sort out how exactly Kant arrives at the conditions that specify what it means for noumena to be identical or distinct, or for their properties to be in harmony or opposed etc. The important point for us is that Kant is on board with the Leibnizians right up to the point where, on account of the misconception, they transform these descriptions of noumena and their properties into general principles that are supposed to be valid for all kinds of objects. So, what, then, are the four amphibolies (for objective comparisons) and the resulting problematic general principles?33 Due to space constraints, the follow­ 31  This is suggested by the formulation in the passage cited above that Leibniz found “no other dif­ ferences than the ones through which the understanding distinguishes its pure concepts from each other” (B 326/A 270). See B 328/A 272; B 335/A 279. 32  See B 337–342/A 280–286. 33  In addition to the passages from the Critique indicated below, the amphiboly is also thematized in R 5552, Ak. 18:218–219; and R 5554, Ak. 18:229.

Anja Jauernig  193 ing presentation will have to be rather brief. For each of the four cases, I will list the relevant respect of comparison, formulate the conditions that define the com­ parison if understood as a comparison of noumena, offer possible, and in some cases somewhat speculative, grounds for understanding the relevant concept of reflection in the indicated way by appeal to the properties and representational capabilities of pure concepts, state the resulting general principle that the Leibnizians, on account of the misconception, regard as valid for both noumena and phenomena, and, finally, identify how phenomena violate the general principle. (1)  Identity and distinctness.34 Noumena are identical if, and only if, they have the same intrinsic properties. [Underlying possible reasons: (a) Concepts are identi­ cal if, and only if, they contain the same non-­relational concepts. (b) It is impossible to represent distinct intrinsically indiscernibles in purely conceptual terms.] General principle: Distinct intrinsically indiscernibles are impossible. Problem: There can be multiple distinct intrinsically indiscernible sensible objects, namely, intrinsically indiscernibles that are located at different places. (2)  Harmony and opposition.35 Two (primitive) properties of a noumenon are in harmony/not opposed if, and only if, they are both positive properties. [Underlying possible reasons: (a) Concepts are opposed to each other if, and only if, they contradict each other. (b) It is impossible to represent an opposition of properties in purely conceptual terms in any other way than by contradictory predicates.] General principles: (i) Two properties are opposed to each other if, and only if, one is the contradictory opposite of the other. (ii) Any thing whose properties are opposed to one another is impossible (since nothing can be both A and not-­A at the same time.) Problem: Two sensible properties can oppose each other without being contra­ dictory opposites. Accordingly, a sensible object whose properties are opposed to one another is possible. For example, it is possible for a sensible object to possess both a moving force that, if not opposed, would have the effect of making it travel to the east with a speed of 5 mph, and a moving force that, if not opposed, would have the effect of making it travel to the west with a speed of 5 mph. (3)  Inner and outer.36 The properties of a noumenal substance are inner, in the sense of essential, (if and) only if they are intrinsic properties. The parts of a noumenal substance are inner, in the sense of essential, (if and) only if they are simple. [Underling possible reason: The concepts that are contained in a concept are inner, in the sense of essential, (if and) only if they are non-­relational.] 34  See B 319–320/A 263–264, B 327–328/A 271–272, B 338/A 282. 35  See B 320–321/A 264–265, B 328–330/A 272–274, B 338/A 282. 36  See B 321–322/A 265–266, B 330–331/A 274–275, B 339–342/A 282–285.

194  Kant and the (Alleged) Leibnizian Misconception General principles: (i) Substances that have no intrinsic properties are impos­ sible. (ii) Substances are individuated by their intrinsic properties alone. (iii) Substances are, or are composed of, simples. Problems: The essential properties of sensible substances include extrinsic properties. Or, even more strongly, sensible substances have only extrinsic properties. For example, a piece of matter is wholly defined in terms of its spatiotemporal and causal relations to all other pieces of matter that coexist with it. Moreover, sensible substances consist entirely of relations, that is, they are infinitely divisible, and, thus, not composed of simple parts. (4)  Matter and form.37 In the noumenal realm, matter is ontologically prior to form. [Underlying possible reasons: (a) The matter of a concept, that is, the concepts that are contained in it, logically precede the concept’s form, that is, the relations of containment that obtain between the concept’s parts. (b) More generally, logical form, as manifested in judgments) presupposes matter, that is, the concepts featured in the judgements in question.] General principle: Space and time, the forms, are ontologically dependent on, and thus ontologically posterior to, things and events, the matter. Problem: Sensible objects depend on space and time, which are thus ontologically prior to them. Fifth problem. The Leibnizians are committed to the view that, if our senses were more acute or our powers of disfusing our perceptions were stronger, we could learn something about things in themselves by experience. This mistake of the Leibnizians is also closely connected to their confusion of noumena with phenomena, and can be seen as the flip-­side of the mistake that we just discussed. The problem at the center of attention in the amphiboly chapter is that the Leibnizians illegitimately extend to appearances what they (allegedly) learned about things in themselves by mere a priori thought; the problem to be discussed presently is that the Leibnizians are committed to saying that, under certain conditions, we could learn something about things in themselves by experience, which, in truth, however, is an appropriate method of inquiry only in the case of appearances. Kant’s clearest presentation of this kind of mistake can be found in On Progress in Metaphysics: But as far as the Leibnizian principle of the logical difference between the indis­ tinctness and distinctness of representations is concerned, when he claims that the former kind of representation, which we have called mere intuition, is ba­sic­ ally just a confused concept of its object, and, thus, that intuition is distinguished from the concepts of things, not specifically, but only according to the degree of consciousness, such that, for example, the intuition of a body with the

37  See B 322–324/A 266–268, B 331–333/A 275–278.

Anja Jauernig  195 consciousness of all representations that are contained in it would amount to the concept of the body as an aggregate of monads. The critical philosopher, how­ ever, will notice that in this way the thesis “bodies are composed of monads” could arise from experience, merely through the analysis of perception, if only we could see well enough (i.e., with the appropriate consciousness of the partial representations).  [FM, Ak. 20:278]

The main argument in this passage appears to be that since, according to Leibnizian metaphysics, bodies are aggregates of monads, and, thus, since the concept “body” is equivalent to the concept “aggregate of monad,” the misconception commits the Leibnizians to the view that a disfused intuition of a body, that is, an intuition of a body whose parts have been individually brought to consciousness, is the same as, or equivalent to, the concept of an aggregate of monads. But this means that if our senses were more acute or our powers of disfusing our perceptions were strong enough, we could confirm the claim that bodies are aggregates of monads by experience, namely, by looking very hard at bodies, or by bringing to consciousness the minute parts that compose our perceptions. Kant rejects this position, on the grounds that, as explicated earlier, on his view, the objects of our perceptions are appearances, not things in themselves, a fact that cannot be changed by any amount of intent staring, or by any kind of disfusion efforts.

3  Leibniz on Confused Perceptions The obvious first question that presents itself to an interpreter who is interested in Kant’s misconception objection from the point of view of Leibniz is whether he indeed conceives of the relation between sensible and intellectual representations in the way that Kant complains about. A fair amount has been written on this question, and the emerging consensus in the literature appears to be that the answer is no.38 Personally, I actually agree with Kant that Leibniz is guilty of the misconception, but I will not argue for this assessment here.39 A second im­port­ ant question that is worth pursuing, even if the first question were to be answered in the negative, is whether Leibniz’s philosophy is beset by one or more of the other problems that Kant sees as intimately bound up with the misconception, problems that could very well arise in a philosophical system that does not count the misconception as one of its doctrines. After all, those problems are why Kant is so concerned about the misconception in the first place. In the remainder of this chapter, I will offer some considerations that bear on this question. We will

38  See n. 2 below.

39  See Jauernig (2019, 59–60).

196  Kant and the (Alleged) Leibnizian Misconception begin with a (severely abbreviated) discussion of Leibniz’s conception of the nature and origin of confused perceptions, and conclude with a few thoughts on whether Leibniz is indeed committed to saying that we could learn something about things in themselves by experience if our senses were acute enough or our powers of disfusing perceptions were strong enough. As several commentators have pointed out, one can find several different meanings of the terms “distinct” and “confused” in Leibniz’s texts.40 There is much to say about these different senses, but we will focus on the main senses in which perceptions can be said to be distinct or confused. The first of these corresponds to how Kant understands distinctness/indistinctness in that it centers on the awareness/lack of awareness of the cognizer of the internal complexity of the representation in question. For Leibniz, a conscious perception is distinct1 if, and only if, the cognizer is individually conscious of some (or most?) of the perceptions that compose it, or, rather, quasi-­compose it, as we should say more precisely, for reasons that will become clear below; it is confused1 otherwise. This is what Leibniz usually has in mind when he calls a perception “confused.” Leibniz’s paradigm examples of confused1 perceptions are sensations, or sensible ideas of secondary qualities. Each soul knows the infinite—knows all—but confusedly. It is like walking on the seashore and hearing the great noise of the sea: I hear the particular noises of each wave, of which the whole noise is composed, but without distinguishing them.  [Principles of Nature and Grace §13, G VI, 604, AG 211] It can be maintained, I believe, that these sensible ideas appear simple because they are confused and thus do not provide the mind with any way of making discriminations within what they contain; just like distant things which appear rounded because one cannot discern their angles, even though one is receiving some confused impression from them.  [RB 120]41

The second sense in which perceptions can be distinct or confused is crucial for distinguishing sensations from the kinds of perceptions that dull monads enjoy, and has something to do with whether, or to what degree, they are conscious or noticed or noticeable. Unfortunately, Leibniz’s texts do not make it immediately clear what exactly he has in mind here. If a perception is more distinct, it makes a sensation. [“Specimen of Discoveries of Marvelous Secrets,” A VI 4, 1625]42

40  See McRae (1976, 36, 128); Brandom (1981, 451, 454–459); Wilson (1999b, 339–340); Simmons (2001, 41 n.); Bennett (2001, 307–309). 41  See RB 54. 42  See Mon 19, GP VI 610; On the Souls of Animals, GP VII 330.

Anja Jauernig  197 Death can only be a sleep, and not a lasting one at that: the perceptions merely cease to be sufficiently distinct; in animals they are reduced to a state of confusion which puts awareness into abeyance but which cannot last forever.  [RB 55]43 We are never without perceptions, but, necessarily, we are often without aware­ ness, namely, when none of our perceptions are distinct. [RB 162; translation slightly amended]44

On the basis of passages like these, one might conclude that a perception is dis­ tinct2 if, and only if, it is conscious or noticed.45 This would mean that a sensation is a conscious or noticed perception, or, as one might say alternatively, the no­ticing of a perception.46 This reading has been criticized on various grounds. For example, it has been pointed out that all monads, including the bare, dull ones that vegetate along in a state of unconscious stupor, are supposed to be distinguished from each other in virtue of the different degrees of distinctness of their perceptions. So, if a perception’s distinctness were to be cashed out as its being conscious, none of the dull monads would have any distinct perceptions and, thus, none of them would be distinguishable from any other, which would constitute a violation of the prin­ ciple of the identity of indiscernibles.47 This line of reasoning is acceptable as far as it goes, but it runs into the problem that in the Monadology Leibniz explicitly states that “if, in our perceptions, we had nothing distinct or, so to speak, in relief and stronger in flavor, we would always be in a stupor. And this is the state of the bare monads” (Mon, §24, GP VI 610, AG 216). That is, Leibniz explicitly endorses the view that bare monads have no distinct perceptions without finding it in any way problematic. This suggests that there must be yet another sense of perceptual distinctness (distinct3) that captures the way in which the unconscious perceptions of bare monads can be said to be distinct. We will return to the question of what sense that might be below. Another reason that has been adduced against the identification of distinct perceptions with conscious perceptions is that, at various places, Leibniz seems to express the view that a perception’s being conscious is the result of its distinct­ ness, which would imply that being conscious cannot be the same as being dis­ tinct.48 Many commentators also appear not to want to ascribe the view to Leibniz that consciousness comes in degrees—not least because Leibniz appears to con­ ceive of conscious perceptions as second-­ order, reflected perceptions, or

43  See “Metaphysical Consequences of the Principle of Reason,” C 15; Mon 21, GP VI 611; RB 113. 44  See RB 179. 45  See Furth (1972, esp. §4); McRae (1976, esp. 36–38); Parkinson (1982, 4–12). 46  Parkinson (1982) argues for the latter view. 47  See Brandom (1981, 451–452); Wilson (1999b, 339). 48  See Simmons (2001, 57). Simmons cites New Essays, RB 55 and RB 113.

198  Kant and the (Alleged) Leibnizian Misconception perceptions of perceptions.49 Given that Leibniz treats distinctness as a property capable of gradations, if “distinct” and “conscious” were equivalent it would follow that perceptions could also be more or less conscious.50 Finally, in a (somewhat cryptic) passage from the New Essays, Leibniz seems to say that we can have sensations, that is, distinct perceptions, without noticing them, which would be impossible if distinct perceptions were perceptions that are noticed.51 In response to these (and other) problems, several alternative proposals have been put forward for how to understand distinctness2. To my mind, the most promising proposal to date is due to Allison Simmons.52 According to Simmons, a perception is distinct2 for Leibniz if, and only if, it is distinctive or stands out. Accordingly, she characterizes sensations as perceptions that are noticeable or apt to be noticed, which allows for sensations that are not actually noticed, as Leibniz suggests in the New Essays.53 As Simmons herself admits, the explanation that perceptions are noticed because they are noticeable is not very illuminating, but she counters that it can be further fleshed out, based on Leibniz’s texts, by specifying what grounds a perception’s noticeability, namely, its size and variabil­ ity, and how many other perceptions are perceived with it.54 This proposal is a step in the right direction, but, to my mind, even more could and should be said about the conditions under which a perception counts as distinct or noticeable. In particular, Leibniz’s repeated indication that the noticeability of a perception depends on its size or strength calls for further clarification. We also still lack an account of what it means for the perceptions of a dull monad to be distinct to a certain degree, in virtue of which it is distinguished from all other dull monads. For a dull monad does not only not notice its perceptions, being dull, it cannot notice them.55 As mentioned earlier, the passage from Monadology, §24 suggests that the sense in which the perceptions of dull monads are distinct to a certain degree, call it “distinct3,” differs from the sense of distinctness that we just dis­ cussed (distinct2), which describes the perceptions of higher monads. So, it should

49  See PNG 4, GP VI 600. 50  See Brandom (1981, 452); Simmons (2001, 57). 51  See RB 54: “Memory is needed for attention: 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 some noise which we have just heard, then we remember it and are aware of just having had some sense of it. Thus, we were not straight away aware of these perceptions, and we became aware of them only because we were alerted to them after an interval, however brief.” See Simmons (2001, 57). 52 Alternative prominent proposals include Brandom (1981, 460–479), and Wilson (1999b, 343–345). 53  See Simmons (2001, 58). 54  See ibid. See RB 53–54. 55  On first glance, it might appear that there are several possible maneuvers that could allow us to extend Simmons’s proposal to the case of dull monads, for example, by saying that a perception P of a dull monad is noticeable to a certain degree if, and only if, there is a conscious monad to which P is noticeable to a certain degree. Upon further investigation, it turns out that none of these maneuvers work. With respect to the example, the proposal leaves undetermined which conscious monad to choose as a reference monad, and, worse, it renders all dull monads indiscernible, since they all have the same perceptions.

Anja Jauernig  199 not be held against Simmons’s proposal, which is concerned with distinctness2, that it does not also explain how to understand “distinct3.” But the fact remains that we still need an account of “distinct3,” which, we might add, ideally would also shed some light on how being distinct3 is related to being distinct2, and, for that matter, on how both of them are related to being distinct1, so as to explain why Leibniz refers to all of these properties by the same term. On my reading, to say that a perception is more or less distinct3 is to say that it is more or less “strong” in a sense to be explicated presently. As I understand Leibniz, one of the essential intrinsic qualities of a perception on his account is what one might call its “strength of proto-­consciousness” or its “strength of non-­ reflective consciousness,” call it “SPC.” In order to be consciously perceivable by, or noticeable to, a monad, the SPC of a perception has to be greater than a certain threshold. That is, in order to be distinct2 (noticeable) a perception has to be distinct3 to a certain degree, namely, to the degree that corresponds to an SPC that exceeds the possible-­consciousness threshold of the monad in question. The perceptions of dull, bare monads are distinct3 in that each one of them has a particular moderate SPC, but none of them has an SPC high enough to render it noticeable (distinct2).56 This account receives support from Leibniz’s theory of “little” perceptions and of how they can be combined to yield perceptions that are noticeable. According to Leibniz, monads are connected in one world in virtue of the fact that each monad perceives the entire universe by perceiving its body that receives impressions from all the other bodies in the universe, and that the perceptions of all monads are in sync.57 Since matter is infinitely divisible (and since the actual physical world is infinitely complex), the total perceptual state of each monad at any given time comprises infinitely many little perceptions. According to Leibniz, these little perceptions are “too small” to be noticed individually, that is, in our terminology, none of them has an SPC that surpasses the possible-­consciousness threshold of any monad. Whether or not the total perceptual state of a monad includes any conscious perceptions in addition to the unconscious little ones depends on what kind of monad it is. There are several procedures through which 56  A related account of the relation between unconscious and conscious perceptions is suggested by Jorgensen (2009, 241–245). He proposes that, on Leibniz’s view, all perceptions are distinct to a certain degree, and those perceptions whose degree of distinctness reaches a certain threshold are conscious. But Jorgensen explicates the relevant sense of distinctness as comparative noticeability with respect to other perceptions, while I understand the relevant kind of distinctness, distinctness3, as an intrinsic property of perceptions, namely, their strength of proto-­consciousness, which accounts for, and grounds, their noticeability. Moreover, on my reading, the threshold in question does not mark a tran­ sition from unconscious to conscious perceptions, as on Jorgensen’s, but from non-­noticeable to noticeable perceptions. Not all perceptions that are noticeable are, in fact, noticed, that is, conscious. I should like to add that the present chapter, which I presented as a paper at two conferences in 2009, was completed before I had a chance to read Jorgensen’s essay. 57  See Mon 62, GP VI 617; “Discourse on Metaphysics,” §33, A VI 4, 1582; letter to De Volder, June 20, 1703, GP II 253.

200  Kant and the (Alleged) Leibnizian Misconception conscious perceptions can come about; they all have in common that they include a certain reconfiguration or transformation of the little perceptions that results in a bundling of SPC in some perceptions that are thereby rendered noticeable. One way for this to occur is through the help of sense organs, or, more precisely, through the help of whatever mental faculties in a monad correspond to the sense organs in its body. Sense organs bundle physical impressions, thereby increasing the strength with which the monad’s body is affected by physical objects. Similarly, the mental faculties in the monad that correspond to the sense organs in its body bundle little perceptions and, with them, their strength of proto-­consciousness into new, emergent perceptions with a “heightened” SPC. If the SPC of the resulting perceptions surpasses the possible-­consciousness threshold, they are noticeable, that is, they are sensations. When a monad has organs that are adjusted in such a way that, through them, there is contrast and distinction among the impressions they receive, and conse­ quently contrast and distinction in the perceptions that represent them [in the monad] (as, for example, when the rays of light are concentrated and act with greater force because of the shape of the eye’s humors), then this may amount to sensation, that is, to a perception accompanied by memory. [“Principles of Nature and Grace,” §4, GP VI 499, AG 208]58

So, one important factor that can influence whether or not a monad has any noticeable perceptions, and, if it does, how many, is the effectiveness of its mental faculties in bundling little perceptions. I say that this is one possible factor, because the possession and number of noticeable perceptions also depends on the monad’s particular threshold of possible consciousness. And it could be that the possible-­consciousness threshold is not absolute but varies between different kinds of monads. The duller the monad, the higher the threshold; and the higher the threshold, the more little perceptions must be bundled to yield a perception whose SPC surpasses it. And so the main reason why dull bare monads do not have any noticeable perceptions may be that their specific possible-­consciousness threshold is so high that even if their mental faculties could bundle all of their little perceptions into one giant confused1 perception, its SPC would still not be high enough to make it noticeable. It is important to highlight that Leibniz quite clearly understands the bundling of little perceptions in our sensations not as mere aggregation, but as a (con-)fusing that results in something new, a whole that, as one might say, is more than the sum of its parts.59 The little perceptions are not genuine parts or components of 58  See RB 53–55, 134; Mon 25, GP VI 610. 59 See “Meditations on Knowledge, Truth, and Ideas,” A VI 4, 592, AG 27: “When we perceive colors or smells, we certainly have no perception other than that of shapes and of motions, though so

Anja Jauernig  201 the emergent perception, but its grounds, which might be said to quasi-­compose it. The new, emergent perception is significantly less complex than the group of little perceptions from which it emerged; in fact, depending on how we carve up and count perceptions, one might even say that some emergent perceptions are perfectly simple in that they represent only one homogeneous sensible quality, for example, the color red. So, having a sensation consists in being in a perceptual state that includes both a multitude (indeed, an infinite multitude) of little per­ ceptions whose SPC does not exceed the threshold of possible consciousness, and a noticeable or noticed emergent perception whose SPC does exceed the threshold. In order to accommodate Leibniz’s oscillating usage of the term “sensation,” it is helpful to distinguish between sensations in the narrow sense and sensations in the broad sense. A sensation in the narrow sense is a noticeable perception that emerges through the (con-)fusion of a multitude of unnoticeable little perceptions; a sensation in the broad sense is a perceptual state that includes both a sensation in the narrow sense and a multitude of unnoticeable, little perceptions from which the sensation in the narrow sense emerged. Recognizing these two senses of sensation allows us to make sense of Leibniz’s insistence against Locke that sensations are not arbitrarily connected with physical objects and events, as Locke claims, but that they represent or “express” them in Leibniz’s technical sense, that is, that there is a kind of isomorphism between the sensations and the physical objects, which implies that our sensations must be complex, namely, as complex as the physical objects that they express.60 In this context, “sensation” must be understood in the broad sense. Sensations in the narrow sense are simple or, at least, not very complex, and, accordingly, cannot represent objects that are infinitely complex. But since a sensation in the broad sense corresponds to a perceptual state that includes both a noticeable, not very complex perception and a multitude of infinitely many unnoticeable little perceptions, it has all the com­ plexity that is needed to represent physical objects.61 The proposed account also explains why our sensations (in the narrow sense) are confused1 (and why, in the characterization of confusion1 above, I used the expression “quasi-­compose”). Recall that for a perception to be confused1 means that the cognizer is not individually conscious of the perceptions that

very numerous and so very small that our mind cannot distinctly consider each individual one in this, its present state, and thus does not notice that its perception is composed of perceptions of minute shapes and motions alone, just as when we perceive the color green in a mixture of yellow and blue powder, we sense only yellow and blue finely mixed, even though we do not notice this, but rather fashion some new thing for ourselves” (my emphasis). See letter to Arnauld, April 30, 1687, GP II 91; RB 403. McRae also understands sensations as emergents; see McRae (1976, 38). 60  See RB 56, 131, 132–133. 61  This answers Simmons’s objection against emergentist readings of Leibniz’s theory of sensations, according to which these readings cannot account for the representationality that Leibniz ascribes to sensations. See Simmons (2001, 64).

202  Kant and the (Alleged) Leibnizian Misconception quasi-­compose it. Sensations (in the narrow sense) are inevitably confused in this sense because the little perceptions that quasi-­compose them are not distinct3 enough, that is, they do not have enough SPC, to be noticeable or consciously perceivable on their own. Fusing them together results in a perception whose SPC exceeds the relevant possible-­consciousness threshold, but the price that must be paid for this noticeability is a loss of complexity and, with it, a loss of detail and information, that is, the detail and information about the underlying physical phenomena that are represented in the little perceptions.62

4  Could We Learn Something about Monads by Experience? Is Leibniz committed to the claim that we could empirically learn something about things in themselves, that is, monads, if our senses were acute enough or our powers of disfusing perceptions were strong enough? In order to answer this question, we need to see whether and how, on Leibniz’s view, we can make our perceptions more distinct—the relevant sense of distinctness being distinctness1, which, as we saw, corresponds to how Kant conceives of distinctness, and which is how “distinct” should be understood from now on. Leibniz appears to hold that it is neither in our power to disfuse our confused perceptions by discerning their minute components through some kind of special introspective effort, nor is it in our power to enhance the acuteness of our senses to such an extent that we can gain information about their underlying complex external causes.63 I take it that this is what Leibniz has in mind when he says that confused perceptions are “simple with respect to the senses” (An Introduction on the Value and Method of Natural Science, L 285), or “simple from our point of view” (RB 297). Moreover, and more importantly, Leibniz also seems to think that these limitations are irre­ mediable, since they are essential to who we are. We are not to blame for the confusion which reigns among our ideas, for this is an imperfection in our nature: to be able to pick out the causes of odors and tastes, for instance, and the content of these qualities, is beyond us.  [RB 256]64

So, one could reply to Kant on behalf of Leibniz that the question of whether we could learn something about things in themselves by experience if our senses 62  See PNG 13, GP VI 604; “Discourse on Metaphysics,” §33, A VI 4B, 1582–1583; Response to Bayle, GP IV 563–564; letter to Arnauld, October 9, 1687, GP II 112–113. 63  Note that, strictly speaking, for Leibniz the only real causes of our confused perceptions are our unconscious little perceptions, which, in turn, are generated by the law of the series that defines each one of us qua monad. But Leibniz acknowledges that there is also a sense in which bodies in the phe­ nomenal world can be said to causally interact. It is in this sense of causation, that is, causation at the phenomenal level of reality, that complex mechanical properties can be said to cause our perceptions. 64  See RB 219.

Anja Jauernig  203 were acute enough or our powers of disfusing perceptions were strong enough, strictly speaking, makes no sense on Leibniz’s view. Setting aside whether it would even be possible to gain any information about things in themselves in this way, a being whose senses or ability to introspectively disfuse perceptions were improved in the relevant ways would not be one of us. More generally, given the account of confused perceptions as emergent representations described in the previous section, it seems plausible to assume that Leibniz would agree that it is impossible for any being, not just us, to discern the little perceptions that (quasi-)compose its confused perceptions by introspective analysis. How could inwardly “staring down” an emergent perception possibly tell anybody anything about the underlying little perceptions, which, by assumption, are unconscious and not themselves apprehended in the emergent perception? On the other hand, Leibniz’s examples also suggest that he allows that there could be other finite beings who differ from us only in that their senses are more acute or that their faculty of apperception is more powerful, whose perceptions would represent at least some of the details and complexity of the physical world that are lost in our confused perceptions. And there is good reason to assume that Kant would find the claim that beings of this kind could learn something about things in themselves by experience just as problematic as the thesis that we could accomplish this feat. After all, the acuteness of our senses (or lack thereof) does not play any role in Kant’s argument, briefly rehearsed above, for the claim that the objects of our experience are appearances, not things in themselves, which, accordingly, should apply equally well to beings who differ from us only in that their senses are more acute. There is another way in which Kant’s objection might turn out to have bite. Although Leibniz firmly denies that we can directly learn something about the underlying causes of our confused perceptions by introspectively analyzing them or by staring more intently, he admits that we can learn something about them indirectly, namely, by means of scientific theorizing and scientific experiments, for example, experiments involving optical instruments. The secret of analysis in physics consists in this one device: the reduction of the confused qualities of the senses (namely, heat and cold in the case of touch, fla­ vors in the case of taste, odors in the case of smell, sounds in the case of hearing, and colors in the case of sight) to the distinct qualities that accompany them, namely, number, size, shape, motion and cohesion, of which the last two are proper to physics. [“Recovery of Confused Qualities to Distinct Ones,” A VI 4, 1961–1962] If we had arrived at the inner constitution of certain bodies, these [sensible] qualities would be traced back to their intelligible causes and we should see under what circumstances they were bound to be present; even though it would

204  Kant and the (Alleged) Leibnizian Misconception never be in our power to recognize their causes sensorily, in our sensory ideas which are the confused effects of bodies acting on us.  [RB 403]65

Given Leibniz’s understanding of representation as a form of isomorphism, this also means that even though we cannot discern the little perceptions that (quasi-) compose our confused sense perceptions by introspective analysis, based on mechanistic science we might be able to infer for at least some of them what they must be representations of. So, the scientific investigation of the (external) causes of our confused perceptions at the same time in effect amounts to what one might call an “ersatz analysis” or an “indirect disfusion” of these perceptions by means of which we can learn something about some of the little perceptions that (quasi-) compose them. Now, if, on Leibniz’s view, this scientific ersatz analysis of our confused perceptions into distinct concepts of their causes could be extended further and further to uncover the constituents of these causes, and the constituents of the constituents of the causes and so on, and if the ultimate constituents could be expected to be found at the level of monads, one could argue that he is committed to a version of the claim that Kant objects to, that is, the claim that we could learn something about things in themselves by experience if our powers of disfusing perceptions were strong enough, even though, in this case, the method of disfusion would be somewhat more complicated than mere introspective analysis. Another noteworthy aspect of Leibniz’s discussion of the scientific ersatz analysis of our confused perceptions is that he sometimes appears to want to say that in this way we can discover what the sensible qualities “really” are that these perceptions represent. For example, Leibniz speculates that heat might consists in “a vortex of very fine dust,” that sound is “produced in air as circles are in water when a stone is tossed in” (“On What Is Independent of Sense and Matter,” GP IV 499, AG 186), and asserts white to be “nothing but an assemblage of a number of small convex mirrors” (“Addition to the explanation of the New System,” GP IV 575–576). This seems to suggest that Leibniz holds that our scientific concepts distinctly represent the same objects and properties that our conscious sense per­ ceptions represent confusedly. That is, it looks as if Leibniz would agree with a central assumption that plays an important role in Kant’s discussion of the miscon­ ception, namely, that disfusing or confusing a representation only makes its con­ tent more or less explicit but does not change its referent. But this, in turn, would mean that, if it were indeed the case that Leibniz’s envisioned ultimate results of the scientific indirect disfusion of our sense perceptions were descriptions of the configuration and properties of monads, we would have reason to conclude that Leibniz is guilty of another mistake that Kant accuses him of, namely, the mistake of claiming that we confusedly perceive things in themselves. 65  See letter to Burnett, January 20/30, 1699, GP III 247; RB 120.

Anja Jauernig  205 We will conclude by briefly looking at the question of whether Leibniz indeed holds the problematic claims that would give grist to Kant’s mill, that is, that (1) a confused perception and its scientifically disfused counterpart represent the same object, (2) the perfect, all-­encompassing scientific theory about the ultimate (external) causes of our confused perceptions would amount to a theory about the configuration and properties of monads, and (3) there are possible beings who differ from us only in that their more acute senses enable them to learn something about things in themselves by experience. Regarding (1), in addition to the passages in which Leibniz identifies sensible qualities like colors, odors, and smells with complex mechanical properties, there are also many passages in which he rejects this mechanist understanding and identifies sensible qualities with the qualia experienced by conscious perceivers instead. For example, he repeatedly asserts that the only way in which one can acquire the idea of heat or of a particular color is by experiencing it, which is why a blind man cannot learn what redness or whiteness is. This assertion would be false if redness and whiteness could be identified with complex mechanical prop­ erties.66 Similarly, Leibniz remarks that our ideas of heat or of particular colors cannot be given nominal definitions, which, again, would be false if the objects of these ideas were complex mechanical properties.67 This tension in Leibniz’s texts has led several commentators to conclude that Leibniz is slightly confused, or, at least, that he is ambivalent about sensory qualities.68 To my mind, there is no need to ascribe confusion or ambivalence to Leibniz. His use of such terms as “heat,” “color,” or “sensible quality” is simply ambiguous. Taken in one sense, they refer to complex mechanical properties; taken in another sense, they refer to qualia experienced by conscious perceivers. Given the account of sensation sketched in the previous section, this ambiguity is hardly surprising. Sensations in the narrow sense,that is, emergent confused perceptions, present sensible qualities understood as qualia. And (some of) the multitudes of little perceptions, which, together with the emergent confused perceptions, compose the perceptual states that are sensations in the broad sense, represent sensible qualities under­ stood as complex mechanical properties. Once this ambiguity in Kant’s use of the term “sensible quality” is acknowledged, the passages in which he identifies par­ ticular sensible qualities with particular complex mechanical properties need no longer be taken as evidence that he regards our confused perceptions and the distinct concepts that result from scientifically investigating their causes as repre­ senting the same objects. The distinct concepts of our scientific theories represent the same objects that are represented in our unconscious little perceptions, which,

66  “Introduction to a Small Book on the Elements of Physics,” A IV 4, 2002–2003; “Meditations on Knowledge, Truth, and Ideas,” GP IV 422–423. 67  See RB 194, 297; “Of an Organon or an Ars Magna of Thinking,” C 432. 68  See Wilson (1999a, 323–331); Jolley (1984, esp. 182–188).

206  Kant and the (Alleged) Leibnizian Misconception however, are different from the objects that are represented in our confused perceptions that emerge through the (con-)fusion of the little ones. In response to this line of reasoning, one might argue that the qualia that our confused perceptions present to us are not entities in their own right but mere appearances of complex mechanical properties. In this case, the distinct concepts of our scientific theories and our confused perceptions could be said to represent the same objects after all. The former represent sensible qualities as they really are, the latter represent them as they appear to us; and while the blind man cannot learn what whiteness is as it appears to us, he can learn what whiteness is in itself, as it were.69 To my mind, this response is not convincing. For starters, given Leibniz’s theory of representation as expression, a perception that has no, or at least not much, complexity cannot represent anything that is as complex as a mechanical property. It does not help to insist that sensations (in the narrow sense) represent the mechanical properties merely confusedly or as they appear to us. Representation requires isomorphism between the represented and the representing; without isomorphism we do not have confused representation, simply because we have no representation at all. Setting this quibble aside, the crucial question that needs to be answered in order to determine which one of the two readings under discussion is preferable is whether Leibniz indeed regards the experienced qualia as mere appearances or mere ways of appearing, or whether he takes them to be entities or realities in their own right. In a revealing passage from the New Essays, Leibniz clearly opts for the latter view. Phil: If our senses were acute enough, sensible qualities such as the yellow color of gold, would then disappear, and instead of it we should see an admirable texture of parts . . . Theo: This is all true, and I said something about it earlier. But the color yellow is a reality, all the same, like the rainbow.  [RB 219, my emphasis]

Leibniz’s comparison of the color yellow with a rainbow is especially telling. The rainbow is his favorite example of a so-­called “well-­founded phenomenon.”70 Phenomena do not belong to the most fundamental level of Leibniz’s tiered ontology, which is reserved for simple substances or monads. But phenomena, at least well-­founded ones, are real to some degree and do exist, namely, at the second main level of Leibniz’s ontological scheme.71 Colors, sounds, and odors are 69  This is how Puryear reads Leibniz. See Puryear (2005, 112–115). 70  See letter to De Volder, 1704 or 1705, GP II 276, AG, 182: “In just the same way a rainbow is not improperly to be said to be a thing, even though it is not a substance, that is, it is said to be a phenom­ enon, [a real or well-­founded phenomenon that doesn’t disappoint our expectations based on what precedes.]” See “First Truths,” A IV 4, 1648. 71  Leibniz’s distinction of reality into several ontological levels is widely acknowledged in the litera­ ture. Most commentators favor some version of a three-­level scheme featuring an ultimately real level of (simple) substances, a phenomenal level of physical bodies, and an ideal level of abstract objects.

Anja Jauernig  207 realities and not mere appearances in the sense that they are part of the ­onto­logic­al furniture at the phenomenal level of reality. A legitimate question that might be raised at this juncture is how it could be that colors and sounds are well-­founded phenomena when, as Leibniz’s texts make clear, the more fundamental mechanical properties that cause our confused perceptions are well-­founded phenomena as well. To answer this question, we have to look briefly at how Leibniz conceives of phenomena and how he thinks about their well-­foundedness. Phenomena are best understood as intentional objects of the perceptions of monads. Not all phenomena in this sense are real or well-­founded. Leibniz proposes several criteria for what it means for phenomena to be real or well-­ founded, including internal coherence, intersubjective coherence, conformity to general necessary truths, conformity to the laws of nature, and coherence with God’s representation of phenomena, among others.72 These various criteria allow us to distinguish different degrees of reality or well-­ foundedness at the phenomenal level, and, accordingly, to differentiate several phenomenal sublevels. In the order of increasing reality, these comprise, first, the level of common sense, whose phenomena correspond to the intentional objects of the shared prescientific worldview of all rational monads, which is based on their conscious, confused perceptions. This level includes colors, sounds, tastes, odors, space and time, and rainbows. The second phenomenal sublevel is the level of physical science proper, whose phenomena correspond to the intentional objects of mechanistic physics. This level includes extended, infinitely divisible corpuscles that are characterized by their number, size, shape, and motion (and, according to Leibniz’s version of mechanism, derivative forces). The phenomena at this level are also represented by some of the little perceptions of some monads, namely, little perceptions that, while still not having a large enough SPC to be noticeable, emerged through the combination of even “littler” perceptions. The most fundamental phenomenal sublevel is populated by the organic bodies of corporeal substances, which are primarily characterized by their forces. These organic bodies, which are also infinitely divisible, do not have precise shapes or motions, but exhibit a quasi-­fractal structure.73 The phenomena at this most real phenomenal level are the objects of God’s entirely distinct representation of the phenomenal world, which determines what counts as ultimate phenomenal

See McGuire (1976); Hartz and Cover (1988); Garber (1985, 203–208); Adams (1994, 254f.); Garber (1995). This three-­level interpretation is acceptable as a rough sketch of the general features of Leibniz’s ontology, but on a more fine-­grained level of examination it turns out that several more ontological sub-­levels ought to be distinguished. 72  See “Of Universal Synthesis and Analysis,” A VI 4, 543–544; “On the Method of Distinguishing Real from Imaginary Phenomena,” A VI 4, 1500ff.; RB 374–375, 392; “Discourse on Metaphysics,” §14, §32, A VI 4, 1550f., 1580–1581; Reply to Bayle, GP IV 569; letter to De Volder, June 30, 1704, GP II 268f. 73  For a helpful discussion of this quasi-­fractal character, see Levey (2005).

208  Kant and the (Alleged) Leibnizian Misconception real­ity.74 These phenomena are also represented in the coherent unconscious very little perceptions that are shared by all monads that coexist in one world; and Leibniz’s new (metaphysical) science of dynamics is intended to provide an account of them as well. This distinction of further sublevels of reality at the phenomenal level explains how both colors, sounds, tastes, and odors, on the one hand, and mechanical properties, which are ontologically more fundamental than colors, sounds, tastes, and odors, on the other hand, can be real and phenomenal. The foregoing considerations allow us to conclude that what our confused sense perceptions represent are colors, odors, and smells, understood as qualia, and not their complex mechanical causes—despite the fact that it is, of course, true that our sense perceptions also indicate the presence of the relevant mechanical causes, and, thus, can be said to signify them in some sense. So, we can record that Leibniz does not endorse (1), that is, the claim that a confused perception and its disfused counterpart represent the same object. This also means that even if it turned out that Leibniz agrees with (2), that is, the claim that the perfect scientific theory about the ultimate causes of our confused perceptions would amount to a theory about the configuration and properties of monads, we could still not conclude that Leibniz is committed to the thesis that we perceive things in themselves. In the context of this chapter, we will not be able to reach a conclusive verdict on the question of whether Leibniz agrees with (2), nor will we be able to conclusively decide whether he is indeed committed to (3), that is, the claim that there are possible beings whose senses are acute enough for them to empirically learn something about monads. The answers to both of these questions depend in large part on how Leibniz conceives of the relation between bodies, monads, and aggregates of monads, which goes beyond the scope of this chapter. But without getting into this discussion, we can already identify a number of reasons that strongly suggest a negative answer in both cases. Regarding (2), although Leibniz believes that in order to explain the most fundamental principles of nature one must resort to metaphysical considerations, he also thinks that natural science and metaphysics have different domains that ought to be kept separate. Metaphysics is concerned with the most fundamental level of reality, physics is concerned with the (middle) phenomenal level. In particular, Leibniz holds that a complete explanation of the phenomena, which is the aim of natural science, is possible in purely mechanistic terms, that is, by appeal to the number, size, shape, motion, and derivative forces of bodies, and that references to simple substances or substantial forms are out of place in the context of natural science.75 This 74  See notes for letter to des Bosses, February 5, 1712, GP II 438; “Discourse on Metaphysics,” §14, A VI 4, 1550; “Theodicy,” §403, GP VI 356; Fifth Letter to Clarke, §87, GP VII 411. 75 See “Specimen Dynamicum,” GM VI, 242–243; letter to Arnauld, July 4/14, 1686, GP II 58; “Introduction to a Small Book on the Elements of Physics,” A VI 4, 2005–2007; “Discourse on Metaphysics,” §10, A VI 4, 1542–1543.

Anja Jauernig  209 means that insofar as the described ersatz-­analyses of our confused sense percep­ tions are supposed to proceed by scientifically investigating their causes, we can say a fortiori that monads will not feature in them, not even if the analyses are part of a perfect all-­encompassing scientific theory. Essentially the same point can be made by emphasizing that monads do not interact or cause anything except their own perceptions,76 and, more specifically, that monads do not stand in the kind of causal relations that obtain between our confused sense perceptions and specific mechanical properties, which belong squarely to the phenomenal level of reality. Whatever the ultimate causes of our confused sense perceptions might turn out to be, monads will not be among them; to expect otherwise amounts to a confusion of ontological levels. Finally, the key to seeing that Leibniz is (very probably) also not committed to (3), that is, the claim that there are possible beings who differ from us only in that their senses are more acute than ours such that they can learn something about things in themselves by experience, is to recall that matter, which is what the perceptions of monads represent, is infinitely divisible on Leibniz’s view. The benefits of more acute senses for a cognizer would be that some of the more detailed, minute features of bodies that are washed out in our confused perceptions would come into view. But due to the infinite divisibility of matter, for any one of these more detailed, more minute features that are represented in the cognizer’s conscious perceptions, there would be many even more detailed, even more minute features, which would be represented in his little perceptions but of which he would not be conscious. And, again, since matter is infinitely divisible, this holds no matter how acute we assume the senses of the cog­ nizer to be. If our eyes became better equipped or more penetrating, so that some colors or other qualities disappeared from view, others would appear to arise out of them, and we should need a further increase in acuity to make them disappear too; and since matter is actually divided to infinity, this process could go on to infinity also.  [RB 219]

In order to conclusively establish from this that Leibniz disagrees with (3), we would also have to show that on his view there can be no monads whose bodies have infinitely acute senses or that matter is not literally composed of monads. On my reading, monads with infinitely acute sense perceptions are ruled out for Leibniz because all monads and their bodies, qua created beings, are essentially

76  See Mon 7, GP VI 607; letter to Arnauld, July 14, 1686, GP II 57; Remarks upon Arnauld’s letter, GP II 46–47; DM 9, §14, A VI 4, 1542, 1550; “New System,” GP IV 484, 485–486; “Addition to the Explication of the New System,” GP IV 577–578; “Conversation between Philaréte and Ariste,” GP VI 591.

210  Kant and the (Alleged) Leibnizian Misconception finite. And, to my mind, Leibniz also holds that being infinitely divisible is incompatible with being literally composed of simple parts, even if the parts are non-­extended and even if there are infinitely many of them, which, in turn, implies that matter is not literally composed of monads. Providing a defense of this reading is the topic for a different paper. But I take it that our modest discussion in this section is sufficient to show that there are good reasons to think that at least one of the underlying concerns that fuel Kant’s passionate and enduring interest in the alleged Leibnizian misconception of the difference between sensible and intellectual representations, namely, that the Leibnizians are committed to the claim that we (or very similar beings) could learn something about things in themselves by experience if our senses were more acute or our powers of disfusing perceptions were stronger, does not apply to at least one prominent Leibnizian, namely, Leibniz himself.

8

Kant’s Amphiboly as Critique of Leibniz Martha Brandt Bolton, Rutgers University*1

An often-­neglected chapter of the Critique of Pure Reason, “On the amphiboly of concepts of reflection,” identifies an ambiguity that arises in context of transcendental idealist doctrine. An object about which some judgment is to be made can be known either by understanding alone or also by sensibility, so there may be ambiguity as to which of the cognitive faculties the object comes under. Without the precaution of “transcendental reflection,” taking account of the faculty by which the object in question is known, a metaphysician is liable to adopt principles governing objects known by one faculty which are appropriate only for objects of another faculty. This error is not just an abstract possibility. Kant charges it against two of his prominent predecessors—Locke and Leibniz. With regard to Leibniz, exposing the fallacy “has the advantage of setting before our eyes that which is distinctive in his theory in all its parts and the leading ground of this peculiar way of thinking, which rests on nothing but a misunderstanding” (A 270/B 326). The claim to identify a single source of Leibniz’s metaphysics, let alone an erroneous one, has considerable interest whether it succeeds or not. Transcendental idealism founds the very possibility of metaphysics as a science on what it is possible to know a priori by cognitive faculties such as ours. But scholars are not accustomed to thinking that Leibniz’s metaphysics is significantly affected by his views about cognition. Kant suggests that several of his main metaphysical principles can be traced directly to this source. Indeed, the remark that “Leibniz constructed an intellectual system of the world” (A 270/B 326) rings true. Every truth and every event is subject to the principle of sufficient reason; all truths are “analytic” in his system. Less promising, however, the discussion of the amphiboly apparently mischaracterizes Leibniz’s views of sense perception and understanding. As Kant portrays them, sensibility is not a “special source of representations which bring with them their own distinctions”; for Leibniz; “sensibility was only a confused kind of representation” which obscures the

*  For helpful discussion, I am indebted to participants in second annual Leibniz Society of North America Conference, “Leibniz and Kant,” in Lexington, KY in 2009 and members of the University of Illinois at Chicago Philosophy Department Colloquium in 2012 especially Dan Sutherland and Sally Sedgewick. I am very grateful for Christian Barth for comments on a draft of this chapter. Martha Brandt Bolton, Kant’s Amphiboly as Critique of Leibniz. In: Leibniz and Kant. Edited by Brandon C. Look, Oxford University Press. © Oxford University Press 2021. DOI: 10.1093/oso/9780199606368.003.0008

212 Kant ’ s Amphiboly as Critique of Leibniz logical form of understanding. “In a word, Leibniz intellectualized the appearances, just as Locke totally sensitivized the concepts of understanding . . .” (A 271/B 327). That is, Leibniz takes sense perception to be nothing but confused conceptual representation. Several scholars have shown this to be incorrect.1 It is true that perception is an inherently confused mode of representation, for Leibniz, but concepts (notions), which are the representations proper to understanding, are “confused” in a different sense. A concept is confused in a semantic sense; it is confused just in case it has a definition but it does not enable us to say what it is. By contrast, perceptions are confused when they represent many contiguous things which are so similar that none stand out from the others.2 Although this is an important and well-­established point, it tells us nothing about what Kant found in Leibniz’s thought to warrant the charge that he intellectualized the appearances. Some explanation of this might be gleaned from scattered texts,3 but Kant suggests the characterization is based on certain well-­ known metaphysical doctrines of his predecessor. Just how is it that these doctrines signal the conflation of sense and intellect to Kant? Even if he misses something important in Leibniz’s theory of cognition, he is on to something right. This chapter is an attempt to identify the insights of Kant’s Critique, on the one hand, and misapprehensions of Leibniz’s views on cognition, on the other. The focus is on getting Leibniz’s theory straight. This should be considered a preliminary inquiry because the Critique, read in light of Leibniz’s actual theories, suggests questions, perhaps difficulties, which cannot be pursued here; moreover, Kant’s criticisms might be pressed on the basis of doctrines that are not out on the table in the arguments which explicitly charge Leibniz with fallacy. The claim that Leibniz’s distinctive tenets “rest on nothing but a misunderstanding” implies some sort of error on Leibniz’ part. It can be taken to be an error which undermines his tenets in the eyes of a neutral critic, e.g. incoherence, inconsistency, explanatory inadequacy or indisputable falsity.4 Errors of this sort disqualify Leibniz’s metaphysics in the eyes of even non-­Kantian metaphysicians. On an alternative interpretation, the error arises specifically in context of

1  McRae (1976, 126–129); Parkinson (1982); Parkinson (1981); Wilson (1999); also see Wilson (1990). Pereboom (1991,53–54) suggests this is insufficient to show that Kant is wrong about Leibniz. 2  On concepts, see “Meditations on Knowledge, Truth and Ideas,” GP IV 422; AG 23–24; on perceptions, see e.g. Mon 19; GP VI 610; Mon 60; GP VI 617; to Arnauld, GP II 90; LA 113. 3  A few texts in which “thought” [pensé] is used to refer to perceptions have been mentioned as possible sources of Kant’s remark (see Parkinson 1981: 305–309). This usage conforms to Descartes’s broad use of “perception” for sensory, imaginative, and purely intellectual cognition. Leibniz tends to follow suit especially in texts meant for a Cartesian audience. But often he reserves “thought” for propositional knowledge and uses “perception” for the sort of cognition to which sense perception belongs; his preference for this more precise usage is explicit in RB 210. RB sometimes follows Locke in using “idea” to refer to both sense impressions and general representations, but the former are more properly called “images”; see e.g. RB 261–262, 451, 392. 4  Scholars who adopt this interpretation include McRae (1976), Parkinson (1981), Pereboom (1991).

Martha Brandt Bolton  213 transcendental idealism. Leibniz goes wrong because his claims are inconsistent with the doctrine of synthetic a priori knowledge and broader argument of the first Critique. The interest of the amphiboly, read in this way, lies in its analysis of a systematic disagreement which may clarify what sets the two philosophers apart. By identifying one cognitive theoretical doctrine from which four signature metaphysical principles of Leibniz can be deduced and from the denial of which his own opposite principles follow, Kant reduces many disagreements to one, to borrow a phrase. Although these notions of error are distinct, they are not incompatible. Kant apparently aims to prove both—to show in a systematic way why he and Leibniz disagree thereby discrediting his philosophy, and also to expose difficulties which do not have their source in the specifically Kantian phil­ oso­phy. To anticipate, this chapter argues that the Critique poses no conclusive disqualifying objection against Leibniz. Moreover, its analysis is flawed by misunderstandings, but even so, it points the way to a direct confrontation of metaphysical systems largely shaped by views about how things can be known. Another matter of concern has to do with the fallacy itself. It is sometimes described as a conflation of faculties, acknowledging only one faculty in a cognizant being which actually has two, understanding and sensibility. It is also said to consist in an illicit assimilation of things as they are represented by pure intellect and sensibility: it is “a transcendental amphiboly” to confuse “the pure object of the intellect and the appearance” (A 270/B 326). This vacillation is harmless in Kant’s eyes because he takes sensibility and understanding to be distinct faculties and pure understanding to be a restricted use of the latter. To Kant, Leibniz will seem to overlook a difference in our cognitive faculties unless he maintains that we cognize object as having aspects which cannot be known by pure intellect. More specifically, because Leibniz holds that all possible objects of knowledge are apprehended by intellect alone, Kant infers he posits no faculty other than intellect. In his view, a second faculty would need to represent objects as something pure intellect does not, but intellect suffices on the view in question. This misconstrues Leibniz’s view, for two related reasons. One is that he individuates a cognitive faculty by the representative relation between its acts and objects they represent and the cognitive functions this enables the faculty to fulfill. Perception represents by a device that severely limits what it can represent objects as being. Intellect uses a medium which is capable of representing all truths about all possible objects. The faculties are distinct in virtue of having different functions in human cognition, as I am about to say. Kant’s second misapprehension is failing to understand that Leibniz contrasts the manner in which all possible objects are known in the understanding of God and the way objects are known by finite rational souls such as ours. Although the former is purely intellectual, the latter is both conceptual and perceptual, and must be for cognitive theoretical reasons. Perception is necessary for human thought; it has a

214 Kant ’ s Amphiboly as Critique of Leibniz function in the generation of thoughts which cannot be performed by intellect.5 Kant is right that for Leibniz, there are ways in which perception serves finite cognizant beings in lieu of the purely intellectual understanding enjoyed by God, but wrong in concluding that he conflates perception and conception. This inference depends on a way of distinguishing faculties which Leibniz does not share and is under no apparent pressure to adopt, or so I will argue. The amphiboly, Kant explains, affects “relations of comparison,” ways of bringing objects together, not under one concept and not by means of a judgment, but rather in respects presupposed by the formation of concepts and judgments. These are: (i) sameness and difference, (ii) agreement and opposition, (iii) inner and outer, and (iv) matter and form. Each comparison is liable to be vitiated by the conflation of cognitive faculties proper to the objects in view. For instance, if objects of sensibility are compared by a principle which pertains solely to objects of pure intellect, the sort of object in question is ambiguous. It may be helpful to review the tenets and terminology of Kant’s theory of cognition which are directly relevant to the treatment of the fallacy although a very brief account will have to suffice. According to Kant, our faculties of sensibility and understanding differ in that each is the source of representations of a type different from the other. Things cognized by understanding alone are called “objects of pure understanding” and things cognized by sensibility alone are “appearances.” For Kant, as for Leibniz, understanding (intellect) is the faculty that uses concepts and forms judgments (propositions). Kant takes knowledge of an object by concepts alone to be an “intellectual cognition which undertakes to determine its object without supplementation by the senses” (A 280/B 336). Such an object is called either “the thing in itself ” or noumenon. Suffice it to say that a thing in itself is constituted independently of human cognition and for that reason, not a possible object of our knowledge. The term nuomenon refers to the concept of an object in general which corresponds to substance in experience, but it is not something of which we can have positive knowledge. By contrast, sensibility is our faculty of intuition, or cognitive receptivity to what there really is. Sensibility gives us particular representations; they are sensations ordered by space and time, the a priori forms of sensibility.6 The appearances we receive from sensibility alone are not, so far, cognized by concepts; and although understanding forms concepts, it cannot provide things with ground in reality to which they can be applied. But there is a function of the soul which combines appearances with appropriate concepts to construct objects of experience, particular things about which we can make judgments and know verifiable truths. From Kant’s point of view, Leibniz and Locke fail to realize that 5  “[W]e could not think even about thought itself if we did not think about something else, i.e. about the particular facts the senses provide” (RB 212). 6  A 20/B 34.

Martha Brandt Bolton  215 neither understanding alone nor sensory cognition alone can yield experience. To assess the strength and accuracy of Kant’s Critique, I will consider each instance of the fallacy in turn.

1  Debatable Claims about How Space and Time Determine Objects In the first two instances, Leibniz is charged with mistakenly taking principles which are true of objects known by concepts alone to apply to things given in sensibility. The comparison of sameness and difference puts the famous principle of the identity of indiscernibles (PII) in question: Leibniz took the appearances for things in themselves, thus for intelligibilia, i.e., objects of pure understanding . . . and there his principle of non-­discernibility (principium identitatis indiscernibilium) could hardly be disputed, but since they are objects of sensibility, and the understanding with regard to them is not of pure but of empirical use, multiplicity and numerical difference are already given by space itself as the condition of outer appearances.  [A 264/B 320]

Like other cases of the fallacy, this one has two parts: “part A,” as I will call it, deduces a metaphysical principle from the theory that objects compared in a certain respect are known by concepts alone; “part B” derives an opposite principle from the theory that things are known by sensibility. A specific difficulty for Leibniz is drawn from the two parts. The argument can be stated as follows. (A1) Assuming that X is an object of pure understanding and so is Y, their numerical sameness or difference can be discovered by nothing but the general concepts that apply to them. Accordingly, X and Y are numerically identical just in case they satisfy exactly the same set of concepts. But they satisfy exactly the same concepts if, and only if, they are qualitatively just alike. PII follows from this.7 On the other hand: (B1) If X is represented by sensibility and so is Y, then X and Y are located in space and time. One part of space is numerically different from another just in case the one is entirely outside the other (i.e. they have no common part).8 So X and Y are numerically distinct just in case they have different positions in space (at a given time): “Without further conditions, the difference in place already makes the multi­pli­ city and distinction of objects as appearances not only possible in itself but also necessary” (A 272/B 328).9 Leibniz is accused of holding a principle of in­di­vidu­ ation for sensible objects that has no grip. On Kant’s theory that space is the a

7  See A 281/B 338.

8  A 264/B320.

9  Also A 272/B 328.

216 Kant ’ s Amphiboly as Critique of Leibniz priori form of outer sense, sensory appearances are, and cannot fail to be, distinct in virtue of their spatial locations considered at a time.10 It is clear enough that Kant’s theory of sensibility subverts PII, but if the argument just stated is meant to discredit Leibniz’s thesis in the eyes of a critic who does not subscribe to the doctrine of space as a priori form of outer sense, the reason is far from clear. For one thing, PII cannot be disproved by empirical means, because even if we discover no qualitative difference between, say, differently located drops of water, they may still differ in some qualitative respect too small to be detected by means at hand. Moreover, Leibniz propounds an alternative basis for the individuation of spatial positions: time and place do not constitute the core of identity and diversity, despite the fact that diversity in time or place brings with it differences in the states that are impressed upon a thing, and thus goes hand in hand with diversity of things. To which it can be added that it is by means of things that we must distinguish one time or place from another, rather than vice versa; for times and places are in themselves perfectly alike . . .  [NE 230]

Unless this theory of spatiality can be refuted on non-­Kantian grounds, PII remains viable if considered independently of Kant’s doctrines. But the third and fourth instances of amphiboly target Leibniz’s theory of spatiality. If they hold up (without aid of transcendental idealist claims), the claim that spatial temporal positions are individuated by intrinsic features of things which hold them gives no support to PII. It would be premature to cite the passage just quoted in an effort to show that Kant’s argument begs the question. Another formulation of the first amphiboly purports to derive the difficulty from a purely logical error. This is the logical absurdity of supposing that what is not contained in a general concept is not contained in the particular concepts under it, “for the latter are particular concepts precisely because they contain more than is thought in the general concept.” It would certainly be absurd to say that because the concept of an animal subsumes that of a human being, the latter contains nothing which is not in the former. “Yet Leibniz’s entire intellectual system is really built on [this] principle; it therefore falls together with it, along with all of the ambiguity in the use of the understanding that arises from it” (A 281/B 337). Applying this to sameness and difference: “The principle of indiscernibles is really based on the presupposition that if a certain distinction is not to be found in the concept of a thing in general, then it is also not to be found

10  The principle of the identity of indiscernibles is explicitly applied to substances in Discourse on Metaphysics and several other texts, but also applied to sensible things such as leaves, blades of grass, drops of milk, and indeed everything in nature (see e.g. “Primary Truths,” A VI 4 1645; AG 32; Fourth letter to Clarke, GP VII 372; AG 327; Mon 9; GP VI 608; AG 214).

Martha Brandt Bolton  217 in the things themselves; . . .” (A 281/B 337). Here the relation between a concept and a thing in its extension is in question. But the logical absurdity pertains to concepts one of which subsumes the other. These are logically different relations which hold between different sorts of things; it is not safe to assume that what is absurd for one is absurd for the other. There is a complete concept which somehow contains all truths about the thing in its extension, according to Leibniz;11 only one concept is involved. It is just that existence does not determine a thing in any respect left indeterminate by a certain concept of it. Against this, Kant can show that no concept determines the numerical identity (or grounds the particularity) of the thing(s) in its extension by invoking sensibility, but not by the logical absurdity alone.12 Still, argument (A1) identifies a constraint on what Leibniz can consistently say. It makes a strong case that if objects are known by concepts alone, then they conform to PII. This has implications for Leibniz because he subscribes to the antecedent. According to him, everything in the actual world is known in the understanding of God as a condition of its actuality. Leibniz subscribes what is now called an “actualist” theory that all possible things are grounded in something actual.13 The actuality of finite things is grounded in what is possible, and finite things are possible, at all, only because their reality is grounded in God’s understanding. They are made actual by divine will which presupposes intellect. God is also invoked to ground the metaphysical necessity of necessary truths: 14 God is . . . the source of . . . essences insofar as they are real, that is, the source of that which is real in possibility. This is because God’s understanding is the realm of eternal truths or that of the ideas on which they depend; without him there would be nothing real in possibilities, but also nothing would be possible. [Mon 43; GP VI 614; AG 218]15

This puts Leibniz’s metaphysics squarely within the scope of (A1); his account of the ground of the possibility of finite things commits him to PII on pain of logical inconsistency. 11  E.g. DM 8–9. 12  The question whether existence is a predicate is closely connected to this aspect of the Critique. It might seem to indicate that Leibniz must take the relation between a substance’s existing in the world and its being conceived as possible in the understanding of God to be primitive, but the issues cannot be pursued here. Another question is how Leibniz might defend the view that a complete concept cannot be divided by more than one particular, as the concept of a human being is divided by numerous particular things. 13  Also Mon 44; GP VI 614; AG 218. 14  See Mugnai (1990); Adams (1994, 177–183). 15  “. . . one should not say . . . that the eternal verities would exist even though there were no understanding, not even that of God. For it is, in my judgment, the divine understanding which gives reality to the eternal verities, albeit God’s will have no part therein. All reality is founded on something existent . . . [W]ithout God, not only would there be nothing existent, but there would be nothing possible” (T 184).

218 Kant ’ s Amphiboly as Critique of Leibniz What does this imply about how we, finite spirits, cognize actual things and represent them as numerically different? Kant answers on the basis of his theory that there is a faculty distinct from pure intellect only if it represents objects as being something that pure intellect does not. It seems to him that Leibniz cannot posit a second faculty which represents things as individual. The intelligibility of PII blinds him to the fact that we cognize particular things by a faculty distinct from intellect: Since he therefore had before his eyes solely their concepts, and not their position in the intuition in which alone the objects can be given, . . . it could not have turned out otherwise but that he extended his principles of indiscernibles . . . to the objects of the senses (mundus pheanomenon) . . .  [A 272/B 328]16

Yet this is not a correct account of Leibniz’s position. He clearly states that we (human spirits) have both a faculty of sense perception and an intellectual fac­ ulty.17 The latter enables us to form (propositional) thoughts about the things we perceive by sense. Propositions contain concepts, which are products of ideas innate in the human mind. Indeed, innate ideas represent possible entities (essences) in God’s understanding.18 But in addition, we have a faculty of perception which represents objects without subsuming them under concepts. In fact, the non-­conceptual basis on which perceptions represent the things they do is logically sufficient to ensure that all objects of perception conform to PII. Leibniz need not invoke the thesis that all possible objects are in known by pure intellect to establish e PII, as Kant assumes. We will look at why this is the case shortly. Before that, I want to introduce the second case of the fallacy, comparison in respect of “agreement and opposition,” because it raises similar issues. Here the objects in view are causes, efficacious tendencies such as physical forces, psychological tendencies, and feelings of enjoyment and pain.19 The notions of “agreement” and “opposition” are tailored to such things. Suppose X and Y are forces tending to cause motion. In Kant’s terminology, X and Y agree provided it is possible for X and Y to be present in, or applied to, the same subject at the same time. X and Y conflict just in case they do not agree, but there are two sorts of conflict. First, X and Y may be mutually contradictory; in this case, the presence of them both in the same subject would render the subject impossible. Second, X and Y conflict if they are opposed. That is, although it is possible for both to exist in, or

16  Also see A 270–271/B 326; A 280/B 336. 17  E.g. RB 172. 18  DM 28; GP IV 453; AG 59; RB 447. 19  A 265/B 321. The text also mentions the morally significant application of the doctrine that all realities agree: “According to this principle, e.g., all ills are nothing but consequences of the limits of created things, i.e. negations, since these are the only opposing things in reality [assuming things are known by concepts alone]” (A 273/B 329).

Martha Brandt Bolton  219 be applied, to the same subject at once, they cancel each other out—each destroys the efficacy of the other.20 The argument is made with regard to concepts which are merely affirmative, containing no negation. Kant argues that if forces are known by such concepts alone, they always agree, yet forces which conflict can be given in sensibility: there is no contradiction at all in the concept of a thing if nothing negative is connected with something affirmative, and merely affirmative concepts cannot, in combination, effect any cancellation. Yet in the sensible intuition in which reality (e.g. motion) is given, there are conditions (opposed directions), from which one had abstracted in the concept of motion in general, that make possible a conflict . . . that produces a zero = 0 out of what which is entirely positive . . .  [A 282/B 339]

We might state the argument this way. (A2) Assume X and Y are physical forces. If objects of cognition are represented by concepts alone, then a comparison of concepts suffices to discover their relations of agreement and opposition. The only conflict discoverable from concepts alone is contradiction. If the concepts of X and Y are purely affirmative, they cannot conflict; so, they cannot be opposed. (If one contains the negation of something contained in the other, they conflict, but because this precludes the possibility of the subject, they cannot be opposed.) On the other side: (B2) Assume X and Y are appearances. Sensibility (outer sense) requires that X and Y be directed toward certain actual bodies, or positions, in actual space. In space, it is possible that two forces act in opposite directions, which makes it possible for forces to act against each other and cancel each other out. Kant evidently thinks this shows that Leibniz holds a metaphysical theory with physical implications which are true of objects of pure understanding but false with regard to objects of experience. Leibniz holds that all possible objects are known by intellect alone but (A2) fails to show that this forces him to deny the possibility of opposed forces. Two issues are at stake: the possibility of forces acting in opposite directions (e.g. forces causing two bodies to collide) and the reality of forces which cancel each other out. If Kant purports to show that Leibniz’s position on either issue is untenable on grounds logically independent of the doctrine of the a priori form of sensibility, his success is questionable. Let us consider each issue in turn. First, although Leibniz does not posit a continuous spatial order in which perceptually cognized objects are given, he holds that space is accessible intellectually. Space, for him, is a uniform continuous ordering relation “which

20 See the wonderful essay “Attempt to Introduce the Concept of Negative Magnitudes into Philosophy” (Kant 1992).

220 Kant ’ s Amphiboly as Critique of Leibniz indicates possibilities beyond any that might be supposed actual.”21 A math­em­at­ ic­al continuum cannot possibly exist in light of the problem of the composition of the continuum, as Leibniz sees it.22 In reality, extended things are composed of extended parts without end; points are real only as terminations of the extension of parts. Continuous space is, then, an abstraction with reality grounded in the understanding of God.23 In us, the idea of space is innate. It is presupposed by ideas (and concepts)24 of determinate spatial configurations. Because space is essentially undifferentiated, and space cannot be in the least part continuous, possible and actual things comprising spatial relations are extrinsic to space and partially realize it. It may be helpful to explain that because all objects of perception are extended, for Leibniz, they are, as we just said, endlessly divided into parts and perception represents them as such. They are densely ordered, not continuous. Still Leibniz maintains that when perceptions are sufficiently distinct for us to be aware of them, there is a continuous homogenous sensory appearance. Speaking very roughly, it is due to several factors according to his theory: the fact that all perceptions of bodies contain perceptions of each of their parts; the fact that a perception of a whole is to some degree more distinct than the perceptions of its parts; the fact that when attention and reflection make us aware of (apperceive) a distinct perception, imagination produces a homogeneous image in which the many confused perceptions contained in the distinct one are somehow expressed by amalgamation.25 In a human soul, then, the sensory appearance of spatial continuity results from a mental operation on endlessly many discontinuous perceptions contained in a distinct perception. For Leibniz, space is the source of truths with regard to spatially determined possibilia; for instance, spatial structure determines the impossibility of a ten-­ sided regular polyhedron, and the possibility of motions in opposite directions which converge at a point. Toward the end of his life, he makes some progress characterizing spatial relations in general,26 but he did not have axioms which fully specify the structure of space. In the absence of that, he relies on the notion

21  RB 154; also 149, 152, 154–156. 22  Leibniz takes the fact that the points on a given line are in one-­one correspondence with the points on any other line, even though some lines are longer than others, to show that the view that extension is composed of points conflicts with the maxim that a whole is greater than its part; he concludes that extension cannot be composed of points; see e.g. “Pacidius to Philalethes: A First Philosophy of Motion,” A VI 3, 549–553; Arthur, 173–181. 23  E.g. RB 149, 154–155; Leibniz’s Fifth Letter to Clarke, GP VII 398; AG 336; see Mugnai (1990, 162–165). 24  In us, ideas are innate as dispositions to form concepts and thoughts of the things they represent. Concepts are not, strictly speaking, “ideas” although Leibniz sometimes uses that word to refer to them. On the distinction, see e.g. DM 27; MKTI uses the strict terminology. 25  E.g. MKTI, last sentence; RB 403, 133; also Bolton (2011, 148–150). 26  See De Risi (2007, ch. 2).

Martha Brandt Bolton  221 of a real definition to prove the possibility of purported spatial determinations.27 The doctrine that space is an undifferentiated ordering relation abstract by comparison with all spatial determinations and real spatially determined things is available from roughly 1700 and it is not idle.28 What we have said is sufficient to put the soundness of (A2) in question. If the concepts of two different spatial relations are entirely affirmative, it may still be the case that a conflict between them can be discovered by concepts alone. For the conflict might be found in the structure of ideal space presupposed by concepts of spatially determined things. Kant apparently assumes that if an object is known by concepts alone, then concepts which are predicates of that object determine all truths about it.29 Yet Leibniz maintains that general truths about, say, a cube are determined by the idea of space, which is not a predicate of a cube but presupposed by its concept. Again, (A2) states that if concepts alone are considered, any two concepts which are not mutually contradictory may be applied to the same thing. Yet Leibniz is in position to say that the structure of space may preclude the possibility that concepts of two different partial instantiations of spatial order apply to the same thing at the same time. For him, it may be due to space, not a contradiction between concepts, that, for instance, a body (a point) cannot move in opposite directions at the same time. Kant seems not to be aware that Leibniz comes to hold a view of concepts and how concepts can be combined in propositions different from the traditional view that two or more concepts can be applied to a possible thing if they are not mutually contradictory. Indeed, Kant may understand two well-­known doctrines of Leibniz in such a way that, in his eyes, they imply a traditional analysis of the form of categorical propositions—the “analytic” theory of truth, in every true (categorical) proposition the concept of the predicate is in some way contained in the concept of the subject, and the doctrine that a necessary truth is either an “axiomatic identity” or can be reduced to one by a finite analysis of its subject and predicate concepts.30 Although I am inclined to think Leibniz’s hierarchy of ideal space in relation to its de­ter­min­ ations is consistent with these doctrines properly understood, the issues are too complex for consideration here. It is enough for my present thesis that the argument in (A2) is inconclusive. Leibniz is not, then, clearly barred from holding that forces acting in opposite directions are possible, and indeed he supposes they are real. Nor is he clearly precluded from granting the possibility of “opposed” forces, in Kant’s sense,

27  A real definition shows that is defindiendum is possible either by analyzing its concept back to simple concepts, or figuring out how to cause it, or finding that it actually exists; e.g. “Meditations on Knowledge, Truth and Ideas,” GP IV 424–425; AG 25–26; RB 294–295, 347. 28  See Hartz and Cover (1988). For use of the doctrine in NE, see Bolton forthcoming. 29  This assumption is explicit in argument (A1) as formulated earlier in this chapter. 30 E.g. De Natura Veritatis, Contingentia et Indifferentiae, A VI 4, 1515–1524; MP 96–105; RB 361–367.

222 Kant ’ s Amphiboly as Critique of Leibniz although he does, in fact, maintain that forces never cancel each other out. The reasons for this would take some time to explain; perhaps most important, created substances, whether corporeal or incorporeal, cannot causally act on each other, as he sees it.31 The principle of continuity is also relevant.32 Neither of these doctrines is mentioned in the second instance of amphiboly, or implicated by it as far as I can see. So, Leibniz’s commitment to the doctrine that all possible objects are known by intellect alone appears not to put him under pressure to hold the theory of forces he does. Second, Leibniz develops a theoretical physics on which the forces which cause bodies to collide, rather than destroying each other, are instead impeded, diverted, and dispersed over a multiplicity of smaller bodies. Roughly put, as he explains it, the force moving a body involved in collision is subsequently directed inward to the parts of that body thereby causing it to deform and eventually rebound away from the point of impact.33 Kant elaborates (B2) in a way which gestures toward a simple empirical refutation of this account: opposition is “unceasingly placed before our eyes by all hindrances and counter effects in nature.”34 There is mention of a rule of empirical mechanics to the effect that if equal forces acting in opposite directions are applied to the same body, their effects are equal to zero. But because the point at issue is a matter of the correct explanatory theory, simple appeals to observation, and rules which generalize from observed cases, do not suffice to settle it.35 The attempt to show Leibniz’s physics to be mistaken is, so far, inconclusive, as is the effort to show he cannot acknowledge the possibility of causal tendencies in opposite directions. Let us return to the point that Leibniz has two ways of showing that PII is true of the actual world: one assumes that all possible beings are known by concepts alone (see A1); the other is based on his theory of perception. The latter hangs on Leibniz’s view that one thing represents another just in case one “expresses” the other.36 This notion of expression is well known, and I will not go into detail.37 Roughly speaking, an expression is isomorphic to the thing it expresses. Examples on offer include a map, a model of a machine, an algebraic equation for a geometrical figure, and a circle which projects onto a plane an ellipse, a parabola, and a hyperbola. The optical geometrical examples are especially pertinent,

31  E.g. DM 14. 32  Specimen Dynamicum, GM 6 234–254; AG 117–138. 33  See Geuroult (1967, 168–171). Leibniz also maintains that psychological tendencies sim­ul­tan­ eous­ly present in a soul do not diminish or destroy one another but only limit the degree to which each realizes its full effect. 34  A 273/B 329. 35  E.g. RB 51, 475; PNG 5; GP VI 601; AG 208–209; “On What Is Independent of Sense and Matter,” GP VI 495–496; AG 190. 36  To Arnauld, October 9, 1687, GP II 112; letter to Masson, GP IV 627–628; A28–29; “What Is an Idea?” GP VII 264; L 208. 37  See Kulstad (1977); Swoyer (1995).

Martha Brandt Bolton  223 because the respect of structural similarity is a spatial arrangement among spatial parts or points.38 In theory of perception, Leibniz exploits the insight that an immaterial monad can comprise a multiplicity of elements mutually related in a way that is, in some respects, the same as the spatial relations among bodies.39 The whole corporeal world is said to be a plenum in which every body is a whole composed of other bodies; that is, given any two bodies, either one is a part of the other or both are parts of a larger whole. In a similar way, a monad has modifications each of which contains modifications as parts and all of which are related as parts and wholes, although these relations are not spatial and the modifications are not extended. Bodily events, also, are endlessly composed of smaller events; and in a monad, every transition from one perceptual modification to another is composed of smaller transitions. Finally, bodies are individuated by their forces, as Leibniz has it.40 Every body has a force which, at every moment, strives to move the body in a certain way relative to other bodies. Corresponding to this, the perceptual state of a monad that expresses a given body at a moment contains an appetite tending to produce perceptions that express the motion to which the body tends at that moment. In short, a monad actively models the corporeal world in virtue of the fact that each has a mereological and causal-­dispositional structure in some ways identical to that of the other.41 Four implications of this doctrine are especially important for our purposes. First, a perceptual modification represents the thing it does in virtue of nothing but a respect of similarity to this thing; it follows that it cannot fail to represent the thing it does just as it is. Related to this, a perceptual state is capable of representing a thing only if that thing is in intrinsic respects unlike everything else. If there were two bodies exactly alike in their qualities, they would be exactly alike in the forces they contain, because qualities reduce to forces on Leibniz’s mechanist physics. Now if bodies were perfectly alike in respect of their forces (qualities), then a perceptual modification that expressed one of the bodies would also express the other; in this case, neither would be perceived. An object is, then, perceived only if it conforms to PII. Moreover, the theory of perception is incompatible with the possibility that physical forces cancel each other out, for that would imply that it is possible for all forces acting in a body to be annihilated, with the result that the body could not be perceived. 38  See Swoyer (1995); De Risi (2007). 39  See letter to Masson, GP IV 627–628; AG 228–229. Leibniz distinguishes bare perception and sense perception, but because the latter is a special case of the former, I neglect the distinction in the present paper (e.g. PNG 4; to Arnauld, GP II 112, LA 144). Leibniz also distinguishes sense perception and apperception, i.e. sense perception of which one is aware (e.g. RB 53–55, 134). This distinction is neglected in the body of this paper because it does not affect the matters at issue between Leibniz and Kant, as far as I can tell. 40  “On Nature Itself,” §13, GP IV 512–514; AG 163–165. 41  Letter to Masson, GP VI 627–628; PNG 3; Mon 56.

224 Kant ’ s Amphiboly as Critique of Leibniz The third point is less obvious: perceptual representation depends on formal features different from those on which conceptual representation depends. A perceptual modification models the things it represents and represents everything it models. By contrast, a concept represents things that satisfy it, and such things do not generally model the concept. To put this differently, perception cannot apprehend things without parts; a perception expresses an object in virtue of the part whole structure they share. But, of course, we have concepts of points, logical connections, etc.42 Evidently perceptions represent on the basis of an iso­morph­ ism very different from that by which concepts represent the things they do.43 There is no need to describe the latter in detail here. Briefly put, a human mind contains innate ideas which are causal tendencies to form concepts of and true propositional thoughts about the objects the ideas express; the network of such causal tendencies and their actual and potential manifestations express (are structurally similar to) possible entities and truths in the understanding of God.44 Not only does he hold that perception and intellect are faculties which represent objects by entirely different structural identities, but he is also able to derive the metaphysical principle targeted by the first instance of amphiboly from his account of perceptual representation alone. The final point is that perception and intellection are distinguished by their disparate roles in the human cognitive system. On Leibniz’s theory, the representational content of our perceptions is so dense, so packed with information, that it comprises every body in the universe and all the endlessly many qualities that individuate it. The vast majority of our perceptions are very deeply confused, but this enables them to represent particular things in all their individual qualitative diversity. Yet, because the distinguishing features of individuals are infinite in number,45 it is beyond the powers of any finite mind to form a distinct, fully defined, concept of a particular thing. The concepts we use in thoughts of particulars are general and insufficient to enable us to cognize them as particular or to recognize them reliably.46 Contrary to what Kant may suppose, Leibniz gives perception a function which cannot be performed in us by the faculty of intellect; that is, cognition of actual particular things. Perception is, in this sense, an “original source” of representations in a human soul.47 42  See Leibniz to Bierling, GP VII 150. 43  All perceptual modifications have parts, but Leibniz maintains that there are simple concepts (notions, ideas) of which all other concepts are composed; e.g. MKTI, A VI 4, 589–590; AG 26; also see Mates (1986, 59 n. 45). 44 DM 28; GP IV 453; AG 59; RB 447; Bolton (2012). “That the ideas of things are in us means . . . nothing but that God, the creator alike of the things and of the mind, has impressed a power of thinking on the mind so that it can by its own operations derive what corresponds perfectly to the nature of things . . . [T]ruths can be derived from [the idea of a circle] which would be confirmed beyond doubt by investigating a real circle” (“What Is an Idea?’, GP VII 264; L 208; also see RB 52, 86). 45  E.g. RB 230, 289–290. 46  RB 289–290. 47  According to Pereboom (1991, 54), what is needed to prove Kant is wrong is “a text which shows that Leibniz distinguishes mutually exclusive types of representations according to role or task.”

Martha Brandt Bolton  225

2  Monads and the Constitution of Matter The third and fourth instances of amphiboly purport to show that Leibniz ascribes to perception representations which he can only regard as conceptual. This is where the thought that he intellectualizes appearances gets hold. In the third instance (i.e. the comparison of inner and outer), substance (substratum) is in the fore. Kant characterizes it in a way both he and Leibniz accept. That is, a substance is a subject of determinations (accidents48) and because a substance is active, its determinations include forces. Part A proceeds in two stages: First, if substance is known by pure intellect, then it cannot contain parts and it must have inner, or non-­relational, determinations. Second, the inner determinations can only be representative states, such as thoughts. So, substances are, as Leibniz says, monads. This is the most developed version of the argument (inserted numbers mark the two stages):49 [1] According to mere concepts the inner is substratum to all relation or outer determinations. If, therefore, I abstract from all conditions of intuition, and restrict myself solely to the concept of a thing in general, then I can abstract from every outer relation, and yet there must remain a concept of it, that signifies no relation but merely inner determinations. Now it seems as if it follows from this that in every thing (substance) there is something that is absolutely internal and precedes all outer determinations, first making them possible, thus that this substratum is something that contains no more outer relations in itself, consequently that it is simple (for all corporeal things are still always only relations, at least of the parts outside one another); [2] and since we are not acquainted with any absolutely inner determinations except through our inner sense, this substratum would not only be simple, but also (according to the analogy with inner sense) determined through representations, i.e. all things would really be monads, or simple things endowed with representations.  [A 283/B 339]

An outer determination is a relational property, such as being taller than Socrates; to have a relation is to have a relational property.50 The first sentence implies that if the inner is represented by concepts alone, it is a sort of entity different in kind from the outer. To be more exact, the general concept of substratum and the general concept of a relation (relational property) represent discrete classes of things. This is because the concept of substratum specifies an entity which has a certain role with respect to outer determinations in general. A nearby passage refers to this role: “Through mere concepts, of course, I cannot think of 48  A 265/B 321. 49  Also A 265–266/B 321–322; A 274/B 330. 50  Van Cleve (1988, 231–247). This article is very helpful for understanding this passage.

226 Kant ’ s Amphiboly as Critique of Leibniz something external without anything inner, for the very reason that relational concepts absolutely presuppose given things and are not possible without these” (A 284/B 340). That is to say, all concepts of relational properties, taken together, do not make the concept of an object; for that, something more is needed. If the requisite item is known by concepts alone, it is given by the concept of substance. This suggests the following construction of the argument. (A3) [1] Assume objects are known by concepts alone. The concept of substance is presupposed by the general concept of a relational property. So, if we think of the world in abstraction from all outer determinations, the concept of substratum remains. So, the general concept of substance represents no outer determination, or relational property, but rather something simple (without parts). Moving to the second stage. (A3) [2] Assume substance is known by concepts alone. Substance must have some determinations (see A 265/B 321).51 They cannot be outer, so they must be inner determinations—absolutely, rather than comparatively, inner, meaning determinations that are not relational properties internal to the substance.52 We are acquainted with no such inner states other than representations. So, a substance is both simple and endowed with de­ter­min­ ations which are representative (i.e. a monad). It may bear repeating that this argument is relevant not because it can be found in Leibniz texts, but because its validity constrains his metaphysical principles.53 By contrast, substances which are cognized by sensibility cannot be monads, Kant argues: The inner determinations of a substantia phaenomenon in space, on the contrary, are nothing but relations, and it is itself entirely a sum total of mere relations. We know substance in space only through forces that are efficacious in it, whether drawing others to it (attraction) or in preventing penetration of it (repulsion and impenetrability); we are acquainted with no other properties constituting the concept of the substance that appears in space and which we call matter.  [A 266/B 322]

To spell this out: (B3) Assume substance is an appearance. Every appearance is in space. Anything in space is a sum total of spatial relations among its parts. Moreover, substance is known by nothing but forces, which are relational properties. So, substance in appearance is constituted by spatial relations and its determinations are nothing but spatial relations.54 One upshot might seem to be 51  “As object of pure understanding . . . every substance must have inner determinations and forces that pertain to its inner reality” (A 265/B 321). 52 On comparative, as opposed to absolute, inner determinations, see also A 277/B 33 and A 285/B 341. 53  Cf. Parkinson (1981, 312). 54  On why an appearance constituted entirely of relations is possible, see A 285/B 341.

Martha Brandt Bolton  227 similar to the results of the first two instances of amphiboly, that Leibniz holds a doctrine of substance unsuited for substances cognized by sensibility. But Kant draws a different, more embarrassing, conclusion: “Thus because [Leibniz] represented them as noumena, taking away in thought everything that might signify outer relation, thus even composition, Leibniz made out of all substances, even the constituents of matter, simple subjects gifted with powers of representation, in a word, monads” (A 266/B 322, emphasis added).55 That ­monads are present in matter as ultimate elements of its constitution (or perhaps composition [Zusammensetzung]) is, for Kant, problematic to say the least. It threatens to contradict Leibniz’s contention that a continuum cannot be composed of extensionless parts. It also calls attention to tensions in Leibniz’s thought regarding the precise dependence relation between monads, on one hand, and bodies, on the other.56 Although some texts indicate that monads are, at least, located in matter, others posit monads as external foundations of the cor­por­ eal realm.57 Our aim is to understand how Kant derives this conclusion from (A3) and (B3). As I understand it, he starts with the doctrine that representations drawn from sensibility are required for knowledge of the existence of particular things outside us, and we represent these things as material. Leibniz agrees that we perceive particulars which exist and the world and that insofar as perceived, they are perceived as material. Kant seems to reason that in order to account for the fact that perception is needed for knowledge of the existence of particulars, Leibniz was led to think particular substances are present in matter. Yet Kant supposes, perception has no cognitive purpose for Leibniz, so he needed to find something else for the senses to do. More generally: “[Leibniz] conceded to sensibility no kind of intuition of its own, but rather sought everything in the understanding, even the empirical representation of objects, and left nothing for the senses but the contemptible occupation of confusing and upsetting the representations of the former” (A 276/B332).58 The argument rests on the assumption that the representational adequacy of pure intellect precludes a faculty of cognition distinct from it. As we saw, there is no mystery about why perception is required for cognition of particular things according to Leibniz. Still perceptual representation is underwritten by structural similarities which are not shared by things without parts such as monads. Monads are represented as immaterial and simple only by

55  Leibniz “assumed that we intuit things as they are (though with confused representations)” (A 267/B 323); he “took appearances for things in themselves, thus for intelligibilia” (A 264/B 320); also A 274/B 331. 56  See Langton (1998, 78–89); also Wilson (1990, 94–95). 57  E.g. from Leibniz’s later years, see letters to Des Bosses, GP II 399; LDB 98–99; and by contrast, to Des Bosses, GP II 450–451; LDB 224–225; to De Volder, GP II 53; AG 178. 58  See also A 44/B 61 and remarks at A 267/B 323 and A 264/B 320.

228 Kant ’ s Amphiboly as Critique of Leibniz concepts and propositional thoughts. A monad’s perceptual modifications might be perceived, but not a monad, as such. If there is a difficulty in the general vicinity to which Kant points, it concerns the coordination of perceptual and conceptual representations. If human perceivers can form concepts of the motions, tendencies and other material features which perception represents bodies as having, then we can recognize forces, actions, etc. and reason our way to monads, as Leibniz does. The immediate question is how non-­conceptual perceptual representations of particular things are brought together with conceptual representations which are satisfied by the particulars as perceived. Kant is not even apprised of this problem. He does not understand that according to Leibniz, sense perception and understanding are distinct faculties that work together to form propositional thoughts of things perceived by sense. This can be seen in the following cryptic passage from Nouveaux Essais: So “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.  [NE 173]

When a human soul has a distinct perception, the faculty of intellect forms a thought, evidently a thought with reference to the object perceived.59 I take this process to be something like Kant’s synthesis of intuition under concepts; unfortunately, to my knowledge, Leibniz says very little about it. So, although he thinks the human soul forms concepts suited to represent things its current perceptual modifications express, I cannot see that he offers an explanation of how it works. Still, he maintains that thoughts of immaterial things naturally arise when we distinctly perceive material things. “So among the notions which come to us with those of material things, there are ideas of things which accompany matter without being corporeal, such as the notions of force, action, change, time, the same, one, true, good, and a thousand others.”60 This bodes well for his general ability to construct an account of how we perceive certain particular bodies and conceive monads, as such, whose modifications express these bodies in particular. Yet the texts focus on the sorts of experience which prompt us to form the concept of substance for the first time. It is by reflecting on ourselves that we acquire the notions of substance, agent, and the like by virtue of ideas which are innate in us.61 How we subsequently apply these concepts to monads other than ourselves is never clearly explained, as far as I can tell. This is disappointing in 59  Suggested at RB 119, 83, 52. 61  DM 27.

60  To Sophie, GP VII 552; also RB 102.

Martha Brandt Bolton  229 face of Kant’s contention that Leibniz cannot explain cognition of particular substances except by the unfortunate conflation of perceptual and intellectual cognition. While it is plain that he does not take that course, he has no fully worked-­out account to offer instead. Even so, I would urge that there is no apparent difficulty standing in the way. At least, the third instance of amphiboly does not point to one. In sum, the third instance neither exposes an incoherence in Leibniz’s view of the relation between monads and matter nor forces him to hold that sensory presentations are nothing but confused conceptual representations.

3  Space and the Intellectualization of Appearances Kant’s misunderstanding of Leibniz also affects the fourth instance of the fallacy, the comparison of form and matter. Matter is abstractly characterized as what is determinable, in general, and form is the corresponding determination; examples are: genus (matter) and species (form); in a judgment, the concepts it contains (matter) and their connection by means of the copula (form); in a being, its essential components (matter) and their connection (essential form). Now the fallacy under this heading does not affect the objects a philosopher regards as matter, rather than form. Kant takes this for granted. Among objects of cognition, substance and its determinations are matter; space and time are form. The fallacy is a confusion about which of the two grounds the possibility of the other.62 As Kant argues: The understanding . . . demands first that something be given (at least in the concept) in order to be able to determine it in a certain way. Hence in the concept of pure understanding matter precedes form, and on this account Leibniz first assumed things (monads) and an internal power of representation in them, in order subsequently to ground on that their outer relation and the community of their states (namely of the representations) on that. Hence space and time were possible, the former only through the relations of substances, the latter through the connection of their determinations as grounds and consequents.  [A 276/B 232-­2]

The notion of one thing’s grounding the possibility of another can be understood in terms of a familiar notion of cognitive priority. We might say that A grounds the possibility of B just in case (i) knowledge of the existence of A is not (does not contain) knowledge of the existence of B and (ii) knowledge of the existence of B

62  Longeunesse (1998,147–149) is very helpful on this.

230 Kant ’ s Amphiboly as Critique of Leibniz is (contains) knowledge of the existence of A. We might, then, state the argument as follows. (A4) Assume things are known by concepts alone. Then knowledge of the existence of substance does not include knowledge of outer determinations, but knowledge of the existence of outer determinations includes knowledge of the substance they determine. This is shown by the previous instance of amphiboly. So, substance is the ground of the possibility of outer determinations. The spatial and temporal relations of a substance are among its outer determinations. So, substance is ground of the possibility of spatial and temporal relations. So, for Kant, substance is the ground of the possibility of space and time. On the other hand, Kant argues: But if it is only sensible intuitions in which we determine all objects merely as appearances, then the form of intuition (as a subjective condition of sensibility) precedes all appearances and all data of appearances, and instead first make the latter possible . . . But since sensibility is an entirely peculiar subjective condition, which grounds all perception a priori, and the form of which is original, thus the form is given for itself alone, and . . . the possibility [of matter, or appearances] presupposes a formal intuition (of space and time) as given.  [A 267–268/B 323–234]63

The argument can be stated this way. (B4) We intuit space and time without appearances and we do not intuit appearances without space and time, according to the doctrine of the Transcendental Aesthetic. So, if substance is an appearance, then space and time are the ground of the possibility of substance. Leibniz is thus shown to be committed to an account of the grounding relation between cognition of space and time, on the one hand, and cognition of substance, on the other, which is opposite the theory of Kant. To Kant’s mind, this exposes an incoherence in “Leibniz’s famous doctrine of space and time.” The alleged error is a disqualifying difficulty for his opponent’s claim that space is an ideal entity, rather than real in the world as we know it. To maintain this doctrine, Leibniz is supposedly forced to the untenable “intellectualization of appearances.” This indictment rests on a tortuous line of reasoning, as I see it. Kant takes (A4) to show that Leibniz “thought of space as a certain order in the community of substances, and thought of time as the dynamic sequence of their states” (A 275/B 331). Grant, for simplicity’s sake, that this correctly describes Leibniz’s position.64 The disqualifying conflation of sensory and conceptual representations is drawn from a further argument: 63  Parenthetical expressions are in the text; the material in brackets is interpolated by the present author. 64  It ignores Leibniz’s distinction between space, an abstract ordering which cannot possibly exist, and actual substances whose respective modifications partially realize the ordering. On the distinction between the ideal and the real, see e.g. to De Volder, June 30, 1704; GP II 268, trans. Lodge.

Martha Brandt Bolton  231 The uniqueness and independence from things, however, which both of these [sc. space and time] seem to have in themselves, he ascribed to the confusion of these concepts, which made of that which is a mere form of dynamical relations be taken for an intuition subsisting by itself and preceding the things themselves. Thus space became the intelligible form of the connection of the things (substances and their states) in themselves. The things, however, were intelligible substances (substantiate noumena). Nevertheless he wanted to make these concepts valid for appearances since he conceded to sensibility no kind of intuition of its own, but rather sought everything in the understanding, even the em­pir­ ic­al representation of objects, and left nothing for the senses but the con­tempt­ ible occupation of confusing and upsetting the representations. [A 276/B 332; italics added]

Kant takes it for granted that space and time appear to be independent of things which they order. Leibniz is said to ascribe this apparent independence to the confusion of the concepts of space and time, which cannot be defined, as Leibniz knows very well. Accordingly, he maintains that space and time are entities known prior to substances. Now space, time, substances, and their determinations are all known by pure intellect, according to Leibniz, and Kant reasons that he wanted their concepts to apply to appearances. I take it this is because our senses are involved in cognition of things as bodies, as having positions and moving in relation to each other. Kant seems to think Leibniz’s “famous theory of space and time” attempts to explicate the desired application. The phrase probably refers to the well-­known account of the concept of space in Leibniz’s fifth letter to Clarke. Leibniz’s stated aim here is to explain “how men come to form to themselves the concept of space.” The process is not unlike that by which we come to form the concept of substance, mentioned earlier. In both cases, Leibniz needs first to say that when we perceive something by sense (or reflection) our intellect (our innate ideas) forms concepts that are satisfied by the thing perceived and thoughts that are made true by it. He can then proceed to explain how reasoning from such thoughts leads intellect to generate the concept in question. In the letter to Clarke, conceptual representations of bodies, distance relations, and motions as perceived by the senses are taken for granted. Only the reasoning is explained. A sketch will due for our purpose. We consider that bodies have a “certain order of coexistence” in respect to their situation or distance from each other. Leibniz derives the concept of “place” from the fact that a succession of different bodies can have exactly similar distance relations to a collection of bodies which are at rest relative to each other. Concepts of multiple places are derived in much the same way. Finally, space is what we consider “when all places are taken together.” But, so far, space comprises nothing but places conceived in terms of sequences of bodies and their particular, but exactly similar, distance relations to the fixed bodies. In face of this, the mind forms the concept of

232 Kant ’ s Amphiboly as Critique of Leibniz something that is “truly the same” in all the particular cases, something which must be conceived as being “extrinsic to the subjects,” an ordering relation in abstraction from relata. This is the concept of space. This is reason enough to say that space “can only be an ideal thing.”65 As we said, its reality is founded in the understanding of God.66 To explain how certain experiences prompt our intellects to form the idea of space, Leibniz needs to move from perception, or what we perceive things as being, to conception, or how we conceive what we perceive and thoughts we have of it. The text of the letter takes this for granted. Leibniz can be accused of hand waving when it comes to explicating this process, but there is a clear place for it in his theory of human cognition. He is under no theoretical pressure to intellectualize the appearances. In conclusion, Kant’s Critique of Leibniz stems from the insight that a single claim about all possible objects of cognition can, with some plausibility, be regarded as the source of several of his metaphysical doctrines; since this connection is not stressed by Leibniz, the Critique offers an intriguing perspective on the unity of his views. Because Kant’s opposite theory about cognitive objects suffices to derive opposite metaphysical principles, the Critique provides a unique opportunity for systematic appraisal of the main tenets of the contrasting metaphysical schemes. Yet an accurate account of the issues on which they disagree, let alone assessment of the tenability of the two sides, requires correct understanding of what Leibniz holds with regard to cognition, both human and divine. As it is, the arguments explicitly adduced in the second, third, and fourth instances of the fallacy and the difficulties they claim to expose are flawed by Kant’s apparent lack of information about what his predecessor actually holds, as I hope to have shown. In fact, Leibniz’s theory of human cognition has much more in common with Kant’s than the latter seems to realize; nor is this brought out in the secondary literature on the amphiboly. This suggests that their points of disagreement are sufficiently sharp and interesting to deserve more lengthy consideration. This chapter should, then, be regarded as a start which points to further questions about Leibniz and further elaboration of arguments laid down in Kant’s Critique.

65  LC V, GP VII 400–401; AG 338. 66  E.g. RB149, 154, 155; to De Volder, January 19, 1706; GP II 82; L 539. See Hartz and Cover (1988).

9 The Teleologies of Leibniz and Kant So Close Yet So Far Apart Paul Guyer, Brown University

1 “The Critique of Pure Reason Might Well Be the True Apology for Leibniz” It goes without saying that Kant conceived of some of the defining features of his “Critical” philosophy as criticisms of the philosophy of Leibniz as it was known to him. Kant’s insistence that the characteristic propositions of mathematics, natural science, and metaphysics are synthetic rather than analytic; his insistence that intuitions are a distinct and indispensable source of knowledge, not just confused representations of concepts, and indeed provide insuperable limits on all use of concepts, as well as his view that space and time are merely phenomenal, but do not depend upon antecedent relations among objects but rather, in this regard like Newtonian absolutes, condition the possibility of all such relations; his insistence that the principle of sufficient reason needs to be proven and can be proven, but only as a condition of the possibility of experience, not of reality as it is in itself; and his argument that we cannot infer the simplicity of the self or the soul from the simplicity of our representation of it by the unanalyzable “I”—all of these characteristic Kantian doctrines were clearly conceived of as alternatives to equally characteristic Leibnizian doctrines. Nevertheless, in his 1790 dispute with J.  A.  Eberhard, who had argued that Kant’s supposedly new critique of pure reason had been rendered superfluous by an older one already found in the works of Leibniz, Kant concluded his self-­ defense by arguing that he and not Eberhard was the true Leibnizian. Kant claimed, first, that Leibniz did not really wish “to have his principle of sufficient reason construed objectively (as a natural law),” but rather that for Leibniz, as for himself, “this principle was merely a subjective one, having reference only to a critique of reason,”1 and, second, that Leibniz, “such a great mathematician!”, did not want to compose bodies out of monads (and hence space out of simple parts),” and instead meant that not “the physical world, but rather its substrate,” is 1 “On a Discovery whereby any New Critique of Pure Reason Is to Be Made Superfluous by an Older One,” in Kant (2002, 8:247–248).

Paul Guyer, The Teleologies of Leibniz and Kant: So Close Yet So Far Apart In: Leibniz and Kant. Edited by: Brandon C. Look, Oxford University Press. © Oxford University Press 2021. DOI: 10.1093/oso/9780199606368.003.0009

234  The Teleologies of Leibniz and Kant composed out of simple monads, although that substrate is “unknowable by us” and “lies merely in the Idea of reason.”2 The first of these claims is not credible, and only the first clause—but not the second clause—of the second claim is credible. What is most interesting, however, is the third way in which Kant attempts to close the gap between Leibniz’s philosophy and his own, namely an extended argument that Leibniz’s conception of “pre-­established harmony” anticipates central points of Kant’s own philosophy. Kant first claims that, “by his pre-­established harmony between soul and body,” Leibniz could not have “understood the accord of two entities that are by nature completely independent of each other, and cannot be brought into community through any powers of their own,” but instead meant that body and soul are two species of “appearances . . . as mere forms of . . . intuition, . . . merely representations, and there the community between understanding and sensibility in the same subject can well be conceived under certain a priori laws,”3 as Kant himself had argued, for example in the fourth “Paralogism of Pure Reason” in the first edition of the Critique of Pure Reason.4 Kant continues that Leibniz had thereby anticipated his own position that “harmony between understanding and sensibility . . . makes possible cognition of universal laws of nature a priori” without which “no experience is possible.”5 In the face of Leibniz’s own claims that all truth is in principle knowable through the analysis of concepts alone and that it is only due to our cognitive limits that humans are reliant on empirical knowledge and therefore on sensory intuitions “in three quarters of what they do,”6 Kant’s claim that Leibniz’s conception of pre-­ established harmony was an anticipation of his own theory that both empirical and especially synthetic a priori cognition are possible only through the co­oper­ ation of understanding and sensibility is just as incredible as his previous claim that Leibniz intended the principle of sufficient reason only as a subjective prin­ ciple; that is, a principle of the possibility of experience. Thus far, then, Kant’s claims seem to obfuscate the fundamental differences between his philosophy and that of Leibniz. However, Kant’s argument about pre-­ established harmony now takes a more interesting turn: But we could still provide no reason why we have precisely such a mode of sens­ ibil­ity and an understanding of such a nature, that by their combination experience becomes possible; nor yet, why, as otherwise fully heterogeneous sources of cognition, they always conform so well to the possibility of empirical cognition in general, but especially (as the Critique of [the Power of] Judgment will in­tim­ ate) for the possibility of an experience of nature under its manifold particular

2  “On a Discovery,” Ak. 8:248. 3  Ibid., 8:249. 4  See esp. (from the A edition) A 370–375. 5  “On a Discovery,” Ak. 8:249. 6  PNG 15, WF 261; see also Mon 28, WF 271–272. In this chapter, I will cite only such texts by Leibniz as had been published during Kant’s lifetime and were presumably known by him.

Paul Guyer  235 and merely empirical laws, of which the understanding teaches us nothing a priori, as if nature were deliberately ordered for our own comprehension; this we could not further explain (and neither can anyone else). Leibniz termed the ground of this agreement, especially in regard to the cognition of body, and thereunder primarily our own, as the middle ground of this relation, a pre-­ established harmony, by which he had obviously not explained this agreement, nor was seeking to do so, but was merely indicating that we would have to suppose thereby a certain purposiveness in the dispositions of the supreme cause, of ourselves as well as of all things outside us; and this indeed as something already lodged in creation (predetermined), albeit a predetermination, not of things existing in separation, but only of the mental powers in us, sensibility and understanding, each in its own way for the other, just as the Critique teaches that for the a priori cognition of things they must stand in a reciprocal relationship to one another in the mind. That this was his true, though not clearly developed opinion, may be surmised by the fact that he extends the pre-­established harmony much further than to the agreement of the soul and body, namely to that between the Kingdoms of Nature and of Grace (the Kingdom of Ends in relation to the final end, i.e., mankind under moral laws), where a harmony has to be thought between the consequences of our concepts of nature and those of our concept of freedom, a union, therefore, of two totally different faculties, under wholly dissimilar principles in us, and not that of a pair of different things, existing in harmony outside each other (as morality actually requires); though, as the Critique teaches, it can by no means be conceived from the constitution of the world, but rather as an agreement that for us at least is contingent, and comprehensible only through an intelligent world-­cause. In this way, then, the Critique of Pure Reason might well be the true apology for Leibniz . . . 7

In this passage, after repeating the thoroughly implausible claim that Leibniz’s conception of the pre-­established harmony anticipates his own doctrine of the necessity of both sensibility and understanding for any knowledge, which he had always presented as an anti-­Leibnizian claim, Kant adds that Leibniz’s conception of the pre-­established harmony between the kingdoms of nature and grace anticipates his own conception of the union between our concepts of nature and our concept of freedom, the central thought of the third Critique, and of the necessary realizability of the kingdom of ends in nature in the form of the highest good because of the common authorship of the laws of both nature and freedom, the culminating thought of the second Critique, although Kant suggests that both of these thoughts were already contained—or at least incipient—in the first Critique.

7  “On a Discovery,” Ak. 8:249–250.

236  The Teleologies of Leibniz and Kant Kant’s suggestion that the central thought of the two later critiques was implicit in the first is not false, even though when he wrote the first Critique he had no intention of writing two more, for his cherished idea that after the critique of all theoretical grounds for metaphysics it can be reconstructed on practical grounds, through the postulates of practical reason necessary for us to conceive of the realizability of the highest good commanded by morality, was first expounded in the “Canon of Pure Reason” of that work’s “Doctrine of Method.” And there too Kant had already connected his own idea of the highest good with Leibniz’s conception of the realm of grace, distinct from the realm of nature: Leibniz called the world, insofar as in it one attends only to rational beings and their interconnection in accordance with moral laws under the rule of the highest good, the realm of grace, and distinguished it from the realm of nature, where, to be sure, rational beings stand under moral laws but cannot expect any successes for their conduct except in accordance with the course of nature in our sensible world. Thus to regard ourselves as in the realm of grace, where every happiness awaits us as long as we do not ourselves limit our share of it through the unworthiness to be happy, is a practically necessary idea of reason.8

So it seems that Kant had thought of his own conception of the highest good in connection with Leibniz’s conception of the principles of nature and grace, expounded in his essay with that title,9 long before his polemic with Eberhard. But as this quotation from the first Critique indicates, the relation between the teleological visions of the two philosophers must be complex, at least in part because Kant’s own conception of the highest good underwent a long evolution that was only beginning in 1781. Kant’s view in the first Critique was that the happiness that must be expected to follow from the worthiness to be happy constituted by being motivated by the moral law10 can be expected to follow it only in “a world that is future for us,”11 and he identified that world with Leibniz’s realm of grace as distinguished from the realm of nature. But, of course, Leibniz had also argued that these two realms are in harmony, and while suggesting that the virtuous would indeed enjoy their happiness fully only in the future he also held that they would enjoy at least a “foretaste” of it in the present world; that is, the realm of nature. Thus Kant’s conception of worthiness and happiness in 1781 seems to separate them more rigidly than Leibniz had separated the realms of nature and grace. But then again, by the time Kant made his remark about Leibniz in “On a Discovery,” he seems to have been well on his way to holding that for morality to be possible we must be able to believe that happiness will accompany worthiness in nature, thus to tying the realms of nature and grace more tightly together than 8  A 812/B 840. 11  A 811/B 839.

9  PNG (1714); WF 258–266.

10  See A 806/B 834.

Paul Guyer  237 Leibniz ever had. Does he thus swing from one misappropriation of Leibniz to another? I will argue that although Kant’s claim that on the issue of the relation between nature and freedom his own philosophy is the true apology for Leibniz is by no means as incredible as his other claims that he, rather than Eberhard, is Leibniz’s true heir, there are nevertheless fundamental differences as well as similarities between Kant’s conception of the ultimate significance of the pre-­established harmony and Leibniz’s. One apparently major difference is that part of Leibniz’s argument for the harmony between the realms of nature and grace is an assertion of the compatibility between the completeness of mechanical explanation “at the level of physical inquiry” with the possibility and necessity of final causes or “the great principle . . . that nothing comes about without a sufficient reason” at the level of “metaphysical” inquiry,12 while Kant’s argument for the necessity of adding a level of teleological explanation to the mechanical explanation of nature turns on the limits to mechanical explanation entailed by our experience of organisms as “natural ends.” A second apparently major difference between the two views is that while Leibniz has no hesitation about wholeheartedly asserting the parallelism of mechanical and teleological explanation as part of what Kant would call “constitutive” and “dogmatic” metaphysics, for Kant the teleological explanation of nature can only be considered a “regulative” principle for the further development of mechanical explanation of natural phenomena. I shall argue that on these two points there is less difference between the two philosophies than initially meets the eye. First, Leibniz can assert the completeness of mechanical ex­plan­ ation and thus that teleological explanation parallels rather than supplementing mechanical explanation at the level of physical inquiry only because he includes as part of mechanical explanation the idea of organisms as self-­moving entelechies, which is the very notion that Kant regards as the limitation of mechanical explanation, encountered in our experience of organisms, that starts us on the path toward teleological explanation; the real difference between Leibniz and Kant on this point is rather only that Leibniz regards everything in nature as ul­tim­ate­ly an organism while Kant recognizes only some things in nature as organisms. Second, Leibniz as well as Kant insists that the teleological conception of nature should be used as a heuristic for the further development of physical explanations, so here too there seems to be less substantive difference between the teleologies of Leibniz and Kant than might be suggested by the contrast between “constitutive” and “regulative” principles. However, any similarities on these two points only mask a profound difference between Leibniz and Kant; namely, that while for Leibniz the “final end” or telos of the teleological explanation of nature is God’s free volition of the best of all possible worlds, for Kant the only candidate for the final end of nature is nothing other than the realization of human freedom 12  PNG 7; WF 262.

238  The Teleologies of Leibniz and Kant and through that human happiness. In the end, Kant supposes that we must postulate the existence of God to explain the possibility of realizing our moral goal in nature, while for Leibniz God’s point in the creation of nature is the satisfaction of his own reason, which we can perhaps to some degree imitate but mostly enjoy only as it were vicariously. In the end perhaps Kant does tie the realms of nature and grace together even more tightly than Leibniz does: for Leibniz the realms of nature and grace must be harmonious, but for Kant, at least by the time he has fully developed his conception of the highest good, they must be identical. In what follows, I will first look at the differences and similarities in the arguments by which Leibniz and Kant arrive at their teleologies, and then look at the similarities and differences in the ultimate contents of those teleologies themselves.

2  Leibniz and Kant on Mechanical and Teleological Explanation In this section I will examine the arguments by means of which Leibniz and Kant motivate the adoption of a teleological conception of nature, in the case of Leibniz as a matter of straightforward metaphysics but in the case of Kant as a matter of “reflective” and “regulative” rather than “determinant” and “constitutive” judgment. A central thesis for Leibniz, asserted in different publications over a period of decades, was that while at one level the behavior of all bodies in nature can be explained in mechanical laws, there must be an underlying explanation of the truth of these laws themselves. In “On Nature Itself; Or, The Inherent Force and Activity of Created Things,” published in the Acta Eruditorum in 1698, he asserts “that mechanism itself has its origin not merely in a material principle or in mathematical reasons, but in some higher and, so to speak, metaphysical source.”13 In the “Principles of Nature and Grace,” sixteen years later, he expanded on this, arguing that both the injection of motion into a system of material bodies and the laws that govern such motion need an explanation that can only come from the existence of a self-­causing, necessary, and wise being outside of that system. First he makes what seems to be the general claim that “the sufficient reason for the existence of the universe can never be found in the series of contingent things, in bodies and their representations in souls.” But then he argues that this is true specifically “because matter itself is indifferent to motion or rest,” or as we would say subject to the law of inertia: Therefore we could never find in matter a reason for motion . . . And since any motion which is in matter at present comes from a previous motion, and that 13  “On Nature Itself,” §3, WF 211.

Paul Guyer  239 too from a previous one, we are no further forward if we go on and on as far as we like . . . Therefore the sufficient reason, which has no need of any further reason, must lie outside that series of contingent things, and must be found in a substance which is the cause of the series; it must be a necessary being, which carries the reason for its existence within itself, otherwise we still would not have a sufficient reason at which we can stop. And that final reason for things is what we call God.14

But though this ultimate cause of motion must itself be self-­caused and necessary, the laws of motion are not necessitated by the very nature of this being, but are rather an expression of its wisdom and its goodness: The supreme goodness of God made him choose in particular the laws of motion which were the most appropriate, and which fitted best with abstract or metaphysical reasoning . . . What is striking is that 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, as logical, arithmetical, and geometrical truths do, but on the principle of compatibility, the choices of wisdom, that is.15

The principles of motion that Leibniz has in mind are summed up in his thesis, held already for many years, that the behavior of a body is governed by the conservation of force rather than the conservation of motion; and his view is that, while the principles of logic, arithmetic, and geometry are necessary truths grounded in the intellect of God, the principle of the conservation of force is not intrinsically necessary but is rather chosen by the wise and benevolent will of God as most consistent with the principle of sufficient reason understood to define the best of all possible worlds by the achievement of maximal variety with minimal means. Thus we go from an explanation of the very possibility of motion in a physical world to the teleological conclusion that a world with motion must be the best of all possible worlds. From an early stage Kant too was moved by the idea that the behavior of created things in nature, not just as the level of physical motions but even at the level of human choices such as when to marry and procreate, is thoroughly governed by laws, but that these laws themselves, expressed in the very possibilities or essences of created things, are chosen by God in accordance with the “rule of the best.” In his first book-­length work of philosophy, the Only Possible Argument in support of a Demonstration of the Existence of God of 1763, Kant argued that God must be posited as the ground of the very possibility of created things, and then

14  PNG 8; WF 262.

15  PNG 11; WF 263.

240  The Teleologies of Leibniz and Kant that God’s choice of what sorts of things even to make possible, which is the same as his choice of what laws should govern the behavior of things, reflect his preference for a unified and harmonious world: “If all this were supposed, then there would obviously inhere in the very essence of things themselves universal relations to unity and cohesiveness, and a universal harmony would extend throughout the realm of possibility itself. Such a state of affairs would fill us with admiration for such extensive adaptedness and natural harmony.”16 Kant’s pos­ ition, clearly inspired by Leibniz, is that God does not achieve his beneficent aims by occasional interventions in nature, which would appear miraculous, but by his original wise choice of a harmonious set of possibilities or laws for the behavior of natural things. Having discussed the laws that govern weather, the movements of liquids, the behavior of fibers in textiles, and so on, he continues that: The reason why I have discussed such humble effects which are so little esteemed and which take their rise from the simplest and most general laws of nature, is this: . . . to show how, from these lowly effects, one must infer the great and in­fin­ ite­ly extended harmony of the essences of things and the important things attributable to that harmony . . . I have also been concerned to show the absurdity of attributing these same harmonies to the wisdom of God as their special ground. The fact that things, which are so beautifully related to each other, should exist at all, is to be attributed to the wise choice of Him who created them on account of that harmony. But that each of these things should, in virtue of simple grounds, contain such an extensive adaptedness to harmony of many different kinds, and that a wonderful unity in the whole should, as a result, be able to be maintained—that is inherent in the very possibility of the things in question.17

Nature can produce harmony without miraculous intervention because the possibility of such harmony is built into the very laws of nature itself, which were chosen by God in his original choice of what sorts of things are even possible. This argument departs from Leibniz in seeing God’s wise and benevolent choice as the ground of all possibilities rather than seeing God as choosing wisely and benevolently from among possibilities that are somehow already present to his intellect, but is otherwise in the spirit of Leibniz’s argument that the laws of nature must themselves be explained by reference to God. Of course, once Kant developed his critique of traditional metaphysics in the Critique of Pure Reason, such an unqualified assertion about the divine ground of the laws of nature was no longer open to him. But rather than simply abandon his 16  The Only Possible Argument in support of a Demonstration of the Existence of God, §2, First Reflection, Ak. 2:296; trans. from Kant (1992b, 140). 17  Only Possible Argument, §2, Second Reflection, Ak. 2:203; Kant (1992b, 146).

Paul Guyer  241 long-­ standing commitment to a teleological perspective upon nature, Kant sought for a way to incorporate it into the Critical philosophy. In the Appendix to his critique of rational theology in the Transcendental Dialectic of the first Critique, Kant proposed that the idea of God’s choice of the laws of nature be regarded and employed simply as a “regulative principle” for the scientific investigation of nature: [T]he idea of the highest wisdom is a regulative one in the investigation of nature and a principle of the systematic and purposive unity thereof in accordance with universal laws . . . For the greatest systematic and purposive unity, which your reason demands as a regulative principle to ground all investigation of nature, was precisely what justified you in making the idea of a highest intelligence the ground as a schema of this regulative principle . . . 18

But this simple solution did not seem to content him, and he returned to the subject of teleology in a third Critique; indeed, in a well-­known letter to Karl Leonhard Reinhold, just as he was finishing the second Critique and contemplating a third, Kant went so far as to make it seem that solving the problem of how to incorporate teleology into a Critical philosophy would be the primary issue for this new work. Kant starts off by saying that he is at work on a “critique of taste,” but then makes it look as if teleology will be its sole subject: My inner conviction grows, as I discover on working on different topics that not only does my system remain self-­consistent but I find also, when sometimes I cannot see the right way to investigate a certain subject, that I need only look back at the general picture of the elements of knowledge, and of the mental ­powers pertaining to them, in order to discover elucidations that I had not expected. I am now at work on the critique of taste, and I have discovered a new sort of a priori principles, different from those heretofore observed. For there are three faculties of mind: the faculty of cognition, the faculty of feeling pleasure and displeasure, and the faculty of desire. In the Critique of Pure (the­or­et­ ic­al) Reason, I found a priori principles for the first of these, and in the Critique of Practical Reason, a priori principles for the third. I tried to find them for the second as well . . . This . . . put me on the path to recognizing the three parts of philosophy, each of which has its a priori principles, which can be enumerated and for which one can delimit precisely the knowledge that may be based on them: theoretical philosophy, teleology, and practical philosophy, of which the second is, to be sure, the least rich in a priori grounds of determination—19

18  A 699/B 727. 19  Letter to Reinhold, December 25 and 31, 1787, Ak. 10:514; in Kant (1999, 272).

242  The Teleologies of Leibniz and Kant but by no means entirely devoid of them. As things turned out, Kant came to recognize that taste and teleology are two different subjects, and the eventual Critique of the Power of Judgment, which was published, not as Kant predicted in his letter to Reinhold at Easter of 1788, but only two years later, was divided into two parts, a “Critique of the Aesthetic Power of Judgment” and a “Critique of the Teleological Power of Judgment,” the relation between which is complicated and controversial. But that issue is not our problem here.20 Our question here is, rather, in what ways does Kant’s critical revision of teleology in the third Critique continue to resemble Leibniz’s teleology and in what ways does it differ from it? What I will now propose is that in the third Critique Kant deploys two different arguments for the necessity of our adoption of a critical but still teleological viewpoint on nature, the first of which differs from Leibniz’s approach on one perhaps small point, the second of which seems radically opposed to Leibniz’s approach but turns out to be less drastically opposed to Leibniz’s position once we fully understand the latter. I refer first to Kant’s argument in the Introduction to the Critique of the Power of Judgment that laws of nature that are more particular than the “most universal” laws of nature such as that “All alteration has its cause” must be able to be regarded by us as necessary (necessarily true) even though they do not follow from the most universal laws,21 and that the only way for us to think of such laws as being necessary is to think of them as part of a system of laws thought by an intelligence like but greater than our own—like our own, because all lawfulness must be a product of understanding according to Kant’s Copernican revolution, but greater than our own, precisely because this intelligence must be the source of the necessity of laws that are left contingent by the most general laws of our intellect.22 Kant states that the principle according to which the particular laws of nature can appear to us to be necessary is this: that since universal laws of nature have their ground in our understanding, which prescribes them to nature (although only in accordance with the universal concept of it as nature), the particular empirical laws, in regard to that which

20  For my original position on this issue, that Kant’s aesthetics and his teleology are more opposed than related to one another, because aesthetic judgment depends upon the appearance of contingency in the beauty of its objects while teleology seeks to abolish the appearance of contingency from the laws of nature, see Guyer (1979, ch. 2); for a more refined account which recognizes some similarities among the two kinds of judgment in spite of this difference, see Guyer (2003a, 1–61); for an approach that argues for a much more intimate connection between aesthetic and teleological judgment than I have been willing to allow, see Zuckert (2007, esp. ch. 2). Zuckert argues that both forms of judgment are informatively unified by Kant’s notion of “purposiveness without a purpose,” while my view has been that there are only superficial similarities between Kant’s uses of this notion in the two parts of the third Critique. 21  See A 127/B 165. 22  I have discussed this argument in more detail in Guyer (1990; cf. 2005, esp. 44–55) and Guyer (2003b; cf. 2005, esp. 60–68).

Paul Guyer  243 is left undetermined in them by the former, must be considered in terms of the sort of unity they would have if an understanding (even if not ours) had likewise given them for the sake of our faculty of cognition, in order to make possible a system of experience in accordance with particular laws of nature.23

To be sure, the now Critical Kant adds that it is “Not as if in this way such an understanding must really be assumed (for it is only the reflecting power of judgment for which this idea serves as a principle . . . ); rather this faculty . . . gives a law only to itself, and not to nature,” and stresses this point again when he argues in the next section that while “the power of judgment must thus assume it as an a priori principle for its own use that what is contingent for human insight in the particular (empirical) laws of nature nevertheless contains a lawful unity, not fathomable by us but still thinkable, in the combination of its manifold into one experience possible in itself ”: this transcendental concept of a purposiveness of nature is neither a concept of nature nor a concept of freedom, since it attributes nothing at all to the object (of nature), but rather only represents the unique way in which we must proceed in reflection on the objects of nature with the aim of a thoroughly interconnected experience, consequently it is a subjective principle (maxim) of the power of judgment . . . 24

Kant’s use of the term “purposiveness” in this last passage indicates that he conceives of the systematicity of the particular laws of nature that we posit to provide them with necessity and explain as due to an intelligence greater than our own as facilitating our own purely cognitive purposes; only in a subsequent argument will he add that we must also conceive of the intelligent designer of nature as having any purpose for his design beyond the satisfaction of our own purely cognitive needs. We will come back to that point, ultimately Kant’s most profound difference with Leibniz’s teleology, in §3 of this chapter. In the meantime, one difference from Leibniz’s version of teleology—namely, that for Kant the postulation of an intelligent author of the laws of nature is only a “subjective principle” or “maxim” for use in our own reflection on nature but not a constitutive principle of theoretical metaphysics--that is, not something that we can simply assert to be true about reality beyond our way of conceiving it—is obvious in this passage. However, Kant’s difference from Leibniz is not that the latter does not recognize the heuristic value of the thought of God’s design of nature in our investigation of it, for he certainly does, but rather that Leibniz does not treat this thought as merely heuristic, instead conceiving of it as a piece of genuine metaphysics: as we 23  Critique of the Power of Judgment, Introduction, Section IV, Ak. 5:180. 24  Ibid., Section V, Ak. 5:183–184.

244  The Teleologies of Leibniz and Kant saw in an earlier quotation from the “Principles of Nature and Grace,” Leibniz introduces the principle of sufficient reason and the creator of nature who honors it in moving from the level of physical inquiry to the level of metaphysical inquiry, and places no qualification on the objective significance of what we discover at the latter.25 Further, a less obvious difference between the two approaches is that, while Leibniz argues that we must see God as selecting one general law for nature (i.e. the law of the conservation of force), he does not stress that we must conceive of God as responsible for a system of particular laws of nature, as Kant does, for he does not worry about the problem of how we could see such laws as necessary when they appear contingent to our intellect, as Kant does. To be sure, Leibniz recognizes, indeed constantly emphasizes, that the human intellect is limited in comparison to the divine intellect, so he could have noticed the necessity of particular laws of nature as one point where the limits of our intellect manifest themselves. But he did not. Nor does he posit the systematicity of laws of nature as a goal of physical inquiry for any other reason. Thus, one difference between Kant and Leibniz is that Kant introduces the idea of an intelligent ground of nature as a merely subjective explanation of the systematicity of its particular laws, while Leibniz frequently emphasizes God’s role as the source of the particular law of the conservation of force, but does not emphasize the systematicity of particular laws of nature, a fortiori does not explicitly assign God a role, whether constitutively or regulatively, as the source of such systematicity. The apparently greatest difference between their arguments for conceiving of God as the source of nature, however, is that while Leibniz appears to argue for the necessity of God as the ground of a nature that is mechanical throughout, in his second argument for the necessity of our adopting a teleo­ logic­al point of view on nature, the argument of the “Critique of the Teleological Power of Judgment” rather than the argument about necessity from the Introduction, Kant argues that we are inevitably led to the thought of a designer of nature by our experience of objects within nature that seem to constitute an exception to the completeness of mechanical explanation, namely organisms (or as Kant calls them, “organized beings”). Here, however, I will argue that the difference between the two philosophers is less than initially meets the eye, because of Leibniz’s characteristic view that even at the level of nature we must suppose that bodies are really composed of organisms. We will see that one of his chief reasons for this doctrine is the very same reason that compels Kant to treat organisms as an exception to mechanical explanation that require teleological explanation, at least from the human standpoint. This time, let us look at Kant’s argument about organisms, and then return to consider whether it is really very different from Leibniz’s position.26 Kant begins 25  See PNG 7; WF 262. 26  I have discussed Kant’s argument elsewhere, so will present it as briefly as possible here. See esp. Guyer (2001b; cf. 2005); see also Goy (2008).

Paul Guyer  245 the “Critique of the Teleological Power of Judgment” by observing that judgments of “external” or “relative purposiveness of nature” would be, on their own, entirely arbitrary: we might be tempted, for example, to judge that the icy seas contain “great sea animals filled with oil” and carry driftwood for the benefit of those who live north of the Arctic Circle, the Greenlanders and Lapps and so on—“But one does not see why human beings have to live there at all,” and unless we can answer that question our thought that “all these natural products” exist for the sake “of an advantage for certain miserable creatures would be a very bold and arbitrary judgment.”27 In the end, Kant’s argument will be that we can and indeed must adopt a teleological view of nature precisely because there is unconditional value in human existence, for which the rest of nature is the condition, but he does not make that argument right away. Instead, he makes a detour, arguing that due to the limits of human cognition, we can only comprehend certain things within nature—namely, organized beings—as “internally purposive” “natural ends,” and then that our conception of organisms as natural ends inevitably leads us to conceive of nature, as a whole, as a system that must have had a designer and a point for its design and creation. This argument differs from the argument of the Introduction that we have already discussed because it concerns nature as a system of objects rather than our understanding of nature through a system of laws, and in this regard it is closer to Leibniz’s conception of the universe as a system of harmonious substances than was the Introduction’s image of a harmonious system of laws of nature. The argument is divided into two main parts: first, the argument that organisms can only be understood as natural ends with internal purposiveness; second, the argument that if organisms must be understood as purposive systems, nature as a whole must be understood as a single purposive system. Of course one might jump off the train after the first stage of this overall argument, or alternatively find a way of arguing for the second stage in­de­pend­ ent­ly from the first. In fact, Kant has already argued for the second stage alone on purely moral grounds, as a postulate of pure practical reason, and so, as we will see when we turn to the second stage of Kant’s argument in §3 of this chapter, the whole argument can only be understood as complementing Kant’s moral phil­oso­ phy rather than as introducing an entirely new idea into his thought. In the first stage of the argument that we are now considering, the limitation of human cognition that requires us to adopt a new perspective in order to comprehend organisms is our restriction to mechanical explanation in natural science. By mechanical causation Kant means “The causal nexus, insofar as it is conceived merely by the understanding, [as] a connection that constitutes a series (of causes and effects) that is always descending,” so that “things . . . which as effects presuppose others as their causes, cannot conversely be the causes of these at the same

27  Critique of the Power of Judgment, §63, Ak. 5:369.

246  The Teleologies of Leibniz and Kant time.”28 As we will see from the subsequent course of the argument, this must mean two different things: first, that the properties of a whole are determined by the properties of its parts, but not vice versa;29 second, that the motion of any whole or system can only be explained by antecedent motion, either of its own parts or of something external to it. In the first case, the second claim may be regarded as a specific version of the first, but in the second case, it is independent of the first. The reason these two aspects of mechanism have to be distinguished is that Kant’s argument that organisms are natural ends or internally purposive systems, the first stage of Kant’s larger argument, itself has two parts. The first part, which employs ideas that come from Buffon but ultimately go back to Aristotle,30 is that there are certain characteristic properties or abilities of organisms, such as their capacity for reproduction, growth, and self-­preservation, that we can understand only as effects of the whole on its parts and not merely as effects of parts on the whole. Thus we can only understand the organism as both cause and effect of itself, the whole as the cause of its parts as well as the whole as the effect of its parts. In reproduction, we conceive of the organism as both cause and effect of itself (though only at the level of the species, not the individual, which makes this argument particularly dubious); in growth, we must understand the organism as both dependent on its nutrients as parts but also as itself as a whole transforming those nutrients into parts; and in self-­preservation, we must understand the health of the whole as dependent upon its parts but also the health of the parts as dependent upon the proper functioning of the whole—the tree cannot live without its roots and leaves properly functioning, but the roots and leaves cannot live without the whole tree properly functioning.31 In all of this, it is our restriction to mechanical explanation as the dependence of parts on wholes or the descent from the former to the latter that is being strained. Kant then argues that the only way we can make sense of the idea of the whole as the cause of its parts, which has been forced upon us by the characteristic properties of organisms, is by analogy with our own characteristically intentional production of artifacts, in which an antecedent representation of the whole is the cause of the parts that are then in turn the cause of the whole, thereby preserving the unidirectionality of mech­an­ ic­al causation. And this requires us to conceive of organisms as the product of design. But Kant also stresses that even this analogy is imperfect, because the artifacts that we produce, such as watches, are not capable of repairing themselves or producing more watches: they are “organized” but not “self-­organizing beings.”32 Thus although we will be led to the idea of a design and designer by our experience of organisms, we will also be led to the idea of a designer who must be greater 28  Ibid., §65, Ak. 5:372. 29  This aspect of what Kant means by mechanical explanation has been stressed by McLaughlin (1990, 152–158) and Allison (1991, 27). 30  See Löw (1980, ch. 1). 31  Critique of the Power of Judgment, §64, Ak. 5:371–372. 32  Ibid., §65, Ak. 5:374.

Paul Guyer  247 than ourselves, as we were earlier led to the same idea by the idea of a system of necessary but particular laws of nature greater than what we ourselves can prod­ uce in the form of our limited set of necessary but completely general laws of nature. But Kant does not take this step to the thought of such a designer immediately. Instead, he next says that the concept of matter “as organized that necessarily carries with it the concept of itself as a natural end . . . necessarily leads to the idea of the whole of nature as a system in accordance with the rule of ends, to which idea all of the mechanism of nature in accordance with principles of reason must now be subordinated.”33 Kant does not explain why we must make this move from some things in nature to all of nature, so we can only conjecture that he is supposing that our reason is always unitary (i.e. seeks an single ultimate explanatory principle34) and that therefore we can only conceive the even greater intellect that we have to postulate as the ground for some things in nature—namely, organisms— as similarly unitary and therefore as creating a nature that is as a whole a single system of ends, not merely a nature that contains some systems within it. But even if that is so, Kant also stresses in the present passage that this idea of nature as a whole as a single system is only an idea by which “to test natural appearances,” “not a principle for the determining but only for the reflecting power of judgment, . . . regulative and not constitutive.”35 This, of course, is an obvious difference between Kant the Critical teleologist and Leibniz the metaphysical teleologist, who recognizes the heuristic value of the idea of the systematicity of nature but does not confine the value of that idea to its heuristic value. But another difference between the two philosophers seems to be that Kant has reached this point in his argument starting from the assumption that there is an essential difference between two kinds of objects in nature, non-­organisms and organisms, and even while inferring from the latter that nature as a whole must be in some sense a single system has not then undermined his original distinction by inferring that every thing within nature is also a system of the kind that individual organisms or even species of them are. Leibniz, however, as we have already seen, seems to insist that everything in nature can be explained mech­an­ic­al­ly and yet that the whole mechanical system of nature must also admit of a teleo­logic­al explanation. Before we conclude that the two philosophers are irretrievably divided on this point, however, we should notice that (still at the first stage of his overall argument) Kant offers a second reason for his claim that organisms require us to adopt

33  Ibid., §67, Ak. 5:378–379. 34  This fact about reason would be expressed by its “maxim” of “homogeneity”; see A 644–645/B 672–673 and A 657–661/B 685–689. Kant’s assumption that reason seeks a single principle in its ex­plan­ation of objects is to be distinguished from the assumption of post-­Kantian philosophers that philosophy seeks a single principle for the explanation of our faculties themselves; on this, see Franks (2005). 35  Critique of the Power of Judgment, §67, Ak. 5:379.

248  The Teleologies of Leibniz and Kant an alternative to mechanical explanations, and this brings out a deep similarity between his argument for teleology and Leibniz’s. This is his claim, made in the course of his review of competing “systems concerning the purposiveness of nature,”36 that organisms cannot be understood mechanically because “lifelessness, inertia constitutes [the] essential characteristic” of matter,37 but that being self-­moving (i.e. not subject to inertia) is the essence of organisms. The systems of the purposiveness of nature are two forms of “idealism”—that is, theories that nature merely appears purposive but is not-- the “accidentality” of Epicurus, which says that everything in nature is a matter of blind chance with no room for purpose and the “fatality” of Spinozism, which in turn says that everything is a matter of blind necessity with no room for purpose—and then two forms of “realism”—namely “hylozoism,” which derives purposiveness in nature either from multiple living matter or a single “animating inner principle, a world-­soul,” within nature, and “theism,” which derives purposiveness “from the original ground of the world-­whole, as an intentionally productive (originally living) intelligent being.”38 Having dismissed the two forms of idealism or denial of purposiveness in nature altogether as incompatible with our comprehension of organisms, Kant also dismisses hylozoism because it requires the supposition of non-­inertial behavior within material nature, which is also inconceivable to us. So we are left with theism as “incapable of dogmatically establishing the possibility of natural ends as a key to teleology” but nevertheless, as “a ground for the reflecting, not for the determining power of judgment,” our only “way of judging the generation of [nature’s] products as natural ends . . . through a supreme understanding as the cause of the world.”39 Kant’s complicated position is thus that we have to think of organisms because we find a power of self-­motion in material nature, but at the same time we cannot think of self-­motion as a real property of anything in nature, because of material nature’s thoroughgoing subjection to the law of inertia, thus we have to think of the motion of organisms as due to an external, intelligent ground of nature, not just because as self-­organizing beings they are too complex for our capacity for mechanical explanation, but also because, as self-­organizing beings, they violate the law of inertia and need a non-­natural source of motion. (Of course, one might argue here that they are no longer really self-­organizing and self-­moving after all.) This argument is why I said earlier that we have to think of Kant’s conception of mechanism as involving not merely the idea that only parts determine the whole, and not vice versa, but also the idea that all motions of bodies are due to previous motions of bodies: the argument now is that we have to infer an external ground of organisms because we cannot explain

36  Ibid., §72, Ak. 5:389. On the competing systems of teleology, see Watkins (2008). 37  Critique of the Power of Judgment, §73, Ak. 5:394. 38  Ibid., §72, Ak. 5:391–392. 39  Ibid., §73, Ak. 5:395.

Paul Guyer  249 their motions as due to the prior motions of bodies, their own or other bodies within nature—even though this new explanation so to speak mimics inertial explanation by supposing in the end that organisms are not really entirely self-­ moving but moved by a ground that is external to nature altogether. The paradoxes inherent in this argument would preoccupy Kant until his dying day, for in the manuscripts known as the Opus postumum he began to argue that everything in nature presupposes a self-­moving ether, which would then undermine the most fundamental distinction between non-­organic and organic matter, though in this uncompleted work Kant never quite reached that conclusion.40 But the point that I want to stress here is that in this second part of the first stage of his overall argument in the “Critique of the Teleological Power of Judgment,” Kant has now backed into a certain rapprochement with Leibniz, for one of the central motivations for Leibniz’s argument that mechanical explanation must be undergirded by teleological explanation is precisely that the law of inertia, although characteristic of mechanism, stands in the way of a complete ex­plan­ ation of nature. Leibniz of course has a number of different arguments that lead him first to accompany the realm of bodies with another level of reality and eventually to reduce the realm of bodies altogether to a realm of phenomena bene fundata of these other realities—namely, monads.41 The argument that there must be simple substances but that nothing presented in space and time is simple, so what is given in space and time must at least be accompanied with if they are not mere appearances of non-­spatio-­temporal simple substances is one;42 and the argument that substances must bear traces of everything that ever has and ever will be true of them but that extended matter cannot do so, thus extended matter must be accompanied if they are not mere appearances of more mind-­like substances, which can bear such traces in the forms of clear or obscure memories or foresight, is another.43 But one of Leibniz’s favorite arguments is that there must be an energy or force within or attached to material substances that originates their motion, thus violating the law of inertia, although that must in turn be due to the original action of God, which, as it were, restores the law of inertia—Leibniz’s

40  For some citations and discussion of this, see Guyer 1991; Friedman (1992, ch. 5); and Förster (1993 and 2000, ch. 4). 41  Naturally there are different chronologies of Leibniz’s transition from dualism to the full-­blown monadology of his last works, ranging from Christia Mercer’s view that Leibniz had arrived at the essentials of his mature philosophy by 1679 to Garber’s view that Leibniz did not settle on the monism of the monadology until the last decade of his work; see Mercer (2001, ch. 10) and Garber (2009, chs. 7–8). Kant clearly read the few earlier works of Leibniz that could have been known to him through the lens of Leibniz’s final monadological works, even if, as seems likely, Garber’s view that Leibniz only settled on the monadology rather than dualism late in his career is correct. 42  See e.g. “New System of the Nature of Substances, and Their Communication,” §3, WF 145, PNG 2; WF 259, and Mon §§1–3; WF 268. 43  This argument appears clearly in Leibniz’s letter to Arnauld of November 28/December 8, 1686, WF 119, this being a text unknown to Kant.

250  The Teleologies of Leibniz and Kant account thereby having the same complexity as Kant’s. For Leibniz, all matter must thereby either be linked to or really consist in organisms, precisely as Kant understands organisms—as self-­moving although created beings—although once again in Kant’s view this is simply how we must conceive of organisms while for Leibniz it is straightforward metaphysical fact. The following passage from “On Nature Itself ” is worth quoting at length because it rejects the idea that motion is due to an internal “world-­soul” on its way to the conclusion that it must be due instead to organisms that have an external ground, just as Kant would argue, and may have been a model for that stage of Kant’s argument, as well as for Kant’s earlier argument that there must be a distinction as well as an analogy between works of nature and works of art: I want to ask two questions. Firstly, what makes up the nature which we normally attribute to things . . . ? Secondly, whether there is any energeia in created things . . . As to the first question, concerning nature itself: . . . I certainly agree that there is no world soul...Rather the whole of nature is, so to speak, the in­geni­ ous handiwork of God, so much so that every natural machine (and this is the true but rarely recognized distinction between nature and art) is made up of an finite number of other organisms [organis], and therefore requires infinite wisdom and power on the part of its creator and ruler . . . I say that it is enough for the machine of things to have been constructed with such wisdom that these wonders come about through its own workings, and in particular, I believe, through organic beings which unfold themselves in accordance with some kind of pre-­arranged plan.44

Leibniz denies that the whole world is a self-­moving organism with energy that needs no further explanation, and instead maintains that although we might refer to it as a single machine it is actually composed of numerous natural machines that are in turn comprised of organisms, which are both self-­moving and yet created. He turns this picture in the next paragraph against the corpuscularian the­ or­ist Robert Boyle, according to whom “we must take nature as being just the mechanism of bodies.” Leibniz responds: In broad terms we can agree with this; but on a closer look we must distinguish the principles of this mechanism and what is derived from them . . . I have already more than once expressed the view (which I think should be useful in preventing mechanical explanations of material things from being carried to far, and to the detriment of piety—as if matter could stand by itself and mechanism needed no intelligence or spiritual substance) that mechanism itself has its

44  “Nature Itself,” §2, WF 210.

Paul Guyer  251 origin not merely in a material principle or in mathematical reasons, but in some higher, and so to speak, metaphysical source.45

Now, any almost mechanist of the seventeenth century, and no doubt the pious Boyle, would have admitted that as either an assemblage of smaller machines or one grand machine, nature needs to be created and set into motion by God; but Leibniz’s position is that nature needs to consist of self-­moving organisms, although these too need to be created by God. He stresses this in the central passage of “On Nature Itself,” in which he indeed introduces the term “monad,” although his position in this paper of 1698 may still be that matter needs to be accompanied by monads rather than being merely a phenomena bene fundata of monads: However, it is indeed true that bodies in themselves are inert, provided that this is correctly understood, as meaning that what is in some respect at rest cannot set itself in motion in that respect . . . And so it must be admitted that extension, or what is geometrical in bodies, if taken by itself contains nothing which can give rise to action and motion. Indeed on the contrary, we must admit that matter resists being moved by a certain natural inertia . . . Now, since [its] ac­tiv­ ities . . . certainly cannot be modifications of primary matter or mass, which is something essentially passive, we can conclude . . . that a first entelechy or first subject of activity must recognized in corporeal substance; that is, a primitive motive force, additional to extension . . . and mass . . . And it is thus substantial principle which is called the soul in living things, and a substantial form in ­others, and insofar as together with matter it makes up a substance which is truly one, or one per se, it forms what I call a monad.46

Leibniz goes on from here to make his characteristic argument that connection to a soul or substantial form is necessary to make any matter into a true unity rather than a mere heap or aggregate. But the point that I want to stress is that before he takes this next step he has argued precisely that because matter as such is subject to the law of inertia, and is thus capable of transmitting but not originating motion, the presence of motion in the created world can only be explained by the existence of self-­moving organisms in it, though those in turn must also be created by God. Leibniz actually omits this argument from “The Principles of Nature and Grace,” there arguing only that because “matter in itself is indifferent to motion or rest, and to this motion or that,” the source of motion “must lie outside that series of contingent things, and must be found in the substance which is the cause of the

45  Ibid., §3, WF 211.

46  Ibid., §11, WF 217.

252  The Teleologies of Leibniz and Kant series” and “must be a necessary being,”47 the position that a mechanist like Boyle would also have accepted. But in the “Monadology” he makes clear that he continues to believe his intermediate step that motion must originate within nature from entelechies or organisms, though these themselves are of course created by God. Thus he writes that “We could give the name entelechy to all simple substances or created monads, because they have within them a certain perfection; there is a kind of self-­sufficiency which makes them sources of their own internal actions”48—only a kind of self-­sufficiency, of course, because they must have been created by God, but still a kind of self-­sufficiency, in comparison to matter were it governed entirely by the law of inertia. Thus Leibniz has not argued that nature is what either his contemporaries or we would regard as a thoroughly mechanical system that nevertheless has a teleo­ logic­al explanation; he has rather argued that there must be within nature self-­ moving objects that violate the law of inertia, namely organisms, although of course these must have their own teleological explanation. Leibniz thus introduced organisms into his picture of nature for the same reason that Kant was later to do so, namely to explain the (apparent) violation or limitation of the law of inertia, and indeed it is hard to imagine, especially given the common argument that the world must contain organisms but not a “world-­soul,” that Kant did not actually have Leibniz’s argument in “On Nature Itself ” in mind in constructing his own argument. There remain significant differences between the two philo­ sophers: Leibniz ultimately holds that all of nature is comprised of organisms or entelechies, although of course not all of them are conscious, like souls,49 while Kant is willing to argue only that some things in nature are organisms, and although our conception of their origin leads us to think of all of nature as a single system, in his view we still should not infer from this that everything in nature is an organism after all. And, of course, for Kant all of this is supposed to be understood regulatively rather than constitutively—although Kant does not explain why the apparent violation of the law of inertia should be due to a mere limitation of our own cognitive powers, which we can overcome only by a regulative idea, rather than constituting an objective fact, which should be reflected in speculative metaphysics. I have now said what I want about the first stage of Kant’s overall argument for his Critical teleology, and for that matter about the first stage of Leibniz’s overall argument for his speculative teleology as well; namely, that in spite of their many differences they share a common view about the opposition between inertia and organisms that is a fundamental motivation for their amplification of a purely mechanical view of nature, though each in his own way. I now want to turn to the second stage of their teleologies, where they move from motivations for

47  PNG 8; WF 262.

48  Mon 18; WF 270.

49  Mon 19; WF 270.

Paul Guyer  253 intro­du­cing a teleological view of nature to an account of the point or as Kant calls it “final end” of the creation of nature. Here I want to reverse the strategy of the present section by arguing that in spite of some obvious similarities between Leibniz’s conception of the harmony between the realms of nature and grace and Kant’s conception of the bridge between the legislations of nature and freedom there is also a profound difference; namely, that while Leibniz is willing to speculate about God’s sufficient reason for the creation of nature, for Kant we must in the end conceive of the final point of nature in strictly human terms (i.e. as the realization of human freedom in human morality).

3  “The Highest Level of Virtue and Goodness Possible” Kant can hardly have failed to remember the conclusions of both the “Principles of Nature and Grace” and the “Monadology” in culminating each of his own three critiques with the argument that complete object of morality, the highest good, requires the postulation of the existence of God as the common author of both the laws of nature and the moral law. In the “Monadology” Leibniz states that “under the perfect government” of God “there will be no good action which doesn’t have its reward, and no bad one without its punishment,”50 which asserts a proportionality between virtue or vice and reward or punishment that seems to be the same as Kant’s idea that there should be a proportion between virtue as worthiness to be happy and actual happiness in the condition of the highest good. Leibniz describes the “City of God, this truly universal monarchy,” as “a moral world within the natural world,”51 language that Kant echoes in his first description of the highest good in the “Canon of Pure Reason” in the first Critique.52 And Leibniz states that the harmony “between the physical realm of nature, and the moral realm of Grace” depends upon the identity of “God considered as designer of the machine of the universe, and God considered as monarch of the divine city of minds,”53 which seems to express the same idea motivating Kant’s argument in the second Critique that “the highest good in the world is possible only insofar as a supreme cause of nature having a causality in keeping with the moral dispos­ ition is assumed.”54 However, there are also crucial differences between Kant’s conception of the highest good and the postulate of the supreme cause of nature and Leibniz’s conception of the perfect government of God, differences that suggest that Kant’s placement of his treatment of the highest good and the postulates of pure practical reason in a position in his own works analogous to Leibniz’s treatment of the harmony between the realms of nature and grace at the conclusion of his is meant to signal criticism of Leibniz as much as homage. There is of 50  Mon 90; WF 281. 53  Mon 87; WF 280.

51  Mon 86; WF 280. 52  A 809/B 837. 54  Critique of Practical Reason, Ak. 5:125.

254  The Teleologies of Leibniz and Kant course the obvious difference that for Leibniz the assurance of the harmony between the two realms is a matter of metaphysical fact while for Kant it is a postulate of pure practical reason or a matter for reflective rather than determinant judgment. But there is also the difference that Leibniz describes our moral duty, in the moral realm of grace, as that of loving God, and happiness as the immediate outcome of such love, while for Kant our duty is to achieve our own freedom, which makes no reference to God, and the connection between the fulfillment of that duty and the realization of happiness is by no means immediate, and not just because it may need to be mediated by God’s authorship of nature. Leibniz’s thought proceeds as follows. First, he moves without elaborate argument from the harmony between mechanical and teleological causation that he has been at pains to establish through his theory of substance to harmony between the natural and the moral realms: Just as we earlier established a perfect harmony between two natural realms, the one of efficient causes and the other of final causes, so we must also point out here another harmony, between the physical realm of nature, and the moral realm of grace; that is, between God considered as designer of the machine of the universe, and God considered as monarch of the divine city of minds.55

Leibniz apparently takes the inference from the presence of final causes in nature to the realization of a moral order in nature to need no further explanation. He also takes it as obvious that once the teleological foundation of the natural order has been established, it follows immediately that the realization of a moral order in nature will come about through natural mechanisms: This harmony ensures that things lead toward grace through the paths of nature itself, so that this globe, for example, must be destroyed and restored by natural means at such times as is required by the government of minds in order to achieve the punishing of some and the rewarding of others.56

In other words, God brings about the moral condition of justice through entirely natural means, for example, presumably, a flood. Such natural means bring about both of the requisite sides of the equation of a fully moral world, the punishment of sinners and the reward of the virtuous: We can also say that God the designer meets the requirements of God the legislator in every respect, and therefore that sins must bring along with them their own punishment through the natural order, and in virtue of the mechanical

55  Mon 87; WF 280.

56  Mon 88; WF 280.

Paul Guyer  255 structure of things themselves. And in the same way good actions will attract their reward through mechanical means in relation to bodies, even though that reward cannot and should not always arrive right away.57

Finally, the conduct of the virtuous is to love God, and pleasure is the natural response to and thus reward for loving: the good are “those in this great state who are not discontented, who trust in providence when they have done their duty, and who love and copy as they should the author of all good.” Thus they “delight in the consideration of his perfections, in accordance with the nature of genuinely pure love, which leads them to take pleasure in the happiness of what is loved.”58 Leibniz reiterates both the first and the last parts of this vision in “The Principles of Nature and Grace.” Here he argues first that “the most perfect state, formed and governed by the greatest and best of monarchs, in which there is no crime without punishment, no good act without its appropriate reward,” is brought about “not by a disturbance of nature, as if what God has laid down for souls might interfere with the nature of bodies, but through the very order of natural things, in virtue of the harmony which has been pre-­established from all time between the kingdoms of nature and grace, between God as architect and God as monarch.”59 Then he reiterates the second step, focusing exclusively on our love for God and the pleasure it brings about—any implication that our duty includes loving other humans or other creatures would have to be drawn from the phrase that we are to “love and copy the author of all good.” The connection between the fulfillment of this duty of love of God and pleasure is immediate: “since true pure love consists in being in a state which enables one to take pleasure in the perfections and the happiness of the person one loves, it follows that when God is its object love must give us the greatest pleasure of which we are capable.”60 Since loving naturally brings pleasure in the happiness of the beloved in its train, and loving God is the greatest love there can be, an entirely natural mechanisms adds the greatest possible happiness to the fullest performance of duty. Leibniz says nothing about whether the performance of duty can itself come about by entirely natural means, although his view that nature is ultimately comprised of monads all but harmoniously expressing their own internal principles implies that it must. How close and yet how far is this from Kant’s conception of the highest good and the conditions for its realization. Kant describes the goal of morality as the transformation of the natural world into a moral world,61 and holds that happiness as well as virtue must be part of the “complete (consummatum)” good for human moral agents.62 But beneath this level of agreement, there are profound differences between Kant’s conception of the highest good and his arguments for it and the Leibnizian model of the harmony between nature and grace we have 57  Mon 89; WF 280–281. 60  PNG 16; WF 265.

58  Mon 90; WF 281. 59  PNG 15; WF 265. 61  A 809/B 837. 62  Critique of Practical Reason, Ak. 5:110.

256  The Teleologies of Leibniz and Kant just described. Since Kant’s thought about the highest good is one of the most complex and in some ways unstable aspects of his philosophy, we will have to touch upon some general questions about it before turning to his specific argument in the Critique of the Power of Judgment and then highlighting Kant’s differences with Leibniz. In the 1793 essay “On the Common Saying: That May Be Correct in Theory But It Is of No Use in Practice,” Kant defines the highest good, as an “end for the human being’s will” “introduced” by the “concept of duty,” toward which each human being is therefore obliged to work, as “universal happiness combined with and in conformity with the purest morality throughout the world.”63 In order to reach that formulation, however, Kant had to resolve ambiguities in his conception of whose happiness is concerned in the highest good, why duty should concern happiness when a desire for happiness is not supposed to be the motive for duty, and when and where the happiness involved in the highest good should be expected to be realized. The first question is whether the highest good links the happiness of the particular moral agent to her morality, or makes the happiness of all human beings the goal of the virtuous agent; the second question is whether happiness is a natural and non-­moral goal of human beings that is to be “conditioned” or constrained by morality, or an end commanded by morality itself even though it is not the motivation for morality, and along with this whether Kant believes, as Leibniz does, that the goal of morality is the punishment of the sinners as well as the reward of the good, or only the maximization of both goodness and happiness; and the third question concerns whether in the realization of the highest good happiness is supposed to be possible in the natural world or only in another world, a world that is “future for us.”64 The answers to these questions are no doubt linked: the moral agent’s own happiness would seem likely to be a nat­ ural end that needs to be constrained by morality, and conversely the happiness of all would not seem to be a natural end of any agent but rather an end set by morality itself; and given the empirical evidence, the individual happiness of the virtuous moral agent might not seem to be an end always realized in the natural course of life, though maybe the happiness of all can be imagined as a possible state of the human species at some point in its natural existence. Prior to “Theory and Practice,” Kant does not seem to take unequivocal positions on these issues.65 Thus, in the Critique of Pure Reason, where as we saw Kant explicitly connects his

63  Ak. 8:279; trans. from Kant (1996, 282). 64  See A 811/B 839. I have previously discussed a number of these questions in Guyer (2000, ch. 10, and 2002). 65  I have offered an account of the stages in Kant’s development of his conception of the highest good in Guyer (2011).

Paul Guyer  257 treatment of the highest good to Leibniz’s treatment of the realms of nature and grace,66 Kant begins by stating that in an ideal world: [in] which we have abstracted from all hindrances to morality (of the in­clin­ ations), such a system of happiness proportionally combined with morality can also be thought as necessary, since freedom, partly moved and partly restricted by moral laws, would itself be the cause of the general happiness, and rational beings . . . would themselves be the authors of their own enduring welfare and at the same time that of others.67

This suggests that moral agents should work for the happiness of all, not just themselves, and that under ideal circumstances universal happiness would thus be the natural outcome of universal morality (and has nothing to say about punishment). But then Kant says that since in fact not everyone lives up to the demands of morality, the agent who does should at least be able to expect that her efforts will be rewarded with her own happiness (although Kant still has nothing to say about the punishment of non-­cooperators),68 and this then becomes the basis for the postulate of personal immortality, because in the obvious absence of evidence that personal virtue is rewarded with personal happiness in this life, the agent has to be able to believe that her virtue will be rewarded with happiness in a future life. The happiness at issue here is obviously personal happiness, not commanded by morality itself but constrained by it. In the Groundwork for the Metaphysics of Morals, four years after the first edition of the first Critique, Kant does not mention the highest good, but introduces the premise that is needed to explain how universal happiness could be the object of morality even when the natural desire for one’s own happiness needs to be constrained by morality: namely, while “the natural end that all human beings have is their own happiness,” there would be “only a negative and not a positive agreement with humanity as an end in itself unless everyone also tries, as far as he can, to further the ends of others.”69 Kant subsequently incorporates this thought into his definition of the “realm of ends” as the final formulation of what we are commanded to seek by morality as “a whole of all ends in systematic connection (a whole both of rational beings as ends in themselves and of the ends of his own that each may set for himself).”70 If morality commands us to promote the real­ iza­tion of the ends of all, insofar as they are compatible with one another and thus that is possible, not because of our natural desire for our own happiness but because that is what it is to treat all as ends in themselves, and if happiness is just 66  A 812/B 840. 67  A 809/B 837. 68  A 810/B 838. 69  Groundwork for the Metaphysics of Morals, Ak. 4:430; Kant (1996, 81). 70  Groundwork for the Metaphysics of Morals, Ak. 4:433; Kant (1996, 83).

258  The Teleologies of Leibniz and Kant what results from the realization of ends, then indeed the maximal possible universal happiness is what is commanded by morality, though not because of the value of happiness as such but because of the value of freely set ends and their realization. Thus morality itself determines the highest good as defined in “Theory and Practice” as its complete end, and the highest good must thus be able to be considered as possible if morality itself is to be possible. (Once again, nothing about the punishment of the vicious follows or is said to follow from such an argument.) But when the Critique of Practical Reason explicitly returns to the topic of the highest good, this clarity may once again be muddied, for it looks as if Kant’s opening argument there, that the highest good is the complete object of morality, combines a natural need for happiness with the moral condition of being worthy of it,71 and Kant’s further argument that both the Stoic and Epicurean conceptions of the highest good erred in treating (each in its own way) morality and happiness as the same thing, “analytically” connected, rather than recognizing that they are “two specifically quite different elements of the highest good” that must be connected by a “synthesis of concepts”72 also suggests that the highest good connects a moral end and a natural end, which can only be one’s own happiness constrained by morality rather than universal happiness commanded by morality. In spite of this, in the second Critique Kant departs from the position of the first by arguing that the postulate of personal immortality is necessary to ground the possibility of perfecting one’s virtue, not belatedly enjoying happiness, and instead treats the happiness component of the highest good as something that must be realizable in nature, a possibility that we can only ground by postulating God as the “supreme cause of nature having a causality in keeping with the moral disposition.”73 Kant does not explain why we must be able to conceive of the happiness component of the highest good as realizable in nature if the virtue component could be perfected only beyond our natural life, but nevertheless henceforth always treats the highest good as a condition we must believe can be realized in nature and in fact makes no further effort to argue that personal immortality is a condition of the possibility of the highest good. (And once again, Kant says nothing about the punishment of sinners.) The Critique of the Power of Judgment links the topic of the highest good to its Critical teleology in a complicated dialectic that does not fully disambiguate Kant’s conception of the highest good but that does make it plain that we must conceive of it as being realizable within a nature which, however, we must regard as the product of an intelligent author—and thus not in a realm of grace separate from the realm of nature. The entire argument can be read as urging “attention to 71  Critique of Practical Reason, Ak. 5:110; Kant (1996, 229). 72  Critique of Practical Reason, Ak. 5:112–113; Kant (1996, 230). 73  Critique of Practical Reason, Ak. 5:125; Kant (1996, 125).

Paul Guyer  259 the ends of nature and research into the inconceivably great art that lies hidden behind its forms in order to provide incidental confirmation from natural ends for the ideas created by pure practical reason”;74 in other words, the chain of teleo­logic­al reflection that has begun with our experience of organisms is supposed to lead us to the same vision of nature as a realm suitable for the realization of the complete object of morality to which reflection on the demands of morality itself has previously led us.75 This chain of thought, the first stage of which has been the move from our experience of organisms to a non-­dogmatic theism already described, continues as follows. It is based on the premise that if we must conceive of a designer and creator of nature then we must also be able to conceive of a goal that designer had and thus a point for its creation: But if we assume that a connection to ends in the world is real and assume that there is a special kind of causality for it, namely that of an intentionally acting cause, then we cannot stop at the question why things in the world (organized beings) have this or that form, or are placed by nature in relation to this or that other thing; rather, once an understanding has been conceived that must be regarded as the cause of the possibility of such forms as they are really found in things, then we must also raise the question of the objective ground that could have determined this productive understanding to an effect of this sort, which is then the final end for which such things exist.76

Kant formally defines a final end in ontological terms as “that end which needs no other as the condition of its possibility,”77 but his argument really turns on the assumption that our conception of the creator of nature and therefore of nature itself must impute to it an end that is final in a normative sense as well; that is, which exists entirely because of its own value and not merely because it is a means to something else.78 Kant does not bother to explain why we must conceive of such a final end for the creator and creation of nature; presumably his assumption is that, since we cannot conceive of our own practical reason without a final end, we certainly cannot conceive of a reason greater than our own acting without a final end.79 On the basis of this assumption, Kant’s argument is then that human happiness as such cannot immediately be conceived of as the final end of nature: not only does nature not seem especially well-­disposed to our happiness, but, perhaps even more importantly, from a strictly natural point of view the “human being is . . . only a link in the chain of natural ends,”80 and thus our happiness would 74  Critique of the Power of Judgment, §86, Ak. 5:444–445. 75  On this point, see Guyer (2001a). 76  Critique of the Power of Judgment, §84, Ak. 5:434–435. 77  Ibid., §84, Ak. 5:434. 78  See ibid., §82, Ak. 5:426. 79  On this issue, see Guyer (2014). 80  Critique of the Power of Judgment, §83, Ak. 5:430–431.

260  The Teleologies of Leibniz and Kant from this point of view be at most a link in a chain of ends, not a natural end, even if nature favored it. Instead, it is only human freedom, as fully realized only in human morality, that can be conceived of as an end for nature that is both ontologically and normatively unconditioned, thus a final end. Kant writes: Now we have in the world only a single sort of beings whose causality is teleo­ logic­al, i.e., aimed at ends and yet at the same time so constituted that the law in accordance with which they have to determine ends is represented by themselves as unconditioned and independent of natural conditions but yet as necessary in itself. The being of this sort is the human being, though considered as noumenon: the only natural being in which we can nevertheless cognize, on the basis of its own constitution, a supersensible faculty (freedom) and even the law of the causality together with the object that it can set for itself as the highest end (the highest good in the world). Now of the human being . . . as a moral being, it cannot be further asked why (quem in finem) it exists. His existence contains the highest end itself to which, as far as he is capable, he can subject the whole of nature . . . only in the human being, although in him only as a subject of morality, is unconditional legislation with regard to ends to be found, which therefore makes him alone capable of being a final end, to which the whole of nature is teleologically subordinated.81

Kant’s assumption is that we must be able to find a final end for nature in nature, and that only human beings, in virtue of their freedom, are candidates for this role: freedom, as spontaneity, is ontologically unconditioned, but as the foundation of morality it is also normatively unconditioned. No doubt we must conceive of God himself as unconditioned, but God is outside of nature, so not himself a candidate for being the final end of nature. Only human beings as moral agents, or human morality itself, can be conceived as the final end of the system of nature. Our experience of organisms gets us started down the path to a teleological view of nature that culminates in conceiving of the realization of our own morality as the only possible point for the creation of nature, and the realm of grace as something that can be achieved only by humans (though perhaps with divine underwriting) and only in the realm of nature. As our last quotation has already suggested, Kant infers from this that we must actually conceive of the complete object of morality—namely, the highest good— as the final end of nature, and this is what is supposed to “incidentally confirm” the position previously arrived at on moral grounds alone that we must able to conceive of the highest good as realizable within nature. But before he takes that step, Kant actually makes another move. It will not have escaped notice that Kant

81 Ibid., §84, Ak. 5:435–436.

Paul Guyer  261 has said that it is the human being or more specifically human freedom as nou­ menon that is the only thing that we can conceive of as the final end of nature, and of course the noumenal is not literally in nature, a point that Kant elsewhere stresses by stating that we must conceive of the empirical character of the human agent, which is subject to natural laws, as the produce of the intelligible character of the human agent, subject to moral laws.82 This might suggest that human freedom properly understood cannot be the final end of nature in nature after all. Part of the way that Kant deals with this problem is by arguing that it is human culture, specifically the culture of discipline, that should be conceived of as the final end of nature in nature, or more precisely the “ultimate end” of nature:83 discipline can be conceived of as something that can be developed entirely within nature, by natural means, and that is not identical to freedom, which must be understood noumenally, but that is a necessary condition for the realization of the ends set by freedom. That is, although we may have to conceive of the free choice to be moral as noumenal, outside of nature, we can also suppose that such a choice can be made effective in the natural world only through the discipline of our natural constitution, with all its otherwise unruly inclinations. (Of course, a choice to be evil could also be made effective only through discipline—discipline is a necessary but not sufficient condition of morality.)84 In spite of this nicety, Kant also conceives of the highest good as a condition that can be realized within nature, and thus as a candidate, indeed the final candidate, for the final end of nature. This conclusion would seem most natural if the highest good were conceived of as commanded entirely by morality itself; that is, if its element of happiness were conceived of not as personal happiness naturally desired and constrained by morality but rather as universal happiness commanded by morality because of morality’s own injunction to promote the ends of all within a realm of ends. Yet in the third Critique, as in the second, Kant often speaks of the highest good as if it consisted merely in the moral constraint of the natural desire for happiness (and therefore the desire for one’s own happiness). Thus in a footnote to the passage from §84 just quoted, Kant says that “happiness can only be a conditioned end, and . . . the human being can thus be the final end of creation only as a moral being; as far as his state is concerned, happiness is connected to it only as a consequence, in proportion to the correspondence with that end as the end of his existence.”85 Although Kant’s remark that happiness is connected to the state of the moral being “as a consequence” is ambiguous, his

82  See A 532–537/B 560–565, and Critique of Practical Reason, Ak. 5:94–100. 83  Critique of the Power of Judgment, §83, Ak. 5:431–432. 84  On this point, see again Guyer (2014). 85  Critique of the Power of Judgment, §84, Ak. 5:436 n. This is one of the places where Kant uses a word that can be translated as “proportion” (namely, Übereinstimmung) in describing the connection between happiness and morality in the highest good, but there is still no suggestion here that the required proportionality has anything to do with making sure that sinners are punished.

262  The Teleologies of Leibniz and Kant previous remark that “happiness can only be a conditioned end” suggests that he is thinking of it as a natural end that needs to be constrained by morality, which is the true final end of nature. Kant’s restatement of the “moral proof of the existence of God” in §86 is similarly ambiguous. First Kant says that: The moral law, as the formal rational condition of the use of our freedom, obligates us by itself alone, without depending on any sort of end as a material condition; yet it also determines for us, and indeed does so a priori, a final end, to strive after which it makes obligatory for us, and this is the highest good in the world possible through freedom.86

Kant’s claim that the moral law determines a final end for us a priori lends itself to the interpretation that he is taking the moral law to command that we seek to promote the happiness of all through its command that we promote the particular ends of all in a realm of ends; thus there would be a direct path from the idea that freedom is the final end of nature to the idea that morality is the final end of nature to the idea that the highest good is the final end of nature—and since happiness is clearly something that can be realized only in nature, even if only at some future stage in the natural life of the human species, the final end of nature would also be within nature, even if it originates with a free choice that is in some sense outside of nature. Paradox could be avoided here by treating the moral component of the highest good, namely worthiness to be happy, as something that is achieved within nature, as the consequence of the extra-­natural choice to be moral. Or this element could be identified with virtue, which is the expression within nature of the noumenal commitment to be moral.87 But Kant immediately clouds this picture by continuing that: The subjective condition under which the human being . . . can set a final end for itself under the above law is happiness. Hence the highest physical good that is possible in the world and which can be promoted, as far as it is up to us, as a final end, is happiness—under the objective condition of the concordance with humans with the law of morality, as the worthiness to be happy.88

Once again Kant seems to be conceiving of happiness as the merely natural desire of human beings, which would of course mean that it is a desire for their own happiness, which has to be conditioned or constrained by compatibility with the 86  Ibid., §86, Ak. 5:450. 87 See Metaphysics of Morals, Doctrine of Virtue, Introduction, section IX, Ak. 6:394–395; Kant (1996, 524–526). Virtue is here described as the strength to overcome the “obstacles” to morality set by “natural inclinations,” and since they are in nature it too must be in nature, perhaps an effect of the noumenal determination of the will but not identical to it. 88  Critique of the Power of Judgment, §86, Ak. 5:450.

Paul Guyer  263 demands of morality, and the highest good would be a hybrid object of the faculty of desire, partly merely a matter of inclination and only partly a matter of morality. Thus it seems that only in “Theory and Practice” does Kant finally resolve the ambiguity in his conception of the highest good, making it clear there that it is only universal happiness commanded by morality, not one’s own happiness merely constrained by morality, that is the happiness included in the highest good. (Of course, universal happiness would include one’s own, so through the concept of the highest good morality would actually command the pursuit of one’s own happiness as part of the pursuit of universal happiness.)89 But even though the Critique of the Power of Judgment has not yet resolved this ambiguity, it has made enough so that we can now clearly see the ultimate differences between Kant’s teleology and Leibniz’s. Both, as we have previously seen, start with the concept of organisms, but while Leibniz argues that we can assert as a matter of speculative metaphysics both that all substances in nature must ul­tim­ ate­ly be organisms and that all such organisms must have the ground of their being in God, Kant starts with the assumption that only some objects in nature are organisms, even though our subjective conception of the condition of their possibility will lead us to think of all of nature as a system, and then adds that we must think of God as the ground of the possibility of this system on regulative and practical grounds, but not as a matter of speculative metaphysics. Further, although both share a vision of the harmony between the realms of nature and grace that is similar at the most general level, and indeed Kant explicitly ac­know­ ledges the inspiration of Leibniz for his own view, there are nevertheless crucial differences between their conceptions of this harmony, or the highest good. Leibniz locates the duty of humankind in loving God, and sees happiness as the immediate result of this love; perhaps we should also love other human beings and derive happiness from this too, in a kind of imitatio of our love for God, but Leibniz barely even hints at such a human dimension to morality. However, while Kant derives the idea that the system of nature must have a final end from an argument that we must conceive of this system as the product of a supremely rational God who would not act without a final end, he immediately turns around and argues that we can only conceive of this final point in human terms—namely, as the development of our own morality, which requires us to express the unconditional value of our own freedom, the only unconditional value within the world, by treating ourselves and others as ends in themselves and for the sake of that also promoting the particular ends freely chosen by ourselves and others, which is what would bring about universal happiness. So, for Kant the harmony between the realm of nature and the realm of grace must, in the first instance, be a harmony between nature and the possibility of the exercise of human freedom in

89  See esp. Metaphysics of Morals, Doctrine of Virtue, §27, Ak. 6:451.

264  The Teleologies of Leibniz and Kant nature. By means of his conception of the highest good, even though that is fully clarified only after the Critique of the Power of Judgment, in “Theory and Practice,” Kant also finally makes it clear that we must conceive of human happiness, although it is the mediate rather than immediate object of morality, as possible within and as the eventual outcome of nature. Since for Kant the laws of the realm of grace demand respect for human beings rather than love of God and happiness must be the consequence of this rather than the future reward for goodness and punishment for sin, for Kant the harmony between the realms of nature and grace must in fact be their possible identity within nature, not a mere harmony, like Leibniz’s. For Kant, finally, God can only be seen as the guarantor of the possibility of this harmony between nature and grace within nature, and not as the object of our duty love for whom immediately produces happiness. So, in spite of the similarity between their conceptions of organism, which provide the first stages of their teleologies, and the connection between the ideas of the harmony between the realms of nature and grace and the concept of the highest good that Kant himself recognizes in the first Critique and claims against Eberhard, the two philosophers’ conceptions of the harmony between nature and grace end up being quite different, Kant’s not only regulative but also morally humanistic while Leibniz’s is not only metaphysical but morally theological as well.

10 Kant’s Theory of Divine and Secondary Causation Desmond Hogan, Princeton University

In this chapter I intend to examine Kant’s theory of the relation between God’s causal activity in the world and so-­called “secondary” causation, the causality of created beings. Kant’s central contributions to a venerable philosophical debate on this topic can be approached along two intersecting metaphysical axes. One issue to which his early work devotes considerable attention is the character of God’s contribution to causal relations among created entities. His pre-­Critical writings develop a distinctive model of this contribution as a component of an interactionist cosmology targeting Leibniz’s pre-­ established harmony and Malebranche’s occasionalism. A second and related problem occupying Kant throughout his life is the mode of divine causal activity in nature more generally— what one might call the general division of causal labor between God and created beings. The central question he faces here is the traditional one for a theistic metaphysics: How does the activity of God viewed as primordial creator and conserver of the world relate in general to the causal activity, if any, of created beings? Kant accepts the view of many predecessors that a theory of God’s causal role in the ordinary course of nature is a prerequisite of any coherent metaphysics of miraculous interventions. His own evolving account of this divine causal role is shaped by his ongoing engagement with three competing theories of divine causation distinguished in late scholastic philosophy, and vigorously debated in the modern period by Malebranche and Leibniz. The first of these theories is occasionalism, whose full-­fledged version holds that God is the only true cause, solely responsible for all effects in nature. There is no genuine secondary causation at all according to occasionalism—God produces everything in nature by his own power, and finite substances and states count as causes only in the attenuated sense of providing occasions for God to act and produce effects in accordance with his own decrees. Malebranche, the most influential occasionalist of the modern period, maintains that “the nature or power of each thing is nothing but the will of God . . . all natural causes are not true causes but only occasional causes.”1 Occasionalism is not an innovation of 1  OCM II 312 (trans. Malebranche 1997a, 448). Desmond Hogan, Kant’s Theory of Divine and Secondary Causation. In: Leibniz and Kant. Edited by Brandon C. Look, Oxford University Press. © Oxford University Press 2021. DOI: 10.1093/oso/9780199606368.003.0010

266 Kant ’ s Theory of Divine and Secondary Causation the modern period—earlier strains of it are seen in medieval Islamic thought in response to perceived tensions between Aristotelian essentialism and scriptural accounts of miracles.2 Another theory of divine causation Freddoso labels mere conservationism (conservationism) is associated by Suarez, Malebranche, and Leibniz with the fourteenth-­century Dominican theologian Durandus of St. Pourçain. On this view, immediate divine action in the ordinary course of nature is limited to God’s action in conserving created beings in existence. These conserved beings are regarded as capable of producing effects through their own powers without any additional divine action. In contrast to occasionalism, mere conservationism thus treats finite substances as genuine causes of natural effects and their actions as their own and not God’s actions. General concurrence (concurrence) is a third theory of divine and secondary causation widely endorsed in late scholastic theology and the early modern period. It agrees with conservationism and occasionalism that finite substances depend completely for their existence on God’s creative and conserving action. It sides with conservationism against occasionalism in holding that there is genuine secondary causation. But concurrence claims that production of any effect whatsoever by creatures requires a divine action going beyond their mere conservation. In particular, the view holds that God must also act as immediate co-­cause of every naturally produced effect through a cooperation or concurrence with the creature’s powers, one in which neither causal contribution alone suffices to elicit the effect. Concurrentism thus posits a more intimate and extensive divine involvement than conservationism in the production of natural effects, while opposing the occasionalist conclusion that God does everything and creatures nothing. While Kant rejects occasionalism throughout his life, his objections to central occasionalist arguments are not always explicitly spelled out. Identifying these objections is one task in evaluating his stance in the traditional debate. Kant’s contributions to the concurrence/conservation dispute are also at the heart of his mature theory of divine causation. In spite of their significance for his theory of grace, these contributions have gone more or less unexamined.3 One reason for the neglect is simply that medieval and modern debates on concurrence and conservationism were until recently poorly understood. Freddoso’s influential scholarship on the late medieval concurrence debate has greatly advanced our understanding of the issue in the modern context.4 Freddoso points out that mere conservationism was not, as one might expect, the default position for medieval 2  For good overviews, see Perler and Rudolph (2000); Nadler (1996). 3  A new exception is Insole (2013), which makes reference to a manuscript version of this chapter. I have added some responses below. 4  See Freddoso (1988, 1991, 1994). The manner of God’s concurring action in the ordinary course of nature, my main topic in this chapter, is traditionally treated separately from the metaphysics of

Desmond Hogan  267 opponents of occasionalism. Conservationism was rejected as philosophically and theologically deficient by nearly all scholastic participants in the divine caus­ ation debate. It continued to be regarded as suspect well into the modern period. It is explicitly rejected by Leibniz, for example, and dismissed by Malebranche as “untenable in every respect.”5 Common objections included a claim that conservationism issues in a deistic metaphysics failing to do justice to the absoluteness of creaturely dependence, and also that it is incompatible with Christian orthodoxy on grace. Kant’s German predecessors generally follow Leibniz in rejecting conservationism in favor of concurrence. Kant himself, though heavily influenced by this theological tradition, rejects concurrence in favor of conservationism. One question this raises is whether the departure indicates sympathy for the more deistic position with which conservationism was traditionally associated. This chapter study sets out from an examination of Kant’s response to an influential argument for occasionalism proceeding from premises regarding creation and conservation apparently close to his own position. §§2 and 3 turn to Kant’s early stance in the conservation/concurrence debate, and also examine his early division of divine action into the ordinary and miraculous. A familiar narrative emphasizes Kant’s increasing discontentment with the metaphysics of the German schools in the period leading up to the Critical turn. The narrative’s influence has tended to overshadow claims, beginning in the 1760s and continuing into the Critical period, that some divine interventions in nature are probably necessary. This neglected doctrine is rooted in a major shift in Kant’s metaphysics of agency in the 1760s, one with important consequences for his subsequent views on divine and secondary causation. §4 investigates the evolution of Kant’s views on divine causation into the Critical period, focusing on his mature opposition to concurrentism. Analysis of the arguments reveals that his own version of conservationism is not motivated by a rejection of theistic orthodoxy regarding the absoluteness of creaturely dependence. Another finding is a considerable continuity between Kant’s mature and early metaphysics of divine and secondary causation. There is reason to conclude that the most significant milestone in the emergence of his mature account does not lie in his transition from pre-­Critical to Critical philosopher, but in an earlier one from necessitarian to libertarian on human agency. The notion that the Critical philosophy just dispenses with or forbids metaphysical theorizing regarding God’s relation to creaturely causes nevertheless remains influential. It is best dispelled by attending to Kant’s writings, which display no scruples about insisting on the correctness of a particular model of the relation. It will thus be necessary to examine how the proposed model relates to grace and miraculous interventions. See Suarez, Metaphysical Disputations [DM] 22.4.11, trans. in Suarez (2002, 223); also Leibniz, Causa Dei, GP VI 440. 5  OCM III 243 (trans. Malebranche 1997a, 680).

268 Kant ’ s Theory of Divine and Secondary Causation transcendental idealism, and how its correctness can supposedly be affirmed given Critical strictures on knowledge.

1  Creation, Conservation, and Secondary Causation Kant’s early writings subscribe to a pluralistic substance metaphysics presenting God as creator and conserver—Urheber und Erhalter—of all finite existence through his will, and the ontological ground of possibility itself through his understanding. This theme of ontological dependence is prominent throughout early writings. It finds expression in a claim that “whatever exists, whether it is possible or actual, is only something insofar as it is given through God” (Only Possible Proof, Ak. 2:151). In particular, Kant accepts the view of Descartes, Malebranche, and Leibniz that finite substances do not depend on God merely at the moment of their creation. The contingency of finite substances is held to entail their complete dependence on divine conserving activity subsequent to creation. This conservation Kant describes as “actuation of continued existence,” as opposed to creation as first “production of the world” (Ak. 28:1307; R 4792, 6173). His early thought does not reject an orthodox Christian view of the universe as of finite age and its creation as an event in time.6 While his mature idealism demands a different understanding of creation, he continues in mature works to describe his metaphysical position as “theism,” and to defend the doctrine of creation (Practical Reason, Ak. 5:101–102; Danziger Rationaltheologie [1783–1784], Ak. 28:1299; R 6173). He also continues to describe created entities as “only perduring through a continuous actum divinum . . . as much power as was required for the creation of substances is needed for their conservation” (Religionslehre Pölitz, Ak. 28:1104; cf. Nova Dilucidatio, Ak.1:414). The strong emphasis on the ontological dependence of creatures in Kant’s early and mature writings raises questions about his lifelong opposition to occasionalism. The opposition is not itself in doubt.7 Created substances are presented in his earliest works as subsistent entities endowed with active and passive forces causing an array of effects in nature.8 What is not so obvious is how 6 “[T]he remaining part of the time sequence of eternity is always infinite and the part already flown by is finite . . . Creation is never completed. Though it did begin once, it will never end” (Universal Natural History [1755], Ak. 1:314). 7  For early affirmations of realism regarding secondary causation, see True Estimation, Ak.1:17–22, 142; Universal Natural History, Ak.1:234; Nova Dilucidatio, Ak. 1:411–412, 415; Physical Monadology, Ak. 1:480; Prize Essay, Ak. 2:287; Dreams of a Spiritseer, Ak. 2:324, 343; Inaugural Dissertation, Ak. 2:397, 409. See also the detailed discussion of Kant’s early causal realism in Watkins (2005, chs. 1–2). 8  Kant rejects Descartes’s definition of substance as a “thing which so exists that it needs no other thing in order to exist” (Principles I, 51) on the usual grounds that it invites a short argument to Spinozism (Ak. 28:563–564; Pölitz Religionslehre, Ak.1104–1105). Baumgarten, whose textbook Kant used in lectures, proposes as a definition applying univocally to God and creatures: “a being which can exist without being a determination of another” [Ens, quod potest exsistere, licet non sit determinatio

Desmond Hogan  269 Kant proposes to respond to Malebranche’s argument that the ontological dependence of creatures already forces acceptance of an occasionalist model of God’s relation to secondary causes. One of Malebranche’s most influential arguments for global occasionalism rests on the principle that divine conservation amounts to continuous creation. This (“CCC”) principle is rooted in the idea, endorsed in one form or other by Aquinas, Suarez, Descartes, Leibniz, and Kant, that the ontological dependence of finite substances subsequent to creation remains such that they would immediately cease to exist were God to withdraw his conserving support. Such lack of existential inertia is presented as an immediate correlate of creaturely contingency, and it motivates an account of creatures’ conservation as metaphysically symmetrical with creation. As Leibniz explains, “the dependence [of creatures] being as great in the sequel as in the beginning, the extrinsic denomination, of being new or not, does not change its nature” (Theodicy §385, GP VI 343–344).9 One of Malebranche’s central arguments for occasionalism sets out from this idea. It begins with the claim that God cannot create a body—in his example a chair—“without at the same time willing that it exist either here or there and without His will placing it somewhere.”10 The idea here is that creation requires the fixing of some or other complete set of properties—it is impossible for God to create a chair “in general.” Since the created entity did not exist prior to its creation, Malebranche concludes that its esse and initial properties must be entirely fixed by God at the moment of creation, leaving nothing undetermined to be caused by the created entity at that instant. He then generates the occasionalist conclusion by extending the argument to all created entities and generalizing it using CCC to all times: Creation does not pass, because the conservation of creatures is—on God’s part—simply a continuous creation, a single volition subsisting and operating continuously. Now God can neither conceive nor consequently will that a body exist nowhere, nor that it does not stand in certain relations of distance to other bodies. Thus, God cannot will that this armchair exist, and by this volition create alterius] (preface to Metaphysica, repr. in Ak. 17:8–9). Kant incorporates the “first subject” requirement, while rejecting the suggestion that mere possibility of existing first subject suffices for substancehood: “substance is that which exists only as subject . . . when some opine that substances could also exist as inherent things [als Inhärenzen]—this is incorrect” (Ak. 28:563). 9  Cf. Descartes, “All the time of my life can be divided into parts, each of which is entirely independent of the others, so that from the fact that I existed a short time ago, it does not follow that I ought to exist now, unless some cause as it were creates me again in this moment, that is, conserves me. For it is quite clear to anyone . . . that the same power and action are needed to preserve anything at each individual moment of its duration as would be required to create that thing anew if it were not yet in existence. Hence the distinction between preservation and creation is only a conceptual one, and this is one of the things that are evident by the natural light” (AT VII 48–49; CSM II 33); also Suarez, DM 22.1.16; OCM XII 156 (trans. Malebranche 1997b, 112). 10  OCM XII 156 (trans. Malebranche 1997b, 112). See Garber (2001, 209–210) on De La Forge’s earlier version of the argument.

270 Kant ’ s Theory of Divine and Secondary Causation or conserve it, without situating it here, there, or elsewhere. It is a contradiction, therefore, for one body to be able to move another.  [OCM XII 160]

The final sentence adds to the thought that there is nothing left undetermined by God at creation the idea that the creature cannot modify or override God’s creative choices. Kant offers no explicit response to this argument, though his writings exhibit some familiarity with Malebranche’s thought.11 On one interpretation, he might be taken as rejecting Malebranche’s extension of the claim that God does everything and creatures nothing beyond the instant of creation. While Leibniz, Wolff, Baumgarten, and Meier accept the CCC principle on which this extension rests, some Kantian texts suggest resistance.12 On the one hand, he affirms even into the Critical period that “there is no distinction between creation and conservation with respect to God; since in God’s case we cannot distinguish beginning and continuation. It is one act; a distinction is made only in the world” (Danziger Rationaltheologie, Ak. 28:1308). The same transcript continues, however, by warning that conservation “cannot be called creatio continua, because creation is the production of the ortus or beginning. A continuous beginning is contradictory. We cannot make any concept of it” (Ak. 28:1308).13 Given the initial affirmation of symmetry, it is not clear whether Kant’s qualification denies that the symmetry is sufficiently robust for Malebranche’s purposes. Such denial would not have been unprecedented. While the CCC principle finds broad support in eighteenth century Germany, Leibniz already records some opposition. In Kant’s day, the principle is rejected by Crusius, a major influence on his early thought.14 Another more promising approach to Kant’s early rejection of Malebranche’s occasionalist conclusion interprets him as rejecting the premise that God does

11  For early references to Malebranche’s occasionalism and doctrine of vision in God, see Nova Dilucidatio, Ak. 1:415; Metaphysik Herder, Ak. 28:887; Dissertation, Ak. 2:410; R 4275. Acquaintance with De la recherche de la vérité is evident in Logik Philippi (1770–1771), Ak. 24:337; Logik Busolt, Ak. 24:613. 12  Leibniz writes that conservation by God “consists in the perpetual immediate influence which the dependence of creatures demands. This dependence attaches not only to the substance but also to the action, and one can perhaps not explain it better than by saying with theologians and philosophers in general that it is a continued creation” (Theodicy §27, GP VI 118). He later qualifies the claim somewhat, writing that “the production of modifications has never been called creation . . . God produces substances from nothing, and the substances produce accidents by the changes of their limits” (Theodicy §§395–396, GP VI 351–352). Cf. Christian Wolff, “Quoniam actio conservationis et creationis quoad Deum minime differunt; conservatio continuata creatio est” (Wolff 1736, §845); Baumgarten, “unde conservatio non male dicitur continuata creatio” (Baumgarten  1757, §951; also Meier 1757, §1023). 13  An undated handwritten note in Kant’s copy of Baumgarten’s Metaphysica likewise asserts that “a continuous beginning is a contradiction” (R 3688; cf. Metaphysik Dohna, Ak. 28:702). 14 See Theodicy §382, GP VI 342. Crusius’s objection appeals to lack of insight into God’s activity: “We cannot understand the internal character of the divine actions, and so it is not justified when the divine action by which the world continues to exist is identified with that with which it is created and conservation called continuous creation” (Crusius 1745, §329).

Desmond Hogan  271 everything and creatures nothing even at the instant of creation. Malebranche rests this key move on an intuition that “only He who gives being could give the ways of being, since the ways of being are nothing but beings themselves, in this or that fashion.”15 Leibniz’s response is to maintain that CCC can be affirmed and yet occasionalism averted by marking distinct orders of causation within a single instant: Let us assume that the creature is produced anew at each instant; let us grant also that the instant excludes all priority of time, being indivisible; but let us point out that it does not exclude priority of nature, or what is called anteriority in signo rationis, and that this is sufficient. The production, or action whereby God produces, is anterior by nature to the existence of the creature that is produced; the creature taken in itself, with its nature and its necessary properties, is anterior to its accidental affections and to its actions; and yet all these things are in being in the same moment. God produces the creature in conformity with the exigency of the preceding instants, according to the laws of his wisdom; and the creature operates in conformity with that nature which God conveys to it in creating it always. The limitations and imperfections arise therein through the nature of the subject, which sets bounds to God’s production; this is the consequence of the original imperfection of creatures.  [Theodicy §388, GP VI 345–346]

There is reason to think Kant is familiar with and endorses this line of thought. His earliest statement of causal realism explicitly acknowledges its debt to Leibniz, and it exhibits good knowledge of the Theodicy (True Estimation, Ak. 1:17–23; cf. R 3750; Ak. 17:236). Kant’s earliest works also describe attractive and repulsive forces as essential to matter, and this requires that both are operative at the instant of creation (Universal Natural History, Ak. 1:230). He also explicitly upholds a division of creation into aspects whose dependence on God is “through the mediation of the order of nature” and those “independent of that order.” The existence or alteration of anything in the first class is described as “sufficiently grounded in the forces of nature,” meaning that these are its “efficient cause” (Only Possible Proof, Ak. 2:103–104). In addition, Kant’s mature philosophy makes essential use of the idea of non-­temporal “priority of nature.” The doctrine of the ideality of time together with an enduring theistic framework demands appeal to non-­ temporal causal orderings. Late discussions of divine providence, for instance, criticize German terms for such divine activity (Vorsehung, Vorsorge) as “misleadingly infected” with connotations of time (Ak. 28:1110). Kant accepts throughout his life the traditional theological view that God’s creative and conserving action cannot be delegated; he thus rejects the idea that

15  OCM XI 160 (trans. Malebranche 1993, 160).

272 Kant ’ s Theory of Divine and Secondary Causation finite substances could depend merely mediately on God.16 He does not mean to deny that creatures depend on each other in various ways. Medieval and modern debates on conservation distinguish ways in which creatures’ existence can depend on each other—for protection, sustenance, omission of destructive acts, and so forth—from the kind of “bottom-­ up” conservation seen as God’s prerogative alone. Suarez describes this latter activity as a “persisting influx or inpouring of the very esse communicated through the [creature’s] production.”17 While Kant thus holds that the immediate ground of substance in divine conservation is God alone, his early works consistently describe substantial states or accidents as having immediate “real grounds” in the finite substances themselves: “The substrate of substances [das substantiale] contains the first real ground of all inhering accidents” (Metaphysik Herder, Ak. 28:25; cf. True Estimation, Ak. 1:17–21; Universal Natural History, Ak. 1:230; Nova Dilucidatio Ak. 1:415; Physical Monadology, 1:480). In this respect his early thought follows self-­consciously in the metaphysical tradition of Leibniz’s pluralistic dynamism.

2  Kant’s Early Theory of Divine Action: Conservation or Concurrence? Kant thus thinks it possible to resist Malebranche’s inference that God as sole giver of being in creation and conservation must therefore also be the sole source of “the ways of being.” His early works subscribe instead to a version of Leibniz’s view that God causes the existence of substances, while these substances are “real grounds” of their states. Though Kant’s early model of the relation of divine and secondary causation acknowledges its Leibnizian debt, it would be rash to identify their positions. For one thing, Kant consistently rejects the preestablished harmony in favor of interactionism.18 What is far less clear is whether he ever accepts Leibniz’s position that states or accidents of substances in the ordinary course of nature depend immediately on God. As noted, this is a key point dividing proponents of general concurrence from mere conservationists. Leibniz rejects the view of Durandus that “God creates substances and gives them the force they need; and thereafter he leaves them to themselves, and does nothing but conserve them, without aiding them in their actions.” He insists instead that “the perpetual immediate influence which the dependence of creatures demands . . . 16 “God doesn’t concur in the existence of [finite substances]; since these substances contribute nothing to their own continued existence, thus cannot contribute as co-­ causes to their own conservation; for otherwise they would be independent of him” (Religionsphilosophie Volckmann, Ak. 28:1208). Cf. Theodicy §249, GP VI 265. 17  Suarez (2002, 21.3.2); cf. Meier (1757, §1024). For discussion, see Freddoso (1991, 564). 18 “An aggregate of substances doesn’t yet amount to a world, rather, the causal interaction (commercium) of substances first produces a world” (Ak. 28:212). See True Estimation, Ak. 1:22–23; Nova Dilucidatio (1755), Ak. 1:410–412; Inaugural Dissertation (1770), Ak. 2:407–408.

Desmond Hogan  273 attaches not only to the substance but also to the action” (Theodicy §27, GP VI 119; Causa Dei, GP VI 440). In rejecting conservationism, Leibniz notes that the view “apparently met with disapproval in the writings of Pelagius” (Theodicy §27, GP VI 118). He seems to sympathize with a traditional theological motivation for concurrence according to which the absoluteness of divine sovereignty favors a theory viewing God as immediately involved in all aspects of the world. Suarez describes as his “best” argument for concurrence the idea that “this manner of acting in and with all agents pertains to the breadth of the divine power” (Suarez  2002, 22.1.13). As Freddoso explains, the central thought here is simply that “theistic naturalists should be antecedently disposed to countenance in nature the maximal degree of divine activity compatible with the thesis that there is genuine secondary caus­ ation” (Freddoso  1991, 577). Leibniz’s language in staking out a middle ground between occasionalism and conservationism in Discourse on Metaphysics §8 suggests sympathy for this traditional motivation. He writes that “it is rather difficult to distinguish the actions of God from those of creatures; for there are some who believe that God does everything; others imagine (d’autres s’imaginent) that he does nothing but conserve the force that he has given to creatures” (GP IV 432). While Kant’s later works explicitly reject the doctrine of concurrence on grounds yet to be examined, it is not so clear where his early thought stands with respect to it. He is perhaps most plausibly read as sympathetic to the doctrine if we accept Robert Sleigh’s interpretation of Leibniz’s model of divine concurrence. On this interpretation, Leibniz views concurrence as a kind of divided effort in which “God produces whatever there is of perfection in the states of creatures; creatures produce whatever there is of limitation in their own states” (Sleigh 1990, 185). The model is suggested by a number of texts discussing the “author of sin” problem. The problem is how God as sovereign creator, conserver, and providential lord of creation can avoid authorship of, and thus responsibility for, creatures’ sins. Leibniz’s response appeals to a favorite analogy of a boat whose motion results from the combination of its own inertia and the river’s current. The inertia is meant to correspond to the creature’s negative contribution to its states by virtue of its “natural imperfection,” the current to God’s production of the perfection or reality in those states. Leibniz explains that “the current is the cause of the boat’s movement, but not of its retardation; and [by analogy] God is the cause of perfection in the nature and the actions of the creature, but the limitation of receptivity of the creature is the cause of the defects in its action” (Theodicy §30, GP VI 120).19 On this model, God produces what is positive or 19 “[J]e me suis servi dans les Essais de l’exemple d’un bateau chargé, que le courant emporte d’autant plus tard que le bateau est plus chargé. On y voit clairement que le courant est cause de ce qui est positif dans ce mouvement, de la perfection, de la force, de la vitesse du bateau, mais que la charge est cause de la restriction de cette force et qu’elle produit la tardité” (Mémoires de Trévoux, July 1712, GP VI 348; cf. GP VI 450).

274 Kant ’ s Theory of Divine and Secondary Causation real in creatures, while sins and imperfections flow from the “original limitation” of creatures’ natures and are imputable to them rather than to God. The model treats states of finite substances as well as the substances themselves as immediately dependent on God. Leibniz writes that, “in acting [as well as existing], all things depend on God, since God concurs in the actions of things insofar as these have some degree of perfection, which at least must come from God” (Causa Dei, GP VI 440). Kant’s early though not mature works accept a similar solution to the problem of the author of sin, and this might be taken as indicating early sympathy for concurrentism. Writings of the 1750s follow Leibniz in tracing moral evil to intellectual limits of creatures leading them to favor the worse above the better. Sinful actions are described as “infallibly” ordained by creatures’ natures. Imputability of actions is nevertheless upheld on the grounds that acting freely requires only doing what one desires “with consciousness” (Nova Dilucidatio, Ak. 1:403). Kant adds that the imperfection of creatures entails no imperfection on God’s part, because “God’s creative act is limited according to the nature of the limited being to be produced” [Limitata enim est actio Dei creatrix, pro ratione entis limitati producendi] (Nova Dilucidatio, Ak. 1:406). Given Kant’s consistent rejection of occasionalism, it may be tempting to infer that his early proximity to Leibniz on the authorship of sin indicates sympathy for divine concurrence interpreted along the lines proposed by Sleigh. I think however that such an inference should be resisted. For one thing, Kant’s early writings never explicitly assert that states of creatures in the ordinary course of nature depend immediately on God. His formulations appear designed to express a view of creatures’ forces as sole immediate causes of their states. Furthermore, it is not clear that he understands Leibniz’s model of divine concurrence along Sleigh’s proposed lines. The interpretation of divine concurrence upheld by followers of Leibniz and later attacked by Kant looks different, as we will see. Adams has suggested that passages in which Leibniz describes creatures as mere founts of imperfection and God as the source of all that is real in their states seem to be focused more on issues of theodicy than the general metaphysics of divine and secondary causation. For other Leibnizian texts suggest different approaches to concurrence, and as Adams points out, Sleigh’s general model seems difficult to square with Leibniz’s view that creatures as well as God possess active as well as passive forces.20 Discussions of transeunt causation in Kant’s early work offer further support for the conclusion that he belongs in the conservationist camp from the

20  See Adams (1994, 94–98). Sleigh concedes that “there appears to be a tension between what we might call Leibniz’s metaphysical exposition of creaturely action and what we might call Leibniz’s theo­logic­al exposition of creaturely action. On the deism-­occasionalism scale, the former seems to be located near the deism end, the latter closer to the occasionalism end” (Sleigh 1990, 185).

Desmond Hogan  275 beginning. Relevant discussions set out from a key premise that the existence of finite substances does not yet entail cosmological relations among them—“finite substances stand by virtue of their mere existence in no relation to each other” (Nova Dilucidatio, Ak. 1:412).21 This is meant as a rejection of Leibniz’s view that substances can belong to one world simply by existing with suitably harmonious intrinsic states.22 Kant claims instead that real causal relations are required for genuine cosmological unity, while dismissing what he labels a “vulgar” interactionism on which the mere existence of finite substances suffices for the existence of these causal relations. His proposal holds that the possibility of causal bonds underwriting cosmological connection rests on a special divine act. He concludes that God could connect finite substances in discrete groups, giving rise to discrete actual worlds.23 The notable point here is his consistent classification of this divine connecting act with God’s conserving activity: The schema of the divine understanding, the origin of existences, is an enduring act (it is called conservation [conservatio]); and in that act [NB], if any substances are conceived by God as existing in isolation and without any relational determinations, no connection between them and no reciprocal relation would come into being. If, however, they are conceived as related in God’s intelligence, their determinations would subsequently always relate to each other for as long as they continued to exist in conformity with this idea. That is to say, they would act and react.  [Nova Dilucidatio, Ak. 1:414]

Kant’s descriptions of this connecting act sometimes have a voluntarist ring—he once says that it is “obviously arbitrary on God’s part and can therefore be omitted or not omitted at his pleasure” (Nova Dilucidatio, Ak. 1:414). Langton infers that his early doctrine of cosmological connection treats intrinsic facts about finite substances as “not constrain[ing] relational facts in any way” (Langton 2001, 119). The claim that the mere existence of substances does not entail their cosmological connection is not however the claim that intrinsic states of such

21 “It is not necessary to [a substance’s] existence that it should stand in connection with other things” (True Estimation, Ak. 1:22); “Composition is only a relation for these [substances] and therefore only a contingent property, which can be removed without affecting their existence” (Physical Monadology, Ak. 1:477). For recent discussions see Langton (2001, 115–121); Watkins (2005, 141–160). 22  “An aggregate of substances doesn’t yet amount to a world, rather, the causal interaction (commercium) of substances first produces a world” (Ak. 28:212). Cf. True Estimation, Ak. 1:22–23; Nova Dilucidatio (1755), Ak. 1:410–412; Inaugural Dissertation (1770), Ak. 2:407–408. 23  “The possibility that, had it pleased God, there might be a number of worlds, even in the metaphysical sense, is not absurd” (Nova Dilucidatio, Ak.1:414); “It is really possible, even taken in a strict metaphysical sense, that God has created many millions of worlds; thus it is undecided whether they really exist or not” (True Estimation, Ak. 1:22). The claim is distinct from Newton’s claim in the Opticks that “God is able . . . to vary the laws of nature and make worlds of several sorts in several parts of the universe” (Newton, 1952 [1730], 31).

276 Kant ’ s Theory of Divine and Secondary Causation substances do not constrain their relations if connected. In Langton’s terms, Kant does reject the proposition that if a certain causal relation R holds between finite substances then these substances instantiate certain intrinsic properties, such that necessarily, if they exist with those properties, R must hold between them. This is a weaker result than Langton supposes, however, since it is compatible with the proposition that substances if connected must be connected in one way only—by means of a certain law, say, or one of a fixed number of laws. That is, Kant’s early endorsement of a divine libertas contradictionis with regard to cosmological relations, God’s freedom to create or omit them, is not an endorsement of an unconstrained libertas contrarietatis in this regard—a complete lack of constraint of things’ relations by their intrinsic properties.24 This becomes especially clear in Kant’s doctrine that harmony of the intrinsic forces of finite substances is required for their connection to be possible (True Estimation, Ak. 1:25; Metaphysik Herder, Ak.28:51–53).25 In contrast to a picture on which causal relations float entirely free of intrinsic properties, his early model of cosmological connection has more in common with a jigsaw puzzle. Though the puzzle might be left undone, intrinsic features of the pieces constrain how they could be connected.26 The significance of this for Kant’s early position on conservationism and concurrence is straightforward. His early metaphysics ascribes an essential role to intrinsic forces of finite substances in making their connection possible, but he never takes the opportunity to describe this as a creaturely concurrence with God’s activity in connecting them. Kant’s language is suggestive of the conservationist’s clean division of labor between God and creature. He locates the possibility of transeunt causation in a voluntary divine representation of interdependence described as belonging to God’s act of conservation. He then speaks of particular transeunt effects in created substances which themselves “act and react” (Ak. 1:414). This suggests a view of God as ground of the possibility of transeunt influence by virtue of the conserving/connecting act, and of creatures as sole causes of actual transeunt effects. Such an interpretation of

24  Langton interprets Kant’s mature doctrine of the unknowability of the thing in itself as asserting the inscrutability of intrinsic states of things affecting us. This inscrutability is derived from Kant’s supposed view that relations of substances are unconstrained by their intrinsic properties. 25 Kant claims that physical influx between substances presupposes a harmony of the laws in accordance with which they could act and “receive effects” from without. The doctrine is discussed in Ameriks (1992, 262); Watkins (2005, 153–154); Hogan (2005, 21–22, 59); Insole (2013, 50–52). There is a potential conceptual difficulty in Kant’s earliest explanation of intrinsic harmonizing forces as a function of distance at which substances act, since he describes space as first given by their connection. See Böhme (1914, 9–11). 26 The analogy is imperfect, since there is an attractive middle ground between Langton’s interpretation on which intrinsic features exert no constraint on substances’ relations and a jigsaw model treating these features as removing most or all leeway regarding connection. Kant’s doctrine of cosmological connection subsequently backs away from a voluntarist element, by suggesting that God must connect substances with the right intrinsic features—though this connection is still said to require interaction (Inaugural Dissertation, Ak. 2:409).

Desmond Hogan  277 Kant’s early position agrees with the Dissertation’s description of God’s act of cosmological connection: The connection which constitutes the essential form of a world is seen as the principle of the possible influences of the substances which constitute it. For actual influences do not belong to the essence but to the state, and the transeunt forces [of creatures], which are the causes of [actual] influences, suppose some principle [i.e. God’s act] by which it may be possible that the states of several things, the subsistence of each of which is nonetheless independent of the others, should be mutually related to one another as states determined by a ground. [Inaugural Dissertation, Ak. 2:390]

While a conservationist position is again strongly suggested here, it is explicitly affirmed in later works. A closer examination of the motivations and mechanics of Kant’s mere conservationism in the Critical setting will provide some further evidence for a reading of his early view as conservationist. Before turning to this, it is necessary to say something about his early position on immediate divine action outside of the ordinary course of nature.

3  Supernatural Action in Kant’s Early Philosophy The conclusion that Kant’s early thought embraces a consistent conservationism might seek to draw further support from his defense in the Universal Natural History and Theory of the Heavens (1755) of a mechanistic cosmogony, one which seeks to derive the large-­scale structure of the cosmos “from the simplest state of nature through mechanical laws alone” (Ak. 1.234). The slogan of this early work is, “Give me only matter and I will build you a world out of it” (Ak. 1:230). It sets out from the hypothesis of a primordial distribution of matter endowed with essential forces of attraction and repulsion in an infinite space. Kant does not extend his attempted explanation of natural order to organisms, rather asserting that we are incapable of understanding these on mechanical grounds (Ak. 1:230). Neither does he show any sympathy for materialism, insisting on the “infinite distance” between mind and matter (Ak. 1:355).27 The work does however present the operation of both minds and matter as both natural and determined. Mechanistic evolutionism is juxtaposed with Leibniz’s spiritual determinism in the claim that “the same unlimited fruitfulness of nature has brought forth the inhabited celestial globes as well as the comets, the useful mountains and the harmful crags, the habitable lands and empty deserts, the virtue and the vice” (Ak. 1:347).

27  This point is rightly emphasized by Ameriks (1982a, 88–89); cf. Schönfeld (2000, 119).

278 Kant ’ s Theory of Divine and Secondary Causation The work’s central project is to advance mechanism beyond Newton’s conclusion that celestial motions require periodic “reformation” through immediate divine action.28 Another aim is to argue that theism is strengthened rather than undermined by acceptance of a fully mechanistic cosmology in which natural forces produce all regularity and order in the universe (Ak. 1:331). Kant rejects the idea that miracles are necessary for cosmological order as suggesting a conception of God as mere “architect” of a world wrestling with recalcitrant preexisting matter—a “refined atheism [feineren Atheismus]” his early and mature writings associate with Greek antiquity (Only Possible Proof, Ak. 2:122; Danziger Rationaltheologie, Ak. 28:1308). He complains that on such a conception, God is, “while great, yet not infinite, while powerful, yet not all-­sufficient” (Universal Natural History, Ak.1:223, 333). The proposal that created natures rather than “God’s immediate hand” effect harmony and order is described as supporting rather than undermining theism and the doctrine of creation ex nihilo. The central argument here is that acknowledgment of nature’s capacity to produce all order in the world supposedly directs us to the ontological dependence of essence itself on God (Ak. 1:333, 336).29 While the Universal Natural History’s rejection of miraculous interventions may appear suggestive of conservationism, caution is required. Contemporary concurrentists including Leibniz and Wolff also tightly limit miraculous interventions, distinguishing God’s general concurrence as an immediate action that is an assumed prerequisite of all natural order and special interventions which violate this order.30 The claim that nature’s ability to produce order testifies to a divine ground of possibility remains prominent in Kant’s writings of the 1760s, but his position on divine action in nature undergoes a shift of considerable importance for his mature theory of divine causation. His Only Possible Proof of 1763 develops a theistic proof from the idea that “it is only because God exists that anything else is 28  Newton describes it as “unphilosophical to pretend that [orbital motion] might arise out of a Chaos by the mere Laws of Nature.” He proposes that God “being in all Places, is more able by his Will to move the Bodies within his boundless uniform Sensorium, and thereby to form and reform the Parts of our own Bodies” (Newton 1952 [1730], 402). 29  “How would it be possible for things of different natures to produce such excellent harmonies and beauty in connection with each other . . . for the benefit of things which in are in a sense outside the realm of dead matter, namely the uses of men and animals, if they did not have a common origin, an infinite intellect in which all things were designed in respect of essential properties?” (Universal Natural History, 1:225; cf. Only Possible Proof, Ak. 2:112). Kant however rejects creation of the eternal truths in favor of Leibniz’s view that essences are grounded in God’s understanding. Here I pass over his proposal that the ontological dependence of essence on God should help ensure that miracles are rarely necessary for cosmological order. 30  Discourse on Metaphysics §16 offers a theory of miracles as not exceeding the forces encoded in substantial forms of created things: “God’s extraordinary [i.e. miraculous] concourse is included in that which our essence expresses, for this expression extends to everything.” These miracles “are always in conformity with the universal law of the general order, even though they may be above the subordinate maxims” (GP IV 441–442). For discussion of divine actions excluded from this sense of “miracle,” see Adams (1994, ch. 2).

Desmond Hogan  279 possible at all” (Ak. 2:112).31 The work includes a summary of the earlier Universal Natural History, whose print run had been mostly destroyed. The later discussion extends Kant’s critique of supernatural explanations, now described as impeding human understanding by “imposing a reverential silence upon reason in its enquiries” (Ak. 2:122). What is generally overlooked is that this methodological critique is accompanied by a new claim that miraculous interventions are probably required for the purposes of providential control. Kant explains as follows: The alterations which occur in the world are either necessary, and necessary in virtue of the initial order of the universe and of the laws of nature, both general and particular—and everything which takes place mechanically in the corporeal world is of this character—or, alternatively, these same alterations possess, notwithstanding, an inadequately understood contingency—a case in point being the actions which issue from freedom and of which the nature is not properly understood. Changes in the world of this latter kind, in so far as they appear to have about them an indeterminacy in respect of determining grounds and necessary laws, harbor within themselves a possibility of deviating from the general tendency of natural things towards perfection. And, for this reason, it can be expected that supplementary supernatural interventions may be necessary, for it is possible that the course of nature, looked at in this light may on occasion run contrary to the will of God. [Only Possible Proof, Ak. 2:210–211]

As the passage indicates, the argument rests on an underlying shift to a libertarian theory of agency. Many texts testify to Kant’s conversion by the mid-­1760s to a robustly libertarian metaphysics.32 He now describes the non-­ determined character of human agency as grounds to assume that miracles will be required for God’s perfect providential control. In one text from this period, he bluntly asserts that harmony of free acts with providential purposes “will not be fully possible through the order of nature”—thus “one must admit” some extraordinary direction (Reflections on Religion 1760–1764, R 8081, Ak. 19:617).33 It is significant that Kant continues to assert even into the Critical period that providential control may require supplements to nature:

31  The argument is already sketched in the 1755 Nova Dilucidatio (Ak. 1:395), and is obviously influenced by Leibniz’s claim that God contains not only the source of existences, “but also of essences in so far as they are real, that is, [God is] the source of what is real in the possible” (Mon 43; GP VI 614). 32  For a textual basis for this shift, see Hogan (2009, 517–518). 33  Kant is probably influenced by his libertarian contemporary Crusius, who offers the same argument: “Since there are in the world infinitely many free acts which on account of their freedom do not follow in accordance with any rule of physical predetermination, it cannot be understood how the mechanical edifice of the world which must by its nature always respect the same rules could be set up so that it would agree everywhere with all of these free acts and divine intentions, without further spontaneous directing on God’s part” (Crusius 1745, §341).

280 Kant ’ s Theory of Divine and Secondary Causation There is nothing at all impossible in the thesis that in a best world natural forces should from time to time require immediate divine cooperation in order to achieve certain important ends . . . Who dares to be so presumptuous as to claim to cognize the possibility that God can achieve everything He has planned for the world through general laws without any extraordinary direction? . . . Such exceptions from the rules of nature may be necessary on the grounds that God wouldn’t otherwise be able to achieve many important ends via the normal course of nature. We must however be careful about desiring to determine whether such extraordinary intervention occurred in this or that instance. [Religionslehre Pölitz, Ak. 28:1112; cf. Religion, Ak. 6:85 n.; Religionsphilosophie Volckmann, Ak. 28:1215]

Kant’s methodological prohibition on supernatural explanations is thus ­supplemented from the 1760s with a metaphysical thesis that full providential control may well demand miraculous supplements to the ordinary course of nature. Subsequent criticisms of supernatural explanations are carefully ­formulated to agree with the metaphysical thesis.34 This finding can help resolve a long-­standing puzzle regarding his earliest motives for transcendental idealism. Textual evidence strongly indicates that the antinomy of freedom and empirical determinism was a major motive in the doctrine’s emergence. But this conclusion has also faced a major hurdle stemming from the causal principle’s status in the 1770 Inaugural Dissertation, the work in which the ideality of space and time is first proposed. The Dissertation describes the principle that “all things in the ­universe in happen in accordance with the order of nature” as a merely regulative principle of philosophical enquiry (Dissertation §30, Ak. 2:418). Commentators have been understandably inclined to conclude that Kant does not yet claim a priori knowledge of the causal principle in the empirical realm.35 The puzzle is then how an antinomy of freedom and determinism could have motivated the emergence of his idealism, as the broader evidence strongly indicates.36

34  “It is a well-­known rule of philosophers or rather of common sense in general that nothing is to be regarded as a miracle or supernatural event unless there are weighty reasons for doing so. This rule implies firstly that miracles are rare; and secondly, that the whole perfection of the universe can in conformity with the will of God and in accordance with the laws of nature be attained without many supernatural influences” (Only Possible Proof, Ak. 2:108). Discussions of Kant on miracles often fail to distinguish his methodological and metaphysical theses regarding such interventions. Schönfeld’s long discussion, for example, asserts that Kant’s view has “sharpened into a criticism of miracles” by the 1760s (Schönfeld 2000, 111). 35  See e.g. the discussion of the passage in Guyer (1987, 19–20). 36  According to a famous note, the antinomies first led Kant to his mature idealism: “I saw this system as though in a twilight. I tried in all seriousness to prove propositions and their opposites, not in order to construct a skeptical philosophy, but because I suspected an illusion of the understanding—to discover, wherein it was concealed. The year ’69 gave me great light” (Ak. 18:69, cf. Kant’s 1798 letter to Garve, Ak. 12:257; also R 6353; R 6344, R 6349; Nachlass, Ak. 20:235).

Desmond Hogan  281 A simple solution presents itself in light of the discussion above. The Dissertation’s regulative principle of philosophical enquiry is followed with a warning that we cannot demonstrate “the impossibility or the very slight hypothetical possibility of supernatural events,” but should remember that “hasty appeal to supernatural events is the cushion of a lazy understanding” (Ak. 2:418). This is plainly a reiteration of Kant’s methodological prohibition on supernatural explanation. That prohibition does not however compete in his mature philosophy with the KrV’s doctrine that determinism is constitutive of empirical order, or with his metaphysical thesis that some divine supplements to nature may be necessary to maintain providential control. The Dissertation’s merely regulative principle of philosophical enquiry can thus be taken simply as a reminder of Kant’s warning that we could never know miraculous interventions to have occurred in “this or that instance.” Since this does not conflict with the claim that appearances are known as deterministically ordered, it becomes possible to interpret him as already committed to this doctrine in the Dissertation itself. Indeed, the work explicitly associates knowledge of causal laws with a priori representations of space and time. Kant asserts that “pure mathematics deals with space in geometry and time in pure mechanics,” and that “all observable events in the world, all motions and all internal changes necessarily accord with axioms which can be known about time” (Ak. 2:397, 402). Broader evidence indicates that the principles here associated with a priori intuition are already viewed as imposing deterministic order in nature.37

4  Mere Conservationism in the Critical Philosophy Readers of the first Critique have often regarded the work as limiting all discussion of causality to the empirical realm, thus as rejecting metaphysical theorizing regarding God’s relation to secondary causes. Historically, this reading has been 37  By the late 1760s Kant maintains that the Principle of Sufficient Reason holds throughout the empirical realm (R 4007, 4012, 4172, 4174, 4225). He also follows Leibniz and Wolff in associating a priori cognizable laws of motion with physical determinism: “Neither through a miracle nor through a spiritual being can a motion be brought about in the world without producing just as much motion in the opposite direction, thus in accordance with the laws of action and reaction . . . Motions cannot begin by themselves, nor through something which wasn’t itself in motion; and freedom is not to be met with among the phenomena, neither are occasional miracles” (R 5997, 1780s). For discussion of Kant’s taxonomy of miracles, see Chignell (Ch. 12 of this volume). Assuming that a priori forms of experience entail a priori principles of experience, including some a priori constraints on matter, miracles that do not suspend these forms must be consistent with such principles. It is less clear how the miracles Kant says may be required for providential control relate to laws not knowable a priori. Chignell sees him as allowing miraculous violations of empirical laws. Since, however, such violations would presumably “be met with among the phenomena,” R 5997 favors a model of miraculous interventions as analogous to free action—viz. as involving noumenal difference making that appears as woven into a fully law-­governed empirical order. It is most likely that Kant never arrived at a clear final position on this issue.

282 Kant ’ s Theory of Divine and Secondary Causation heavily influenced by the work’s doctrine that theoretical knowledge of causal relations rests on categories of cause-­ effect and community “containing the ground of the possibility of all experience in general from the side of the understanding” (B 167–168). In fact, Kant’s restriction of theoretical causal knowledge to a transcendentally ideal domain of experience in which the categories play a constitutive role is not a rejection of supersensible causation. He maintains that the “unschematized” category of causation—the category in abstraction from conditions of its empirical application—allows for supersensible application, and such application is central to his theory of freedom, doctrine of noumenal affection, and moral theology (A 88/B 120, 166–167; A 254/B 309; KpV Ak. 5:55–57, 134–141).38 This point notwithstanding, the KrV’s epistemic strictures and its attack on traditional theistic proofs have been widely understood as implying that the Critical philosophy simply dismisses debates—of such importance to Descartes, Malebranche, Leibniz, and Berkeley—on the metaphysics of God’s relation to secondary causes. Perhaps the most efficient way to dispel this common misunderstanding is by turning to the positive theory of divine and secondary causation proposed and defended in Kant’s Critical writings. The KrV describes as one of its central aims the “destruction of the roots of materialism, fatalism and atheism” (B xxxiv). Kant explains in a 1790 letter that his philosophical efforts, “so far concerned with critique, are by no means intended to work against the Leibniz-­ Wolffian philosophy  .  .  .  but are rather intended to lead this philosophy through a roundabout route . . . to the same end, but only through the combination of theoretical philosophy with the practical—an intention that will become clearer if I live long enough to put metaphysics into a coherent system, as I plan to” (Correspondence, Ak. 11:186). A late summary of his thought argues that “the ultimate purpose, to which the whole of metaphysics is directed” lies in addressing the questions of the existence of God, the freedom of the will and the immortality of the soul (Progress, Ak. 20:260). Kant’s moral arguments serve this purpose, and the “practico-­dogmatic metaphysics” they ground is held to vindicate “reason’s proper claim to knowledge [Erkenntniß] of the supersensible” on combined practical and theoretical grounds (Progress, Ak. 20:310).39 Writings and lectures from the Critical period remain heavily engaged with the rationalist theology of the German schools. In particular, Kant continues to debate the metaphysics of 38  See esp. Erdmann (1878, 44–47, 73–74); Adickes (1924b, 4–12, 28–37, 50–51); cf. Van Cleve (1999, 137); Adams (1997b, 820–821); Hogan (2009, 501–532). 39  “It was not until the moral laws unveiled the supersensible in man, namely freedom . . . that reason made proper claim to knowledge of the supersensible [gerechten Anspruch auf Erkenntniß des Übersinnlichen], though only when confined to its use in the latter [practical] capacity” (Progress, Ak. 20:310); “Transcendental philosophy, the doctrine of the possibility of a priori knowledge as such . . . has as its purpose the founding of a metaphysics envisaging as an aim of pure reason the extension of the latter from the limits of the sensible to the field of the supersensible” (Progress, Ak. 20:272–273; cf. B xxvi n.; Practical Reason, Ak. 5:121–148; Judgment, Ak. 5:469).

Desmond Hogan  283 divine and secondary causation under Baumgarten’s headings of creation, conservation, concurrence, and omnipresence. As we will see, he also commits himself to a positive account of such causation. The philosophical theology of the Critical period is accurately summarized by Wood as the product of a mind “fundamentally unable to conceive of the human situation except theistically, and unable to conceive of God in any terms except those of the scholastic-­rationalist tradition.”40 The question to be addressed in the remainder of this discussion is how the Critical philosophy can insist upon a positive account of God’s relation to secondary causes in light of Critical epistemological strictures. Kant’s reasoning, which sets out from and so presupposes the general theistic conclusion of his moral argument, is best examined on an issue-­by-­issue basis. It is striking that his continued rejection in the Critical period of occasionalist models of divine and secondary causation appeals in part to the very epistemological strictures sometimes assumed to exclude all transcendent theorizing. The Critique of Judgment describes occasionalism in the terms the pre-­Critical philosophy uses in rejecting explanatory appeals to miracles as “imposing a reverential silence upon reason in its enquiries” (Ak. 2:122). Kant presents occasionalist explanations as ad hoc, a charge he also levels against Cartesian and Crusian theories of a priori knowledge resting on supposed divine illumination: “If we assume occasionalism for the production of organized beings, then all nature in this production is lost entirely . . . We may therefore assume that anyone who is at all concerned to do philosophy will not adopt this system” (Judgment, Ak. 5:422).41 Now it is one thing to reject a priori justifications resting on purported divine illumination, and another to rule out a positive noumenal model of causation. We can nevertheless see a straightforward sense in which the rejection of occasionalism and divine illumination expresses the one preference for broadly naturalist explanation, where this naturalism is held to exclude appeals to occasionalism as a theory of the relation of noumenal and empirical reality. Practical grounds also play an important part in the opposition to occasionalism. Kant obviously accepts Leibniz’s view that creaturely agency presupposes the

40 Wood (1978, 17). An important debate in the early twentieth century asked whether Kant’s mature theology requires a “conservative-­metaphysical” or admits of a “fictionalist” reading. Vaihinger admits that Kant affirms the transcendent existence of God as the metaphysical reading asserts, but sees a tension with another account of God as mere rational fiction subserving theoretical and prac­ tical ends (Vaihinger 1927a). Adickes’s patient analysis of Vaihinger’s textual case reaches the verdict that there is “not one single passage which can and should be interpreted fictionally [i.e. as incompatible with the metaphysical reading] when examined systematically in narrower or wider context, as duty clearly demands” (Adickes 1927, 290). 41  Kant’s grounds for rejecting occasionalism in the production of organisms evidently generalize. Cf. his rejection of Crusius’s “preformation” theory of a priori knowledge at Ak.10:130 (1772); A 92/B 125–126; Ak. 4:319; R 4473.

284 Kant ’ s Theory of Divine and Secondary Causation rejection of occasionalism and the reality of secondary causation.42 His moral philosophy makes heavy appeal to powers and forces, struggles, and obstacles. Virtue itself is defined as “moral disposition in the struggle [Kampfe]” to obey the law; it is elsewhere described as “the deployment of one’s forces [Kräfte] in the observance of duty” (Practical Reason, Ak. 5:84; Religion, Ak. 6:201; Metaphysics of Morals, Ak. 6:484). Kant’s mature philosophical theology is intended as a moral theology, providing what he evidently sees as powerful grounds to reduce live options for God’s relation to secondary causes to versions of conservationism and concurrence.43 Kant’s rejection of divine concurrence in favor of conservationism in Critical writings provides further instructive illustration of his combining of epistemological, practical and conceptual arguments to justify a positive metaphysics of ­supersensible causation even against the backdrop of Critical strictures. I will focus in what follows on his arguments against Leibniz and German followers on the issue of divine concurrence. Several objections to concurrence repeated often in texts from the Critical period are summarized in a footnote to the late Perpetual Peace essay. Kant writes: As for the concept, current in the schools, of a divine intervention or col­lab­or­ ation (concursus) toward an effect in the sensible world, this must be given up. For to want to pair what is disparate (gryphes iungere equis) and to let what is itself the complete cause of alterations in the world supplement its own predetermining providence (which must therefore have been inadequate) during the course of the world, is first, self-­contradictory. For example, to say that, next to God, the physician cured the illness, and was thus his assistant in it, is in the first place self-­contradictory. For causa solitaria non iuvat [a solitary cause does not assist]. God is the author of the physician together with all his medicines and so the effect must be ascribed entirely to him, if one wants to ascend all the way to that highest original ground, theoretically incomprehensible to us. Or one can ascribe it entirely to the physician, insofar as we follow up this event as belonging to the order of nature and as explicable in terms of the order of nature, within

42  Leibniz holds that Malebranche’s occasionalism leads directly to Spinozism; it makes “created things disappear into mere modifications of the one divine substance, since that which does not act . . . can in no way be substance” (On Nature Itself, GP IV 515, AG 166–167). Kant also associates causal activity and nature, defined as “the connection of the determinations of a thing according to an inner principle of causality” (A 419–420/B 466–467). Cf. “[I]f we assume occasionalism for the production of organized beings, then all nature in this production is lost entirely” (Critique of Judgment, Ak. 5:422). 43  There remains room, of course, for a Cartesian-­style multiplication of occasionalist and realist causal models in body–body, mind–body (etc.) cases. Kant’s pre-­Critical philosophy stands in the tradition of Leibniz’s rehabilitation of substantial forms in upholding metaphysical continuity between spiritual and physical causation. We will see that the Critical philosophy does propose different models of God’s causal relation to free and determined causes, though both models are ­ anti-­occasionalist.

Desmond Hogan  285 the chain of causes in the world. Second, such a way of thinking [concurrentism] also does away with all determinate principles for appraising an effect. But from a morally practical point of view (which is thus directed entirely to the supersensible), as, e.g., in the belief that God, by means incomprehensible to us, will make up for the lack of our own righteousness if only our disposition is genuine, so that we should never slacken in our striving toward the good, the concept of a divine concursus is quite appropriate and even necessary; but it is self-­evident that no one must attempt to explain a good action (as an event in the world) through this concursus, which is a futile theoretical cognition of the supersensible and is therefore absurd.  [Perpetual Peace, Ak. 8:361 n.]

In speaking of concurrentism as “current in the schools,” Kant does not refer to Aquinas or Suarez but to proponents of the theory in the German rationalist tradition including Leibniz, Wolff, Baumgarten, and Meier.44 One objection above, that concurrence “does away with all determinate principles for appraising an effect,” reprises the familiar methodological rejection of supernatural explanation. Suarez and Leibniz describe divine concurrence as necessary for production of every effect in nature and can therefore respond that parity of reasoning should lead to the rejection of creation and conservation. It is important to emphasize that Kant refuses to take such a step; he continues to affirm creation and conservation as essential components of his practical-­dogmatic theism. The Critique of Practical Reason asserts that the proposition that “God as universal original being is the cause also of the existence of substance [is] a proposition that can never be given up without also giving up the concept of God as the being of all beings and with it his all-­sufficiency, on which everything in theology depends” (Ak. 5:101–102).45 Kant’s continued commitment to the doctrine of creation in the Critical period calls for an account of the creative act consistent with his Critical metaphysics. His official doctrine is that creation is non-­temporal and directed at underlying noumena rather than phenomena. He writes, “if existence in time is only a sensible way of representing things which belongs to thinking beings in the world and consequently does not apply to them as things in themselves, then the creation of these beings is a creation of things in themselves, since the concept of a creation does not belong to the sensible way of representing existence or causality but can only be referred to noumena” (Practical Reason, Ak. 5:102). Kant rejects Baumgarten’s description of creation and conservation as a causal “influx,” reserving this term for production of properties rather than substrates (Religionslehre Pölitz, Ak. 28:1107). What is most striking in mature discussions 44 See Theodicy §§ 27–32, GP VI 118–121; cf. Wolff (1736, §875); Baumgarten (1757, §§ 954–959); Meier (1757, §§1027–1028). 45 Cf. Danziger Rationaltheologie (1783–1784), Ak. 28:1104, 1299; R 6173.

286 Kant ’ s Theory of Divine and Secondary Causation of creation and conservation is how little they differ from pre-­Critical writings. The concept of God is of course now described as “merely thinkable” from the perspective of theoretical reason alone—it is “assertorically declared to have a real object [only] because practical reason indispensably requires this existence for the possibility of its absolutely practically necessary object, the highest good, and theoretical reason is thereby justified to assume it” (Practical Reason, Ak. 5:134). The main metaphysical difference between pre-­Critical and Critical accounts of creation, however, concerns the non-­spatiotemporal character of the substratum created and conserved in the later theory. Regarding the relation between created noumena and empirical reality, the Critical philosophy maintains that “no one can have the slightest knowledge” whether “the supersensible underlying the appearance of a body is composite or simple as thing in itself ” (Ak. 8:209 n.). What is essential to Kant’s practical-­dogmatic metaphysics is that distinct moral agents at the empirical level are noumenally distinct.46 That Kant’s moral theology does not require more does not entail sympathy for Berkeley’s claim that there are only minds, and indeed it is quite clear that he lacks sympathy for such a doctrine (Prol, Ak. 4:493; Correspondence, Ak. 11:395). Whether this is rightly interpreted as reflecting a fundamental and unargued “realist bias” in his thought, as Adickes argues at length, can be bracketed here.47 What is important for present purposes is that the Critical philosophy continues to describe conservation as broadly symmetrical with creation and both as directed towards the causally active substratum of empirical reality. Lectures presented after the appearance of the KrV describe conservation as a divine action “in the innermost component [das innerste] of substances, that is, in the substantial [das Substantiale] or first inner principle of action of substances” (Danziger Rationaltheologie, Ak. 28:1310). We are told that “the omnipresence of God is the most intimate presence: that is, God conserves das Substantiale, the inner nature [das Innere] of substance . . . and without God’s unceasing inner and essential actuation of the Substantiale of things in the world, these things must cease to exist” (Religionslehre Pölitz, Ak. 28:1107).48 While Kant continues to present creation and conservation as essential components of his moral theology, the Perpetual Peace passage cited above shows that he regards divine concurrence as competing with naturalism regarding secondary causation in a way in which creation and conservation supposedly do not.49 His concurrentist predecessors would respond that this objection conflates

46  See B xxvii–xxviii; Ak. 5:104, 114; Adickes (1924b, 115–116, 159); Heimsoeth (1924); Adams (1997b, 822). 47  This is a central claim of Adickes’s classic Kant und das Ding an Sich. For a similar proposal in recent times, see Ameriks (2003, 33–40). 48  On Kant’s notion of the “substantiale” and his view of noumena as causally active, see Heimsoeth (1929, 124–128). 49 Cf. Religionslehre Pölitz, Ak. 28:1106; Danziger Rationaltheologie, Ak. 28:1309.

Desmond Hogan  287 a general metaphysical condition for the efficacy of creaturely powers with mir­ acles or violations of the natural order. Kant evidently sees no conflation; he claims repeatedly that “all concursus is miraculous,” that “every case in which God acts immediately [in nature] is an exception to the laws of nature.”50 From the concurrentist’s perspective, this begs the question against the doctrine that secondary causation rooted in creatures’ natures requires concurrence as condition of its natural efficacy. In effect, Kant charges the concurrentist’s doctrine of immediate divine action in nature with the error Leibniz attributes to occasionalism— that of appealing to “miracles which are no less miraculous for being continual.”51 Adjudication of this dispute requires consideration of further arguments on both sides. Kant is unlikely to be moved by an argument offered by Suarez in favor of concurrence proceeding from the mechanics of so-­ called “contra naturam” miracles—those in which the miraculous effect goes against a natural disposition in creatures. Suarez notes that the concurrentist can, for example, explain God’s holding back the effects of the Babylonian furnace in the book of Daniel as a withholding of concurrence with the fire’s natural powers, without which cooperation the fire can effect nothing. The mere conservationist lacks this mechanism and Suarez argues he must describe God’s action in contra naturam miracles as thwarting the natural activity of creatures—as it were from without. Suarez rejects such acting against creatures as unbecoming to the absoluteness of divine sovereignty. He argues that the doctrine of concurrence gains an important advantage here by admitting a metaphysical symmetry between God’s ability to annihilate creatures merely by withholding his conservation and his ability to “deprive a created entity of its natural action merely by withholding concurrence” (Suarez 2002, 22.1.11).52 Kant does consider and reject another argument for concurrence often employed in the schools. Its central idea is that a proper understanding of conservation reveals the necessity of an immediate divine action in all natural effects. We saw a similar claim in one of Malebranche’s arguments for occasionalism. Malebranche argues that God’s giving of being must involve “the ways of being, since the ways of being are nothing but beings themselves, in this or that fashion.”53 The concurrentist seeks to infer divine immediacy in all natural effects on these grounds while avoiding Malebranche’s occasionalist conclusion that God produces everything and creatures nothing. Leibniz gestures at such an argument when he writes that “one cannot say what ‘to conserve is,’ without reverting to the general [i.e. concurrentist] opinion. It must be taken into account

50  Danziger Rationaltheologie, Ak. 28:1308; Religionsphilosophie Volckmann, Ak. 28:1213. 51  Letter to Arnauld, April 30, 1687, GP II 92. 52 For an attempted defense of Suarez here from a theistic perspective, see Freddoso (1991, 572–577). 53  OCM XI 160.

288 Kant ’ s Theory of Divine and Secondary Causation that the action of God in conserving should have some reference to that which is conserved” (Theodicy §27). Arguments from demands of conservation to concurrence are also seen in Suarez, Wolff, Baumgarten, and Meier.54 Suarez argues that, “if it is not the case that all things are effected immediately by God, then neither is it the case that they are conserved immediately, given that an entity is related to its esse in the same way that it is related to its being-­made. For an entity’s esse cannot depend more on an adequate cause after it has come to be than it did while it was coming to be” (Suarez 2002, 22.1.7). His idea is that if a creature can produce any effect whatsoever without God’s immediate assistance, then that effect does not depend on an immediate divine conservation at the moment of its production—contrary to Suarez’s understanding of divine conservation. Kant would simply grant the conclusion, since his official view is that God conserves accidents only mediately, by virtue of an immediate conservation of substances in which they inhere.55 Both Baumgarten and Meier likewise claim that the conservation of creaturely forces puts God into an immediate causal relation with natural effects of these forces—thus with all effects of secondary causes.56 On one recent interpretation of Leibniz’s model of divine concurrence, he has something similar in mind. Adams suggests that Leibniz’s association of concurrence with demands of conservation might be understood as claiming that “what God (directly) produces, we may say, is not just the creature’s nature and its affections and actions, but the creature’s nature ‘operating’ and thus producing its affections and actions. In thus producing the creature’s producing, God’s conserving activity has a direct causal relation to the creature’s actions, but without excluding the productive agency of the created nature” (Adams 1994, 97).57 The main problem for this proposal is the danger of collapsing concurrence into mere conservationism. Suarez objects that Durandus’s conservationism fails to maximize divine sovereignty in denying a further action of God “in and with all agents” (Suarez  2002, 22.1.13). Concurrence is standardly presented as something more than mere conservation—Suarez explains that “it is clear that even though God can be said to act through the secondary cause to the extent that he gives and conserves its power to act and to the extent that he instituted it 54  Wolff 1737, §394; Baumgarten 1757, §958; Meier 1757, §§1027–1029. 55  See Freddoso (1991, 566–569), for an attempted defense of Suarez here. 56 “Conservatio virium huius universi quarumcumque, in ipso earundem actu, est CONCURSUS DEI PHYSICUS, isque, quia et quatenus ad singulas singularum substantiarum actiones extenditur, GENERALIS (universalis) dicitur” (Baumgarten 1757, §958). Cf. Meier (1757, §1027). 57  In this way Adams offers a possible resolution of Leibniz’s apparent dithering on the immediacy of God’s action in nature. In opposing deism, Leibniz insists that immediate divine action in all effects of creatures is a condition of conservation (Theodicy §27, GP VI 118). Where occasionalism is his concern, he seems to restrict God’s immediate contribution in creation and conservation to production of substance: “The production of modifications has never been called creation . . . God produces substances from nothing, and the substances produce accidents by the changes of their limits” (Theodicy §§395–396, GP VI 351–352).

Desmond Hogan  289 for acting, nonetheless, to the extent that he immediately cooperates with the creature, he does not, properly speaking, act through the creature but acts instead through himself and through his own power and strength” (Suarez 2002, 22.1.21). If the immediacy of God’s action in concurring reduces to God’s producing creatures’ producing, then Durandus is at worst guilty of failing to grasp an immediacy of divine action that his own doctrine of conservation entails. Kant evidently accepts the characterization on which divine concurrence is supposed to be something over and above mere conservation. On this understanding, the individual concurring causes, including the divinely conserved secondary cause, are regarded as individually insufficient for the effect. The concurrentist does not deny God could produce the effect alone if he wished, but rather claims that the action by which God actually concurs is insufficient to produce it without the creature’s contribution. This provides the basis for Kant’s objection that “conservation is not concursus, because concursus is the causality of an insufficient cause providing a complementum ad sufficientiam” (Danziger Rationaltheologie, Ak. 28:1309). In light of the traditional characterization of concurrence as something more than mere conservation, part of the point of Kant’s physician example is surely to record what he sees as the confusion of Baumgarten and Meier in seeking to derive concurrence from conservation. He objects that “God is the author of the physician together with all his medicines and so the effect must be ascribed entirely to [God], if one wants to ascend all the way to that highest original ground . . . Or one can ascribe it entirely to the physician, insofar as we follow up this event as belonging to the order of nature and as explicable in terms of the order of nature, within the chain of causes in the world” (Perpetual Peace, Ak. 8:361 n.).58 The passage offers a conceptual objection to the very idea of divine concurrence with those secondary causes operating by a necessity of nature. Kant writes, “to allow what is itself the complete cause of alterations in the world supplement its own predetermining providence (which must therefore have been inadequate) during the course of the world, is first, self-­contradictory . . . for causa solitaria non iuvat” (ibid.). This line of argument against concurrence is repeated in many mature texts (Religionslehre Pölitz, Ak. 28:1105–1106; Danziger Rationaltheologie, Ak. 28:1308–1309; Metaphysik K2, Ak. 28:811; Metaphysik Dohna, Ak. 28:648).59 Leibniz and Suarez would respond by rejecting the sense of “predetermining providence” implicit in Kant’s argument. The objection presents divine concurrence

58 A referee suggests that this passage evinces sympathy for occasionalism, but occasionalism involves immediate and not just “entire” ascription of effects to God. The tradition agrees that ascription of all worldly effects entirely to God as their “highest original ground” is compatible with the reality of secondary causation. 59  An early statement of this argument is also found in a text dated to 1760–1764—further evidence that Kant’s pre-­Critical metaphysics is consistently conservationist (Reflections on Religion, R 8083, Ak. 19:627).

290 Kant ’ s Theory of Divine and Secondary Causation as God’s overdetermination of a natural effect through cooperation with a secondary cause which would have produced the effect in any case. Insofar as concurrentist opponents present concurrence as a condition, above and beyond conservation, of the efficacy of secondary causes, the idea is rather that created natures are incapable of producing anything without divine cooperation. This point is admittedly muddied by arguments in which opponents suggest that concurrence follows from demands of conservation. But once it is made clear, it is evident that the concurrentist does not understand God’s “predetermining providence” to mean that effects would come about in any case; that is, even without the concurring act. As Suarez explains, in the case of causes acting by a necessity of nature, God concurs with them “in the manner of a nature”—which is to say that God’s decision to create and conserve such creatures brings with it a defeasible commitment “to concur with these same entities in their actions according to their capacities” (Suarez  2002, 22.4.3–4). God’s predetermining providence thus includes this defeasible cooperation in the ordinary course of things—it is not supplemented by it. Kant’s conceptual objection might be seen as a response to the weight contemporaries place on arguments for the necessity of concurrence from demands of conservation. It might also be taken as a rejection of the coherence of divine concurrence as traditionally understood. As Freddoso explains, models of divine concurrence were traditionally viewed as having to avoid a number of pitfalls representing significant metaphysical constraints on proposals for its operation. Two important pitfalls involve splitting the effect and splitting the action, in the sense that a distinct part of the naturally produced effect or a distinct action is attributed to the creature alone. Any model of concurrence involving such splitting is in danger of violating the concurrentist’s fundamental doctrine that the creature can do or produce nothing without divine cooperation.60 Kant offers very limited discussion of particular models of concurrent causation, and it is unclear whether he is aware of traditional worries concerning the satisfiability of the no-­ splitting constraint. Part of Malebranche’s case for occasionalism is that divine concurrence as an alternative account of God’s relation to secondary causes “appears the more incomprehensible the more effort is spent to understand it.”61 Kant’s conceptual objection might be read as reflecting agreement with Malebranche that there is simply no coherent concurrentist

60  Durandus notes that “it is one thing to say that God immediately produces something [in the effect and the creature something else] . . . It is quite another thing to say that God immediately produces each thing that a creature produces,” as concurrentism requires. Petri Lombardi sententias theologicas commentariorum libri IIII, §6, trans. in Freddoso (1994, 144). Durandus argues that the requirement makes it impossible to offer a coherent model of divine concurrence—hence his mere conservationism. 61  OCM III 243 (trans. Malebranche 1997a, 680).

Desmond Hogan  291 middle ground between mere conservationism and an occasionalist attribution of effects solely to God. One apparent problem with this reading is that Kant does not reject concurrent causation per se. He allows, of course, concurrence among secondary causes—this does not run afoul of his conceptual objection, since the co-­causes are not “completely subordinated” to each other (Ak. 19:627).62 But drawing on his mature libertarian metaphysics, Kant also allows for the possibility of divine concurrence with free causes. In this case, as he repeatedly argues, the creaturely causes as free are not “completely subordinated” to God’s “predetermining providence,” and so his conceptual objection to concurrence with causes acting by a necessity of nature simply does not apply (Religionslehre Pölitz , Ak. 28:1110; Metaphysik K2, Ak. 28:811; Danziger Rationaltheologie, Ak. 28:1311). Discussion of this issue brings us into the ambit of Kant’s theory of grace, which I can only touch on here. As noted, Leibniz and Suarez distinguish the metaphysics of divine and secondary causation in the ordinary course of nature from the metaphysics of grace as a special variety of concurrence. The former is the concern here but the issues are related, and Leibniz notes the old association of Durandus’s conservationism with the Pelagian heresy (Theodicy §27, GP VI 118). Kant’s conservationism has the interesting consequence for his metaphysics of grace that he can ignore the no-­splitting-­of-­action/effect constraint on divine concurrence as traditionally understood. For we have seen that this constraint is motivated by the concurrentist’s basic idea that creatures can do or produce nothing without divine assistance beyond mere conservation, and this is a doctrine Kant has given up. It follows that his admission of the possibility of divine concurrence with free acts is less of a concession to concurrence than might at first appear, since such cooperation need not meet traditional constraints on such cooperative causation. There remains room to interpret Kant as agreeing with Malebranche that divine concurrence as conceived of in the schools is unintelligible. That Kant admits the possibility of divine concourse with free acts does not contradict a traditional verdict that his mature theory of grace is Pelagian in rejecting or at least strongly downplaying prevenient grace—that divine assistance through which the finite agent first turns towards the good—in favor of an emphasis on human effort and freedom.63 He writes in the Religion: Granted that some supernatural cooperation is also needed to his becoming good or better, whether this cooperation only consist in the diminution of

62  The KrV speaks explicitly of concurrence of secondary causes: “an effect that arises from the concurrence of many acting substances is possible if this effect is merely external (as, e.g., the movement of a body is the united movement of all its parts)” (A 352). 63  Cf. Adams (1998, xxi–xxii).

292 Kant ’ s Theory of Divine and Secondary Causation obs­tacles or be also a positive assistance, the human being must nonetheless make himself antecedently worthy of receiving it, and he must accept this help (which is no small matter), i.e. he must incorporate this positive increase of force into his maxim: in this way alone is it possible that the good be imputed to him, and that he be acknowledged a good human being.  [Religion, Ak. 6:44; cf. 191]

There remains a crucial gap between Kant’s doctrine that divine cooperation with free acts is metaphysically possible and Christian orthodoxy according to which grace is absolutely essential to all moral agency.64 Some of Kant’s most positive pronouncements regarding divine concurrence belong in contexts where his problem is not how the fallen agent can right himself morally, but rather the engineering issue of providential control in a libertarian setting.65 By contrast, he often appears least inclined to allow divine concourse with free acts where his focus is the acts’ moral character. The Religion describes “the concept of a supernatural intervention into our moral though deficient faculty” as, “very risky and hard to reconcile with reason; for what is to be accredited to us as morally good conduct must take place not through foreign influence but only through the use of our own powers” (Religion, Ak. 6:191). Lecture transcripts employ even stronger language and provide further support for viewing Kant’s position on grace as Pelagian.66 We have now seen, at any rate, how Kant thinks it possible to justify a conservationist metaphysics of divine and secondary causation even in the face of his Critical strictures on knowledge of noumena. Such strictures do not rule out acceptance of the reality of finite agency on practical grounds, and neither do they exclude appeals to a broadly conceived naturalism in opposing the continuous interventions of occasionalism. Kant also argues for conservationism by elimination, with the aid of conceptual arguments against concurrence, though neither Leibniz nor Suarez would be very impressed by the arguments he offers. Both would regard his objection that concurrence violates naturalism as conflating what they present as a metaphysical condition on all secondary causation and miracles in the sense of violations of the natural order. Kant’s defense of conservationism over concurrence is weakened by his failure to make this basic distinction. 64  See the canons of the Council of Carthage against the Pelagians (Denzinger and Schönmetzer 1997, §§104–105); cf. canons of the Council of Trent (Denzinger and Schönmetzer 1997, §§811–814). 65  “So far as moral concourse or free cooperation of God in free actions is concerned, this cannot be understood due to the nature of freedom but shouldn’t be seen as impossible either. Since it is presupposed that rational beings can spontaneously act freely and independently of natural necessity even against God’s plan, it is quite possible that God should cooperate as concausa with their moral acts [zu ihrer Moralität] in order to keep their use of freedom in line with his highest will” (Pölitz Religionslehre, Ak. 28:1110; cf. Danziger Rationaltheologie, Ak. 28:1311). 66 We read in one case that “divine concourse with free acts is not conceivable . . . If God is their determining cause, the acts aren’t free . . . If God concurs in moral action [zur Moralität], the human has no moral worth, since [the action] cannot be imputed to him” (Danziger Rationaltheologie, Ak. 28:1309; cf. Religionsphilosophie Volckmann, Ak. 28:1209).

Desmond Hogan  293 His efforts to press the conservationist line on broadly naturalist grounds also remain in some tension with his own continued commitment to the dependence of all finite existence on divine acts of creation and conservation.

5 Conclusion Kant’s mature theory of God’s relation to creaturely causality involves by his own admission a fundamental tension absent from his earliest thought. Alongside his enduring commitment to the complete ontological dependence of creatures, Kant comes to embrace a libertarian doctrine of agency on which free acts are not completely fixed by God’s “predetermining providence.” Late writings openly acknowledge the resulting tension: It is totally incomprehensible to our reason how beings can be created to use their powers freely, for according to the principle of causality we cannot attribute any other inner ground of action to a being which we assume to have been produced except that which the producing cause has placed in it . . . the possibility of beings who are thus [morally] called is for speculation an impenetrable mystery. [Religion, Ak. 6:142] For to be a creature and, as a natural being, merely the result of the will of the creator; yet to be capable of responsibility as a freely acting being (one which has a will independent of external influence and possible opposed to the latter in a variety of ways); but again, to consider one’s own deed at the same time also as the effect of a higher being—this is a combination of concepts which we must indeed think together in the idea of a world and of a highest good, but which can be intuited only by one who penetrates to the cognition of the supersensible (intelligible) world and sees the manner in which this grounds the sensible world. [On the Miscarriage of all Philosophical Trials in Theodicy, Ak. 8:264]67

Leibniz loosens this knot by rejecting the libertarian theory in favor of a universally “predetermining providence.” Kant’s early writings follow him in treating God’s creative act as “the well or bubbling spring from which all things flow with 67  These texts seriously undermine Insole’s central thesis that Kant views transcendental idealism as “solving” the problem of the consistency of createdness and freedom (Insole  2013, 77, 89, 112, 178–179, 184, 192, 223). The second Critique does claim that transcendental idealism makes space for a possible reconciliation unavailable to the realist, but adds that this “hardly allows of a lucid presentation” (KpV, Ak. 5:100–102; cf. Brewer and Watkins 2012). Kant’s considered position I take to be that of the Religion and Theodicy texts: while transcendental idealism is a necessary condition of the possibility of freedom, freedom’s consistency with creation remains an insoluble mystery. Insole cites both texts in his final paragraphs, reading them not as inconsistent with his main thesis but as suggesting rapprochement with theological compatibilism (Insole 2013, 243–244). Given Kant’s emphasis on freedom’s incomprehensibility, however, he is hardly envisaging a return to a compatibilist position.

294 Kant ’ s Theory of Divine and Secondary Causation infallible necessity down an inclined channel” (Nova Dilucidatio, Ak. 1:403). Jesuit libertarians including Molina and Suarez seek to unravel the knot with the help of a sophisticated theory of divine concurrence with free acts. Their theory is designed to preserve robust dependence of such acts on God without destroying creatures’ freedom with regard to omission or specification. Suarez’s version employs Molina’s doctrine of middle knowledge and an ingenious model of “conditional” divine concurrences with free acts (Suarez  2002, 22.4.10–39). It is not explored by Kant, and it is not clear how much he understands of it.68 What is clear is that the Critical philosophy expresses deep unease about proposals to reduce the tension between dependence and freedom by strengthening the role of grace. It is tempting to conclude that Kant is thereby forced to a more deistic position on which the bonds of creaturely dependence on God are relaxed. I believe it is important for an understanding of the Critical philosophy to recognize his firm rejection of this conclusion. God’s mere conservation of causes acting by a necessity of nature is described as a complete ontological dependence of such causes and their effects.69 Kant denies that concurrentism strengthens creaturely dependence on God in this case, arguing that the position is unintelligible. Even those free acts not subordinated to “predetermining providence” continued to be described by Kant as “effects” of God. He writes that one must “consider one’s own deed at the same time also as the effect of a higher being” (Ak. 8:264). His libertarian theory of freedom is never presented as justifying a deistic dilution of creaturely dependence. Instead Kant simply acknowledges the tension between demands of ontological dependence and those of morality, while holding out for the possibility of a resolution beyond human understanding.

68  This theory is overlooked by Insole when he describes concurrentism as involving a compatibilist approach to divine causality and creaturely freedom (Insole 2013, 221–223). That might be said of the Thomist account of concurrence (Insole’s main model), but Molina dismisses that account precisely on the grounds that it “subverts freedom of choice” (Concordia 4.53.2.20). According to Suarez, divine concourse with created freedom is not intrinsically efficacious or inefficacious, since acts also depend on creaturely contributions of which God is not “total” cause. On Kant’s stance vis-­à-­vis Molinism see Hogan (2014). 69 “God is the author of the physician together with all his medicines and so the effect must be ascribed entirely to him, if one wants to ascend all the way to that highest original ground, the­or­et­ic­ al­ly incomprehensible to us.” Perpetual Peace, Ak. 8:361 n.

11

The Development of Kant’s Conception of Divine Freedom Patrick Kain, Purdue University*

Lewis White Beck once astutely suggested that the conception of a holy divine will “is an important conception, for, by comparison with it, Kant brings out more clearly the peculiarities of the human . . . practical faculty” (Beck  1960, 50 n.).1 J.  B.  Schneewind, and before him Dieter Henrich, suggested that Kant’s early engagement with Leibniz’s philosophical theology led him to a conception of the divine will that helped to motivate many of the distinctive features of Kant’s mature moral psychology and moral philosophy (Henrich  1963; Schneewind 1996, 1998). Indeed, Kant regularly invoked comparisons and contrasts with the divine holy will in order to clarify his conception of imperatives and his theory of obligation, to clarify his conception of “respect” as a morally worthy incentive to action, to explain the nature of virtue, and to illuminate his account of our relation to the highest good.2 Once you start looking, references to a holy or divine will can be seen to litter the Groundwork and the second Critique, for example.3 In his lectures, Kant even suggested to his students that the freedom of a divine holy will is “easier to comprehend than that of the human will” (Ak. 28:609).4 Nevertheless, Kant’s conception of the divine will has remained almost completely ignored, politely avoided as a quaint or distracting relic of a theological tradition toward which Kant himself showed ambivalence. Even worse than neglect, however, is the fact that when commentators have deigned to mention Kant’s conception of such a divine will or holy will they have often quickly dismissed it as insignificant, irrelevant, mysterious, or even incoherent. (To take

*  I would like to thank Karl Ameriks, Andrew Chignell, Jan Cover, Eric Watkins, and audiences at Notre Dame, Houghton College, University of Wisconsin-­ Madison, University of California-­ Riverside, and University of Kentucky for helpful discussion of earlier versions of this chapter. 1 Schneewind makes a complementary point about the neglected significance of Kant’s claims about all rational beings in the Groundwork preface (Schneewind  1996, 34; 1998, 510–511; 2010, 215–216). 2  I have suggested elsewhere that the necessary volition of the divine will provides a way to make sense of Kant’s suggestion that we have a necessary end, happiness—which grounds his account of prudence (Kain 2001). 3  A quick search turns up at least twenty references to a divine or holy will in these two works. 4  For similar claims, see also Ak. 28:700, 806, 1279; R 4787, 4788, Ak. 17:728.

Patrick Kain, The Development of Kant’s Conception of Divine Freedom. In: Leibniz and Kant. Edited by Brandon C. Look, Oxford University Press. © Oxford University Press 2021. DOI: 10.1093/oso/9780199606368.003.0011

296  Development of Kant ’ s Conception of Divine Freedom talk of it “littering” Kant’s text too literally!) Even Schneewind and Henrich, among the few to devote any sustained attention to the issue, stop short of making sense of Kant’s conception of the divine will. There are many ways in which the details of Kant’s account of the divine will seem confusing or obscure, especially in relation to questions about his puzzling, if somewhat more familiar, account of human wills operating under obligation. The concept of an agent that necessarily complies with rational norms (which is part of how Kant thinks of the divine will) has struck many contemporary philosophers as incoherent and certainly incompatible with Kant’s own deeply “libertarian” convictions about rational agency. Likewise, the very differences Kant emphasizes between divine and human wills would seem to make the details of his account of the divine will irrelevant to the human case, which most interests the majority of contemporary scholars. Even if one were to accept such appearances, ignoring Kant’s conception of the divine will comes at the historical cost of blinding us to the full range of Kant’s own intellectual commitments on matters of significant interest and to potentially salient features of Kant’s historical context. Contrary to these appearances and contentions, however, I will argue that Kant had a robust and arguably coherent conception of the divine will, and I  will suggest that this conception is of significant value for our understanding some important features of his complex theory of free rational agency. After reviewing some of Kant’s familiar claims about the will (in general), and about the divine holy will in particular, I consider how those claims give rise to some initial objections to that conception. Then I defend an interpretation of Kant’s conception of the divine will, and of its historical development in relation to Leibniz and Spinoza, which identifies the content, origin, and role of God’s representations in a way that is responsive to some of the historical and contemporary problems. Finally, I trace a few of the implications of this account of the divine will for our understanding Kant’s account of freedom more generally, including human freedom.

1  Thinking of a Holy Will A will or “faculty of desire,” Kant tells us, is a particular kind of causal faculty, “a being’s faculty to be by means of its representations the cause of the reality of the objects of these representations” (Ak. 5:9 n.; cf. 6:211).5 An agent’s will is an efficient 5  Relatedly, in the Groundwork, a will is described as “the capacity to act in accordance with the representation of laws, that is, in accordance with principles” (Ak. 4:412). In what follows, translations of Kant’s works are based, when possible, on those of Mary Gregor in Practical Philosophy; Allen Wood in Religion and Rational Theology; David Walford and Ralf Meerbote in Theoretical Philosophy, 1755–1770; Curtis Bowman, Paul Guyer, and Frederick Rauscher in Notes and Fragments; Gary Hatfield in Theoretical Philosophy after 1781 (all in the Cambridge Edition of the Works of Immanuel

Patrick Kain  297 cause of what he brings about through his representations. Following Ameriks, we might say that the agent’s representations function as the formal cause of the actions and/or of the end they effect, insofar as the actions and/or the end are brought about in virtue of the content of the representation. For example, I may represent to myself a (not yet existent) sandwich and an act of sandwich construction, and I may will that sandwich-­making. When I do so successfully, I am the efficient cause of the act and of the sandwich, and am the cause of them in virtue of the content of my representations. A free will involves something more specific: the capacity to have certain normative representations (e.g. representations of goodness) and to act spontaneously through those representations. In one of his earliest works, the 1755 Nova dilucidatio, Kant put part of the point this way: “spontaneity is action which issues from an inner principle. When this spontaneity is determined in conformity with the representation of what is best it is called freedom” (Ak. 1:402). Once Kant came to focus on the particular im­port­ ance of pure, rational representations and on the centrality of merely formal practical principles (e.g. the moral law) as the key to the conception of the good, he claimed specifically that a free will is a capacity to be an efficient cause in virtue of one’s representation of the moral law.6 He also came to insist, relatedly, that a free will “can be efficient independently of alien causes determining it” (Ak. 4:446, cf. Ak. 5:95, 101). The central question about the coherence of Kant’s account of divine freedom, then, is whether he can provide an account of the divine will that satisfies these conditions (and any other necessary conditions) for freedom. In the Groundwork, Kant claims that in a perfectly good being “reason infallibly determines [unausbleiblich bestimmt] the will [and] the actions . . . that are cognized as objectively necessary are also subjectively necessary, that is, the will is a capacity to choose only that which reason independently of inclination cognizes as practically necessary, that is, as good” (Ak. 4:412). The will of such a holy being cannot be “necessitated” (genöthigt) to action because the will “of itself, by its subjective constitution [von selbst nach seiner subjectiven Beschaffenheit], can be determined [bestimmt] only through the representation of the good”; his “volition is of itself necessarily in accord with the law” (das Wollen schon von selbst mit dem Gesetz nothwendig einstimmig ist) (Ak. 4:414) and his “maxims necessarily harmonize [nothwendig . . . zusammenstimmen] with the laws of autonomy” (Ak. Kant); and that of Thomas Hilgers, Uygar Abaci, and Michael Nance in Observations on the Feeling of the Beautiful and Sublime and Other Writings (in the Cambridge Texts in the History of Philosophy). In each case, I have modified the translations as I have found necessary. 6  As Ameriks has helpfully noted, Kant demands that representations, esp. specifically rational representations, can be both the efficient cause and formal cause of our actions (Ameriks 2003, 252). We might put the latter point in terms of the agent’s representations serving as efficient causes in virtue of how their content informs the actions, events, and objects to which they give rise. Often, at least, the agent represents a producible end and that end is a final cause of the action (Ak. 6:381, 385). For present purposes, I will set aside questions about other kinds of ends and their suitability for the role of final cause.

298  Development of Kant ’ s Conception of Divine Freedom 4:439, cf. 453–454).7 On a quick read, it may seem that Kant comes close to conceding the incoherence of the concept of a free holy will when in the Groundwork, for example, he says that “reason infallibly determines” the holy will (Ak. 4:412) and “no imperatives hold for the divine will and in general for a holy will: the ‘ought’ is out of place here” (Ak. 4:414). This may seem to suggest that the divine will is not “free,” and that any normative claims about the divine will involve a category mistake. Inspection of the relevant passages, however, reveals that Kant is making a point about specific concepts like “ought,” “obligation,” “necessitation,” “duty,” and “imperative.” He clearly insists that the practical laws of the good “apply to,” “include,” and guide the activity of a self-­sufficient, perfect agent, though not in an imperatival form. A holy will is “above all practically restrictive laws and so above obligation and duty” but “not indeed above all practical laws” (Ak. 5:32). A holy will acts “through the representation of the good” (Ak. 4:414). Kant insists that the moral law (or the laws of the good) must be valid for and guide all rational beings. He is by no means conceding the absence of normative representations in a holy will or the incoherence of the idea of a divine, free, holy will. Yet, Kant’s references to the divine will may still seem at best an unnecessary distraction and, perhaps more likely, the embarrassingly uncritical indulgence of some dogmatic metaphysical and theological baggage. Some might even object that any mention of such a concept sits quite uneasily with Kant’s alleged “Critical” insistence upon respecting the limitation of all knowledge and even meaningful judgment to the “bounds of sense.” The first thing to say is that Kant’s Critical epistemology does not prohibit legitimate “thoughts” about supersensible objects; indeed, it purports to warrant “practical knowledge” of the moral law and transcendental freedom, and practical “belief ” or “faith” in the existence of God and immortality (Kain 2010). The second thing to say is that the concept of a “perfectly free” being is of particular interest to Kant throughout the Critical period. Moreover, it plays a crucial role in the central argument of the Groundwork that is intended to be compatible with Kant’s Critical philosophy. Early in the third section of the Groundwork, the section in which he offers a “deduction” of the categorical imperative, Kant introduces a rather compact argument for the oft-­misunderstood claim that “a free will and a will under moral laws are one and the same. If, therefore, freedom of the will is presupposed, morality together with its principle follows from it by mere analysis of its concept” (Ak. 4:447.6–9). Careful textual analysis reveals that what Kant is attempting to analyze at this point is the concept of a perfectly free, self-­sufficient will, the will of a being in whom reason “is practical without hindrance” (Ak. 4:449.18), or, what is equivalent, the will of a

7  Similarly in Ak. 5:32, 79.

Patrick Kain  299 free being considered in abstraction from or as lacking all affections, needs and desires. Kant claims that such a will can act, and can only act, solely on the principle of morality, and that this is an analytic truth—this is what Schönecker calls Kant’s “Analyticity Thesis” (Schönecker 1999; Schönecker and Wood 2002; Timmermann 2003).8 Since Kant’s Critical philosophy contains no embargo on analytic claims that transcend the boundaries of sense, this analyticity thesis poses no special problem for Kant’s Critical epistemology. Although Kant came to reject his deduction of the categorical imperative, this does not appear to be due to critical concerns about this premise or the conception of a holy will. The precise epistemic status of, and the legitimate methodology for determining and using, judgments (esp. “meaningful” judgments and existence claims) within Kant’s philosophical theology remain complicated and controversial. I will not attempt to settle those issues here, aside from noting that Kant himself did not insist upon, nor did he abide by, an absolute embargo on such matters, and that Kant had a particular and sustained interest in some of them, including the concept of a divine holy will.9 What follows is an attempt to better understand Kant’s account of this concept, and defend its coherence against one particular sort of objection.

2  The Blind Mechanism Objection Many contemporary philosophers, including metaphysicians, action theorists, and philosophers of religion, have expressed serious doubts about the coherence of the concept of a rational agent whose behavior is necessarily perfect, or more generally, about the concept of any “agent” whose behavior is either physically necessary or metaphysically necessary.10 As Lavin has recently noted, it is widely held that “the way in which agents are directed by principles is radically different from the way in which mere bits of stuff are directed by physical laws” (Lavin 2004, 444). We might add: it is also widely held that the way agents are directed by principles must be radically different from the way in which metaphysically 8  That this is true, much less analytic, is certainly controversial, especially since it may appear to suggest that God is “necessitated” by the moral law or constrained by something other than himself. In what follows, we will explore aspects of Kant’s contentions about the divine will and how it is not “necessitated,” is not obligated, and how it is not constrained by anything other than himself. 9  In what follows, I set aside in general questions about the status of predications about God, whether they are to be understood as simple predications of God’s intrinsic nature or as analogical predications of inscrutable grounds in him of particular relations to us, and questions about their proper use in theoretical and/or practical philosophy. On some of these questions, see e.g. Ak. 8:400 n.; Ak. 5:465, 465 n.; Ak. 20:362. 10  Questions about the nature of divine perfection and divine freedom have attracted significant attention in recent philosophy of religion (Morriston 1985; Morris 1987; Yandell 1988; Wierenga 1989; Yandell  1999; Rowe  2002; Wierenga 2002; Bergmann and Cover  2006; Rowe  2006; Wierenga  2007; Pereboom 2009).

300  Development of Kant ’ s Conception of Divine Freedom necessary truths (at least typical, or arbitrarily chosen, ones such as the principle of non-­contradiction or the truths of mathematics), are made true. For “leeway incompatibilists,” especially those who take the “principle of alternate possibilities” to be fundamental to freedom and moral responsibility, the point may be obvious, but doubts extend to those who allow that some forms of physical necessity may be compatible with responsible agency. “Source incompatibilists” emphasize that an agent must play a “crucial” “causal role” in the production of her actions, being the “source” of her action (Pereboom 2006, 542–544). Many insist that it matters how “the person’s own mental activity is involved in [the action’s] production.” The actions of a genuinely rational agent must be “the expression of her own mental activity,” “motivated by her own recognition of the appropriate conceptual connection[s]” and “not merely the result of the operation of beliefs and desires in her” (Korsgaard 1997, 221, 236). Genuine rational agency must be more than “brute luck,” more than merely “accidental conformity” to or compliance with some principle (Pettit and Smith 1996, 445; Korsgaard 1997; Lavin 2004, 451). The agent must actively “participate” in her own compliance with the relevant rational norms. Her action must be “self-­guided.”11 While “unswerving compliance,” or more generally, physical necessity or metaphysical necessity, may seem to be anything but “accidental,” it seems only accidentally related to the exercise of rational reflection, and thus, is often taken as an “indication” of a mere blind mechanism, rather than of an agent’s active “participation” in the light of reasons (Lavin 2004, 445).12 The point seems at its strongest when unswerving compliance is metaphysically necessary: such behavior seems to fail even standard tests for “guidance control” offered by contemporary semicompatibilists. If moral truths are necessary and overriding, it is not possible for them to be otherwise, nor is it possible for there to be sufficient reason to act contrary to them; and if it is metaphysically necessary that the agent’s kind of “mechanism” behave as it does when moral reason are present, there cannot be the requisite counterfactual manifestations of “reasons-­ responsiveness.”13 Thus, “[A] would-­ be perfectly rational will—something whose will is in a state of perfection and thus which can’t really go wrong—must amount to no more than a strange sort of mechanism or automaton” (Lavin 2004, 443). 11  “It is in the nature of activities, as opposed to mechanical processes, that one who engages in them is self-­guided . . . ” (Korsgaard 1996, 236 n.; Lavin 2004). 12  Lavin suggests that this idea lies behind what he calls the “imperatival interpretation” of an “error constraint” on agency, which he finds in Bennett, Brandom, Darwall, Korsgaard, McDowell, Railton, Searle, Wallace, Williams, and others. Lavin himself withholds judgment about “imperativalism.” 13  For a seminal discussion of “guidance control,” which requires a possible scenario in which there is sufficient reason to act otherwise, see Fischer and Ravizza (1998). It seems difficult to design a counterfactual (but not counterpossible) test for reasons-­responsiveness that would cover this case without triviality. At least in Kant’s case, concentrating on the kind of mechanism will not be sufficient: the divine mechanism is unique precisely in its type of determination and its choices may lack counterfactual variability. An insistence upon such a possibility distinguishes what Lavin calls the “imperatival interpretation” from the “logical interpretation” of the error constraint on agency.

Patrick Kain  301 If the truths describing an alleged agent’s behavior are metaphysically necessary, then those truths will be entailed by other truths and laws, that may make no reference to the alleged agent’s rational capacities or their exercise. In this case, the form of explanation of these behavioral truths may seem indistinguishable from the form of mechanical explanations of the behavior of mere bits of matter or the form of explanation of familiar metaphysically necessary truths (or an assertion of their bruteness): the behavior appears to be sufficiently explained, independently of the agent’s alleged reasons and powers, by the relevant necessities.14 The behavior of such an alleged agent appears indistinguishable from the mechanistic behavior of a mere bit of matter or from the brute, impersonal necessity of some arbitrarily chosen necessary truth. In such a substance, alleged rational capacities and their alleged relation to rational principles may seem, at best, “practically impotent,” irrelevant to the genesis of its behavior; and occurrent thoughts of such principles, if there are any, may be merely epiphenomenal.15 Such behavior, it seems, cannot be considered the action of a genuinely rational agent at all. We can call this line of objection to the coherence of a Kantian holy will the “blind mechanism objection.” A refutation of every version of this objection to Kant’s conception of divine freedom is beyond the scope of a single chapter, but a defense of Kant’s conception of divine freedom is possible, if we can explain how a proper account of divine activity might include a form of divine rational agency, even if that behavior is metaphysically necessary. We need to understand why Kant thinks divine activity can be understood as an expression of divine mental activity, activity in which the representation of the good or the moral law might function as a guide, even though it may be impossible that the divine will choose to violate or resist those norms. What is needed is an account of how the divine will might have and act on normative representations: an account that identifies those representations and renders them something more than accidental,

14  Not that entailment, or even relevant entailment, is always sufficient for explanation; but it may appear so in these contexts. Many take this sort of entailment to be incompatible with, or to create a strong presumption against, the kind of explanation, or lack thereof, that is constitutive of rational or responsible agency. Some insist that one cannot be free or morally responsible for anything “entailed by the truth about the past, the laws of nature, and the laws of logic” (Yandell  1988, 334–340; Yandell 1999, 315–321). Van Inwagen, for example, suggests that “hardly anyone besides Descartes is willing to” reject the principle that from necessarily P we may infer “no one has or ever has had any choice about whether” P. This is his “inference rule alpha” (Van Inwagen 1983, 93–96). If “having a choice about whether P” entails “it is possible that one render P false,” this may be obvious; but if it only entails “having the power to render P false” as Van Inwagen may intend (66–67) (or, weaker still, if it only relevantly entails “having the power to bring about P”), it may be disputed, at least in the theological case, and not only by radical modal voluntarists such as Van Inwagen’s Descartes. Talbott is one contemporary example (Talbott 1988). Morris proposes a subtly more modest formulation of the “Principle of Avoidance” according to which an act cannot be free if conditions which render it “necessary, or unavoidable, in a broadly logical sense, . . . by doing so in fact bring it about” (Morris 1987, 28). 15  “Practically impotent” is Lavin’s phrase (Lavin 2004, 445).

302  Development of Kant ’ s Conception of Divine Freedom epiphenomenal, or explanatorily irrelevant, and that reveals that the volitional activity of which they are a part is self-­ guided. The intelligibility of Kant’s conception of the divine will turns on a specification of the content, origins, and role of the relevant representations, and how they serve to distinguish divine behavior from that of a mere blind mechanism or brute, impersonal metaphysical necessity.16 It turns out that Kant had such a specification, even if it was not designed specifically to address the particular concerns of these contemporary discussions.

3  God, All-­Sufficiency, and the Problem of the Divine Will To properly understand Kant’s account of divine freedom, it is helpful to begin by sketching some important features of Kant’s account of the concept of God and to see some of the ways that Kant tries to situate his views in relation to those of Spinoza and Leibniz. First, we will consider Kant’s account of God as the material ground of all possibility, then we will proceed to his account of God as efficient cause of the contingent beings that exist, and third, we will focus on his particular emphasis upon God’s “all-­sufficiency.” In an early work, The Only Possible Argument in Support of a Demonstration of the Existence of God (1763), Kant argued that “real” possibility and necessity must have some actual ground, and that this point provides “the only possible” metaphysical basis for a demonstration of the existence of God.17 Within Kant’s theory of real possibility, the presence of a contradiction within a concept is sufficient for its impossibility and the absence of a contradiction within a concept is sufficient for its logical possibility; yet the absence of a contradiction within a concept is not itself sufficient for that concept’s real possibility.18 To be the thought of something really possible, a concept must also include some material element or content; there must be some positive quality that might be predicated of an object by means of the concept in question (Ak. 2:77–78). Following a long tradition in Platonic-­inspired thought, Kant conceived of the positive qualities, properties, or determinations of things as “realities,” “degrees,” or “magnitudes” of reality, being,

16  Cf. Korsgaard in a slightly different context: “if what we mean when we say that the person’s actions are motivated by reasons is that the person is caused to act by his recognition of certain considerations as reasons, then we must say what it is that he recognizes” (Korsgaard 1997, 243). In a similar vein, see the preliminary speculations about the content of the divine will in Pereboom (2009). 17  There are several important recent discussions of this book, the “possibility” argument, and Kant’s later attitude toward it (Fisher and Watkins  1998; Adams  2000; Logan  2007; Chignell  2009; Stang 2010; Chignell 2012). 18  Real possibility, even “absolute” real possibility, may be somewhat weaker than metaphysical possibility in the contemporary sense, since it is a species of “internal possibility,” which excludes consideration of the possible “grounds for the existence of things,” particularly when that grounding “must be sought in a free choice” or any “decision of a will.” (Ak. 2:100–101, cf. 91)

Patrick Kain  303 or perfection (Ak. 2:31; Ak. 28:1004–1005, 1013, 1146–1148).19 Kant argued that, if anything is really possible, then there must be an actual being that must itself “embrace” all possible realities, either having them as its own “determinations” or as its “consequences,” and can thereby serve as the single, necessarily existent, material ground of all possibilities (Ak. 2:85–87). (In more Aristotelian terms, we can say that this being serves as the formal cause of real possibilia.) Such a being, Kant argued further, would be God, conceived of along Christianized neo-­ Platonic lines as the ens realissimum, a “most real” being that exemplifies or possesses all fundamental positive realities without limitation, including having an understanding and will, and who serves as the material ground of those real­ ities.20 The possibilia are materially grounded, directly or indirectly, in God’s nature, rather than in acts of God’s will. Kant’s contention in the Beweisgrund essay was that only such an ens realissimum could be the material ground of all real possibility.21 In the Critical period, Kant’s attitude toward this argument for the existence of God is conflicted, complicated, and controversial—he came to insist that only a moral argument for God’s existence was fully adequate—but the core concept of God as ens realissimum, regardless of how we might establish his existence, remains central to Kant’s thought. “Everything in theology depends” upon “the concept of God as the being of all beings and with it his all-­sufficiency” (Ak. 5:100).22

19  The basic elements of this ontology appear in the first Critique, as well, e.g. A 143/B 182; A 575–578/B 603–605. In a seminal discussion of this topic, Wood points out that there are many respects in which Kant’s version of this ontology remains ambiguous (Wood 1978, 28–34). For Kant’s hesitations about the concept of “perfections,” see Ak. 2:90. Adams provides a helpful discussion of Leibniz’s version of this ontology (Adams 1994, ch. 4). 20  Since we regard our own psychological capacities of intellect and will to be limited perfections, we may infer that God, the material ground of this possibility, possesses a perfect intellect and will (Ak. 2:88). However, as we will see, Kant prefers an argument for the divine will based upon efficient causality; and he came to think that the most decisive considerations in favor of divine moral perfection come from arguments with specifically moral premises. This is a Christianized or theistic neo-­Platonism because it insists that all of the fundamental real­ ities or perfections are united in one personal being, God. A doctrine of creation, rather than em­an­ ation, will further distinguish this conception from other forms of neo-­Platonism. In the contemporary literature, this conception of God is often dubbed “Anselmian” (cf. Morris 1987). I hesitate to employ that term here because it may misleadingly suggest that Kant endorsed an Anselmian ontological argument, emphasized the concept of “perfection,” or conceived of himself as inheriting this conception from Anselm, as opposed to say, Descartes or Wolff. 21  The roots of this conception of possibility can be found in Leibniz, e.g. in Monadology §38f. (Adams  1994, ch. 4). Part of what Kant thought was innovative about his own treatment is how it emphasized that possibilities are not merely ideas in God’s mind, but limited possible exemplifications of God’s own reality or perfections, and how he argues for God’s existence explicitly from nature of possibility. On neo-­Platonism in Leibniz’s conception of God’s creative emanation, see Fouke 1994. 22  “The transcendental concept of God, as the ens realissimum, cannot be circumvented in philosophy, however abstract such a concept it may be; for it pertains to the union, and at the same time the elucidation, of everything concrete that may subsequently enter into applied theology and theory of religion” (Ak. 8:400 n.). Kant repeatedly invoked this perfect being conception in the Dialectic of the second Critique, the Dialectic of the first Critique, and in Religionslehre Pölitz lectures, for example. Cf. “The precise concept of God is the concept of a most perfect thing” (Ak. 28:1008).

304  Development of Kant ’ s Conception of Divine Freedom In addition to his articulation of the conception of God as ens realissimum and as the material ground of all real possibility (or, as I described it, as the formal cause of such possibilities), Kant reflected upon the conception of God as efficient cause of everything else that exists.23 Kant insisted that, if there are to be a plurality of substances, God must have a will. One of Kant’s targets on this point is Spinoza, who explicitly denied the antecedent (as well as the consequent, as we will see). Following Wolff, Kant considered substance monism to be Spinoza’s core doctrine and he regarded it, first and foremost, as the consequence of Spinoza’s arbitrary and erroneous definition of substance (Allison  1980; Morrison 1993).24 Kant also confronted the doctrine of substance monism with an argument based upon our awareness of our own self-­conscious thought and our awareness of our own finitude (Ak. 28:52, 458, 601, 1041–1042, 1052, 1269; Ak. 2:90–91; Allison 1980, 205–207; Ameriks 1992, 263). Kant had a second target in his sights here, too: an allegedly distinct hypothesis that God could be the efficient cause of a plurality of finite substances “by the necessity of God’s nature” (nach or durch der Nothwendigkeit seiner Natur) as “a blindly working eternal root of all things, a natura bruta” (Ak. 28:1054, 1061, 1068).25 Kant objected that a doctrine of “blind” emanation collapses back into substance monism, since the blind emanation of a distinct substance is not pos­ sible (Ak. 28:1061, 1092).26 In order for there to be a world external to God yet caused by him, Kant insists, there must be a fundamental difference between

23  I concur with Stang that Kant’s references to the “consequences arising from” God’s de­ter­min­ ations suggest a role for God’s powers in the “material grounding” of some (but not all) real pos­si­bil­ ities (Stang 2010). At the same time, it is important to note that, at least in this text, Kant’s conception of real possibility is a species of “internal possibility,” and should not include a general requirement that real possibility involves the possibility of being efficiently caused. Thus, I do not see Kant agreeing with Crusius’s general contention that the concept of possibility must include the concept of (efficient) causality, a conclusion I share with Chignell (Chignell  2009). For example, the real possibility of a finite mind depends upon God, the material ground of all possibility, himself having a mind; but this real possibility does not depend materially upon God having the power to create finite minds (which he no doubt also possesses). 24  See Spinoza, Ethics I.14. 25  Cf. Ak. 8:401 n.; A 633/B 661; Ak. 2:89. 26  In contrast to Leibniz, Kant takes emanation to be blind by definition. Cf. Leibniz’s DM 14; Causa Dei §9; Mon 47. “Without understanding it would have no faculty at all for relating itself, its own subject, to something else, or for representing something external to itself; and yet it is only under this condition that anything can be the cause of other things external to itself. From this it follows that an all-­sufficient being can produce things external to itself only through will and not through the necessity of its nature” (Ak. 28:1061). This system of emanation “has one ground of reason opposed to it, which at once overthrows it. This ground is taken from the nature of an absolutely ne­ces­sary being and consists in the fact that the actions which an absolutely necessary being undertakes from the necessity of its nature can never be any but those internal actions which belong to the absolute necessity of its essence. For it is unthinkable that such a being should produce anything outside itself which is not also absolutely necessary. But how can something produced by something else be thought of as absolutely necessary? Yet if it is contingent then how could it have emanated from a nature which is absolutely necessary? Every action performed by such a being from the necessity of its nature is immanent and can concern only its essence. Other things external to it can be produced by it only per libertatem, otherwise they are not things external to it but belong to the absolute necessity of its own essence and are therefore internal to it” (Ak. 28: 1092–1093).

Patrick Kain  305 what is contained in or follows immediately from God’s nature and what follows, even if necessarily and fully determinately, only via God’s understanding and will. “The absolute necessity of his nature and his essence [seiner Natur und seines Wesens] does not make his actions absolutely necessary. The absolute necessity of his essence is completely different from the determination of action according to his will [Willkühr]” (Ak. 28:335, cf. 342).27 Any consequence of God’s essence not mediated by his understanding and will is part of God’s nature and thus internal to God, rather than external to him. Thus, God can only be the efficient cause of substances distinct from himself via his will. And contra what he calls theological “fatalism,” Kant insists that it must be through God’s free will—that is, through his cognition of goodness—that God acts. One thing that is striking is that, especially early in his development, Kant is greatly concerned about both substance monism and theological fatalism, but not with necessitarianism per se.28 To this point, Kant’s resistance to Spinozistic monism and to fatalism is more or less Leibnizian or Wolffian. For example, Kant seems employ Leibniz’s distinction between an absolute necessity, brute necessity, or blind “fate” (which Leibniz went to great lengths to deny), and the “complete determination” of the divine will, considered as a free (i.e. a value cognizing) and completely determined will (which Leibniz endorsed and characterized as a “moral necessity” of the divine will) (Theodicy §§168–240)29 Working within this framework, Kant proceeds to consider what such a divine will could be like and to offer his own, distinctive account of what could ground God’s decision to create something distinct from himself. The primary features of Kant’s account of God’s will can be found, in compressed form, in a footnote in his 1793 “Theory and Practice” essay: 27  There is an important point here, underexplored by Wood and Pereboom: Kant’s denial that God’s action is “necessitated by his nature” has a technical sense, and is specifically contrasted with being necessarily “determined according to his will.” Pereboom is right that even here, Kant insists that nothing can cause or “necessitate” the divine volition, yet Wood would be on the right track to observe that God is allegedly “necessarily determined” “from within by reason” (if not “necessitated” by it) and that Kant is closer to Leibniz here than he is to many contemporary incompatibilists (Wood 1984, 82; Pereboom 2006, 543; 2009). What follows is an attempt to explore such aspects of Kant’s account. “Necessitation” can only occur in the absence of “necessary determination” (Ak. 4:413). 28  Kant does deny that God’s actions are absolutely necessary (Ak. 28:335, cf. 342). As Stang suggests, it is easy for Kant to deny that the best possible world is absolutely necessary (since it is distinct from and dependent upon the absolutely necessary being) (Stang 2010, 283). Kant can deny that the best possible world (which does not itself contain God or truths about his choices) is “internally” really necessary, and he can maintain that other possible worlds are “internally” really possible. However, in contrast to Leibniz, Kant does not make much of an effort to explain why he may have technical grounds to resist related versions of theological necessitarianism. As we will see, Kant seem comfortable affirming that with respect to creation, God’s will is necessarily determined, determined “of itself ” (Ak. 4:414), to precisely one object. 29  For Kant’s use of distinctions between the “blind necessity of [God’s] nature” and “moral necessity,” “practical necessity,” or “subjective necessity,” see also Ak. 28:806, 1068, 1092, 1276, 1280–1281 (cited below); R 3911, 4125, 4128, 4129, 4218, 4738, 4739. Without explanation, Kant himself refrains from deploying such distinctions at 4:344 n.: “[God’s] action is determined in his eternal reason, hence in the divine nature.”

306  Development of Kant ’ s Conception of Divine Freedom [The Deity] although subjectively in need of no external thing, still cannot be thought to shut himself up within himself but rather [must be thought] to be determined to produce the highest good beyond himself just by his consciousness of his all-­sufficiency; and this necessity in [Nothwendigkeit am] the supreme being (which in [beim] the human being is duty) cannot be represented by us other than as a moral need.  [Ak. 8:280 n.]30

The first thing to note is the negative principle at the heart of Kant’s account of the divine will (which we also saw in the Groundwork’s “analyticity thesis”). God is “subjectively in need of no external things” and can have no needs, no incentives, no interests, and no inclinations (Ak. 4:413 n., 434; Ak. 5:34, 72, 79–83, 131).31 We must think of “a will that is directed to objects upon the existence of which its satisfaction does not in the least depend” (Ak. 5:137). As Leibniz would say, “God is never moved by anything outside himself, nor is he subject to inward passions.”32 God’s omnipotence and independence entail that he must be “determined to action only by himself.”33 Despite being indebted to Leibniz on this point, Kant was quite concerned that Leibniz’s own account failed to do justice to the principle involved, a concern Kant first made clear in his “Reflections on Optimism” (1753–1754). On Leibniz’s account, as Kant understood it, possibilities are “regarded as a necessary object of Divine Wisdom” (Ak. 2:151).34 But the possibilities, nonetheless, also seem to possess a certain kind of “independence” from God. “All possibility is spread out before God, God beholds it, considers it, examines it . . . ” (Ak. 17:237). The thought that God might be choosing amongst possibilities that are not themselves explicitly materially grounded in the fundamental realities that God exemplifies was already enough to raise Kant’s suspicions.35 But on Leibniz’s account, it also 30  I have amended the Gregor translation as necessary. Schneewind cites this footnote, highlighting its suggestion of some “analogy with the divinity,” but without examining the account it provides of the divine will (Schneewind 1998, 512). For an early expression of this account, see Kant’s remarks (1763–1764) in his personal copy of his Observations: “The supreme reason to create is because it is good. From this it must follow that since God, with his power and his great cognition, finds himself good, he also finds it good to actualize everything that is possible thereby. Second, that he has satisfaction in everything that is good for something, but the most [satisfaction], however, in that which aims at the greatest good. The first is good as a consequence, the second as a ground” (Ak. 20:33.13–34.3). 31  Timmermann insists that holy wills cannot act on maxims, citing Ak. 5:79 and emphasizing Kant’s use of the subjunctive at Ak. 5:32; but Timmermann seems to overlook Ak. 4:439 (Timmermann  2003). Kant actually makes some conflicting claims about the possibility of divine desires, maxims, and ends. In these and other passages (esp. Ak. 5:79) Kant sometimes claims that God has no desires and no maxims or ends (at least in the ordinary or “psychological” sense of those terms), but in other places he insists that God does have desires and maxims, or at least something suitably analogous to them. 32 Leibniz, Theodicy §228. (In what follows, translations of Leibniz come from H and L.) 33 Leibniz, Causa Dei §6. 34  “For it is, in my judgment, the divine understanding which gives reality to the eternal verities, although God’s will have no part therein” (Theodicy §184, cf. §189, 335). Similar expressions may be found in Leibniz’s Mon 43 and in Wolff ’s vernunftige Gedanken §975 (Chignell 2009, 170). 35  Leibniz did hold that God possesses all fundamental perfections (DM 1; Mon 40–41). Leibniz may actually have entertained a view, closer to Kant’s, which does not end the explanation of possibilia

Patrick Kain  307 seems that “the rules, which aim at perfection, [conflict] with each other in their application”; some “necessary fatality” or “alien ground” outside of God seems to cause “the essential determinations of things to conflict with each other when they are combined together” (Ak. 17:237; Ak. 2:151). “[Leibniz] recognizes the action appropriate to the Supreme Wisdom by the fact that it chooses on the side of the best, just as the sailor sacrifices part of his cargo in order to save the ship and the rest of the cargo” (Ak. 17:236–237).36 While Kant was willing to concede that the choice of the best possible under such circumstances would still be wise and good, he thought such an account, including trade-­offs and internal conflict, capitulated on God’s “infinity and independence” (Ak. 17:236–237).37 Here Kant might have quoted Spinoza to make his point, arguing that Leibnizians: who hold that God does everything with the good in mind . . . seem to posit something external to God that does not depend upon him, to which in acting God looks as if it were a model, or to which he aims, as if it were a fixed target. This is surely to subject God to fate; and no more absurd assertion can be made about God, whom we have shown to be the first and the only free cause of both the essence and the existence of things.38

The charge is that the Leibnizian account fails to adequately defend the claim that a self-­sufficient being could have a rational will and be the creator of the world.39 Kant’s suggestion is that Leibniz (and, as we shall see, Leibniz’s opponents) failed to adequately consider that God, the ens realissimum, is “all-­sufficient” (allgenugsam): “this thought, of all thoughts the most sublime, is still widely neglected, and mostly not considered at all” (Ak. 2:151). An ens realissimum, a being that could be the material ground of all possibility, must itself possess (or in God’s understanding: “The dependence of everything upon God extends to all that is possible . . . the possibility of things, even of those that have no actual existence, has itself a reality founded in the divine existence” (Causa Dei §7–8). A similar thought can be found in Leibniz’s May 1671 letter to Wedderkopf: “these possibilities or ideas of things coincide rather with God himself.” But these are suggestions which Leibniz hardly developed; in particular, he never developed their implications in the way that Kant did. 36  For Leibniz’s own discussion of this classic example of “mixed action,” see Theodicy §324. 37  These concerns stay with Kant. See Ak. 28:1058, 1065, discussed at n. 49 below. 38 Spinoza, Ethics I.33 Scholium 2. (In what follows, translations of Spinoza come from Shirley in Spinoza 1982.) Leibniz wrestled with similar suggestions from Arnauld and Bayle, in Theodicy §218, ff. Leibnizians would insist that the divine ideas are not “external” to God, but Kant’s contention is that if they are not materially grounded in God, they are either materially grounded in something external to God or they are materially ungrounded, neither of which is acceptable. That representational thoughts require an actual “material” ground does not seem patently absurd; this basic requirement may come easily in the case of phenomenal qualities, whose exemplification may consist in their presence within a consciousness. There are, however, serious questions about whether the realities exemplified by God are varied enough or mundane enough to ground, directly or indirectly, the content of every real property found in creatures, and whether Kant has the resources to defend his claims about metaphysical harmony (Chignell 2012). 39  For an illuminating discussion of these issues at the heart of seventeenth-­century debates, see Nadler (2011).

308  Development of Kant ’ s Conception of Divine Freedom exemplify) all fundamental realities or perfections, possessing them necessarily and completely from itself. It lacks nothing, depends on nothing external to itself, and yet is, in itself, entirely sufficient for all other things. “Whatever exists, whether it be possible or actual, is only something insofar as it is given through him . . . ” Possibilities are not only the “necessary object of Divine Wisdom,” they must themselves be “a consequence of this Incomprehensible Being” (Ak. 2:151). If, as Kant maintains, the possibilities themselves are all rooted in God as their common material ground of possibility, then each must already be internally harmonious and they must be remarkably harmonious amongst themselves, regardless of appearances to the contrary (because all of the fundamental “realities” at issue are coexemplified by a single, perfect being). Moreover, they cannot be considered something “materially” external to God himself that might constrain him. God’s will, properly described, is not subject to disharmony or to any external constraint. This concept of God’s all-­sufficiency remained of crucial importance to Kant (Ak. 5:100). It is the strong conception of divine self-­sufficiency that underwrites Kant’s idea of “a will that is directed to objects upon the existence of which its satisfaction does not in the least depend” (Ak. 5:137). The negative point here may seem to be so strong as to be troublesome, however. Insofar as divine self-­ sufficiency excludes expected pleasure or satisfaction in the existence of the object as potential motives and requires that the necessary objects of divine wisdom be (or be derived from) God’s own realities, what could serve to ground and make intelligible God’s will to create (R 6059, 6068; Ak. 18:440, 441)? Spinoza contended that there could be no suitable answer here and that an adequate conception of divine self-­sufficiency is simply incompatible with and must exclude the idea of a divine will.40 In perhaps the only sustained discussions of this problem in Kant, Schneewind and Henrich each posited an elusive, though developmentally significant, Kantian response. Schneewind suggests that Kant: thinks that there must be a principle at work in the divine nature, and indeed only one principle; but [Kant] gives us no idea of what it might be [Ak. 2:125–126]. Whatever it is, however, it must allow God’s creation to be the outcome of his own inmost nature. It must explain how God acts – and God acts autonomously.  [Schneewind 1998, 496–497, emphasis added]41

40  Spinoza had argued that the idea of a self-­sufficient being with a will was incoherent. “If God acts with an end in view, he must necessarily be seeking something that he lacks” which would be contrary to his perfection (Ethics I, Appendix). Although Spinoza considers God to be a “thinking thing,” he denies both intellect and will to God (II.1; I.17 Scholium; I.33 Scholium 2). 41  Henrich also seems to conclude his investigation of this question with the thought that “die Auseinandersetzung Kants mit Leibniz’ natürlicher Theologie . . . nötigte dazu, dem Willen eigene, ihm selbst immanente Prinzipien zuzusprechen, die nicht auf den Intellekt zurückgeführt werden können” (Henrich 1963, 413). Henrich, too, avoids articulating what could serve as such a principle.

Patrick Kain  309 Schneewind is right that Kant’s conception of the divine will anticipates important elements of his mature conception of autonomy (even though the details of the latter remain controversial). Unfortunately, Schneewind’s interpretation of Kant’s account of the divine will fails, on its own terms, to really make rationally intelligible the divine volition to create.42 Schneewind’s interpretation eschews the identification of any rational grounds for God to create, and it isolates only pro­ ced­ural norms for divine volition, should it occur. While one can understand how a finite being, with needs and desires, may be directed to act by the moral law’s demand for universalization, or how, once finite beings exist, the law’s demand to respect existent finite ends-­in-­themselves can prompt action, it is unclear how either of these lines of thought could render intelligible the volition of an infinite, self-­sufficient being to create. It is left completely mysterious how God’s “inmost nature,” “autonomy,” or governance by the moral law might prompt God to will or create anything at all, much less to create this world in particular. In short, the account does not really help to make sense of divine volition, much less to make sense of it as a form of rational agency. In its historical context, this account would fail to deliver a substantive response to Spinoza’s objection to the idea of a rational divine will, and to creation, substantial pluralism, and divine providence (as opposed to fatalism). In systematic terms, the account would provide little to help against the contemporary “blind mechanism objection”: in the absence of a positive account of the divine will involving a rational principle, there is little to distinguish divine action from the behavior of a mechanism or the impersonal necessity of familiar necessary truths. This account of the divine will, were it Kant’s, would possess limited developmental significance: it might have influenced the development of Kant’s thoughts about rational agency in general and human agency in particular; but the gap at its core would leave Kant’s very development in this direction mysterious and its inadequacy would cast a shadow of suspicion over the account of autonomy it helped to birth.

4  Kant’s Positive Model of the Divine Will A careful re-­examination of Kant’s corpus, however, reveals a significantly more robust account of the divine will, one which renders intelligible the divine will to create and which suggests a cogent response to the blind mechanism objection. The passage from “Theory and Practice,” cited above, goes beyond the negative point already emphasized, to suggest a positive account of the divine will to create:

42  As Schneewind notes, on his interpretation, Kant ends up uncomfortably close to radical voluntarists, such as Descartes (Schneewind 1998, 496). Elsewhere, I argue that Schneewind also misreads a semantic claim Kant makes about the concept of perfection, misinterprets Kant’s theory of possibility, and mistakenly suggests that the account excludes all “preexistent goodness.”

310  Development of Kant ’ s Conception of Divine Freedom God’s “consciousness of his all-­sufficiency” leads him to “produce the highest good beyond himself ” (Ak. 8:280 n.). This suggestion can be filled out, with the help of notes from Kant’s lectures on rational theology and metaphysics.43 The key to unpacking Kant’s model can be found where his emphasis on divine all-­sufficiency is integrated with his innovative tripartite psychology.44 One of Kant’s proudest contributions to psychological reflection was the development of a three-­way distinction between the faculty of cognition, the faculty of being pleased or of satisfaction, and the faculty of desire or will, which he insisted are not reducible to a single faculty or power (Frierson  2005).45 In cognition, an object is represented by a subject. In satisfaction or dissatisfaction, the object is represented in relation to the subject’s condition as a whole. In desire and volition, the object is represented in relation to the subject’s causality. In Kant’s psychological theory, action springs from desire or volition, which is preceded by satisfaction or dissatisfaction, which is preceded by cognition. Not every cognition need give rise to satisfaction or dissatisfaction and not every satisfaction or dissatisfaction need give rise to a desire, but each desire or volition presupposes some satisfaction or dissatisfaction, which in turn must presupposes some cognition. It might appear that this trichotomy of faculties makes matters worse: not only must one explain the idea of the will of a self-­sufficient being, but now one must also explain how a self-­sufficient being could experience pleasure or satisfaction. Yet Kant insists that the latter must be discussed before the former if there is to be an ad­equate account of the divine will (Ak. 28: 1060). Kant’s model begins with God’s cognition of himself (Ak. 28: 1054). God represents himself to himself and is thus aware that he possesses all fundamental realities or perfections, is completely independent, and is lacking nothing. This self-­cognition grounds God’s original self-­satisfaction or complete well-­pleasedness 43  In the rest of this section, I rely primarily on the Religionslehre Pölitz, which is based on student notes of Kant’s lectures, most likely in 1783–1784. This manuscript has the advantage of being relatively unified, containing significant discussion of several of the key points, and stemming from the mid-­1780s. As I hope some of my cross-­references reveal, the account found there fits well with Kant’s published remarks and is also well attested in Kant’s Reflectionen and other sets of lecture notes. In particular, this account lies behind some of the reasoning in the second Critique, e.g. Ak. 5:131. 44  Again, there are important questions about the content and status of many of the judgments involved, which I cannot settle here. My goal here is to unpack Kant’s proposed model and examine it for a certain kind of coherence. 45  For the distinction in general, see e.g. Ak. 5:177; Ak. 28:245; 674–675; Ak. 29:877–878. Often Kant refers to the second faculty merely as “the feeling of pleasure or displeasure,” though he recognizes that such a name can be misleading in both the human and the divine case (Ak. 28: 1059, 1275, 335–336, 608, 798). Henrich notes that criticism of the monism of the Wolffian psychology can be found in works by Rudiger and Hoffmann (Henrich 1963, 413). The basis of a three-­faculty psych­ ology can be found in Baumgarten (Hilgard 1980). Part of Kant’s novelty is in his insistence upon the irreducibility of the three faculties. Henrich helpfully noted Kant’s insistence upon the presence of distinctively affective elements within theological psychology, found as far back as in the Nova dilucidatio and “Optimism” remarks, and also how this signaled a significant modification of the monistic, cognitive Wolffian psychology (Henrich 1963, 410–413, 424). Much of Henrich’s evidence from Nova dilucidatio may be specific to the human case (esp. Ak. 1:401), but the “Optimism” remarks seem more decisive (Ak. 2:34).

Patrick Kain  311 with himself, which is a contentment or acquiescence in himself (acquiescentia in semetipso), blessedness (Seligkeit), or beatitude (beatitudo) (Ak. 28: 1060, 1065, 1089; 1275–1276; 335–336; 608; 798; cf. Ak. 5:25, 118; Ak. 8:399 n.). This is a kind of satisfaction uniquely appropriate to a self-­sufficient being. Since God is the ens realissimum and thus the ground of all possibilities, his self-­cognition and his well-­pleasedness with himself can also serve as the basis of the cognition, satisfaction, and volition sufficient for his creation of something distinct from himself: The self-­sufficiency of God, connected to his understanding, is all-­sufficiency. For in cognizing himself, he cognizes everything possible which is contained in him as its ground. The well-­pleasedness [Wohlgefallen] of a being with itself as a possible ground for the production of things is what determines its causality. The same thing can be expressed in other words by saying that with God, the cause [Ursache] of his will that things external to him shall exist consists precisely in his highest self-­contentment [Selbstzufriedenheit], insofar as he is conscious of himself as an all-­sufficient being. God cognizes himself by means of his highest understanding as the all-­sufficient ground of everything possible. He is most well-­pleased with his unlimited faculty as regards all possible things, and this self-­ contentment is indeed the cause [Ursache] why he makes these possibilities actual. Hence it is this which is God’s desire [Begierde] to produce things external to himself. The product of such a will will be the greatest whole of everything possible, that is the summum bonum finitum, the most perfect world . . . In a being which is independent and thus self-­sufficient as well, the ground [Grund] of its volition and the desire that additional things, external to itself, should exist is precisely that it cognizes its own faculty of actualizing things external to itself.  [Ak. 28:1061–1062]46

God’s representation of a possible (good) thing, eminently contained within his representation of himself, and his representation of his capacity to create something distinct from himself are the ground for God’s volition to create. Since the fundamental realities or perfections that possible good things might exemplify to a limited degree are fully exemplified by God, and since God is well-­pleased with himself, he may also be well-­pleased with the intrinsic nature of the possible goods, the goodness represented “in the idea” of the possible good things.47 Kant is insistent that, given God’s self-­satisfaction, this well-­pleasedness is a sufficient ground of God’s volition to create. No further well-­pleasedness, no interest, need, or “subjective relation” is necessary:

46  Here I make several minor modifications to the Wood translation. 47  Indeed, as Kant argued in The Only Possible Argument, the striking harmony and unity within and among the possibilities themselves is a result of their possessing a common, harmonious and unified material ground of possibility in God himself (esp. Ak. 2:110–112, 126 n.).

312  Development of Kant ’ s Conception of Divine Freedom [God’s] pleasure which he has in the perfection of an object in the idea, combined with the consciousness of himself as a sufficient ground of every ­perfection, already determines his causality.  [Ak. 28: 1101]

In one sense, God is quite indifferent to the existence of anything distinct from himself: because he is independent and self-­sufficient, he does not need its existence and its existence cannot really affect him. Yet precisely because of his complete independence, self-­contentment, and omnipotence, the absence of any additional, positive interest or subjective relation between God and the represented good poses no barrier to his willing its existence. A finite agent, in addition to being well-­pleased in its judging the goodness of a possible good, must also be well-­pleased in its choosing to actualize one alternative in the face of others. Every finite rational being has such “subjective conditions” for end setting; their cognition and appreciation of a possible good is not by itself a sufficient ground for willing (Ak. 5:450). Such is not the case with God. For God, the representation of and satisfaction with a possible good, already eminently contained in his representation of and satisfaction with himself, are all that is necessary for his volition to create. The grounds of his willing are his understanding of and well-­pleasedness with the possible goods (of which he would be formal and efficient cause) and his capacity to produce them. He wills them and actualizes them in virtue of their goodness in the idea, not in virtue of any distinct, anticipated goodness of their existence, nor in virtue of any anticipated satisfaction that God would derive from their existence.48 Kant contends that there is a unique, optimal sum of compossible finite goods and that such a best possible world, the finite highest good, will be the product of God’s cognition, satisfaction and will. Only a world chosen in accord with the “rule of the best” would be “worthy” of God’s choice (Ak. 2:109, 111; Ak. 1:404). Again: “The product of such a will will be the greatest whole of everything possible, that is the summum bonum finitum, the most perfect world” (Ak. 28: 1061). Kant proceeds to provide moral grounds for claiming that such a world

48  On Kant’s account, insofar as God cognizes the goodness “in the idea” of the best possible world, he does will to create it. Is this a sufficient explanation of God’s reason to create? Why could God not just enjoy the idea of such a perfect world without creating anything? Kant thinks that this must be sufficient reason to create: requiring, or even allowing, anything more would compromise God’s independence, blessedness, and perfection. God’s “pleasure in the perfection of the thing in its idea alone” must suffice; “the perfection of the thing it wills to produce is by itself sufficient motive [Bewegungsgründ] for him actually to put it into practice.” God need not “hesitate,” and has no need for “subjective incentives” to choose; no “further subjective pleasure in the existence of this thing had to be added” (Ak. 28:1101; also R 6071, Ak. 18:442). Thanks to John Hare and Andrew Chignell for raising questions on this point. (This need not entail that it is not better, in some sense, that a perfect or maximally good world exist than merely be conceived or represented; but it does seem to entail that God’s volition is not in response to such a further value. Kant does think that the will of a creator that chose to create something other than the best possible world would be less good than God’s will; but it is not clear that this claim plays any role in God’s decision to create the best possible world either.)

Patrick Kain  313 must involve the union and harmony of virtue and happiness in proportion to it (Ak. 28: 1099–1102).49 It is useful to pull together the several moments involved in this model of the divine will. (i) God, the ens realissimum and summum bonum originarium, cognizes himself, cognizing his self-­sufficiency and the way in which he contains, even exemplifies, every fundamental perfection or reality and is the material ground of everything possible and how he can be the formal and efficient cause (and probably a final cause) of everything actual; that is, he cognizes his all-­ sufficiency. This self-­cognition grounds (ii) God’s well-­pleasedness with himself, his appreciation of his own perfect goodness, as self-­sufficient and all-­sufficient. It is this well-­pleasedness with himself, including his well-­pleasedness with the goodness in his representation of the things for which he represents himself as the possibly sufficient cause (rather than any anticipation of well-­pleasedness that could arise for his creatures or for him from his creatures or their existence), that grounds or constitutes (iii) his willing of the greatest sum of possibilities, and this willing is the ground of actuality of (iv) the world, indeed the most perfect world, a world in which the highest good is (or can be) realized. I submit that this is the model Kant has in mind when he writes that God must be thought “to be determined to produce the highest good beyond himself just by his consciousness of his all-­sufficiency” (Ak. 8:280 n.). As Kant suggests in the second Critique, the “highest original good,” God, through his representation of himself, is the formal and efficient cause of the “highest derived good,” a world in which the harmony of virtue and happiness can be realized (Ak. 5: 125,131, 132). Kant thinks this model defuses charges of fatalism, at least with respect to God. (As we will see more clearly below, it seems to simultaneously commit Kant to a certain kind of necessitarianism with respect to God.) Pace Schneewind, Kant does isolate a rational principle that renders intelligible God’s creating and Kant provides an account of its relation to God himself: God’s rational principle for creating is himself. God’s cognition of himself, which eminently contains the goodness of the best possible world, provides his sufficient reason to create. It is this explicit account of God, the ens realissimum, as himself the unified material ground of all possibility who creates out of his all-­sufficiency, that Kant thinks is the decisive elaboration of or amendment to the traditional rationalist conception 49  Kant persistently maintains that we cognize a priori that a perfect creator would create the (single) best possible world—which is the highest finite (moral) good—even though it is only on moral grounds that we can cognize that there is a perfect creator and even though it is difficult to reconcile this claim with our experience of the empirical world (e.g. Ak. 8:279, 5:434–436, 442–445; 28: 701, 809, 1097, 1102). What Kant comes to insist is that his form of optimism falls short of the objectives of a theodicy. Kant also insists that God’s choice of the best possible world does not involve compromising between competing values, as he took Leibniz’s account to do. There are no trade-­offs imposed by anything external to God, and there is no collision within the divine will (Ak. 28: 1058, 1065). I take this to be one of the key points dating back to the “Reflections on Optimism.” (It is debatable whether Leibniz himself really intended to suggest otherwise.)

314  Development of Kant ’ s Conception of Divine Freedom of God, that renders God’s creative action sufficiently rationally intelligible so as to avoid fatalism and substance monism, while upholding the commitment to God’s independence. Let us return to consider the implications of this account for the blind mechanism objection. Necessary perfection or “unswerving compliance,” it was suggested, seems indistinguishable from the behavior of a mere blind mechanism or the bruteness of an arbitrarily chosen metaphysically necessary truths, and thus fails to meet the requirement that genuine rational agents must actively participate in their compliance with rational norms. What Kant maintains, however, is that a proper account of metaphysical modality is tied up with an account of divine all-­sufficiency, that substance pluralism requires divine volition, and that a coherent account of such divine volition makes ineliminable reference to God’s guidance by God’s own normative representations (e.g. his representations of goodness, goodness which he himself exemplifies). On Kant’s account, possibility and necessity are not, at least not in general, dependent upon God’s will; but some necessities are indeed so-­called “moral necessities,” which are dependent upon or constituted by God’s will (which may itself be, at least in certain important respects, noncontingent).50 Whatever the precise epistemic status of this model of the divine will, it motivates a distinction between divine agency and mere blind mechanism or brute impersonal necessity, insofar as God’s normative representations are not “practically impotent,” epiphenomenal, or ex­plana­ tor­ily irrelevant. Without any appeal to an ability to do otherwise, Kant has provided a model of how a perfect and self-­sufficient will could be normatively guided. (Indeed, as we will see more clearly below, Kant explicitly denies God has any ability to do otherwise.) A self-­sufficient ens realissimum may spontaneously guide itself in light of its full appreciation of its own reality, a goodness which extends to the goodness of its possible object. Indeed, Kant contends, creation (and substance pluralism) cannot be conceived at all, absent such normative self-­guidance. Within this conception there is little room for complaints about a “mechanism” or “alien cause” or merely “accidental” conformity to an “external standard” to gain a foothold. Absent a showing that this conception of rational agency is logically incoherent, it seems to provide a robust defense of the idea of a necessarily perfect self-­sufficient rational agent against the blind mechanism objection.

50  In general, “internal” possibility does not depend upon acts of God’s will. “Moral necessity” is identified with what God necessarily wills. Without additional precision about the nature of “internal” possibility and necessity, the precise modal status of divine volitions may be difficult to determine. Specified one way, “internal necessity” may exclude God’s volitions by definition; specified another way it may trivially entail the internal possibility and/or internal necessity of any potential divine vol­ ition. For another potential route to universal necessitarianism, see e.g. Ak. 5:402; R 6020, Ak. 18:425.

Patrick Kain  315

5  Human Freedom, the Problem of Dependence, and Alternate Possibilities Having fleshed out Kant’s conception of the divine will and having addressed the “blind mechanism objection,” we may now briefly consider a few ways that this conception of the divine will sheds light on the nature and development of Kant’s conception of freedom more generally, and of human freedom in particular. As we have already noted, in one of his earliest works, Kant suggested that the power to be formal and efficient cause through representations of the good is necessary for freedom (Ak. 1:402). In fact, in the Nova dilucidatio, Kant had insisted that this must suffice for freedom and argued that there can be no “liberty of indifference,” and no genuine alternate possibilities for any free being. Kant maintained that both divine actions and human actions meet this condition and are, thus, free, yet he argued that every contingent thing, including every action, human or divine, must be fully and certainly determined by antecedent grounds and its opposite impossible and excluded by those antecedent determining grounds. This is part and parcel of the “principle of the determining ground” which Kant had insisted upon at that point (Ak. 1:397–400). Kant soon came to insist, however, that this cannot constitute an adequate account of freedom, or at least of human freedom. By the early 1760s, Kant reversed himself on the “principle of the determining ground” and suggested that some changes within the world (e.g. human free actions), while “not properly understood,” “appear to have about them an indeterminacy in respect of determining grounds and necessary laws” (Ak. 2:110). “In human beings, the chain of determining causes is in every case cut off ” (R 3855, Ak. 17:314). “The human power of choice is not determined by any grounds” (R 4226, Ak. 17:465). As he suggested in the 1780s, “The human being could always decide something else, e.g., instead of being benevolent in this case, could also not be that” (Ak. 28:1068). “The human being can also do otherwise than he does” (Ak. 28:1280, cf. Ak. 6:50 n.). When Kant came to insist that all events in the empirical world are determined according to natural laws, he also insisted that noumenal indeterminacy remained theoretically possible and was indeed required by morality. It may be tempting to think that Kant adopted a form of “leeway incompatibilism,” or a fundamentally indeterministic libertarianism, with respect to noumena or things-­in-­themselves. Yet, Kant continued to consider his original, “necessitarian” conception of the divine will to be a more or less sufficient (if not fully explicit) account of divine freedom. Not only is it not possible that God perform an action that is morally wrong, there seems to be no possibility of God not acting precisely as he in fact does. As Kant had put it in his Nova Dilucidatio (1755): “The act of creation . . . is so certainly determinate [in God] that the opposite would be unworthy of God, in other words that the opposite could not

316  Development of Kant ’ s Conception of Divine Freedom be ascribed to Him at all” (Ak. 1:400, cf. Ak. 2:34). In Nova dilucidatio Kant had been quite insistent that this necessity is compatible with divine freedom: “the action is free, for it is determined by those grounds, which, insofar as they incline His will with the greatest possible certainty, include the motives of His infinite intelligence, and do not issue from a certain blind power of nature to prod­uce effects” (Ak. 1:400). This is a point he steadfastly maintained: “The divine power of choice is determinate . . . determined secundum [rationes] intellectuales” (R 4226, Ak. 17:465).51 “The divine will cannot decide otherwise than as it does . . . Here there is in God a practical necessity” (Ak. 28:1280).52 As Kant suggested in the Groundwork, “reason infallibly determines” the divine will (Ak. 4:412). The point is also reiterated in the quick sketch from “Theory and Practice” (1793): “God [must be thought] to be determined to produce the highest good beyond himself just by his consciousness of his all-­sufficiency” (Ak. 8:280 n.). While Kant insists that God is “determined” but not “necessitated,” and that God’s creative action is not absolutely necessary, these are technical points about God’s nature and the metaphysics of modality, not a rejection of all forms of ne­ces­si­tar­ ian­ism. Kant seems to be suggesting that God and his volition to create are what explains why it is metaphysically necessary (as we might say) that he create, rather than that God’s reason to create is that it is metaphysically necessary (as we might say) that God create. In many respects, the necessity is just as “strong” as and is as fully grounded in God as “absolute” and “internal” necessity is, even if it is a distinct, and differently grounded modality. Yet God’s rational volition is meta­phys­ ic­al­ly foundational, rather than impotent or epiphenomenal.53 So, Kant came to think that human freedom requires that humans can do other­wise, but divine freedom precludes the possibility of God doing otherwise.54 51  Cf. R 4337, Ak. 17:510. 52  “God’s will is a free will. God wills the best and cannot do the contrary . . . The impossibility of doing something that would be contrary to the law of the good is the highest freedom. The divine necessity of action is not a natural necessity, but rather moral necessity, that is, it springs from absolute spontaneity” (Ak. 28:806, cf. related modal claims at Ak. 28: 1061, 1062, 1068, 1097). This seems to express an unqualified determination of God’s power of choice or decision (even if his other powers may be conceived to have a broader scope). Kant seems to foreclose any alternate possibilities, in an unqualified sense, for God’s choice. 53  Kant’s distinction between the “absolute necessity of God’s nature” and the “moral necessity” or “practical necessity” of his free will is primarily a distinction concerning the role of representation in God’s causality. One might worry that this is not a distinction that can make enough of a modal difference—in particular enough of a difference to secure the ontological distinction between God and the world he supposedly creates. At this point, Kant may be in roughly the same boat as Leibniz and Wolff, who also appeal to the distinction between blind necessity of nature and a moral necessity with respect to “things possible in themselves.” Yet, if Kant can coherently reconcile noumenal indeterminism of the human will with divine necessitarianism (esp. by excluding human choices from divine control), then he has a further set of resources for resisting both universal necessitarianism and monism. This position is, in any event, in marked contrast with the views of Crusius, who insisted that there can be no best possible world, and that God has both libertas contrarietas and contradictionis and could even have created the world at a different time (Crusius 1745). Kant alleges that Crusius’s conception of freedom, which allows and involves divine volition without sufficient rational grounds, is another version of fatalism (Ak. 28: 1281). 54  Kant does refuses to define freedom in terms of an ability to do otherwise or to act contrary to the law (Ak. 6:226), and he sometimes characterizes the capacity to act contrary to reason as a “lack of

Patrick Kain  317 The question is: why did Kant come to adopt and maintain this differential account of free agency, and what, if anything, unifies this account? Kant’s account of the divine will helps to shed some light on these questions. A significant answer focuses on a long-­standing concern of Kant’s, which we find expressed in the second Critique: insofar as the inner determining ground of a creature’s action is placed in him by his creator, insofar as “his existence and the entire determination of his causality absolutely depend” upon God as “cause of the existence of substance,” and the all-­sufficient “being of all beings,” the determining ground of a creature’s actions would seem to be “altogether beyond his control,” determined by an alien cause, and not free (Ak. 5:100–101; Ak. 6:142).55 The thought seems to be that the presence of “alternate possibilities” is needed to defend the possibility of genuine creaturely self-­control and responsibility in this theological context: If the human being is to be a free creature and responsible for the development and cultivation of his abilities and predispositions, then it must also be within his power to follow or shun the laws of morality. His use of freedom has to depend upon him, even if it should wholly conflict with the plan God designed for the moral world.  [Ak. 28:1113]

On this line of thought, creatures endowed with only one principle of action, were that possible, would fail to meet the relevant “control” condition for freedom and responsibility, and this would be so regardless of the content of that single principle of action. While such beings could be the proximate cause of their actions or behaviors, their actions would be attributable to and the responsibility of their creator, the being who implanted that single principle of action in them, rather than themselves. God, and not the creature, would be the source of the creature’s behavior: If God rules the determinations of the power of choice, then he acts; if the charms of things necessarily determine it, then they necessitate; in both cases the action does not arise from me, rather I am only the means of another cause. [R 4225, Ak. 17:464]56

freedom” (Ak. 28: 1068). Nonetheless, in many of the same texts (e.g. Ak. 28: 1068, 1113), he suggests that free creatures must have this capacity. What is a bit harder to discern is the possible significance of this point. 55  As Kant had explained in the mid-­1770s, “if I assume: [the soul] is a being derived from another , then it appears to be quite probable that it is also determined by this cause in all its thoughts and actions . . . that it indeed acts freely according to the inner principle, but is determined by a cause” (Ak. 28:268). The outlines of this problem can be found at least as early as R 3857, 3872 (mid-­1760s); and R 4218–4221, 4225–4226 (1769–1770). 56  Kant may have had a similar thought in mind when he suggested that freedom, the capacity to act “according to our own will,” must be the basis of our dignity, not reason alone (Ak. 27:1322).

318  Development of Kant ’ s Conception of Divine Freedom The presence of creaturely needs and their significance within a creaturely agent’s practical reasoning help to satisfy a necessary condition of free and responsible creaturely agency.57 For a creature, self-­control, agency, and responsibility are made possible by the room left for us to choose our character, to choose to prioritize our dignity and our “own nature” (or “proper self ”), or to prioritize our needs and wants instead (Ak. 4:435, 457, 461; Ak. 5:87; Ak. 6:21–22; Ak. 8:263–264; Ak. 28:1077–1078, 1287–1288). In contrast, for an “independent” and self-­sufficient being (Ak. 5:100), there can be no question about volition being in anyone or anything else’s control or influenced by anything alien or external; there is nothing external to him that is not the result of his will. Nor is there any criterion of goodness outside of God himself. The “grounds” of God’s choice are rational, internal, and not causally influenced by anything external to God. No alternate possibilities are needed here; indeed, Kant reasoned that the idea that God could do or could have done otherwise may be completely excluded. There need be nothing incoherent in having such a differentiated account of divine and human freedom.58 Underlying this difference is a common, core concept of freedom that applies to both: a free rational agent has the capacity, within its own control, to be an efficient cause, independently of all “alien causes,” through its representations of the good or of the moral law. The difference between divine freedom and human freedom (or, more generally, the freedom of a creature) is a reflection of the way that the requirement of self-­control must be satisfied differently in the cases of creator and creature. In the divine being, freedom requires, and is thus compatible with, complete and ultimate determination; in a creature such as a human being, freedom requires the absence of complete and ultimate determination (although again, Kant thinks, for reasons not explored here, that creaturely freedom is compatible with complete empirical determination by natural causes). How such beings may be thought to coexist raises many further questions that must be left for another occasion.59 57  For a distinct, yet related, point about duality and blame, see R 3872. 58  Visser and Williams have recently argued for a similar claim about Anselm: that the insistence upon alternative possibilities within Anselm’s “doctrine of the two affections” is necessary for created (but not divine) freedom, in order for creaturely actions to be genuinely self-­initiated. They have suggested that conjoining a conception of divine freedom that excludes alternate possibilities with a conception of human freedom requiring alternate possibilities presents an intriguing compromise between considerations supporting and opposing the “principle of alternate possibilities” (Visser and Williams 2001). My suggestion is that Kant found himself defending such a combination of views, for this sort of reason. I have seen no indications Kant derived this position directly from Anselm or from his own predecessors. Neither Wolff nor Crusius, for example, held such a position. 59  Within this account of creaturely agency, Kant intends to leave conceptual room for some kind of dependence within creaturely agency and for the possibility of divine concurrence and divine [fore]knowledge of human free actions, however mysterious those things must remain (see e.g. Ak. 8:263–264; 8:361–362 n.; Ak. 5:448–449 n.; Ak. 28:347, 1106, 1110, 1113; R 6057, etc.). Questions about these issues come to the surface in the second Critique (Ak. 5:100–103), the Religion (Ak. 6:142) and the Metaphysics of Morals (Ak. 6:280 n.).

Patrick Kain  319

6 Conclusion I have argued that Kant’s conception of free divine rational agency can be considered an advance within its historical context (esp. in the Leibnizian battle against Spinozism) and that it is responsive to the contemporary blind mechanism objection. This conception sheds important light upon Kant’s account of freedom and rational agency more generally, just as Kant suggested.

12

Leibniz and Kant on Empirical Miracles Rationalism, Freedom, and the Laws Andrew Chignell, Princeton University*

Miracles are therefore not against nature, but rather against what is known of nature. Augustine, City of God xxi, 81

1 Introduction Most monotheists join everyone else in regarding created nature as a stable and efficient structure: its laws do not require tweaking, and its states do not capriciously alter. Many monotheists also orient their picture of the world, however, by texts depicting a deity that is willing to intervene and suspend nature’s normal operations on certain occasions: Then Moses stretched out his hand over the sea, and the LORD drove the sea back by a strong east wind all night and made the sea dry land, and the waters were divided. And the people of Israel went into the midst of the sea on dry ground, the waters being a wall to them on their right hand and on their left. [Exodus 13: 21–22]

As a result of texts like this, biblical theists (laypeople as well as professional theologians) may find it reasonable to hope or even believe that an occasional em­pir­ic­al miracle—that is, a physical event at variance with the normal causal order—has occurred or will occur. Both Leibniz and Kant were heirs of this tradition. But both were also explanatory rationalists about the empirical world: more committed than your average *  Thanks to participants in the New York–New Jersey Early Modern Colloquium, the PrincetonBucharest Seminar in Early Modern Philosophy a workshop at the University of Miami, and the Leibniz: Reception and Relevance conference in Lisbon for feedback on drafts of portions of this chapter. I am also grateful to Jan Cover, Alexander Englert, Don Garrett, Brendan Kolb, Brandon Look, Colin McLear, Eleonore Stump, Joshua Watson, and Eric Watkins for additional conversations and ­correspondence. It would be an empirical miracle if there were no remaining errors; those I claim as my own. 1  “Portentum ergo fit non contra naturam, sed contra quam est nota natura.” Andrew Chignell, Leibniz and Kant on Miracles: Rationalism, Religion, and the Laws. In: Leibniz and Kant. Edited by Brandon C. Look, Oxford University Press. © Oxford University Press 2021. DOI: 10.1093/oso/9780199606368.003.0012

Andrew Chignell  321 philosopher to its thoroughgoing intelligibility. (Leibniz was also an explanatory rationalist about the non-­empirical, fundamental world; on that issue, given his commitments to transcendental freedom, Kant famously demurred.) On its face, such intelligibility would seem to preclude any intervention from outside the system that would lead to a violation of the natural laws. These dual allegiances—to supernaturalist tradition and to empirical rationalism—thus generate a powerful tension across both philosophers’ systems, one that is most palpable in their accounts of empirical miracles. Neither philosopher was unaware of the tension. Of the two, Leibniz makes the more concerted effort to accommodate the traditional portrait of God as miraculously intervening in the natural world. Leibniz’s goal is effectively to save the appearances of the traditional religious doctrine of miracles, or at least avoid direct conflict with biblical orthodoxy, without giving up his arch-­rationalist principles. My claim in this chapter is that his effort is at best a partial success. Leibniz’s way of marrying these two commitments is coherent—he can have his miracle doctrine and eat his deterministic cake too. However, the marriage comes at the cost of a certain amount of epistemic inhospitableness, at least for finite minds. This raises questions about whether it could really obtain in the best possible world—one whose bestness is partly measured by the “happiness” of such minds. Kant, by contrast, did not say all that much about how miracles could be integrated into his natural philosophy, and this has led many commentators to assume that he was not seriously endorsing their possibility. In the second part of the chapter, however, I draw on some of his critical notes and lectures to sketch a way in which he might think the integration goes. If the sketch is accurate, then Kant (surprisingly enough) offers a view of God’s relationship to nature that is similar to Leibniz’s in important ways, and moreover does not face the same obstacles in accommodating the doctrine of miracles that Leibniz does. A secondary goal throughout the chapter is to show that I conclude with the even more surprising suggestion that the structure of Kant’s system leaves room the idea that our finite freedom, too, may be the transcendental ground of ­empirical miracles. A comparative exam­in­ation of the status of the doctrine of empirical miracles in Leibniz and Kant provides a deeper understanding of their respective philosophies of nature in general.

2  Miracles, Wonders, Signs In order to proceed, we need a working conception of a miracle in the natural, empirical world. There are treacherous debates in this area, many of which have to do with whether such events count as “violations” of the natural laws or not.2 J. L. Mackie offers a formulation that sidesteps some of these issues: an empirical 2 For a sampling, see Swinburne (1970), Mavrodes (1985), Earman (2000), Luck (2011), and Twelftree (2011).

322  Leibniz and Kant on Empirical Miracles miracle, he says, is an event that occurs “when the world is not left to itself, when something distinct from the natural order as a whole intrudes into it.” By “natural order” Mackie means the order described by natural laws or “regularities”—the principles that “describe the ways in which the world—including, of course, human beings—works when left to itself, when not interfered with” (Mackie 1982: 19–20). In addition to the characteristically tetchy phrasing, Mackie’s formulation has significant virtues: it captures a lot of what religious people mean when they talk of miracles (although they would avoid terms like “intrude”, which make God sound like an impolite dinner guest), and it is consistent with the broadly Humean conception of miracles as (roughly) exceptions to nomological regularities. This is no surprise, since Mackie was a fairly orthodox Humean.3 But the formulation also fits with older conceptions of miracles as events in nature that are beyond the productive power of any natural thing. It also allows us to remain neutral about what the “laws” of that nature consist in, and how, if at all, they might be violated. Best of all, the formulation is endorsed not only by Mackie, a well-­known critic of theism, but also by a contemporary religious philosopher/apologist, Timothy McGrew, who cites Mackie’s formulation approvingly in a survey article on mir­ acles (McGrew 2019, 4–5). So, it is an irenic conception, as well as a philo­soph­ic­ al­ly interesting one. In light of these virtues, I propose to use the following Mackie/McGrew Miracle (MMM) analysis as our working conception: (MMM)  An empirical miracle obtains when a being that is not part of the order described by the laws of nature purposively intervenes to produce an event that counts as an exception to at least one of those laws. A few clarifications: first, “empirical” is intended to restrict the domain to events or states in the part of the world that empirical science seeks to describe and explain (via observation and causal inference, for instance). There may be moral, soteriological, or eschatological miracles too but, unless they make an empirical difference, they are not our focus here.4 Second, it is important to insist (following Mackie and McGrew) that the intervention produces an exception to the laws of nature; otherwise, ordinary divine concurrence with natural events would also count. We will see in a moment that Leibniz regards such concurrence as miraculous in some sense, whereas Kant does not. But this is not part of the folk concept of an empirical miracle, and most

3  Hume goes on to say that nomological regularities must be regarded as exceptionless regularities, but that inference is widely contested. See Earman (2000) and Fogelin (2005). 4  I have written about Kant’s conception of a “moral miracle” elsewhere: see Chignell (2013). See also Ameriks (2014). On Leibniz’s view of moral miracles, see Paull (1992).

Andrew Chignell  323 philosophical discussions (including those of Mackie and McGrew) follow Kant in excluding it. Third, it is worth noting that Latin terms like “miraculum” and “portentum” are ambiguous: they elide key distinctions we find in other languages between “miracle,” “wonder,” and “sign” or “portent.” For one thing, lots of wonders—that is, extraordinary natural events—can cause extreme psychological effects in n ­ ormal observers: astonishment, shock, awe, reverence, etc. (in German these would be  “Bewunderungen”). But such wonders need not count as genuine miracles (“Wunder”): even the Seven Wonders of the World, impressive as they are, ­presumably came about through quite explicable natural processes Conversely, although many miracles count as wonders, others may be undetectable, and still others may be so commonplace that we cease to wonder at them. Thus, even though many luminaries in the t­ radition (Aquinas, Hobbes, Clarke5) build psy­cho­logic­al (“wonder”) or even informational components (“signs and portents”) into their analyses of a miracle simpliciter, it is better to follow Leibniz and Kant here in keeping them apart.6 At bottom, as our Mackie/McGrew (MMM) conception indicates, the category of the miraculous is an ontological rather than a psy­cho­ logic­al one. A bona fide empirical miracle, as Kant puts it, “interrupts [unterbricht] the order of nature” (Ak. 2:116).7 With MMM in place as our working conception, we can turn directly to Leibniz and Kant and see how they incorporate the possibility of empirical mir­ acles into their rationalist-­determinist pictures of the empirical universe.

3 Leibniz 3.1  Five Miracle Concepts Leibniz’s ontology is notoriously complex: one scholar detects six (!) different ­levels of reality in the system: from monadological bedrock to phenomenal illusion.8 Leibniz scholarship is further complicated by debates about chron­ology—not 5  See Aquinas 1265–1274: I.110.a4. Hobbes speaks of miracles as “signs supernatural” (Hobbes 1651: I.xii.28) and Clarke says that they are “unusual” events produced by God “for the Proof or Evidence of some particular Doctrine, or in attestation to the Authority of some particular Person” (Clarke 1823 [1704]: 2.701). 6  See LC V 89 (Leibniz 2000, 57) for the distinction between “perpetual wonder” and genuine “perpetual miracle.” Leibniz explicitly rejects Clarke’s attempt to include a psychological component in the basic concept; he also rejects Clarke’s claim that miracles must be “unusual” or infrequent (cf. LC V 110 [Leibniz 2000, 62] and Clarke’s Fifth Reply [Leibniz 2000, 82]). In a paper from July 1698, there is a reference to the distinction between “a rare and wonderful thing,” which may still be naturally produced, and a genuine miracle “which exceeds the powers of created being” (L 494). 7  This is from a pre-­Critical essay of 1763, but there are similar descriptions in various Critical lectures. See V-­Met-­L1 (Ak. 28:217ff.), V-­Met-­Mron (Ak. 29:870ff.), V-­Met/Dohna (Ak. 28:667). 8  Anja Jauernig, in conversation.

324  Leibniz and Kant on Empirical Miracles only about whether Leibniz was a full-­blown metaphysical idealist, but also about when he became one, if at all.9 These debates in turn are riddled with orthographical disagreements about whether and when Leibniz wrote a particular text or letter. It would be quixotic for me (especially qua Kant scholar) to try to establish a position on these issues here. Instead, I will simply assume (on the good authority of some Leibnizean friends) that it is acceptable to speak in terms of two main levels in Leibniz’s mature ontology.10 The first is the fundamental level comprising unified substances in mutual relations of expression and perception. For the later Leibniz of Monadology (1714), at least, these substances are immaterial, simple, psychological unities (“monads”) whose successive states are pre-­established by God to “express” the entire history of the universe, though at varying degrees of distinctness. Monads aggregate in various ways, but there are no genuine causal relations between them. The second main level is the derivative, physical one comprising objects of our experience as well as the particles that make them up. These objects—namely, bodies—fill up space and are related by forces in lawful ways that are somehow a function of the expression relations that hold at the monadic level. The question before us, then, is how (if at all) an empirical miracle could fit into this scheme. Given the (MMM) analysis, such an event would clearly have to manifest at the derivative level—it would have to make a difference to the way things are in the empirical world. But how could that be the case if the world’s pre-established harmony makes it the best one possible? In order to answer this question, we have to look at the way Leibniz characterizes miracles in various texts. The relevant passages, in my view, reveal Leibniz working with at least five distinct concepts, although the last three have the same extension. I will briefly sketch each concept before going on to examine how they might have instances in the empirical world as Leibniz envisaged it.

3.1.1  First Rank In Discourse on Metaphysics (1686), Theodicy (1710), and the correspondence with Clarke (1715–1716), Leibniz says that miracles of the “first rank” or “highest order” are the result of God’s immediate action and involve no creaturely contribution. He cites three paradigmatic examples: creation, annihilation, and incarnation.11 I will consider creation and annihilation first, and then say something about the trickier case of incarnation.

9  See e.g. Garber (2009). 10  Leibniz’s own references to two “kingdoms” (of nature and of grace) reinforce this strategy. See also Bennett (2005), McDonough (2008). 11  DM 32 (AG 63–64); Theodicy §249, H 280; LC IV 44 (Leibniz 2000, 27).

Andrew Chignell  325 God alone bears genuine causal relations to other substances in the actual world, for Leibniz, and thus God alone can be involved in creating and annihilating. Such acts do not count as interventions into the order of nature, however, since they effectively ground that order. In other words, because the natural order is just a function of the natures of the substances in the world, the choice to create and annihilate certain substances is tantamount to the choice of that order. If God had created different substances—or created but then annihilated some of them along the way—God would have ipso facto selected a different world with a different order. These two miracles of the highest rank, then, are not empirical mir­ acles in our Mackie/McGrew sense.12 What about God’s role in keeping the world in being? There is controversy about whether the early Leibniz was a “mere conservationist,” but by the mid-­1680s he was clearly committed to the doctrine that God has ongoing causal responsibility for producing finite substances and their states. He speaks of “concurrence” (concours) throughout the Discourse and, later, of God’s “continual production” of the world and the divine “assistance” that all “natural powers” require.13 There are at least three major conceptual obstacles, however, to counting or­din­ ary concurrence as a miracle of the first rank. 1. For starters, is not clear how God can count as producing the effect in question immediately, since concurrence involves the production of a finite effect in concourse with creatures. Perhaps we can say, following Robinet and Adams, that God concurs with , and thus counts as immediately concurring with that entire state of affairs.14 But stacking the effects in this way still threatens either to subsume the creature’s agency into God’s, or to imply that producing is again a cooperative affair rather than something that God does immediately. (Insofar as I can think about this clearly at all, I find it hard to believe that I am not involved in the production of the state of affairs !) A possible solution to this puzzle invokes the Leibnizean point that finite creatures are only virtual and not real causes at all—they have appetitions for and perceptions of effects in other substances, but do not provide any of the causal oomph. So perhaps we can say that God is indeed immediately producing the complex effect by way of the pre-­ established

12  See Sleigh (1990, 58–67) as well as Adams (1997a, 277ff.). Elsewhere, Adams (1994, 99ff.) discusses the question of whether God could, in some broadly logical sense, decide to annihilate a substance at some point “later” than creation, and thereby change the order of things. Whatever the correct answer, it is clear that in the best possible world this will not occur, given God’s commitments to harmony and the best. 13  LC V 88 (Leibniz 2000: 57) and §112 (Leibniz 2000: 62). See also Specimen Dynamicum (AG 124) and Theodicy §27 (H 90). 14  See Adams (1994, 97–98) and Robinet (1986, 440). Adams cites Robinet here.

326  Leibniz and Kant on Empirical Miracles harmony, but the creaturely natures determine which effects God immediately produces (perhaps by offering God reasons).15 2. Setting that issue aside, the concurrence doctrine is also hard to square with Leibniz’s oft-­repeated charge against occasionalists that their deity is a fussy micro-­manager, unfittingly engaged in “perpetual miracles.”16 For if God concurs with every finite exercise of every power, and if every such act counts as a miracle of the first rank, then perpetual miracles seem unavoidable on his view as well. Two further considerations blunt the edge of the latter objection. First, Leibniz would surely have seen that concurrence involves constant activity on God’s part; the fact that he happily endorses it indicates that his real problem with occasionalism is not that God perpetually acts, but rather that creatures have no role in the production of their own states. Leibniz’s central anti-­occasionalist commitment is that creatures have active powers of their own—the very powers with which God concurs. This is part of what underwrites the ontological distinction between creatures and God, and thus also part of what fends off Spinozism (see e.g. De Ipsa Natura in AG 155ff.). Second, there is an important sense in which ordinary concurrence is nothing over and above creation for Leibniz. Since God actualizes the best collection of compossible substances at creation, and since every truth about every substance is derivable a priori from its essence—for the infinite intellect, anyway—God’s choice of these substances just is the choice to actualize the states intrinsic to them over time (cf. Lee (2004)). Thus, Leibniz refers to God’s ongoing activity in the Theodicy as a kind of “continued creation” (Theodicy §27, H 139). In a letter to Clarke from the same period he says that “natural things” are not the result of “perpetual miracle” but rather the “effect or consequence of an original miracle worked at the creation of things,” even if they are the occasion for “perpetual wonder” (LC V 89, in Leibniz 2000, 57). 3. Even if we set these complications aside and count ordinary concurrence as a divine act that is distinct from creation, however, it still will not involve a change to the order described by the natural laws. On the contrary: concurrence is both consistent with that order and a condition of its obtaining. And thus it will not be an empirical miracle in the MMM sense.17 Something similar can be said about the third and final kind of first-­rank mir­acle—namely, incarnation. Although this miracle involves an individual substance exemplifying two kind-­ natures (divinity and humanity), the joint exemplification of these kind-­natures will be part of the individual nature of the 15  See Lee (2004) for an articulation of the “offering reasons” view. For an extended discussion see Jorati (2017). 16  See e.g. Theodicy §207, H 257. 17  See the discussion of “conservation as continued creation” in Adams (1994, 95–99), as well as Lee (2004).

Andrew Chignell  327 substance in question. In other words, in the incarnation a human being with two kind-­natures is produced and conserved as part of the best possible world. But creating a substance that has the nature of both humanity and divinity does not clearly require an interruption or change to the normal natural order. In sum: miracles of the first rank do not in themselves count as empirical mir­ acles in the Mackie/McGrew (MMM) sense. They do not (in themselves18) make an empirical difference, and thus do not generate the tension with Leibniz’s explanatory rationalism with which we are presently concerned.

3.1.2 Comparative At the other end of the spectrum are “miracles only by comparison” to what human beings can do (Theodicy §249, H 280). These feats are performed “through the ministry of invisible substances, such as the angels,” and many of the biblical miracles are said to fall in this category (ibid.). There is scholastic precedent here: Aquinas claims that “although the angels can do something that is outside the order of corporeal nature, yet they cannot do anything outside the whole created order, which is essential to a miracle” (Summa Theologica, Aquinas 1955: I.110. a4). Leibniz, however, is not willing to allow even angelic acts to surpass the laws of corporeal nature; rather, he views biblical episodes such as Peter walking on water, the water-­to-­wine wonder at Cana, or the mysterious movement of the pool at Bethesda as in accordance with the laws of bodies—just “bodies more rarefied and more vigorous than those we have at our command” (Theodicy §249, H 280). It would be nice to know how water changes to wine without a suspension of the natural laws (do the angels move so fast that they can replace the water with wine without the guests noticing?). Clearly, however, if we allow that such events are physically possible, they will not pose a problem for Leibniz’s explanatory rationalism about the empirical world.19 So there is no conflict here between Leibniz’s rationalist commitments and his effort to save the appearances of bib­ lical religion. 3.1.3  Beyond Nature’s Power, But Still Within Nature Where the conflict does seem to lie is in the most prominent conception of mir­ acle in Leibniz’s middle and later writings. This is the one that he calls his “philosophical” conception—it is simply that of an event that “exceeds the powers of 18  This qualification is important, given what comes below, since God does of course create and concur with any events that count as miracles, and so in that sense creation can lead to an empirical miracle. The point here is simply that the event’s miraculous status is not a function of the fact that it is an effect of divine concurrence. 19  See LC IV 44 (Leibniz 2000, 27); LC V 117 (Leibniz 2000, 63). Joshua Watson provides a sophisticated account of how Leibniz seeks to accommodate miracles of this sort “semantically” even while rejecting the idea that they are miracles “in metaphysical rigor.” See Watson 2012a and 2012b).

328  Leibniz and Kant on Empirical Miracles created beings” (L 494). Leibniz tells Arnauld that “strictly speaking, God performs a miracle when he does a thing that exceeds the forces that he has given to and conserves in creatures” (LA 116). Likewise, in a letter to Conti, Leibniz defines a miracle as “any event that can only occur through the power of the cre­ ator, its ground not being in the nature of creatures” (Leibniz 1899, 277). This conception, like the previous one, has scholastic roots: Aquinas says that “a miracle properly so called is when something is done outside the order of nature” (Aquinas 1955: I.110.a4).20 Leibniz is careful to note that by “nature” in this context he means not just our natures, or the natures of the substances we know about, but rather “all limited natures”—including those of angels, demons, rarified bodies, and so on (DM 16; AG 49; cf. LC III 17; Leibniz 2000, 17). So, they are not merely comparative miracles. These passages indicate that it is too strong to say, with Nicholas Jolley, 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.  [Jolley 2005, 125]

In fact, Leibniz envisions continuity between the realms of nature and grace such that many miracles simply surpass the productive power of nature insofar as they result in some change “outside” of empirical nature but not within it.21 Miracles of the first rank are also miracles of that sort. The question we are asking now, however, is whether there are events that go beyond nature’s power while still remaining within nature. If there are, then Jolley’s bi-­conditional would hold with respect to them, since any empirical event that is beyond the causal powers of finite beings would indeed constitute an exception to the natural laws (which are, after all, just descriptions of what created beings can do). I will discuss this ­concept further in §3.2 of this chapter.

3.1.4  Contrary to the “Subordinate Maxims” A distinct though related concept of miracle is found most prominently in the Discourse on Metaphysics of 1686 and other writings from that period.22 Leibniz repeatedly claims there that miracles are “above the subordinate maxims” of God’s

20  Marilyn McCord Adams points out that “outside” (praeter) here means something like via a different route. So, for God to act outside of nature is “to produce effects that nature can produce, but not that way.” She also notes that Aquinas anticipates Kant’s view that creation and other acts that “lie entirely outside the range of natural causal powers” are not properly-­speaking miraculous (Adams 2013, 12–13). 21  This is not to say that all acts of grace are beyond the productive power of nature. Again, Leibniz views nature and grace as on a kind of continuum. Thus, moral punishment and reward—even in the afterlife –might be accomplished through the “mechanical” effects of our physical behavior over time (Mon 88, AG 224). 22  Garber draws our attention to the 1686 unpublished essay “De natura veritatis” (translated in Leibniz 1973). He also points out that Leibniz seems to vacillate even during this period on whether the principle of the equality of cause and effect (from which follows the law of conservation of force) is “genuinely inviolable” or not. See Garber 2009, 254–5.

Andrew Chignell  329 will—that is, above the contingent laws of nature—even though they are still “in conformity with the universal law of the general order” (DM 16; AG 48–49). Elsewhere in this work it becomes clear that at least some of these miracles are not just above but positively “contrary” to the “subordinate maxims which we call the nature of things” (DM 7; AG 40). He repeats this formulation in a letter to Arnauld of July 14, 1686: “miracles are contrary to some subordinate maxims or laws of nature” (LA 57, my emphasis). If we take this talk of contrariety seriously, then this fourth kind of miracle appears to be slightly different from the previous one in virtue of necessarily involving an exception to the laws. It is the sort that the Scholastics called “contra naturam”: [A miracle] is called contra naturam when there remains in nature a disposition that is contrary to the effect that God works, as when he kept the young men unharmed in the furnace even though the power to incinerate them remained in the fire, and as when the waters of the Jordan stood still even though gravity remained in them. [Aquinas, De Potentia q.5, art.2, ad.3; trans. Freddoso 1991, 573]

In the natural world, however, this concept and the previous one will have the same extension. They both pick out all and only the events within nature that are beyond nature’s power, contrary to the way nature normally works, and thus exceptions to the natural laws. Regarding these, again, Jolley’s biconditional is accurate. The fact that empirical miracles are exceptions to the natural laws conceived as subordinate maxims does not mean that they are entirely unlawful, according to Leibniz. As we saw above, the “true” or “most general order” of things—which he sometimes calls the “essential law of the series”—describes what actually does and must happen, and any miracles, as well as the subordinate maxims, will a fortiori be “derived” from it (Leibniz  1973, 99–100; cf. DM 7, 16). So even if there are occasional violations of the natural laws, they pose no genuine threat to determinism, much less to rationalism. But, of course, the law of the series is not ac­cess­ible to finite minds, and so our inquiries (even the most precise sciences) are limited to describing the “laws” qua subordinate maxims. I will return to this issue, too, in §3.2 of this chapter.

3.1.5  Extraordinary Concurrence Leibniz’s fifth and final concept of miracle invokes God’s “extraordinary and miraculous concurrence” with the powers of creatures.23 On this conception, God concurs with something in creatures to produce an event that is an exception

23  DM 16 (AG 48–49). See also the letter to Arnauld of April 30, 1687 (LA 115), Theodicy §3 (H 74–75), and the letter to Caroline of 1715 (L 675).

330  Leibniz and Kant on Empirical Miracles to the laws set out by the natures of those very creatures. This seems, at first glance anyway, to be in direct opposition to the third concept according to which a mir­ acle is “beyond” the powers of creatures altogether. In order to grasp the tension here, consider the biblical case that Aquinas invokes in the passage just quoted. Leibniz was presumably aware of the in­ter­ pret­ation put forward by Aquinas—as well as Molina and Suarez after him— according to which, when the three captive Israelites enter the Babylonian furnace, the dispositions of fire to burn (and of human hair, skin, and flesh to be burned) do not receive God’s ordinary concurrence. It is because such concurrence is a necessary condition of the usual combustive effects that the three young men are not combusted.24 For the scholastics, this is more or less the whole story: the active disposition of fire to burn and/or the passive dispositions of skin and hair to be burned are rendered inert, because God declines to concur with them. Leibniz, however, views this as robbing creatures of power and portraying God, unfittingly, as in conflict with his own creation. God never withholds concurrence altogether, according to Leibniz; rather, God actively concurs with the powers of creatures in an extra­or­ din­ary way. Is there is anything further to say here about what it is for finite natures to receive extraordinary concurrence? Does the fire in the furnace (on a Leibnizean account) have an active, albeit rarely activated, disposition to cause, say, spring-­ breeze sensations rather than painful burning sensations in human minds? If so, then are the fires in that particular furnace the only ones that have the extra­or­din­ ary powers in question? Or do all fires have them, even though God concurs with them in just a very few cases? There is something unpalatable in each of these alternatives. If Leibniz allows that the extraordinary powers are not really in the finite substance at all, then in those cases, at least, he is departing from his anti-­occasionalist, anti-­Spinozist principle according to which the forces responsible for creaturely states are at least partly in the creatures themselves, rather than wholly in God. If he says, on the other hand, that the extraordinary powers are in the creatures, but only in those through which miracles actually occur, then the account looks rather ad hoc. Finally, if he says that the extraordinary powers exist in every creature of the relevant kind, then we are left with a bloated ontology: vast arrays of powers strewn across numerous different species and individuals, even though most of them are never activated. He also faces questions about how the active powers exercised in miracle cases count as extraordinary—apart from the comparatively trivial fact that God does not usually concur with them.

24  For discussion of Molina and Suarez on this issue, see Freddoso (1991).

Andrew Chignell  331 It is unclear which of these alternatives Leibniz could swallow, or how it could be made more palatable. What is clear is that he does not regard the mere stat­is­ tic­al infrequency of an event as sufficient to make it miraculous: he insists in a letter to Clarke that there is a “real difference between a miracle and what is nat­ ural.” Leibniz also says that this difference must be “internal” to the creature somehow, and not merely an “extrinsic denomination” in God (LC V 110–112; Leibniz 2000, 62; see also LA 116 and A VI 4, 587). This seems to rule out any suggestion according to which God allows the exercise of the ordinary powers but then “blocks” their effects.25 A more promising suggestion, in my view, is that extraordinary concurrence involves God taking one or more of the preexisting powers of a creature and strength­ ening or increasing it.26 Perhaps the fire stays the same as it was, but the ordinary, flame retardant powers of Abednego’s skin are strengthened to the point where it can resist the Babylonian inferno. This is a slight variation, but it makes the account look less ad hoc: extraordinary concurrence builds on the natural powers already present in creatures but also goes “beyond” them. Human skin does have a very limited power to resist flame, even in ordinary circumstances; in the Babylonian furnace, God augments that power, making it go beyond the dermatological norm. This is an appealing line of thought. A remaining concern, however, is that the increase to a creature’s ordinary powers itself lacks a positive explanation or ground in the creaturely natures themselves. This would again run afoul of Leibniz’s general principles that everything in creatures—every perception and every change—must be entirely explained by their natures, and that the difference between the ordinary case and the extraordinary case cannot be a mere “extrinsic denomination” in God. The tension here—between the concept of miracles as beyond the productive power of nature and yet grounded somehow in the natures of things—is a slight but real one. We will return to it in the next section. Having sketched Leibniz’s five concepts of miracle, and having set aside the first and second for present purposes, we can now consider whether the last three concepts can coherently apply to empirical events in the best possible world. Can Leibniz maintain that the best and most fully intelligible world could contain events that are beyond nature’s power to produce, and perhaps even contrary to nature’s powers?

3.2  Nature vs. Essence: Three Strategies Recall that at the derivative, physical level, the world for Leibniz is composed of aggregates of matter and the dynamic relations between them--relations that are 25  This suggestion is from Eleonore Stump, in conversation. She views this as Aquinas’ position. 26  Don Garrett proposed this way of interpreting extraordinary concurrence in correspondence.

332  Leibniz and Kant on Empirical Miracles governed by the laws of nature. This fact itself is no bar to miracles: we have already seen that they would be exceptions to those laws considered as the sub­or­ din­ate maxims or “customs” of the divine will (DM 7; AG 40).27 The main obs­ tacles to empirical miracles arise, rather, from Leibniz’s long-­standing opposition to occasionalism and his commitment to the principle of perfection. The former pushes him, as we have seen, to say that there is a ground in the finite things— most fundamentally in their substantial forms—of the presence of every state that they exemplify. But the latter is also problematic: how can the intelligible natural laws that we seek in scientific inquiry admit of exception? For would not the supremely rational, competent, and benevolent engineer make the elegant laws that we (approximately) grasp also be the laws that really govern the series, especially given the fact that Leibniz explicitly ties the perfection and happiness of minds to their ability to understand the phenomena?28 There are passages in the Discourse that appear to indicate that Leibniz recognized these problems and endorsed the position that there cannot be any mir­ acles, if by “miracle” we mean changes to the natures of finite beings: When we include in our nature everything that it expresses, nothing is supernatural to it, for our nature extends everywhere, since an effect always expresses its cause and God is the cause of substances.  [DM 16; AG 49]

Appearances are misleading here, however, since Leibniz goes on to say that it would be better to use “essence or “idea” to refer to the collection of all of a substance’s properties, and reserve “nature” in the strict sense for that collection of properties that a substance “expresses more perfectly” and “in which its power consists.” In other words, Leibniz proposes to think of the nature of a substance as a function of its active powers—powers that are themselves “limited,” of course— whereas its overall essence contains “many things that surpass the powers of our nature and even surpass the powers of all limited natures” (ibid.). Likewise, in the Theodicy we are told that “the distinguishing mark of miracles (taken in the strictest of senses) is that they cannot be explained by the natures of created things”— here “nature” is presumably being used in the strict sense (Theodicy §207, H 257). Leibniz is not consistent about this terminology, and in many places before, after, and even within the Discourse, he uses “nature” to refer to the broader essence as a whole. If we keep the terminology straight, however (as I will try to 27  Unless the natural laws are construed as conditionals whose antecedents explicitly invoke God’s ordinary concurrence. On such a view, the antecedent is satisfied in ordinary situations, and unsatisfied in miraculous cases; either way there is no exceptions to the laws. See Watson 2012a for an in­ter­ pret­ation of Leibniz along these lines. 28  See e.g. “A Specimen of Discoveries” (1686?) in Leibniz (1973, 83), as well as the further discussion below.

Andrew Chignell  333 do here), the nature/essence distinction may make room for empirical miracles that are beyond the productive power of finite substances, but still included in their essences.29 In order to exploit the distinction in the manner just described, we would need an account of how the properties included in our finite natures count as being “within our power” in a way that the features of our broader essences are not. Leibniz clearly does not mean to say that we consciously choose whether or not to exemplify our natures—many features attach to us without conscious ­volition or appetite, and there are some substances that have no conscious ­appetites at all at a given time (slumbering monads). “Within our power” also cannot mean that these features are changeable somehow: for Leibniz, each of a substance’s properties is essential—or at the very least intrinsic—to it as the individual that it is.30 It’s worth noting that in “Primary Truths” he says that the “concurrence of grace” is inscribed into the essences of finite things but not into their “nature” in the narrower sense (AG 32). But it would be nice to have a deeper account of this.

3.2.1  Wholly Passive Power One approach to this problem would be to say that the properties of our nature are “within our power” because at some level we want to have them. Leibniz says that the primary force or active power of a substance is directed to the series of states in its nature only via conscious or unconscious appetition. This allows him to say that the substance is wholly passive with respect to the remaining extra-­ natural properties of the broader essence. This proposal (call it Wholly Passive) would explain why Leibniz inserted the words “or idea” after “essence” in the second edition Discourse’s discussion of this issue (DM 16; AG 48); it also coheres with his general doctrine that miracles are willed directly by God alone. But as we saw earlier, Wholly Passive threatens to eliminate the positive ground in creatures for the contents of the broader essence: the ground now lies completely in God’s idea of the creature.31 And this is difficult to square with texts in which Leibniz indicates that all of our states, and not just the states of a narrowly circumscribed nature, are the partial result of our primary, active force. For example: I believe that there is no natural truth in things whose ground ought to be sought directly from divine action or will, but that God has always endowed 29  Elsewhere he makes it clear that what he calls the “concurrence of grace,” is inscribed into the essences of finite things but not into their “nature” in the narrower sense (“Primary Truths,” AG 32). 30  See Sleigh’s discussion of the difference between the conventional “superessentialist” reading of Leibniz and his own “superintrinsicalist” reading in (1990, ch. 7). See also Adams’ critical notice of Sleigh’s book (Adams 1997). 31  Adams presents this view without endorsing it at (1994, 87ff).

334  Leibniz and Kant on Empirical Miracles things themselves with something from which all of their predicates are to be explained. [Specimen Dynamicum, AG 125]

Advocates of Wholly Passive could emphasize that Leibniz says here that “there is no natural truth in things” whose ground is solely in God’s will, but then argue that this leaves room for the occasional supernatural truth that is so grounded. It is not clear what they would say, however, about passages where he leaves out such qualifications:32 In my system every simple substance (that is, every true substance) must be the true immediate cause of all its actions and inward passions; and, speaking strictly in a metaphysical sense, it has none other than those which it produces. [Theodicy §400, H 362]

It is possible that Leibniz is speaking very generally here and thus bracketing the special case of miracles. But it would be nice to find an interpretive strategy that could accommodate both kinds of text. Moreover, there are passages in the Discourse that indicate that finite creatures must always have some degree of active force moving them towards a given perception: The soul must actually be affected in a certain way when it thinks of something and it must already have in itself not only the passive power of being able to be affected in this way (which is already wholly determined) but also an active power, a power by virtue of which there have always been in its nature marks of the future production of this thought and dispositions to produce it in its proper time. [DM §29; AG 60, my emphases]

If we interpret “thought” here as referring to all of our psychological states— including those that are in our broader essence but not our nature strictly speaking—then this passage makes it difficult to see how any state of the substance could be wholly a result of the active powers of other substances.

3.2.2  Degrees of Active and Passive Power A better approach, I submit, is to take seriously the thought that all the states of a substance are somehow grounded in its active powers, but then exploit the fact that such powers come in degrees. Call it the Degrees of Power proposal. A few pages before the passage just quoted from Specimen Dynamicum, Leibniz says that substances have both active power (virtus) and passive power, and that in 32  Another example: “[I]n my opinion it is in the nature of created substance to change continually following a certain order which leads it spontaneously (if I may be allowed to use this word) through all the states which it encounters” (L 493, my emphasis).

Andrew Chignell  335 finite creatures they are “found in different degrees” (AG 119). Likewise, much later in Monadology Leibniz characterizes active force as a perfection and passive force as an imperfection, and indicates that both come in degrees (Mon §48–52; AG 219). Applying Degrees of Power to the problem at hand, we can say that miracles “surpass the power of nature” insofar as finite substances have only a very limited degree of active power with respect to those events.33 Indeed, perhaps this is true of all the states or events that are only in the essence: they are the result of a very limited degree of creaturely active power, a degree much lower than the degree of passivity that the same states have vis-­à-­vis God. The advantage here is that we can maintain that there is no state of a substance that is not to some degree a function of its active powers, while also making use of the nature-­versus-­essence distinction to account for miracles. A principled line demarcating the states of our nature (“within our power”) from the states that are only in the broader essence (“outside our power”) would be hard to draw precisely, but it would have to fall comfortably beyond the point where the degree of active power directed towards the relevant states is exceeded by the degree of passive power vis-­à-­vis God.34 A lingering problem for both Wholly Passive and Degrees of Power, however, is that for Leibniz a change in a substance that is either wholly or largely a result of its passive powers is ipso facto a change from the more perfect to the less perfect (see again Mon §49–50; AG 218). But if a miracle involves such a change, then its occurrence will decrease a finite substance’s overall perfection by “demonstrating its weakness” and causing it “some pain” (DM 15; AG 48). It is clear that the weakness and pain that Leibniz has in mind here is primarily cognitive—it involves a decrease in the happiness that a finite mind takes in understanding the order of things: Thus, to speak more clearly, I say that God’s miracles and extraordinary concourse have the peculiarity that they cannot be foreseen by the reasoning of any created mind, no matter how enlightened, because the distinct comprehension of the general order surpasses all of them. On the other hand, everything we call natural depends on the less general maxims that creatures can understand. [DM 16; AG 49]

33  Thanks to Anja Jauernig (conversation) for emphasizing the utility of an appeal to the degree of an active power in this context. For further discussion of degrees of power, see Look (2007). 34  Note that this would have to be a necessary but not a sufficient condition for a state’s counting as miraculous; otherwise, as Sam Newlands pointed out in conversation, the states of a slumbering monad (all or most of which, at least during the slumber, are the result of its passive rather than its active powers) would have to count as miraculous. Other necessary conditions in (MMM), then, are that the state or event occurs in the empirical world of bodies and also counts as an exception to the natural laws.

336  Leibniz and Kant on Empirical Miracles The claim that no finite mind, “no matter how enlightened,” could “foresee [a miracle] by reasoning” indicates that no inference based on known laws could lead to the conclusion that the miracle will occur. This feature of miracles—their unintelligibility to finite minds—is a consequence of the ontological facts: the divine ground of the miraculous event is too complex for us to understand. An initial concern about this unintelligibility doctrine is crudely empirical. Could not someone (Abednego and co., for instance, or even Nebuchadnezzar himself) reasonably expect that a miracle might well occur in a certain case, given their previous dealings with Yahweh? In other words, could not the unintelligibility of the event be made intelligible by what someone knows about God? Perhaps probabilistic prediction such as this would not count as “foreseeing by reasoning” in Leibniz’s strict sense. But even granting that it is not a sufficient basis for a deduction, it is not clear why this does not count as gaining some kind of limited foresight—a sort of “understanding,” at least in our contemporary sense, that is deeper and better than a lucky guess. Another and more substantive concern here is that the unintelligibility doctrine is, again, in tension with Leibniz’s principle of perfection. For recall that the happiness of finite minds is one of the primary variables which determine the goodness of a world, and that such happiness is at least partly a function of how well these minds understand the phenomena: It is clear that minds are the most important part of the universe, and that everything was established for their sake; that is, in choosing the order of things, the greatest account was taken of them, all things being arranged in such a way that they appear the more beautiful the more they are understood. So it must be held certain that God has taken the greatest account of justice and that just as he sought the perfection of things, so he sought the happiness of minds. [“Specimen of Discoveries,” MP 83]

If our happiness as intelligent beings consists partly in understanding the beautiful order of the universe, then, oddly enough, the perception of a miracle will leave us frustrated and discontent at some level. In other words, miracles will lead, again in the language of DM 15, to more “weakness” and (epistemological) “pain”, even if they on balance contribute to the good of the whole (AG 48). This sort of trade-­off is theologically unattractive, and may explain why Leibniz often appears to downplay even the possibility of empirical miracles.35 There is, finally, a related concern about how the miracles doctrine (on either of the proposals sketched here) coheres with Leibniz’s overall claim that the world is arranged in the simplest possible way. This is a concern about the metaphysical elegance of things as opposed to the effect that inelegance has on finite minds. 35  On the trade-­offs between metaphysical and moral perfection, see Lin 2011.

Andrew Chignell  337 Robert Adams raises this problem with reference to the passage in DM 16 according to which miracles make the general order so complex that it is incomprehensible to creaturely minds: “How that is consistent with the preeminent simplicity of the actual general order, Leibniz does not explain, so far as I am aware” (Adams 1994, 86).36

3.2.3  Nature and Essence Collapsed? In light of these sorts of concerns, some commentators opt to collapse the Discourse’s distinction between natures and essences altogether. Donald Rutherford, for instance, claims that for Leibniz “any substance is endowed with an intrinsic force or power sufficient to determine all of its own states or modifications” (Rutherford 1995, 302, my emphasis). There is no talk of degrees of active and passive power here, and no suggestion that these degrees could be used to demarcate a substance’s nature (strictly speaking) from its broader essence. In a cowritten piece, Rutherford and Jan Cover likewise claim that: Leibniz’s naturalism is specifically intended to rule out the possibility that phys­ ic­al 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 nat­ ural phenomena can be explained in terms of the action of powers inherent in the natures of created substances.  [Cover and Rutherford 2005, 7, my emphases]37

If by “natural phenomena” in the second sentence here the authors mean phenomena that can be subsumed under the laws of nature, then this is correct but trivial. But the first sentence indicates that they mean that no “physical or psy­cho­logic­al” phenomena whatsoever could be produced by special intervention on God’s part, and that all such states are a result of the active powers of natures, with which God simply concurs in the ordinary way. That in turn indicates that by “natural phenomena” these authors mean all of the phenomena that do or can occur in the natural world. This is a non-­trivial claim, and it seems to rule out empirical miracles. In support of their model, Rutherford and Cover could cite the passages considered earlier from Specimen Dynamicum or Discourse §29, for instance, or the following: In every substance there is nothing other than the nature or primitive force from which follows the series of its internal operations. This series, i.e. all of its past 36  See also Gregory Brown’s discussion of the tension between miracles and human happiness (Brown 1995, 24ff.). 37  These passages are hard to square with Rutherford’s earlier claim that “genuine miracles” count for Leibniz as “an important class of exceptions” to his overarching “principle of intelligibility” (Rutherford 1995, 240–241). It is possible that Rutherford’s view has changed here.

338  Leibniz and Kant on Empirical Miracles and future states, can be recognized from any state of the substance, i.e. from its nature.  [A VI 4, 1672–1673]38

Passages such as these suggest that all the phenomena in nature, and all the states of a substance, are completely determined by the active forces which constitute natures strictly speaking. There is no room, on this view, for an appeal to the broader essence. Although there are attractions to this collapse of the nature/essence distinction, the case for it is not compelling. For one thing, the Degrees of Power proposal may be able to handle the text just cited by maintaining that the entire “series” of phenomena does follow from the primary force of a finite substance, provided we take into account even the smallest degree of force. Moreover, we have seen numerous other texts in which Leibniz links the natural powers of creatures to the laws of nature, and then allows that other sources of change in those creatures are possible. Such a change would be accounted for by the general law of the series but not by the subordinate maxims/laws of nature—in other words, it would be an empirical miracle. This is true not only in the relatively early Discourse, but also in the Letters to Arnauld (LA 116), the New Essays of 1704 (1996: 66) and the very late Theodicy: Thus it is made clear that God can exempt creatures from the laws he has prescribed for them, and produce in them that which their nature does not bear by performing a miracle. [Theodicy §3, H 7, my emphasis]39

In each of these contexts, Leibniz insists that there is a “nature” (strictly speaking) in creatures with respect to which certain divinely instituted changes would be wholly external. The collapse of the nature/essence distinction thus faces serious textual challenges.

3.3 Summary The goal of this part of the chapter was to see whether Leibniz has a way to in­corp­or­ate empirical miracles into his overall picture of nature, and thus to eliminate the apparent tension between traditional biblical monotheism and

38  Joshua Watson cites this passage on behalf of the view that empirical miracles are not actual for Leibniz. He departs from Cover/Rutherford, however, in holding that they are at least metaphysically possible. See Watson 2012a, ch. 5). 39  Indeed, in the Theodicy Leibniz still expresses some openness to the view that the infusion of reason in a merely “sentient soul” is miraculously performed through “some special operation, or (if you will) through a kind of transcreation” (Theodicy §91, H 173). See Brown (1995, 28ff.) for a discussion of this issue.

Andrew Chignell  339 rationalist-­determinism about the physical universe. Recall the Mackie/McGrew conception of empirical miracle with which we are working: (MMM)  An empirical miracle obtains when a being that is not part of the order described by the laws of nature purposively intervenes to produce an event that counts as an exception to at least one of those laws. The puzzles we have been considering can be summarized in the form of a dilemma: (A)  If empirical miracles (construed in the MMM sense) are included in creatures’ essences but not in their natures, then we are left with no positive ground of the relevant states in the substances themselves: any miracles become “extrinsic denominations,” which seems contrary to the doctrine that all changes in substances have a ground in the active powers of those substances. (B)  If essences are collapsed into natures by construing both as fully grounded in the active powers with which God simply concurs, then Leibniz’s system is unable to accommodate the possibility of empirical miracles, and the many texts in which he continues to speak of them must be regarded as confused or disingenuous. The Wholly Passive proposal sketched above, according to which miracles are solely the result of God’s activity, amounts to an embrace of the first horn of the dilemma. Degrees of Power, however, seems to offer a way between the horns. Again, natures and essences can be demarcated by considering the degree of active power involved in a given change of state, and if a finite substance’s degree of active power with respect to some change is less than its degree of passive power vis-­à-­vis God, then the change may count as an empirical miracle. In other words, the state is outside the creature’s nature strictly speaking and thus not “within” its power, even though it is grounded in a very limited degree of creaturely active power. The Degrees of Power proposal also explains how Leibniz can regard empirical miracles as surpassing the power of natures (in the strict sense), and as exceptions to the laws qua subordinate maxims, and yet as the result of extraordinary concurrence with some degree of active power in creatures. The proposal does not, however, resolve the epistemological puzzle regarding how the decrease in intelligibility of the world and happiness of rational creatures is consistent with its overall perfection. Perhaps the best thing to say on that issue, as usual, is that it is at least conceivable that all the other worlds God might have created would have been even less perfect overall. But it would surely be preferable from a religious point of view if Leibniz could accommodate the thought that the inclusion of a genuine miracle adds to a world’s perfection rather than detracting from it. So

340  Leibniz and Kant on Empirical Miracles even if Leibniz can squeeze empirical miracles of the MMM sort into his picture of nature, some of the tension between his rationalism and his religion remains. It is high time to turn to a discussion of Kant’s views on these issues. We will find that although Kant’s view of the connections between empirical events and creaturely natures differs from Leibniz’s in important respects, there is also a striking similarity between their respective models of empirical miracles. Furthermore, Kant’s two main departures from Leibniz—the doctrines of transcendental freedom and noumenal ignorance—allow his model to avoid some key problems that we have encountered in that of his predecessor.

4 Kant 4.1  Kant on First Rank and Comparative In order to do full justice to a comparison between Leibniz and Kant on this issue, we would need to look at Wolff, Baumgarten, and the other proximate Leibnizians (as well as non-­Leibnizians like Lessing, Reimarus, and Herder) with whom Kant was interacting. For the purposes of systematic comparison, however, it will save time and introduce no significant distortions, I think, simply to consider how Leibniz’s own conceptual scheme (as laid out above) relates to the model Kant developed some eight decades later.40 My argument will be that the models are structurally very similar: both involve an outside agent producing events that violate the normal natural order, even though they are in keeping with a deeper order that is mostly inscrutable to us. Kant does not face the same concerns that Leibniz does regarding the unhappiness of minds in the face of such inscrutability, since he never promised that we would know very much about things-­in-­ themselves in any case. His model also seems to leave open the possibility that not just divine but also finite free agents can perform actions that appear as events that violate the normal natural order. As unKantian as it sounds, I think Kant’s view commits him to the surprising claim that we too are in principle capable of breaking the “laws of nature.” Like Leibniz, Kant envisions God as not only the metaphysical ground of all possibility, but also the causal ground of all finite being. In other words, Kant 40  Quotations from Kant’s works are cited according to the Ausgabe der königlich preussischen Akademie der Wissenschaften (Kant  1902– ), with the two editions of the Critique of Pure Reason cited by the standard [A/B] pagination, and all other works cited as [Abbreviation, volume: page]. Here I have typically though not always used the translations in the Cambridge Edition of the Writings of Immanuel Kant, the general editors of which are Paul Guyer and Allen Wood. A list of further abbreviations can be found at the beginning of this volume. I have presented some of this material in a previous paper (previous but also subsequent, since that paper was actually written after this chapter was first completed; however, it was published much more swiftly). See Chignell (2014). I am grateful for permission to reuse portions of that paper here.

Andrew Chignell  341 views God as the creator of finite things-­in-­themselves, and thus of the features of those things of which the empirical world somehow counts as an appearance. Unlike Leibniz, however, Kant does not regard creation itself as a miracle: “what happens outside the world . . . is not a miracle, e.g. creation is no miracle.” A mir­ acle, rather, is “that which happens contrary to the order of nature in the world” (V-­Met-­Vron, Ak. 29:870); a miracle must “interrupt (unterbricht) the order of nature” (BDG, Ak. 2:116).41 Because creation is a condition of the existence of the order of nature, it “cannot be admitted as an occurrence among the appearances” (A 206/B 251–252), and is thus not an interruption of that order. This coheres nicely with our Mackie/McGrew conception of miracles. What about God’s role in keeping the world in being over time? There are obvious difficulties here given that Kant thinks we must consider not only God but all the other supersensible things-­in-­themselves as non-­temporal.42 All the same, Kant is willing to talk of “conservation” (Erhaltung) in this context, remarking in a lecture from the Critical period that “the same power required for the creation of substances is also needed for their conservation” (Pölitz 28:1104). Whether and how this conservation doctrine ultimately differs from Leibnizean concurrence is a matter of some dispute.43 Either way, however, Kant clearly does not regard such activity as miraculous: “Just as little [as creation] is conservation a miracle.—It is no event in the world” (V-­Met/Dohna, Ak. 28:667). As for the other miracle of first rank—incarnation—Kant spends the second “piece” (Stück) of his Religion book developing an account according to which it is not miraculous at all. Kant says more about Leibniz’s second conception—comparative miracles performed by finite spirits—than one might expect given his general opposition to “enthusiasm” about spirits. In various lecture discussions, as well as in the lengthy “Kiesewetter” fragment—called “On Miracles”—from the late 1780s, he distinguishes a “miraculum rigorosum, which has its ground in a thing outside the world (thus not in nature)” from a “miraculum comparativum, which to be sure has its ground in nature, but in one whose laws we do not know; of the latter sort are the things we ascribe to spirits” (Ak. 18:321; see V-­Met-­L1, Ak. 28:219, V-­Met/

41  This second phrase is from a pre-­Critical essay of 1763, but there are similar descriptions in various Critical lectures. See V-­Met-­L1, Ak. 28:217ff., V-­Met/Dohna, Ak. 28:667, Pölitz 28:1109. 42  “For in God only one infinite act can be thought, a single, enduring force which created an entire world in an instant and preserves it in eternity. Through this act, many natural forces were poured out, as it were, in this world-­whole, which they gradually formed in accordance with general laws” (Pölitz 28:1096; cf. Pölitz 28:1104). 43  A few pages after the passage just quoted, Kant is recorded as saying that “[i]n the same way there takes place no concursus of God with natural occurrences. For insofar as they are supposed to be natural occurrences, it is presupposed that their first proximate cause is in nature itself, and it must be sufficient to effect the occurrence, even if the cause itself (like every natural cause) is grounded in God as the supreme cause” (Pölitz 28:1106). See Hogan (Ch. 10 of this volume) as well as Brewer and Watkins (2012) and Insole (2013) for discussions of Kant on concurrence, freedom, and theological determinism. Lehner argues that for Kant God does not concur with events in nature but does concur with our free actions (Lehner 2007, 316 n.). For more on moral concurrence, see Chignell (2013).

342  Leibniz and Kant on Empirical Miracles Dohna, Ak. 28:667–668).44 In the Religion, too, Kant talks sincerely about “angelic” and “diabolical” miracles and seems to think that they are possible, though not easy to identify (Rel. Ak.6:86). But for Kant—as for Aquinas, Leibniz, and Newtonians like Samuel Clarke—comparative miracles (Wunder) are really just objects of wonder (Bewunderung): dazzling but fully naturalistic events caused by finite beings according to specific empirical laws with which we are not (and perhaps cannot become) familiar. It is in this context that Kant says that: when we ask what is meant by “miracle” (Wunder) (for us, that is in our practical use of reason) then one can say that they are events (Begebenheiten) in the world, whose laws of effects (Wirkungsgesetze) we are ignorant of, and must remain so. [Ibid.]

But this “practical” account applies just as much to various wonders as it does to genuine empirical miracles.

4.2  Kant on Empirical Miracles: Textual Issues Let us turn now to Leibniz’s third, fourth, and fifth concepts of miracle: they are, again, the concept of a miracle as beyond nature’s power but still within nature, as contrary to the subordinate maxims, and as product of extraordinary concurrence with natural powers. These are all empirical miracles, and as we saw earlier their extensions are the same. Is there room in Kant’s system for these? It is familiar Kantian lore that the Causal Principles of the Second and Third Analogies guarantee that every alteration in nature occurs in accordance with a rule, and that every spatiotemporal substance existing at t is in reciprocal causal relations with every other spatiotemporal substance existing at t. These principles are known a priori, and are the transcendental basis of the lawfulness of nature. They can also be further specified in relation to what Kant calls the “empirical concept of matter”: the result is the set of dynamical and mechanical laws that he outlines in the Metaphysical Foundations of Natural Science. Taken together, these principles constitute the “metaphysics of corporeal nature” (Ak. 4:472). Kant makes it clear in the “On Miracles” essay that “no alteration in the world (and so no beginning of that motion) can arise without being determined by causes in the world according to general laws of nature [Naturgesetzen überhaupt], thus not through freedom or a miracle proper” (Kiesewetter Ak.18:320, my em­phasis). For reasons that will become obvious later, I think we need to interpret “general laws of nature” here as referring to a priori metaphysical prin­ciples, rather than to more specific or “particular” empirical laws. This is supported by 44 For an impressively detailed discussion of Kant’s pre-­ critical account of miracles, see Peddicord (2001).

Andrew Chignell  343 Kant’s equation, later in the very same sentence, of “law of nature” with “causality” simpliciter: “appearances according to the law of nature (of causality) [die Erscheinungen nach dem Gesetze der Natur (der Causalität)] are what determine time” (ibid.). If this is right, then Kant’s negative claim in this passage is simply that there can be no events (free, miraculous, or otherwise) among bodies that fail to adhere to the a priori causal, dynamical, and mechanical laws established by the metaphysics of corporeal nature. This does not mean that they are not exceptions to the much more specific empirical laws that we study in natural science. The other main place to look for Kant’s view on miracles is in Religion— especially in the second of the four “parerga to religion within the boundaries of pure reason.” The parerga doctrines do not belong to a religion of pure reason, but “yet border on it” and are thus worthy of discussion (Rel, Ak. 6:52).45 In the main body of Part II, Kant had already asserted that we do not have either a the­ or­et­ic­al basis or a practico-­religious need to postulate the virgin birth or the bodily resurrection. In the parergon attached to that Part, Kant starts off in the same vein by claiming that a moral religion (“the heart’s disposition to observe all human duties as divine commands”) is such that any miracles connected with its inception are completely dispensable. Belief in historical empirical miracles, in other words, is a ladder that might in principle be kicked away if we ever come to accept the authenticity of a moral/religious teaching on other grounds. Indeed, it would manifest an immoral form of “unbelief” (Unglaube), Kant says, to insist that we can only accept morality’s dictates if they are authenticated by miracles. So far, then, the texts seem to favor attributing a firm and unqualified rejection of Mackie/McGrew miracles to Kant.46 However, in spite of his commitment to the inviolability of the Causal Principles and the dispensability of miracle stories, Kant also manifests—in the Religion, the Kiesewetter text, and various lectures and notes from the Critical period—an astounding openness to the real possibility and even historical actuality of empirical miracles. In Religion, for instance, he says that “reason does not dispute the possibility or actuality” of miracles (Rel, Ak. 6:52) and that it is “entirely conformable to the ordinary human way of thinking” for a new religion—even one based on “the spirit and the truth (on moral disposition)”—to announce or “adorn” (ausschmücken) its introduction with dazzling feats (Rel, Ak. 6:84). He goes on to suggest that it is plausible that the work of a “prophet” or “founder” of a new religion would be full of mir­acles (thus helping to win adherents from the old religion), and that the historical testimony to these miracles itself would be miraculously arranged and preserved: “It may well be (es mag also sein),” Kant writes, that the founder’s “appearance on earth, as well as his transition (Entrückung) from it, his eventful life and his passion, are all 45  See Chignell (2010) for an expanded discussion of these issues in Part II.2 of Religion. 46  See Guyer (Ch. 9 in this volume) and Huxford (2018) for examples of the many commentaries that take this to be the whole story. In the end, that view may have a philosophical advantage. My goal here, however, is to see whether there is a way to make sense of all the pro-­miracle passages within the framework of Kant’s theoretical philosophy of nature.

344  Leibniz and Kant on Empirical Miracles miracles—indeed that the history that should testify to the account of these mir­ acles is itself a miracle” (Rel, Ak. 6:84–85). 47 Similar claims are scattered throughout the Critical lectures on religion and metaphysics. Kant obviously had Leibniz and Wolff in mind, for instance, when he taught that God sometimes determines in accord with his aims that individual occurrences should not correspond to the order of nature. For it is not at all impossible, even in the best world, that the powers of nature may sometimes require the immedi­ ate cooperation of God in order to bring about certain excellent ends. It is not impossible that the Lord of Nature might at times communicate to it a comple­ mentum ad sufficientiam in order to carry out his plan. For who would be so presumptuous as to want to cognize how God can achieve everything he has planned for the world in accordance with universal laws and without his extraordinary direction? . . . Such exceptions to the rules of nature [Ausnahmen von den Regeln der Natur] may be necessary because without them God might not be able to put many great aims to work via the normal order of nature [nach dem gewöhnlichen Laufe derselben].  [Pölitz 28:1112, my emphasis]

Similarly: No world can be thought without deficiencies, without certain negations and ­limitations, and thus to make up the defect of nature, miracles are possible in the best world also, and even probable according to the concept of God’s goodness and truth. [V-­Met-­Vron, Ak. 29:871; see also V-­Met/Dohna, Ak. 28:667ff.; Anon-­K2, Ak. 28:732ff.; and BDG, Ak. 2:210–211]

Kant typically describes miracles this way in his lectures—as highly unusual events involving a “complement” from outside of nature that, together with the ordinary powers of finite things, is sufficient to produce the events that accomplish divine purposes. “God’s miracles in the physical world” thus result from God’s “cooperation with occurrences in the sensible realm [Mitwirkung zu den Begebenheiten in der Sinnenwelt Wunder Gottes in der physischen Welt sind]” (Pölitz 28:1106). Without such divine complementation, the normal, natural powers of finite creatures would be insufficient to produce the intended effects.48

47  For the suggestion that there is some Humean “sarcasm” in Kant’s comment here, however, see Ameriks (2014). 48  “A concursus of God with events in the world is not impossible, however; for it is always conceivable that a natural cause be insufficient in itself to accomplish the bringing forth of a certain effect. In this case God would give it a complementum ad sufficientiam, but insofar as he does that, he eo ipso does a miracle [Wunder]; thus we call it a miracle when the cause of an event is supernatural, which it would be if God as concausa cooperated in the bringing forth of the effect” (Ak. Pölitz 28:1209).

Andrew Chignell  345

4.3  Kant on Empirical Miracles: Philosophical Issues Now that we have a sense of the textual situation before us, we can return to our original question: how can any of this fit with the broader Kantian picture of nature as a deterministic system governed by the Causal Principles and the mechanical laws? The “On Miracles” fragment represents Kant’s most detailed attempt to answer this question. We have seen that he starts by saying that the “general laws of nature” are indeed unexceptionable. He goes on to distinguish, however, between two species of miraculum rigorosum: the “material” and the “formal.” A material miracle would be an “immediate effect of the divinity,” whereas a formal miracle has a cause in the world, but one whose “determination takes place outside the world.” Kant’s meaning here is hardly transparent, but he does offer this illustration: If one holds the drying of the Red Sea for the passage of the children of Israel to be a miracle, it is a miraculum materiale if one takes it to be an immediate effect of the divinity, but a miraculum formale if one lets it be dried out by a wind, but a wind sent by the divinity.  [Ak. 18:321–322]

Material miracles are immediately dismissed on the grounds that they would involve the direct introduction of new motion (force), and that this would be opposed to the third law of mechanics; that is, the application of the Third Analogy principle to our empirical concept of matter (Ak. 4:544): Now if a motion were effected by a miracle, then, since it would not stand under the law of effect and counter-­effect, the centrum gravitatis of the world would be altered by it, i.e., in other words, the world would move in empty space; however, a motion in empty space is a contradiction, it would be a relation of a thing to a nothing [eines Dinges zu einem Nichts], for empty space is a mere idea. [Ak. 18:321; cf. Ak. 18:419; R 5997]

Given this rejection of immediate or “material” miracles, it seems clear that Kant’s account of miracles will be no more satisfying than Leibniz’s to enthusiasts, literalists, and others who suggest that God inserts new spatiotemporal events into the world ex nihilo, without regard to the Causal Principles or the mechanical-­ dynamical specifications of them. “Movements,” Kant says, “cannot begin from themselves, and also not from something, that was not itself previously moved” (Refl 5997, Ak. 18:420). All the same, it looks like Kant also explicitly remains open to “formal” mir­acles in the Kiesewetter fragment. The idea, it seems, is that God sets up the world in advance (this is what he calls a “pre-­established” formal miracle), or even intervenes on a particular occasion (an “occasional” formal miracle), such that “the

346  Leibniz and Kant on Empirical Miracles power is in the world, but its determination takes place outside the world” (Ak 18:321.).49 Kant emphasizes that such occurrences must be rare: it would be a serious imperfection in the world if providence had to add its “complement” to lots of finite secondary causes in order to get the world that it wants. Still, like Leibniz, Kant is willing to say that the world might be such that, on rare occasions, the exercise of certain finite powers is accompanied by an extraordinary complement from “outside of nature”; that is, a “determination” that exceeds anything in the powers of the relevant substances, but one that is both necessary and sufficient to accomplish the divine purpose. Thus, for instance, a wind that would normally cause a few whitecaps can be “complemented” in such a way that the entire sea parts; likewise, the fire-­retardant powers of skin and hair that would normally resist fire only briefly can be “complemented” in such a way they survive even the Babylonian furnace. Note that Kant’s way of telling the story explicitly retains the “form” of lawfulness: all events or alterations do have empirical causes, and all spatiotemporal objects are indeed in reciprocal interaction.50 But in these extraordinary cases, the natural powers of finite things are only part of the total cause; the other part is the complementary determination—the extra boost—that comes from outside the empirical nexus. Only the total cause—the natural powers together with the supernatural complement—is “sufficient” for the effect (see Ak. 28:1209). And as a result, we remain ignorant of the laws of their effect (Wirkungsgesetze) (Rel. Ak. 6:86). What should we make of this model? For starters, it seems to entail that the Causal Principles do not guarantee that all alterations have empirical causes that are sufficient by themselves to produce them. For, again, on these extraordinary occasions a complement from outside of nature is required to achieve the effect. This is still consistent with the letter of the Second Analogy law, however, which says simply that empirical alterations follow from their causes in accordance with a rule (see A 188). Perhaps Kant’s idea is that in extraordinary cases natural phenomena are part of the total cause, and there is a rule involved, but the rule also refers to the complementary boost (“determination”) that the empirical cause receives from “outside the world.” The fact that it makes such reference is presumably why Kant also says we cannot even in principle grasp the “laws” by which mir­ acles occur.51 In this way, they differ from merely comparative miracles, which are naturalistic events whose laws we do not in fact but in principle could understand. 49  It is unclear what we should make of the difference between “pre-­established” and “occasional” in a transcendental idealist context. My best guess is that a pre-­established miracle is one that is willed prior to consideration of the choices of finite agents, while an “occasional” one is performed subsequent to or in response to those choices. But note that in R 5997, Kant seems to deny the possibility of “occasional” miracles altogether in favor of preestablished ones (Refl. Ak. 18:420). For more discussion of Kant’s account in the lectures see Bonaccini (2015). 50  For the distinction between the form of causal lawfulness and the empirical “matter” of particular moving forces, see A 207/B 252. 51  “In general, an event in the world whose laws human reason cannot at all cognize is a miracle” (V-­Met/Dohna, Ak. 28:667; though as noted earlier in Religion Kant dubs this the “practical” account (Rel Ak. 6:86)).

Andrew Chignell  347 But even if the Second Analogy principle is untouched, we might still worry that there is a tension between the present model and the principle of the Third Analogy, especially when it is extended to the concept of matter in the third mechanical law: “in all communication of motion, action and reaction are always equal to one another” (Ak. 4:544). It is simply left unclear how a miracle could occur without introducing action that has no reaction into the system. In other words, even if its occasion is the exercise of finite natural powers, God’s addition of a complementary “determination” (e.g. strengthening the power of that Egyptian wind) seems to threaten the mechanical law just as much as God’s directly parting the sea by fiat would. The complement is not, presumably, a mere change in the direction of the winds (a là Descartes’s immaterial mind changing the “direction” of the pineal gland’s vibrations without adding new motion); rather, it is a substantive Mitwirkung that adds something new. Apart from what was quoted earlier about the centrum gravitatis, the Kiesewetter fragment and other texts pass over these issues in silence. Perhaps we can suggest on Kant’s behalf, however, that God sets things up such that, on the occasion of a formal miracle of this sort, a reaction equivalent to the quantity of motion or force contributed by the complement is also simultaneously added such that the overall principle is preserved (and the centrum gravitatis of the world remains unmoved!). In effect, the divine addition to the action of finite powers would be offset by a complementary reaction (“the Lord giveth, and the Lord taketh away”). Kant at least hints at this in a note from the 1780s: Neither through a miracle nor through a mental being can a motion be brought about in the world, without producing just as much motion in the opposite direction, in accordance with the laws of action and reaction in matter. [Ak. 18:419; R 5997, my emphasis]

It sounds like Kant is open to empirical miracles here, and merely putting a transcendental constraint on what they would involve. This kind of deus ex machina move (not unheard of in the early modern period, of course) would allow the model to adhere to the letter of the a priori laws governing matter. If the model is coherent,52 then neither the transcendental principles of the Critique nor their applicability to matter in the “metaphysics of nature” would be threatened. 52  See Reichl (2019) for an illuminating challenge to this suggestion (as articulated in the article on which this chapter is partly based, Chignell (2014)). Reichl argues that “further conditions with respect to space in relation to the application of the third law pose a problem for the manner in which Chignell envisions such a divine complement” (109). More specifically: “Chignell’s strategy would address the problem of the potential imbalance in the total quantity of motion within a system, [but] it does not seem to apply to the more specific problem of the impossibility of causal interventions into relative space from empty space, which is what Kant states miracles would require” (120). Given the production schedule for this volume, it was not possible for me to add a to reply to Reichl’s defense of a wholly “disenchanted” picture of Kantian nature. But I thank Reichl for the provocation and hope to return to it in future work.

348  Leibniz and Kant on Empirical Miracles

4.4  Kant’s Leibnizean Model of Miracles, and How it May Improve on Leibniz’s Own Even granting that the transcendental laws are safe, we still might wonder what we are to make of the apparent conflict between Kant’s explanatory rationalism about the empirical world as depicted by the Causal Principles, and his insistence in Religion and elsewhere that “formal” miracles, at least, are possible and even probable? He says little by way of helping us resolve the puzzle, and many commentators have taken his positive remarks as little more than hat-­tipping to the theological authorities. I want to suggest now that, whatever his rhetorical intentions, there may still be a coherent way for Kant to incorporate empirical mir­ acles, and that the resulting picture looks a lot like Leibniz’s two-­level model described above. Recall that Kant grounds the fundamental lawfulness of the empirical world in transcendental arguments for the necessary and universal truth of the Causal Principles, as well as the application of those principles to the empirical concept of matter in general. Significantly, though, Kant does not insist that these arguments show that all of the much more specific or particular “laws” that we seek in scientific inquiry would be able to account for all of the events in the empirical world. Nor does he claim that these particular laws—even the ones that would be described in an ideal science—have the same universal, necessary, and exceptionless status as the a priori ones. My suggestion, then, is that there may be room in Kant’s system for a variation on Leibniz’s distinction between the general or fundamental order (the “law of the series”) that describes how nature genuinely operates, and the more specific or particular generalizations (what Leibniz calls the “subordinate maxims”) that typically hold, but to which there can be the occasional exception. The latter gen­er­al­iza­tions are still called “natural laws” by both Leibniz and Kant: they comprise the best system of graspable, simple, and strong (if not comprehensive) gen­er­al­iza­tions, and they are adequate to what happens in the vast majority of cases.53 The fundamental law of the series, however, is neither simple nor grasp­able in detail for Leibniz: we know only that it holds without exception. “Rigorous formal mir­acles” in this context would be events that accord with the general “form” or law of the series, but still count as exceptions to the specific, sub­or­din­ate “laws of nature.” In support of the extension of this picture to Kant, consider the following passage, this one from the Dohna lectures of 1792–1793: A miracle strictly defined is called rigorous. [How] is such a thing possible? Because there is an extramundane cause that has produced this order of 53  For a discussion of the effort to balance both simplicity and strength, see Luck (2011, 138–140) and Lewis (1973, 74ff.).

Andrew Chignell  349 things, and thus can produce another. A miracle is therefore possible in itself in­tern­al­ly . . . In general, an event in the world whose laws human reason cannot at all cognize is a miracle.  [V-­Met/Dohna, Ak. 28:667]

The claim here is that empirical events typically appear in an order that we can and do cognize, but that “this order of things” may at times be suspended in favor of “another” order whose “laws” are not humanly cognizable. A more Leibnizean way to put this is to say that there is one true empirical order or law of the series from which all events follow, but that this true order is not epistemically accessible to us in all its details. This is consistent with saying that we know that it obtains and has the basic structure underwritten by various metaphysical principles (for Leibniz) or the categories (for Kant). It also leaves room for the idea that the true empirical order may differ (in terms of the events that it entails) from the specific empirical “laws of nature” that we seek in everyday life and natural science. In Religion Kant says likewise that miracles are “events in the world, whose causes [Ursache] are such that their laws of action [Wirkungsgesetze] are absolutely unknown to us and must remain so” (Rel, Ak. 6:86). It is important to emphasize that the hierarchy here is epistemological: metaphysically speaking, there is only one true, inviolable order. And while empirical events are typically arranged according to relatively simple, general patterns that we can cognize, that “order of things” is only an approximation. On occasion, and for reasons that are typically obscure to us, it gives way to events that are part of the deeper order—the one whose “laws of action” are necessarily unknown to us. Again, this is consistent with saying that we know that the latter order obtains and, for Kant, that it has the basic inviolable structure described by his metaphysics of experience. But it also leaves plenty of room for that order occasionally to differ (in terms of the events that it entails) from the order described by the usually reliable particular “laws” that we seek in empirical investigation. In short, for Kant just as for Leibniz (and Malebranche before him), there may be times when God “determines in accord with his aims that individual occurrences” that do “not correspond to the [subordinate] order of nature” must be “worked into the course of the world [in dem Laufe der Welt gewirkt] in order to bring about some necessary aim of his” (Pölitz 28:1110–1112). Provocatively enough, Kant even suggests in Religion that we may all be witness to some of these miracles, though not under that description: Nobody can have so exaggerated a conceit of his insight as to make bold to assert definitely that, for instance, the most admirable conservation of the species in the plant and animal kingdom, where every spring a new generation once more displays [the species] original and undiminished, with all the inner perfections of mechanism, and even (as in the vegetable kingdom) with all the always-­ delicate beauty of color, without the forces of inorganic nature, otherwise so

350  Leibniz and Kant on Empirical Miracles destructive in the bad weather of autumn and winter, being able to harm the seed at that point—no one can assert that this, I say, is a mere consequence of natural laws, and pretend to grasp that the creator’s direct influence is not rather needed for it each time.  [Rel, Ak. 6:89 n., original emphasis]

Note, again, that the model does not assimilate empirical miracles to “comparative miracles.” The latter adhere to the ordinary specific “laws,” our ignorance of which is a contingent matter. Genuine empirical miracles, by contrast, do not adhere to the ordinary “laws” at all: they are exceptions to those laws that involve the “creator’s direct influence”—thus, their “laws of action” are necessarily beyond our ken. With this sketch of Kant’s version of the Leibnizean model before us, we can now complete the comparison. Recall that there were two main concerns for Leibniz about accommodating empirical miracles. First, he seems to have no clear account of how the extraordinariness of the miracle might be grounded, at least in part, in the natures of creatures themselves—that is, no clear story about what in creatures God “extraordinarily” concurs with. We saw that this is in tension with Leibniz’s anti-­occasionalist doctrine that all the states of a finite substance must be grounded in its active powers somehow. We looked at two proposals in response to the problem (Wholly Passive and Degrees of Power), and noted some of their benefits and costs. Second, there was the concern about why the general order is so complex as to be unintelligible to finite rational minds, and how this coheres with the doctrine that this is the best possible world in terms of both metaphysical economy and the happiness of finite minds. Kant faces neither of these problems. Regarding the first, Kant simply declines to talk of God’s extraordinary concurrence with creaturely powers, and seems to resist any commitment to concurrence in all but its weakest, conservationist form.54 That said, every event in the empirical world, including a miraculous one, is a product of the active powers of finite substances on the Kantian model, at least in part. That is because the fundamental empirical order is itself, at least in part, the result of the spontaneous activity of finite apperceiving subjects. An empirical miracle would therefore also be the result of the activity of such subjects, at least in part, and so in a transcendental idealist context, no event or state in the empirical world—nothing that is among the phenomena—could be a completely “extrinsic denomination” grounded solely in divine powers. Regarding the second concern, Kant is able to offer a more principled ex­plan­ ation of why the true general empirical order is beyond our ken. If the fundamental order of nature is partly a result of the spontaneous synthesizing activity of the

54  Though, again, see Hogan (Chapter 10 of this volume).

Andrew Chignell  351 apperceiving mind, and if we grant Kant’s general claim about our ignorance of the specific details of how this activity works (that is, details that go beyond the application of the categories), then we cannot expect to have epistemic access to all of the detailed nomological results of these activities either. Again, the famous arguments of the Critique and Metaphysical Foundations prove that there is a true, general order of empirical phenomena—one that is structured in accordance with the categories and is indeed “universal and necessary.” But those arguments provide no guarantee that even an idealized scientific account of the specific em­pir­ ic­al laws will reliably map that general phenomenal order. Kant can thus retain an analogue of Leibniz’s gap between the “subordinate” empirical laws and the true “general order” of nature: for Kant the gap is between what we can know about specific laws on the basis of empirical inquiry, and what we do not or perhaps cannot know about the general but still phenomenal order underlying them—the order that results from noumenal affection and the deep structuring activities of mind.55

4.4  Maxims of Judgment Kant almost inevitably moves from acknowledging the possibility of empirical miracles to emphasizing the utter uselessness of appeals to them. “Reason does not contest the possibility or the actuality of the objects of these ideas; it just cannot incorporate them into its maxims of thought and action” (Rel. Ak. 6: 52). In other words, even though such events are possible and actual, our necessary ig­nor­ance of their laws entails that we have no good “positive criterion” for them—a criterion that would reliably tell us when a miracle has occurred. This leaves reason “paralyzed”: “Nowhere in experience can we recognize a supersensible object, even less exert influence upon it to bring it down to us” (Rel, Ak. 6:174).56 In order to avoid such paralysis, Kant says, those who would proceed 55  Although it is beyond the scope of this chapter to go further into details, the model does seem to require the rejection of interpretations of Kant’s philosophy of science according to which all of the particular empirical laws are entailed by the metaphysical laws deduced in the Critique and the Metaphysical Foundations. The fact that there seems to be a gap between the fundamental order of nature (deduced in the Critique and the Metaphysical Foundations) and the specific empirical laws sought by natural scientists has been noted by other commentators as well. Regarding e.g. the Opus Postumum, Eric Watkins remarks that “As Kant struggles with the problems that result from trying to account for much more specific features of matter, it is unclear that (or how) the categories are supposed to be of help in structuring Kant’s argument” (Watkins 2009, 21). 56  See Watkins (2010); Byrne (2007, 158ff.). In Religion we are given the “negative criterion” that something “cannot be a divine miracle despite every appearance of being one” if it is “directly in conflict with morality” (Rel, Ak.6:87). The V-­Met-­L1 lectures (mid-­1770s) are interesting in that Kant is reported to have floated a corresponding positive criterion: “The condition under which it is allowed to assume miracles is this: the course of nature does not coincide with moral laws. Thus imperfection is in the course of nature; it does not agree with the conditions which should concur as motives for the moral laws. Miracles are possible in order to complement this imperfection” (L1 28:219).

352  Leibniz and Kant on Empirical Miracles scientifically must ignore the possibility of miracles and presuppose that any particular event is not the result of a special intervention into the causal nexus. In other words, for scientific and practical purposes we should presume that every empirical event has its total cause in the empirical world (Rel, Ak. 6:88).57 Thus, Kant continues the passage quoted earlier about the wonders of spring in this way: But these are experiences; for us therefore they are nothing other than effects of nature, and ought never to be judged otherwise. For this is what modesty requires of reason’s claims, and to transcend these bounds is presumptuousness and immodesty, even though in asserting miracles people often purport to demonstrate a humble and self-­deprecating way of thinking. [Rel, Ak. 6:89 n., ori­gin­al emphasis]

A few pages earlier, Kant likewise says that “sensible human beings” who “do indeed theoretically believe in miracles” should not count on them in “practical affairs,” and judges must not take them into account in courtroom situations (Rel, Ak. 6:85–87). And while governments and churches may find it useful to teach that revelations and miracles have occurred in ancient times, they must also advise that it is unwise to expect them now. The motive behind these injunctions is baldly pragmatic: old stories about miracles will not cause much uproar, but rumors of new miracle workers could lead to serious civil unrest: “to want to per­ ceive heavenly influences is a kind of madness [Wahnsinn]” (Rel, Ak. 6:174). The discussion in Part Two of Religion concludes with the claim that there are only two principled maxims regarding miracles: we should either accept that they occur all the time “though hidden under the appearance of natural occurrences,” or we should accept they do not occur at all. The first maxim is “in no way compatible with reason” and so we must adopt the second. But, again, note that this is just a “maxim of judgment,” not a “theoretical assertion”: Kant leaves open that it is really possible that empirical miracles occur.58 The only claim about miracles that we must “dispute with all our might” is that they authenticate true religion, and that belief in them is somehow meritorious or pleasing to God (Rel, Ak. 6:85). This combined openness to the real possibility of empirical miracles and skepticism about our ability to identify them is Kant’s consistent position ­ throughout the lectures, notes, and written materials in the Critical period. It is not much changed since the pre-­Critical period: in 1763, he argued that, for 57  Cf. with the “first Rule concerning Miracles” laid down by the seventeenth-­century Newtonian, Thomas Burnet: “That we must not flie to miracles, where Man and Nature are sufficient” (Burnet 1691, section III, ch. viii; qtd. in Harrison 1995, 538). 58  The claim that our commitment to the exceptionless character of the natural laws is a mere maxim of judgment will seem scandalously weak to readers who extrapolate from Kant’s convictions about the universal and necessary status of the Causal Principles to an assumption about the status of specific empirical laws. In the context of the models sketched here, however, the scandal dissipates.

Andrew Chignell  353 scientific and practical reasons, exceptions to the “laws of nature” must be viewed as possible but “rare” and that, in general, philosophy and common sense indicate that “nothing is to be regarded as a miracle or as a supernatural event, unless there are weighty reasons for doing so” (BDG, Ak. 2:108; see also V-­Met-­L1 [from the very early Critical period 1770s]; Ak. 8:217ff.).59

5  Conclusion: Can We Perform Empirical Miracles? There is more to be said, but this sketch of one version of the Kantian model of miracles suggests that Kant fares at least as well as Leibniz in terms of explaining the commitments they share: the commitment to the possibility of empirical miracles and the commitment to our ignorance of the general order that entails them. Kant lacks the metaphysical principles that make it hard for Leibniz to keep the gap between natures and essences from collapsing, and he lacks the arch-­ rationalist principles that lead Leibniz to concede that miracles decrease the perfection of the world by reducing the happiness that finite minds take in the intelligibility of things. Kant can also utilize his version of the gap between the specific empirical laws and the true, general but still phenomenal order to account for the possibility of empirical miracles. And at the transcendental level he can appeal to our ignorance of the details of the spontaneous structuring activities of the mind—as well as the noumenal affection to which they respond—to explain why parts of this fundamental order are inaccessible to us, in particular, the “laws of action” (Wirkungsgesetze) of the causes of miracles. A final point: Kantian readers may have noticed something significant way back at the beginning of the chapter related to our Mackie/McGrew account of empirical miracles: (MMM)  An empirical miracle obtains when a being that is not part of the order described by the laws of nature purposively intervenes to produce an event that counts as an exception to at least one of those laws. What Kantian readers may have noticed is that this concept seems applicable, in principle, to some of the results of finite transcendental freedom too. The suggestion strikes all but the most extreme enthusiast as bizarre and outlandish—and perhaps it is: how could we produce events that are exceptions to the particular laws of nature? But from an interpretive point of view, I think, we must not entirely recoil. For whereas Leibniz merely flirted with the (Cartesian) thought

59  A. T. Nuyen takes comments like these as grounds for interpreting Kant as a wholesale “em­pir­ ic­al skeptic” about particular miracles, and goes on to focus on what he regards as the “miracles” of teleology in nature as a whole, and of the highest good. See Nuyen (2002).

354  Leibniz and Kant on Empirical Miracles that the exercise of free will produces an exception to the laws,60 Kant, even after the Critical turn, suggests that we can believe (for practical reasons) or even practically cognize that our transcendentally free choices contribute to making nature what it is. The traditional interpretation of Kant’s compatibilism takes this to mean that our noumenal choices somehow play a role in determining which specific laws and initial conditions characterize the phenomena, and of course which specific events obtain (G Ak. 4:450ff.).61 But on the model sketched above, what happens in the empirical world is not always governed by the specific em­pir­ic­al laws. No doubt Kant would not want to ascribe material miracles to a finite will any more than to the infinite one. But a formal miracle—an empirical expression of the quality of a finite will that makes the fundamental phenomenal order different than it otherwise would have been—is, at the very least, something that the account has to leave open. Is there some other reason to restrict the phenomenal correlates of transcendentally free acts to what is described by the ordinary, subordinate, specific “laws”? The question hangs, it seems to me, on how we interpret the purposiveness condition in (MMM). According to Kant, our basic transcendentally free choice is simply for or against the dictates of the moral law; our purpose in that context is not to produce any particular empirical event, much less a miraculous one. But could the empirical expression of such a purposive choice still include an event that is an exception to a specific empirical law? And would that count as a purposive production of that event such that it would satisfy the Mackie/McGrew definition? It is impossible to imagine Kant endorsing this idea, although it is also surprisingly hard to build a textual case against it. His only clear recommendation is to remain agnostic. “[F]or theoretical purposes, as regards the causality of freedom (and equally its nature) we cannot even formulate without contradiction the wish to understand it” (Ak. 6:144). Kant’s coyness here notwithstanding, I suggest that it is part of his picture that we can regard ourselves, from a practical point of view at least, as responsible for effects in the empirical world, even while the “laws” by which we are so responsible remain an “impenetrable mystery” (Ak. 6:143). And so perhaps our finite ­freedom, too, is capable of empirical miracles. 60  “Free or intelligent Substances . . . are not bound to any certain subordinate Laws of the universe, but act spontaneously from their own power alone, as if by a sort of private miracle, and by looking to some final cause they interrupt the connection and course of efficient causes operating on their will” (Leibniz 1973,100, my emphasis). 61  Benjamin Vilhauer, for instance: “It must be noted, however, that according to the traditional interpretation of Kant’s theory of free will, it is not absurd to suppose that we have the power to causally affect the laws of nature. Choices of maxims by agents qua noumena are the ontological substrates of both (1) the empirical-­psychological events that constitute the choices of agents qua phenomena and (2) the particular causal laws that necessitate those empirical-­psychological events. If choices of maxims by agents qua noumena had been different, then they would have had different appearances— that is, the empirical-­psychological events that constitute the choices of agents qua phenomena would have been different, and the particular causal laws necessitating them would have been different too” (Vilhauer 2004, 727).

Bibliography Adams, M. 2013. “Powers versus Laws: God and the Order of the World according to Some Late Medieval Aristotelians.” In: E. Watkins (ed.), Divine Order, Human Order, and the Order of Nature, New York: Oxford University Press, pp. 3–26. Adams, R. M. 1994. Leibniz: Determinist, Theist, Idealist, Oxford: Oxford University Press. Adams, R. M. 1997a. Review of Sleigh’s “Leibniz and Arnauld: A Commentary on Their Correspondence.” Nous 31: 266–277. Adams, R. M. 1997b. “Things in Themselves,” Philosophy and Phenomenological Research 57: 801–825. Adams, R.  M. 1998. “Introduction.” In: Religion within the Boundaries of Reason Alone. Cambridge: Cambridge University Press, pp. xxi–xxii Adams, R. M. 2000. “God, Possibility, and Kant,” Faith and Philosophy 17: 425–440. Adams, R. M. 2013. “Powers versus Laws: God and the Order of the World According to Some Late Medieval Aristotelians.” In: E.  Watkins (ed.), Divine Order, Human Order, and the Order of Nature. New York: Oxford University Press, pp. 3–26. Adickes, E. 1924a. Kant als Naturforscher. Vol. 1. Berlin: De Gruyter. Adickes, E. 1924b. Kant und das Ding an Sich. Berlin: Pan Verlag Rolf Heise. Adickes, E. 1927. Kant und die Als-Ob Philosophie. Stuttgart: Frommann. Allais, L. 2004. “Kant’s One World: Interpreting ‘Transcendental Idealism.’” British Journal of the History of Philosophy 12: 655–684. Allison, H. 1980. “Kant’s Critique of Spinoza.” In: R. Kennington (ed.), The Philosophy of Baruch Spinoza. Washington: Catholic University of America Press, pp. 199–277. Allison, H. 1983 (2nd ed., 2004). Kant’s Transcendental Idealism, New Haven: Yale University Press. Allison, H.  E. 1991. “Kant’s Antinomy of Teleological Judgment.” Southern Journal of Philosophy, Supp. Vol. XXX: 25–42. Ameriks, K. 1982a (2nd ed., 2000). Kant’s Theory of Mind. Oxford: Clarendon Press. Ameriks, K. 1982b. “Recent Work on Kant’s Transcendental Philosophy” American Philosophical Quarterly 19: 1–24. (Repr. Ameriks 2003.) Ameriks, K. 1992. “The Critique of Metaphysics: Kant and Traditional Ontology.” In: P. Guyer (ed.), The Cambridge Companion to Kant. Cambridge, Cambridge University Press, pp. 249–79. Ameriks, K. 2003. Interpreting Kant’s Critiques. Oxford: Oxford University Press. Ameriks, K. 2014. “Kant, Miracles and Religion, Parts One and Two.” In: G.  Michalson (ed.), Kant’s Religion within the Bounds of Mere Reason: A Critical Guide. New York: Cambridge University Press, pp. 137–155. Anderson, R. L. 2004. “It Adds Up After All: Kant’s Philosophy of Arithmetic in Light of the Traditional Logic.” Philosophy and Phenomenological Research 69: 501–540. Anderson, R. L. 2015. The Poverty of Conceptual Truth: Kant’s Analytic/Synthetic Distinction and the Limits of Metaphysics. Oxford: Oxford University Press. Antognazza, M.  R. 2009. Leibniz: An Intellectual Biography. Cambridge: Cambridge University Press.

356 Bibliography Aquila, R. 1979. “Things in Themselves and Appearances: Intentionality and Reality in Kant.” Archiv für Geschichte der Philosophie 61: 293–307. Aquinas, T. 1955. Summa Theologica. Ed. D.  Sullivan. Chicago: English Dominicans, Britannica. Arthur, R. 1994. “Space and Relativity in Newton and Leibniz.” British Journal for the Philosophy of Science 45: 219–240. Arthur, R. 1998. “Infinite Aggregates and Phenomenal Wholes: Leibniz’s Theory of Substance as a Solution to the Continuum Problem.” Leibniz Society Review 8: 25–44. Arthur, R. 2013. “Leibniz’s Theory of Space.” Foundations of Science 18: 499–528. Arthur, R. 2014. Leibniz. Cambridge: Polity Press. Baumgarten, A. 1757. Metaphysica. 4th ed. Halle: Karl Hermann Memmerde. Baumgarten, A. 2011. Metaphysica/Metaphysik. Ed. G.r Gawlick and L.  Kreimendahl. Stutgart-Bad Canstatt: Frommann-Holzboog. Beck, L. W. 1960. A Commentary on Kant’s Critique of Practical Reason. Chicago: University of Chicago Press. Beck, L.  W. 1969 (2nd ed. 1996). Early German Philosophy: Kant and His Predecessors. Cambridge, MA: Harvard University Press. Beck, L. W. 1993. “From Leibniz to Kant.” In: R. C. Solomon and K. Higgins (eds.), The Age of German Idealism. London and New York: Routledge, pp. 5–39. Beiser, F. C. 1987. The Fate of Reason: German Philosophy from Kant to Fichte. Cambridge, MA: Harvard University Press. Beiser, F.  C. 2009. Diotima’s Children: German Aesthetic Rationalism from Leibniz to Lessing. Oxford: Oxford University Press. Bell, J.  L. 2008. A Primer of Infinitesimal Analysis. 2nd ed. Cambridge: Cambridge University Press. Belot, G. 2011. Geometric Possibility. Oxford: Oxford University Press. Bennett, J. 2001. Learning from Six Philosophers: Descartes, Spinoza, Leibniz, Locke, Berkeley, Hume, Volume 1, Oxford/New York: Oxford University Press. Bennett, J. 2005. “Leibniz’s Two Realms.” In: J. A. Cover and D. Rutherford (eds.), Leibniz: Nature and Freedom. New York: Oxford University Press, pp. 135–155. Bergmann, M and J.  A.  Cover. 2006. “Divine Responsibility without Divine Freedom.” Faith and Philosophy 23: 381–408. Bianco, B. 1989. “Freiheit gegen Fatalismus. Zu Joachim Langes Kritik an Wolff.” In: N.  Hinske (ed.), Zentren der Aufklärung I.  Halle: Aufklärung und Pietismus. Wolfenbütteler Studien zur Aufklärung 12. Heidelberg: Schneider, pp. 111–155. Bilfinger, G.  B. 1723. De harmonia animae et corporis humani, maxime praestabilita, ex mente illustris Leibnitii, Commentatio Hypothetica. Leipziz and Frankfurt: Mezler. Bilfinger, G.  B. 1725. Dilucidationes Philosophicae de Deo, Anima Humana, Mundo, et Generalibus Rerum Affectibus. Stuttgart: Cotta. Bilfinger, G.  B. 1728. “De viribus corpora moto insitis, et illarum Mensura.” In: Commentarii Academiae Scientarum Imperialis Petropolitanae. Vol. 1. St. Petersburg, pp. 43–120. Bilfinger, G.  B. 1741. Dilucidationes philosophicae de Deo, anima humana, mundo, et generalibus rerum affectionibus. Tübingen: Cott. [= GW III 18] Bodemann, E. (ed.). 1895. Die Leibniz-Handschriften der königlichen öffentlichen Bibliothek zu Hannover. Hanover and Leipzig: Hann’sche Buchhandlung. Böhme, G. 1914. Die Abhängigkeit der Raumauffassung Kants in der ersten Phase der vorkritischen Periode von seiner Auffassung des Newtonschen Attraktionsgesetzes. PhD Dissertation (Leipzig).

Bibliography  357 Bolton, M. 2011. “Leibniz’s Theory of Cognition.” In: B.  C.  Look (ed.), Companion to Leibniz. London: Continuum, pp. 136–158. Bolton, M. 2012. “Metaphysics and Psychology in Leibniz’s Theory of Ideas.” In: H. Breger, J. Herbst, and S. Erdner (eds.), Natur und Subjekt: IX. Internationaler Leibniz-Kongress. Vol. 4. Hannover: Akademie Verlag, pp. 10–19. Bolton, M. 2017. “Locke’s Essay and Leibniz’s Nouveaux Essais: Competing Theories of Universals.” In: S.  di Bello and T.  Schmaltz (eds.), The Problem of Universals in Early Modern Philosophy. Oxford: Oxford University Press, pp. 220–251. Bonaccini, J.  B. 2015. “Kant’s Account of Miracles in his Lectures on Metaphysics”. In: B. Dörflinger, C. La Rocca, R. Louden, and U. Rancan de Azevedo Marques (eds.), Kant’s Lectures/Kant’s Vorlesungen. Berlin: De Gruyter, pp. 247–260. Brandom, R. 1981. “Leibniz and Degrees of Perception.” Journal of the History of Philosophy 19: 447–479. Brecht, M. (ed.). 1995. Geschichte des Pietismus. Vol. 2: Der Pietismus im achtzehnten Jahrhundert. Göttingen: Vandenhoeck & Ruprecht. Brewer, K. and E. Watkins. 2012. “A Difficulty Still Awaits: Kant, Spinoza, and the Threat of Theological Determinism.” Kant-Studien 103: 163–187. Broad, C.  D. 1981 [1946]. “Leibniz’s Last Controversy with the Newtonians.” In: R. S. Woolhouse (ed.), Leibniz: Metaphysics and Philosophy of Science. Oxford: Oxford University Press, pp. 157–174. Brown, G. 1995. “Miracles in the Best of All Possible Worlds: Leibniz’s Dilemma and Leibniz’s Razor.” History of Philosophy Quarterly 12: 19–39. Brucker, J. 1742–1744. Historia critica philosophiae. 5 vols. Leipzig. Burge, T. 1979. “Individualism and the Mental.” Midwest Studies in Philosophy 4: 73–121. Burnet, T. 1681. Telluris Theoria Sacra, or Sacred Theory of the Earth. London: Walter Kettilby. Buschmann, C. 1989. “Wolffianism in Berlin.” In: W. Förster (ed.), Aufklärung in Berlin. Berlin: Akademie Verlag, pp. 73–105. Byrne, P. 1978. “Miracles and the Philosophy of Science.” Heythrop Journal 19: 162–170. Byrne, P. 2007. Kant on God. London: Routledge. Campo, M. 1939. Cristiano Wolff e il razionalismo precritico. Milan: Società Editrice “Vita e Pensiero.” Carl, W. 1992. Die transzendentale Deduktion der Kategorien in der ersten Auflage der Kritik der reinen Vernunft. Frankfurt am Main: Vittorio Klosterman. Cassirer, E. 1907. Das Erkenntnisproblem in der Philosophie und Wissenschaft der neueren Zeit. Volume 2. Berlin: Cassirer. Cassirer, E. 1998 [1932]. Die Philosophie der Aufklärung. Hamburg: Meiner. Châtelet, É. du. 1740. Institutions de Physique. Paris: Prault. Chignell, A. 2009. “Kant, Modality, and the Most Real Being.” Archiv für Geschichte der Philosophie 91: 157–192. Chignell, A. 2010a. “The Devil, the Virgin, and the Envoy: Symbols of Moral Struggle in Religion II.2.” In: O. Hoeffe (ed.), Immanuel Kant: Religion innerhalb der Grenzen der blossen Vernunft. Munich: Oldenbourg—Akademie Verlag, pp. 111–129. Chignell, A. 2010b. “Real Repugnance and Belief about Things-in-Themselves: A Problem and Kant’s Three Solutions.” In: J.  Krueger and B.  B.  Lipscomb (eds.), Kant’s Moral Metaphysics. Berlin: De Gruyter. Chignell, A. 2011. “Real Repugnance and Our Ignorance of Things-in-Themselves: A Lockean Problem in Kant and Hegel.” Internationales Jahrbuch des Deutschen Idealismus 7: 135–59.

358 Bibliography Chignell, A. 2012. “Kant, Real Possibility, and the Threat of Spinoza.” Mind 121: 635–675. Chignell, A. 2013. “Rational Hope, Moral Order, and the Revolution of the Will.” In: E.  Watkins (ed.), Divine Order, Human Order, and the Order of Nature. New York: Oxford University Press, pp. 197–218. Chignell, A. 2014. “Can Kantian Laws be Broken?” Res Philosophica 91: 103–121. Clarke, S. 1823 [1704]. A Demonstration of the Being and Attributes of God. Glasgow: Richard Griffin and Co. Clemm, H. W. 1764. Mathematisches Lehrbuch. 2 Parts. Stuttgart: Metzler. Consentius, E. 1902. Lessing und die Vossische Zeitung. Leipzig: Avenarius. Corr, C.  A. 1972. “Christian Wolff ’s Treatment of Scientific Discovery.” Journal of the History of Philosophy 10: 323–334. Corr, C. A. 1975. “Christian Wolff und Leibniz.” Journal of the History of Ideas 36: 241–262. Correia, F. and B.  Schnieder (eds.). 2012. Metaphysical Grounding: Understanding the Structure of Reality. Cambridge: Cambridge University Press. Cover, J. A. 1989. “Relations and Reduction in Leibniz.” Pacific Philosophical Quarterly 70: 185–211. Cover, J.  A. and J.  Hawthorne. 1999. Substance and Individuation in Leibniz. New York: Cambridge University Press. Cover, J.  A. and D.  Rutherford (eds.). 2005. Leibniz: Nature and Freedom. New York: Oxford University Press. Crockett, T. 1999. “Continuity in Leibniz’s Mature Metaphysics.” Philosophical Studies 94: 119–138. Crockett, T. 2008. “Space and Time in Leibniz’s Early Metaphysics.” The Leibniz Review 18: 41–79. Crusius, C.  A. 1745. Entwurf der nothwendigen Vernunft-Wahrheiten, wiefern sie den zufälligen entgegen gesetzt warden. Leipzig: Johann Friedrich Gleditsch [1964]. Crusius, C.  A. 1747. Weg zur Gewißheit u. Zuverlässigkeit der menschlichen Erkenntnis. Leipzig: Johann Friedrich Gleditsch. De Risi, V. 2007. Geometry and Monadology: Leibniz’s Analysis Situs and Philosophy of Space. Basel/Boston/Berlin: Birkhäuser. Denzinger, H. and A.  Schönmetzer. 1997. Enchiridion symbolorum definitionum et declarationum de rebus fidei et morum. Freiburg, Basel, Rome, and Vienna: Herder. Diderot, D. 1975–. Oeuvres complètes. Ed. H. Dieckmann, J. Proust, and J. Varloot. Paris: Hermann. Döring, D. 1999. Die Philosophie Leibniz’ und die Leipziger Aufklärung in der ersten Hälfte des 18. Jahrhunderts. Abhandlungen der Sächsischen Akademie der Wissenschaften, Bd. 75, H. 4. Stuttgart: Hirzel. Duchesneau, F. 2010. Leibniz, le vivant et l’organisme. Paris: Vrin. Dyck, C. 2011. “A Wolff in Kant’s Clothing: Christian Wolff ’s Influence on Kant’s Accounts of Consciousness, Self-Consciousness, and Psychology.” Philosophy Compass 6: 44–53. Earman, J. 1979. “Was Leibniz a Relationist?” Midwest Studies in Philosophy 4: 263–276. Earman, J. 1989. World Enough and Space-Time. Cambridge: MIT Press. Earman, J. 2000. Hume’s Abject Failure: The Argument against Miracles. New York: Oxford University Press. Eberhard, J. A. 1787. Neue Apologie des Sokrates. 2nd ed. 2 vols. Frankfurt and Leipzig (no publisher). Eberstein, W. L. G. F. von. 1794. Versuch einer Geschichte der Logik und Metaphysik bey den Deutschen von Leibnitz bis auf gegenwärtige Zeit. 2 vols. Vol. 1. Halle: Ruff.

Bibliography  359 École, J. 1979. “En quel sens peut-on dire que Wolff est rationaliste?” Studia Leibnitiana 11: 45–61. École, J. 1991. “De la connaissance qu’avait Kant de la métaphysique wolffienne, ou Kant avait-il lu les ouvrages métaphysiques de Wolff?” Archiv für Geschichte der Philosophie 73(3): 261–276. École, J. 1998. “War Christian Wolff ein Leibnizianer.” Aufklärung 10: 29–46. Edelmann, J. C. 1740. Moses mit aufgedecktem Angesichte. Edelmann, J.  C. 1969. Sämtliche Schriften in Einzelausgaben. Stuttgart-Bad Canstatt: Frommann-Holzboog. Eliot, T. S. 1916. “The Development of Leibniz’s Monadism.” The Monist 26: 534–556. Erdmann, B. 1876. Martin Knutzen und seine Zeit: ein Beitrag zur Geschichte der Wolfischen Schule. Leipzig: Leopold Voss. Erdmann, B. 1878. Kants Kritizismus in der ersten und zweiten Auflage der Kritik der reinen Vernunft. Leipzig: Leopold Voss. Euler, L. 1746. “De la force de percussion et sa veritable mésure,” In: Mémoires de l’Académie Royale des Sciences et des Belles Lettres de Berlin, Année 1745. Berlin: Haude & Spener. Euler, L. 1911–. Opera omnia. Ed. Swiss Academy of Sciences. Basel: Birkhauser/Springer. [Cited by series, volume and page number] Euler, L. 2003. Lettres à une princesse d’Allemagne sur divers sujets de physique et philosophie. Ed. S. D. Chatterji. Lausanne: Presses polytechniques et universitaires romandes. Finster, R. 1986. “Leibniz’ Entwurf der phänomenalen Welt im Hinblick auf Kants ‘Kritik der reinen Vernunft’.” In: A.  Heinekamp (ed.), Beiträge zur Wirkungs- und Rezeptionsgeschichte von Gottfried Wilhelm Leibniz, Studia Leibnitiana Supplementa, Vol. 16. Stuttgart: Steiner, pp. 188–197. Firnhaber, F. J. 1737. Die gerettete Ehre und Würde der Heiligen Schrift auf Veranlassung der zu Wertheim gedruckten freyen Übersetzung der 5 Bücher Mosis . . . zur Bezeugung seiner schuldigen Amtstreue, dargeleget [with a preface of Christian Münden]. Frankfurt and Leipziz: Heinrich Ludwig Brönner. Fischer, H.-P. 1985a. “Eine Antwort auf Kants Briefe vom 23. August 1749,” Kant-Studien 76: 79–89. Fischer, H.-P. 1985b. “Kant an Euler,” Kant-Studien 76: 214–218. Fischer, N. (ed.) 2004. Kants Metaphysik und Religionsphilosophie. Hamburg: Meiner. Fischer, J.  M. and M.  Ravizza. 1998. Responsibility and Control: A Theory of Moral Responsibility. Cambridge: Cambridge University Press. Fisher, M. and E. Watkins. 1998. “Kant on the Material Ground of Possibility: From ‘The Only Possible Argument’ to the ‘Critique of Pure Reason’.” Review of Metaphysics 52: 369–395. Fogelin, R. 2005. A Defense of Hume on Miracles. Princeton: Princeton University Press. Formey, J. L. S. 1741. La Belle Wolfienne. 6 vols. La Haye: Le Vier. [= GW III 16] Formey, J. L. S. 1760. Histoire Abrégé de la Philosophie. Amsterdam: Schneider. Förster, E. 1993. “Kant’s Third Critique and the Opus postumum.” Graduate Faculty Philosophy Journal 16: 345–358. Förster, E. 2000. Kant’s Final Synthesis: An Essay on the Opus postumum. Cambridge, MA: Harvard University Press. Fouke, D. 1994. “Emanation and the Perfections of Being: Divine Causation and the Autonomy of Nature in Leibniz.” Archiv für Geschichte der Philosophie 76: 168–194. Franks, P. 2005. All or Nothing: Systematicity, Transcendental Arguments, and Skepticism in German Idealism. Cambridge, MA: Harvard University Press.

360 Bibliography Freddoso, A. J. 1988. “Medieval Aristotelianism and the Case against Secondary Causation in Nature.” In: T. V. Morris (ed.), Divine and Human Action. Ithaca: Cornell University Press, pp. 74–118. Freddoso, A.  J. 1991. “God’s General Concurrence with Secondary Causes: Why Conservation is not Enough”, Philosophical Perspectives 5: 553–585. Freddoso, A.  J. 1994. “God’s General Concurrence with Secondary Causes: Pitfalls and Prospects.” American Catholic Philosophical Quarterly 67: 131–156. Friedman, M. 1992. Kant and the Exact Sciences. Cambridge, MA: Harvard University Press. Frierson, P. 2005. “Kant’s Empirical Account of Human Action.” Philosophers’ Imprint 5: 1–34. Furth, M. 1972. “Monadology.” In: H.  Frankfurt (ed.), Leibniz: A Collection of Critical Essays. Garden City, NY: Anchor Books, pp. 99–135. Futch, M. J. 2008. Leibniz’s Metaphysics of Time and Space. Berlin: Springer. Garber, D. 1985. “Leibniz and the Foundations of Physics: The Middle Years.” In: K. Okruhlik (ed.), The Natural Philosophy of Leibniz. Dordrecht: Reidel, pp. 27–130. Garber, D. 1987. “How God Causes Motion: Descartes, Divine Sustenance, and Occasionalism.” Journal of Philosophy 84: 567–580. (Repr. Garber 2001, 189–202.) Garber, D. 1994. “Leibniz: Physics and Philosophy.” In: N.  Jolley (ed.), Cambridge Companion to Leibniz. Cambridge: Cambridge University Press, pp. 270–352. Garber, D. 1995. “Leibniz: Physics and Philosophy.” In N.  Jolley (ed.), The Cambridge Companion to Leibniz, New York: Cambridge University Press, pp. 270–352. Garber, D. 2001. Descartes Embodied. Cambridge: Cambridge University Press. Garber, D. 2009. Leibniz: Body, Substance, Monad. Oxford: Oxford University Press. Goldenbaum, U. 1999. “Die Bedeutung der öffentlichen Debatte über das Jugement der Berliner Akademie für die Wissenschaftsgeschichte. Eine Sichtung hartnäckiger Vorurteile.” In: H. Hecht (ed.), Pierre Louis Moreau de Maupertuis. Eine Bilanz nach 300 Jahren. Berlin: Verlag Arno Spitz, pp. 383–417. Goldenbaum, U. 2004. Der Appell an das Publikum. Die öffentliche Debatte in der deutschen Aufklärung 1687–1796. mit Beiträgen von Frank Grunert, Peter Weber, Gerda Heinrich, Brigitte Erker und Winfried Siebers. Berlin: Akademie Verlag. Goldenbaum, U. 2008. “Lessing ein Wolffianer?” In: J.  Stolzenberg and O.-P.  Rudolph (eds.), Christian Wolff und die Europäische Aufklärung. 5 vols. Vol. 4. Hildesheim, New York: Olms, pp. 267–281. [= Christian Wolff: Gesammelte Werke, Abt. III: Materialien und Dokumente, vol. 104] Goldenbaum, U. 2011a. “Spinoza—ein toter Hund? Nicht für Christian Wolff.” Zeitschrift für Ideengeschichte 5: 29–41. Goldenbaum, U. 2011b. “Mendelssohn’s Spinozistic Alternative to Baumgarten’s Pietist Project of Aesthetics.” In: R.  Munk, Moses Mendelssohn’s Metaphysics and Aesthetics. Studies in German Idealism 13. Dordrecht: Springer, pp. 299–327. Goldenbaum, U. 2015. “The Geometrical Method.” Internet Encyclopedia of Philosophy. ISSN 2161–0002, http://www.iep.utm.edu/geo-meth/. Goldenbaum, U. 2016. “How Theological Concerns Favor Empiricism over Rationalism.” In: Arnaud Pelletier (ed.), Leibniz’s Experimental Philosophy. Studia Leibnitiana, Special Issue 46, Wiesbaden: Steiner, pp. 37-63 Gottsched, J. C. 1733–1734. Erste Gründe der gesamten Weltweisheit. 2 vols. Leipzig: (no publisher name). Gottsched, J.  C. 1755. Historische Lobschrift des . . . Freyherrn von Wolff. Halle: Renger, pp. 33–34, 87–88. (Repr. Christian Wolff Biographie in GW I 10.) Gottsched, J. C. 1983. Ausgewählte Werke, vol. 5, part 1, ed. by P. M. Mitchell. Berlin: De Gruyter.

Bibliography  361 Gottsched, J. C. 2007–. Briefwechsel, unter Einschluß des Briefwechsels von Luise Adelgunde Victorie Gottsched. Ed. D. Döring and M. Rudersdorf. Berlin: De Gruyter. Gottsched, L. A. V. 1763. Sämtliche kleinere Gedichte. Leipzig: Breitkopf and Sons. Goy, I. 2008. “Die Teleologie der organischen Natur.” In: O. Höffe (ed.), Immanuel Kant: Kritik der Urteilskraft. Berlin: Akademie Verlag, pp. 223–239. Gueroult, M. 1967. Leibniz, dyamique et métaphysique. 2nd ed. Paris: Aubier-Montaigne. Guyer, P. 1979. Kant and the Claims of Taste. Cambridge, MA: Harvard University Press. Expanded edition, Cambridge: Cambridge University Press, 1997. Guyer, P. 1987. Kant and the Claims of Knowledge. Cambridge: Cambridge University Press. Guyer, P. 1990. “Kant’s Conception of Empirical Law,” Proceedings of the Aristotelian Society, Supp. Vol. 64: 221–242. (Repr. Guyer 2005, 38–55.) Guyer, P. 1991. “Kant’s Ether Deduction and the Possibility of Experience.” In: G. Funke (ed.), Akten des Siebenten Internationalen Kant-Kongresses. Vol. II(1). Bonn: Bouvier Verlag, pp. 119–132. (Repr. Guyer 2005, 74–85.) Guyer, P. 2000. “From a Practical Point of View: Kant’s Conception of a Postulate of Pure Practical Reason.” In: Kant on Freedom, Law, and Happiness. Cambridge: Cambridge University Press, pp. 333–372. Guyer, P. 2001a. “From Nature to Morality: Kant’s New Argument in the ‘Critique of Teleological Judgment’.” In: H.  F.  Fulda and J.  Stolzenberg (eds.), Architektonik und System in der Philosophie Kants. Hamburg: Felix Meiner Verlag, pp. 375–404. (Repr. Guyer 2005, 314–342.) Guyer, P. 2001b. “Organisms and the Unity of Science.” In: E. Watkins (ed.), Kant and the Sciences. Oxford: Oxford University Press, pp. 259–281. (Repr. Guyer 2005, 86–111.) Guyer, P. 2002. “Ends of Reason and Ends of Nature,” Journal of Value Inquiry 36: 161–186. (Repr. Guyer 2005, 169–197.) Guyer, P. 2003a. “Kant’s Principles of Reflecting Judgment.” In: P. Guyer (ed.), Kant’s Critique of the Power of Judgment: Critical Essays. Lanham: Rowman & Littlefield, pp. 1–61. Guyer, P. 2003b. “The Systematicity of Nature: Two Puzzles.” History of Philosophy Quarterly 20: 277–295. (Repr. Guyer 2005, 56–73.) Guyer, P. 2005. Kant’s System of Nature and Freedom. Oxford: Clarendon Press. Guyer, P. 2011. “Kantian Communities.” In: C.  Payne and L.  Thorpe (eds.), Kant and Community. North American Kant Society Publication Series Rochester: University of Rochester Press, pp. 88–137. Guyer, P. 2014. “Freedom, Happiness, and Nature: Kant’s Moral Teleology (Critique of the Power of Judgment, §§83-4, 86-7).” In: I. Goy and E. Watkins (eds.), Kant’s Philosophie der Biologie. Berlin: Walter de Gruyter, pp. 221–238. Hansch, M. G. 1728. Godefredi Guilielmi Leibnitii Principia Philosophiae More Geometrico Demonstrata. Leipzig and Frankfurt: Monath. Hare, P. 1988. Doing Philosophy Historically. Buffalo: Prometheus. Harig, G. 1987. “Galileis ‘Dialog Uber die beiden hauptsächlichen Weltsysteme’ – alte und neue Wissenschaft im Widerstreit.” In: Galileo Galilei: Schriften, Briefe, Dokumente. 2 vols. Ed. A. Mudry. Berlin: Rütten & Loening. Harnack, A.von. 1900. Geschichte der Königlich-Preussischen Akademie der Wissenschaften zu Berlin. Vol. 2. Berlin: The Royal Prussian Academy of Sciences. Harrison, P. 1995. “Newtonian Science, Miracles, and the Laws of Nature.” Journal of the History of Ideas 56: 531–553. Hartmann, G. V. 1737. Anleitung zur Historie der Leibnitzisch-Wolffischen Philosophie und der darinnen von Hn. Prof. Langen erregten Controvers. Frankfurt and Leipzig: Cuno. Hartz, G. and J. A. Cover. 1988. “Space and Time in the Leibnizian Metaphysics.” Nous 22: 493–519.

362 Bibliography Hegel, G.  W.  F. 1986. Werke. 21 vols. Ed. E.  Modenhauer et al. Frankfurt am Main: Suhrkamp. Heimsoeth, H. 1924. “Persönlichkeitsbewußtsein und Ding an sich.” In: Immanuel Kant: Festschrift zur zweiten Jahrhundertfeier seines Geburtstages. Leipzig: Dietrich. Heimsoeth, H. 1929. “Metaphysische Motive in der Ausbildung des Kritischen Idealismus,” Kant-Studien 29: 121–159. Heimsoeth, H. 1956. Studien zur Philosophie Immanuel Kants. Metaphysische Ursprünge und ontologische Grundlagen. Cologne: University of Cologne Press. Heinekamp, A., W. Lenzen, and M. Schneider (eds.). 1990. Mathesis rationis. Festschrift für Heinrich Schepers. Münster: Nodus. Henrich, D. 1963. “Über Kants früheste Ethik.” Kant-Studien 54: 404–431. Henrich, D. 1973. “Die Beweisstruktur von Kants transzendentaler Deduktion.” In: G. Prauss (ed.), Kant: Zur Deutung seiner Theorie von Erkennen und Handeln. Cologne: Kiepenheuer und Witsch, pp. 90–105. Herder, J.  G. 1994. Werke. 10 vols. Ed. G.  Arnold et al. Frankfurt am Main: Deutscher Klassiker Verlag. Hermann, J. 1728. “De mensura virium corporum.” Commentarii Academiae Scientarum Imperialis Petropolitanae 1: 1–42. Hilgard, E. 1980. “The Trilogy of Mind: Cognition, Affection, and Conation.” The Journal of the History of the Behavioral Sciences 16: 107–117. Hinrichs, C. 1971. Preußentum und Pietismus: Der Pietismus in Brandenburg-Preußen als religiös-soziale Reformbewegung. Göttingen: Vandenhoek & Ruprecht. Hißmann, M. 1777. “Bemerkungen über einige Regeln für den Geschichtschreiber philosophischer Systeme; über Dutens Untersuchungen; und über die angebohrnen Begriffe des Plato, Descartes und Leibnitz.” Teutsche Merkur 4th Quarter: 22–52. Hobbes, T. 1994. Leviathan: with selected variants from the Latin edition of 1668. Ed. E. Curley. Indianapolis: Hackett Pub. Co. Hoffmann, P. 1996. “The Being of Leibnizian Phenomena.” Studia Leibnitiana 28: 108–118. Hogan, D. 2005. Rationalism and Causal Realism in Kant’s Metaphysics. Unpublished PhD Dissertation (Yale). Hogan, D. 2009. “Noumenal Affection.” Philosophical Review 118: 502–532. Hogan, D. 2014. “Kant on Foreknowledge of Contingent Truths.” Res Philosophica 91(1): 47–70. Hooker, M. (ed.). 1982. Leibniz: Critical and Interpretative Essays. Minneapolis: University of Minnesota Press. Hunter, I. 2001. Rival Enlightenments: Civil and Metaphysical Philosophy in Early Modern Germany. Cambridge: Cambridge University Press. Huxford, G. 2018. “Evil, the Laws of Nature, and Miracles,” Kant Yearbook 10: 43–62. Insole, C.  J. 2013. Kant and the Creation of Freedom: A Theological Problem. New York: Oxford University Press. Jachmann, R. B. 1804. Immanuel Kant geschildert in Briefen an einen Freund. Königsberg: Nicolovius. Jacobi, F. H. 1787. David Hume über den Glauben; oder Idealismus und Realismus. (Repr. Jacobi (1815). Werke. Leipzig: Gerhard Fleischer.) Jacobi, F. H. 1987. Briefwechsel 1782–1784. Ed. P. Bachmeier et al. Stuttgart-Bad Cannstatt: Frommann-Holzboog. Jauernig, A. 2008. “Kant’s critique of the Leibnizian philosophy : Contra the Leibnizians, but pro Leibniz.” In: D. Garber and B. Longuenesse (eds.), Kant and the Early Moderns. Princeton, NJ: Princeton University Press, pp. 41–63.

Bibliography  363 Jauernig, A. 2010. “Disentangling Leibniz’s Views on Relations and Extrinsic Denominations.” Journal of the History of Philosophy 48: 171–205. Jauernig, A. 2011. “Kant, the Leibnizians and Leibniz.” In: B.  C.  Look (ed.), Continuum Companion to Leibniz. London: Continuum, pp. 289–309. Jauernig, A. 2019. “Finite Minds and Their Representations in Leibniz and Kant.” Internationales Jahrbuch des Deutschen Idealismus/International Yearbook of German Idealism 17: 47–80. Jauernig, A. 2021. The World according to Kant: Appearances and Things in Themselves in Critical Idealism, Oxford/New York: Oxford University Press. Jolley, N. 1984. Leibniz and Locke: A Study of the New Essays on Human Understanding, Oxford: Clarendon Press; New York: Oxford University Press. Jolley, N. 1986. “Leibniz and Phenomenalism.” Studia Leibnitiana 18: 38–51. Jolley, N. 2005. “Leibniz and Occasionalism.” In: J.  A.  Cover and D.  Rutherford (eds.), Leibniz: Nature and Freedom. New York: Oxford University Press, pp. 121–134. Jorati, J. 2017. Leibniz on Causation and Agency, Cambridge: Cambridge University Press. Jorgensen, L. 2009 “The Principle of Continuity and Leibniz’s Theory of Consciousness.” Journal of the History of Philosophy 47: 223–248. Juškevič, A. P. (ed.). 1961. Die Berliner und die Petersburger Akademie der Wissenschaften im Briefwechsel Leonhard Eulers. Berlin: Akademie Verlag. Kain, P. 2001. “A Preliminary Defense of Kantian Prudence.” In: V.  Gerhardt, R.P. Horstmann, and R. Schumacher (eds.), Kant und die Berliner Aufklärung: Akten des IX. Internationaler Kant-Kongresses. Vol. III. Berlin, de Gruyter, pp. 239–246. Kain, P. 2010. “Practical Cognition, Intuition, and the Fact of Reason.” In: B. Lipscomb and J. Krueger (eds.), Kant’s Moral Metaphysics: God, Freedom, and Immortality. Berlin, de Gruyter, pp. 211–230. Kant, I. 1902–. Kants gesammelte Schriften. Berlin: Königlich preussischen Akademie der Wissenschaften (now Walter de Gruyter). Kant, I. 1992a. Lectures on Logic. Ed. and trans. M.  J.  Young. Cambridge: Cambridge University Press. Kant, I. 1992b. Theoretical Philosophy, 1755–1770. Ed. and trans. D. Walford. Cambridge: Cambridge University Press. Kant, I. 1996. Practical Philosophy. Ed. and trans. M.  Gregor. Cambridge: Cambridge University Press. Kant, I. 1997. Lectures on Metaphysics. Ed. and trans. K.  Ameriks and S.  Naragon. Cambridge: Cambridge University Press. Kant, I. 1998. Critique of Pure Reason. Ed. and trans. P. Guyer and A. W. Wood. Cambridge: Cambridge University Press. Kant, I. 1999. Correspondence. Ed. and trans. A.  Zweig. Cambridge: Cambridge University Press. Kant, I. 2000. Critique of the Power of Judgment. Ed. and trans. P. Guyer and E. Matthews. Cambridge: Cambridge University Press. Kant, I. 2002. Theoretical Philosophy after 1781. Ed. and trans. H.  Allison and P.  Heath. Cambridge: Cambridge University Press. Kant, I. 2005. Notes and Fragments. Ed. and trans. P.  Guyer, C.  Bowman, P.  Guyer, and F. Rauscher. Cambridge: Cambridge University Press. Kant, I. 2007. Anthropology, History and Education. Ed. and trans. G. Zöller and R. Louden. Cambridge: Cambridge University Press. Kant, I. 2011. Observations on the Feeling of Beauty and the Sublime. Ed. P. Frierson and P. Guyer. Cambridge: Cambridge University Press.

364 Bibliography Khamara, E. 1993. “Leibniz’s Theory of Space: A Reconstruction.” Philosophical Quarterly 43: 472–488. Kirchner, J. 1928. Die Grundlagen des deutschen Zeitschriftenwesens mit einer Gesamtbibliographie der deutschen Zeitschriften bis zum Jahre 1790. Part 1. Leipzig: Hiersemann. König, S. 1748. Dissertatio inauguralis de optimis Wolffianae et Newtonianae philosophiae methodis earumque amico consensu. Hanover: Franeker. König, S. 1751. “De Universali Principio Aequilibrii et motus in Vi viva reperto, deque nexu inter Vim Vivam et Actionem, utriusque Minimo, Dissertatio.“ Nova Acta eruditorum. Leipzig March 1751, part I, pp. 125–135, and part II, pp. 162–176. Reprint in: L. Euler, Opera omnia. Basel, Zürich: Birkhäuser, Orell/Füssli. Series 2, vol. 5, pp. 317–324. Korsgaard, C.  M. 1996. The Sources of Normativity. Cambridge: Cambridge University Press. Korsgaard, C.  M. 1997. “The Normativity of Instrumental Reason.” In: G.  Cullity and B. Gaut (eds.), Ethics and Practical Reason. Oxford, Clarendon Press, pp. 215–254. Kuehn, M. 2001. Kant: A Biography. Cambridge: Cambridge University Press. Kulstad, M. 1977. “Leibniz’s Conception of Expression.” Studia Leibnitiana 9: 55–76. Lamarra, A., R. Palaia, and P. Pimpinella. 2001. Le primi traduzione della “Monadologia” di Leibniz (1720–21): introduzione storico-critica, sinossi dei testi, concordanze contrastive. Florence: Olschki. Lange, J. 1723. Modesta Disquisitio novi philosophiae systematis de Deo, mundo et homine, et praesertim de harmonia commercii inter animam et corpus praestabilita . . . Halle: Orphonotrophei. Lange, J. 1735. Der philosophische Religionsspötter, Halle: Walther. Langton, R. 1998. Kantian Humility. Our Ignorance of Things in Themselves. Oxford: Oxford University Press. Lavin, D. 2004. “Practical Reason and the Possibility of Error.” Ethics 114: 424–457. Laywine, A. 1993. Kant’s Early Metaphysics and the Origins of the Critical Philosophy. North American Kant Society Studies in Philosophy 3. Atascadero: Ridgeview. Laywine, A. 2003. “Kant on Sensibility and the Understanding in the 1770s.” Canadian Journal of Philosophy 33: 443–482. Laywine, A. 2010. “Kant and Lambert on the Role of Geometrical Postulates in the Reform of Metaphysics.” In: M.  Domski and M.  Dickson (eds.), Discourse on a New Method: Reinvigorating the Marriage of History and Philosophy of Science. Chicago: Open Court, pp. 113–134. Laywine, A. 2020. Kant’s Transcendental Deduction: A Cosmology of Experience. Oxford: Oxford University Press. Leduc, C. 2011. “La Doctrine leibnizienne de l’idée.” In: H. Breger, J. Herbst, and S. Erdner (eds.), Natur und Subjekt: IX. Internationaler Leibniz-Kongress. Vol. 3. Hannover: Akademie Verlag, pp. 582–590. Lee, S. 2004. “Leibniz on Divine Concurrence.” Philosophical Review 113: 203–248. Lehner, U. 2007. Kants Vorsehungskonzept auf dem Hintergrund der deutschen Schulphilosophie und -theologie. Leiden: Brill. Leibniz, G.  W. 1720. Lehrsätze über die Monadologie. Trans. H.  Köhler. Frankfurt and Leipzig: Johann Meyer. Leibniz, G.  W. 1765. Oeuvres philosophiques, latines et françaises, de feu Mr. De Leibniz, tirées des ses manuscrits, qui se conservent dans la bibliothèque royale à Hanovre. Ed. E. R. Raspe. Amsterdam and Leipzig.

Bibliography  365 Leibniz, G. W. 1768. Opera omnia, nunc primum collecta . . . 6 vols. Ed. L. Dutens. Geneva: Fratres de Tournes. (Repr. Hildesheim: Olms, 1989.) [Cited by volume and page number.] Leibniz, G. W. 1849–1863. Leibnizens mathematische Schriften. 7 vols. Ed. C. I. Gerhardt. Halle: Schmidt. (Repr. Hildesheim: Olms, 1971.) [= GM followed by volume and page number] Leibniz, G. W. 1860. Briefwechsel zwischen Leibniz und Christian Wolff. Ed. C. I. Gerhardt. Halle: Schmidt. Leibniz, G.  W. 1875–1890. G.  W.  Leibniz: Die philosophischen Schriften. 7 vols. Ed. C. I. Gerhardt. Berlin: Weidmann. (Repr. Hildesheim: Olms, 1965.) [= GP followed by volume and page number] Leibniz, G. W. 1899. Der Briefwechsel von Gottfried Wilhelm Leibniz mit Mathematikern. Ed. C. I. Gerhardt. Berlin: Mayer and Müller. (Repr. Hildesheim: Olms, 1987.) Leibniz, G. W. 1903. Opuscules et Fragments Inédits de Leibniz. Ed. L. Couturat. Paris: Félix Alcan. (Repr. Hildesheim: Olms, 1988.) Leibniz, G.  W. 1923 –. Gottfried Wilhelm Leibniz: Sämtliche Schriften und Briefe. Ed. Deutsche Akademie der Wissenschaften. Darmstadt and Berlin: Akademie Verlag. [= A followed by series, volume, and page number] Leibniz, G. W. 1956. The Leibniz-Clarke Correspondence. Ed. H. G. Alexander. Manchester: Manchester University Press. [= LC] Leibniz, G. W. 1967. The Leibniz-Arnauld Correspondence. Ed. H. T. Mason. Manchester: Manchester University Press. [= LA] Leibniz, G. W. 1969. Philosophical Papers and Letters. Ed. and trans. L. E. Loemker. 2nd ed. Dordrecht and Boston: Reidel. [= L] Leibniz, G.  W. 1973 (2nd ed., 1995). Philosophical Writings. Ed. M.  Morris and G. H. R. Parkinson. London: Dent. [= MP] Leibniz, G. W. 1985. Theodicy. Ed. and trans. E. M. Huggard. La Salle, IL: Open Court. [= H] Leibniz, G.  W. 1989. Philosophical Essays. Ed. and trans. R.  Ariew and D.  Garber. Indianapolis: Hackett. [= AG] Leibniz, G.  W. 1995. La Caractéristique géométrique. Ed. J.  Echeverría. Trans. M. Parmentier. Paris: Vrin. Leibniz, G. W. 1998. Philosophical Texts. Ed. R. S. Woolhouse and R. Francks. New York: Oxford University Press. Leibniz, G. W. 2000. Leibniz–Clarke Correspondence. Ed. and trans. R. Ariew. Indianapolis: Hackett. Leibniz, G. W. 2001. The Labyrinth of the Continuum: Writings on the Continuum Problem, 1672–86. Ed. and trans. R. T. W. Arthur. New Haven: Yale University Press. Leibniz, G.  W. 2005. Confessio philosophi. Papers Concerning the Problem of Evil, 1671–1678. Ed. and trans. R. SleighJr. New Heaven and London: Yale University Press. Leibniz, G. W. 2007. The Leibniz-Des Bosses Correspondence. Ed. and trans. B. C. Look and D. Rutherford. New Haven: Yale University Press. [= LDB] Lessing, G. E. 1751a. Kleinigkeiten. Leipzig and Frankfurt: Metzler. Lessing, G. E. 1751b. “Verse on Kant.” In: Neuestes aus dem Reiche des Witzes, July 1751. Berlin: Voss, p. 32. Lessing, G. E. 1886. Lessings Sämtliche Schriften. Ed. K. Lachmann and F. Muncker. Vol. I. 3rd ed., Stuttgart: Gösschen, p. 41. Le Sueur, A. 1896. Maupertuis et ses correspondants. Lettres inédites du Grand Frédéric, du Prince Henri de Prusse . . . d’Euler, de Kaestner, de Koenig, de Haller, de Condillac. Montreuil-sur-Mer: Notre-Dame des Prés.

366 Bibliography Levey, S. 1998. “Leibniz on Mathematics and the Actually Infinite Division of Matter.” Philosophical Review 107: 49–96. Levey, S. 2005. “Leibniz on Precise Shapes and the Corporeal World.” In: J. A. Cover and D. Rutherford (eds.), Leibniz: Nature and Freedom. New York: Oxford University Press, pp. 69–94. Lewes, G. H. 1855. The Life and Works of Goethe. London: David Nutt. Lewis, D. K. 1973. Counterfactuals, Oxford: Basil Blackwell. Lin, M.  2011. “Leibniz’s Philosophical Theology.” In: B.  Look (ed.), The Continuum Companion to Leibniz. London: Continuum. Locke, J. 1755. Johann Lockens Anleitung des menschlichen Verstandes zur Erkäntniß der Wahrheit nebst desselben Abhandlung von den Wunderwerken, trans. by G.  D.  Kypke, Königsberg: Hartung. Lodge, P. 2001. “Leibniz’s Notion of an Aggregate.” British Journal of the History of Philosophy 9: 467–486. Logan, I. 2007. “Whatever Happened to Kant’s Ontological Argument?” Philosophy and Phenomenological Research 74: 346–363. Longuenesse, B. 1998. Kant and the Capacity to Judge. Trans. C.  Wolfe. Princeton: Princeton University Press. Look. B. C.1999. Leibniz and the “Vinculum Substantiale”. Studia Leibnitiana Sonderheft 30. Stuttgart: Steiner. Look, B.  C. 2007. “Perfection, Power and the Passions in Spinoza and Leibniz.” Revue Roumaine de la Philosophie 51: 21–38. Look. B. C.2010. “Leibniz’s Metaphysics and Metametaphysics: Idealism, Realism and the Nature of Substance.” Philosophy Compass 5(11): 871–879. Look, B. C. (ed.). 2011a. The Continuum Companion to Leibniz. London: Continuum. Look, B. C. 2011b. “Grounding the Principle of Sufficient Reason: Leibnizian Rationalism versus the Humean Challenge.” In: C. Fraenkel, D. Perinetti, and J. E. H. Smith (eds.), The Rationalists: Between Tradition and Revolution. Dordrecht: Springer, pp. 201–219. Look, B. C. 2017. “The Ideal and the Real in Leibniz’s Metaphysics.” In: T. Leinkauf (ed.), Harmony and Reality in Leibniz’s Late Metaphysics. Studia Leibnitiana Sonderheft 51. Stuttgart: Steiner, pp. 41–54. Look, B. C. 2018. “Baumgarten’s Rationalism.” In: C. Fugate and J. Hymers (eds.), Alexander Baumgarten’s Metaphysics: Sources, Interpretation and Influence. Oxford: Oxford University Press, pp. 10–22. Löscher, E. V. 1735–1742. Frühaufgelesene Früchte der Theologischen Sammlung von Alten und Neuen. Leipzig: Braun. Löw, R. 1980. Philosophie des Lebendigen: Der Begriff des Organischen bei Kant, sein Grund und seine Aktualität. Frankfurt am Main: Suhrkamp Verlag. Luck, M. 2011. “Defining Miracles: Violations of the Laws of Nature.” Philosophy Compass 6: 133–141. Ludovici, C. G. 1737. Ausführlicher Entwurf einer vollständigen Historie der Leibnizischen Philosophie zum Gebrauch seiner Zuhörer. 2 vols. Leipzig: Löwe. (Repr. Hildesheim: Olms, 1966.) Mavrodes, G. I. 1985. “Miracles and the Laws of Nature.” Faith and Philosophy 2: 333–346. McDonough, J. K. 2008. “Leibniz’s Two Realms Revisited,” Noûs 42: 673–696. McGrew, T. 2019. “Miracles”. In: E.  N.  Zalta (ed.), Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/archives/spr2019/entries/miracles/. McGuire, J. E. 1976. “‘Labyrinthus Continui’: Leibniz on Substance, Activity, and Matter.” In: P.  K.  Machamer and R.  G.  Turnbull (eds.), Motion and Time, Space and Matter:

Bibliography  367 Interrelations in the History and Philosophy of Science. Columbus: Ohio State University Press, pp. 291–326. Mackie, J. 1982. The Miracle of Theism: Arguments for and against the Existence of God. New York: Oxford University Press. McLaughlin, P. 1990. Kant’s Critique of Biological Explanation: Antinomy and Teleology. Lewiston: Edwin Mellen Press. McRae, R. 1976. Leibniz; Perception, Apperception, and Thought. Toronto: University of Toronto Press. Malebranche, N. 1958–1984. Oeuvres complètes de Malebranche. A.  Robinet (ed.), Paris: Vrin. [= OCM followed by volume and page number] Malebranche, N. 1993. Treatise on Ethics. Trans. C. Walton. Dordrecht: Kluwer. Malebranche, N. 1997a. The Search for Truth and Elucidations of the Search for Truth. Ed. T. Lennon and P. Olscamp. Cambridge: Cambridge University Press. Malebranche, N. 1997b. Dialogues on Metaphysics and on Religion. Ed. N.  Jolley and D. Scott. Cambridge: Cambridge University Press. Martin, G. 1966. “Kants Auseinandersetzung mit der Bestimmung der Phänomene durch Leibniz und Wolff als verworrene Vorstellungen.” In: F.  Kaulbach and J.  Ritter (eds.), Kritik und Metaphysik. Festschrift für Heinz Heimsoeth. Berlin: De Gruyter, pp. 99–105. Mates, B. 1986. The Philosophy of Leibniz: Metaphysics and Language. New York: Oxford University Press. Meier, G. F. 1755–1759. Metaphysik. 4 vols. Halle: Johann Justinus Gebauer. Mercer, C. 2001. Leibniz’s Metaphysics: Its Origins and Development. Cambridge: Cambridge University Press. Messina, J. and D. Rutherford. 2009. “Leibniz on Compossibility.” Philosophical Compass 4: 962–977. Molina, L.  de. 1988. On Divine Foreknowledge. Part IV of Concordia. Ed. and trans. A. Freddoso. Ithaca: Cornell University Press. Morris, T. V. 1987. Anselmian Explorations: Essays in Philosophical Theology. South Bend, IN: University of Notre Dame Press. Morrison, J.  C. 1993. “Christian Wolff ’s Criticism of Spinoza.” Journal of the History of Philosophy 31: 405–420. Morriston, W. 1985. “Is God 'Significantly Free?’” Faith and Philosophy 2: 257–263. Mugnai, M. 1990. “Leibniz’s Nominalism and the Reality of Ideas in the Mind of God.” In: A. Heinekamp, W. Lenzen, and M. Schneider (eds.), Mathesis Rationis. Münster: Nodus, pp.153–167. Mugnai, M. 1992. Leibniz’ Theory of Relations. Studia Leibnitiana Supplementa 28. Stuttgart: Franz Steiner Verlag. Mugnai, M. 2010. “Leibniz and ‘Bradley’s Regress.” The Leibniz Review 20: 1–12. Mugnai, M. 2012. “Leibniz’s Ontology of Relations: A Last Word?” Oxford Studies in Early Modern Philosophy 6: 171–208. Mylius, C. 1754. Vermischte Schriften. Collected by G. E. Lessing. Berlin: Johann Carl Spener. Nadler, S. 1996. “‘No Necessary Connection’: The Medieval Roots of the Occasionalist Roots of Hume.” The Monist 79: 448–466. Nadler, S. 2011. “Spinoza, Leibniz, and the Gods of Philosophy.” In: C.  Fraenkel, D.  Perenetti, and J.  E.  H.  Smith (eds.), The Rationalists: Between Tradition and Innovation. Dordrecht: Springer, pp. 167–182. Nannini, Alssandro. Forthcoming. “Impetus aestheticus. Baumgarten on Physics and Aesthetics.” O.  Matei and T.  Prunea-Bretonnet (eds.), From Natural History to the Metaphysics of Nature. Perspectives on Sciences.

368 Bibliography Newton, I. 1952 [1730]. Opticks. New York: Dover. Newton, I. 1999. The Principia. Mathematical Principles of Natural Philosophy. Ed. I. B. Cohen and A. Whitman. Berkeley: University of California Press. Newton, I. 2004. Philosophical Writings. Ed. A.  Janiak. Cambridge: Cambridge University Press. Nisbet, H. B. 2008. Lessing. Eine Biographie. Munich: Beck. Noël, F. 1711. Sinensis Imperii Libri Classici Sex. Prague: Kamenicky. Nuyen, A. T. 2002. “Kant on Miracles.” History of Philosophy Quarterly 19: 309–323. Oelrich, J.  C.  C. 1782. “Tagebuch einer gelehrten Reise 1750.” In: J.  Bernoulli (ed.), Sammlung Kurzer Reisebeschreibungen. Vol. 5. Berlin-Dessau. Parkinson, G.  H.  R. 1981. “Kant as a Critic of Leibniz: The Amphiboly of Concepts of Reflection.” Revue Internationale de Philosophie 35: 302–314. Parkinson, G.  H.  R. 1982. “The Intellectualization of Thought and Appearances.” In: M. Hooker (ed.), Leibniz: Critical and Interpretative Essays. Minneapolis: University of Minnesota Press, 3–20. Paton, H. 1969. “Kant and the Errors of Leibniz.” In: L. W. Beck (ed.), Kant Studies Today. Lasalle, IL: Open Court, pp. 72–87. Paull, R. 1992. “Leibniz and the Miracle of Freedom.” Nous, 26: 218–235. Peddicord, C. 2001. Die Wunderkritik Immanuel Kants. Giessen: Geistwissenschaften Online-Publicationen. Pereboom, D. 1991. “Kant’s Amphiboly.” Archive für Geschichte der Philosophie 73: 50–70. Pereboom, D. 2006. “Kant on Transcendental Freedom.” Philosophy and Phenomenological Research 73: 537–567. Pereboom, D. 2009. Review of William L. Rowe, Can God be Free? Philosophical Review 118: 121–127. Perler  D. and U.  Rudolph. 2000. Occasionalismus. Theorien der Kausalität im arabischislamischen und im europäischen Denken. Göttingen: Vandenhoeck & Ruprecht. Pettit, P. and M.  Smith. 1996. “Freedom in Belief and Desire.” Journal of Philosophy 93: 429–449. Puryear, S.  M. 2005. “Was Leibniz Confused about Confusion?” The Leibniz Review 15: 95–124. Putnam, H. 1975. “The Meaning of ‘Meaning’.” In: K.  Gunderson (ed.), Language, Mind and Knowledge. Minnesota Studies in the Philosophy of Science VII. Minneapolis: University of Minnesota Press. Radner, M. 1998. “Unlocking the Second Antinomy: Kant and Wolff.” Journal of the History of Philosophy 36: 413–441. Reichl, P. 2019. “Leaving the Enchanted World Behind: Kant on the Order of Nature, Empirical Space and the Possibility of Miracles,” Kantian Review 24: 103–25. Robinet, A., 1986. Architectonique Disjonctive, Automates Systémiques et Idéalité Transcendantale dans l’œuvre de G.W. Leibniz. Paris: J. Vrin. Rosen, G. 2010. “Metaphysical Dependence: Grounding and Reduction.” In: R. Hale and A.  Hoffmann (eds.), Modality: Metaphysics, Logic and Epistemology. Oxford: Oxford University Press. Ross, G.  M. and A.  McWalter 1990. Kant and His Influence. Bristol: Thoemmes Antiquarian Press. Rowe, W. L. 2002. “Can God Be Free?” Faith and Philosophy 19: 405–424. Rowe, W. L. 2006. Can God Be Free? Oxford: Oxford University Press. Russell, B. 1900 (2nd ed., 1937). A Critical Exposition of the Philosophy of Leibniz. London George Allen & Unwin.

Bibliography  369 Russell, B. 1945. A History of Western Philosophy. London: Routledge. Rutherford, D. 1994. “Leibniz and the Problem of Monadic Aggregation.” Archiv für Geschichte der Philosophie 76: 65–90. Rutherford, D. 1995. Leibniz and the Rational Order of Nature. New York: Cambridge University Press. Rutherford, D. 2004. “Idealism Declined: Leibniz and Christian Wolff.” In: P. Lodge (ed.), Leibniz and His Correspondents. Oxford: Oxford University Press, pp. 214–237. Rutherford, D. 2008a. “Leibniz as Idealist.” Oxford Studies in Early Modern Philosophy 4: 141–190. Rutherford, D. 2008b. “Leibniz on Infinitesimals and the Reality of Force.” In: U.  Goldenbaum and D.  Jesseph (eds.), Infinitesimal Differences: Controversies between Leibniz and his Contemporaries. Berlin: De Gruyter, pp. 255–280. Schepers, H. 1959. Andreas Rüdigers Methodologie und ihre Voraussetzungen. Ein Beitrag zur Geschichte der deutschen Schulphilosophie. Kant-Studien. Erg.-Heft 78. Cologne: Universitäts Verlag. Schmidt, J. L. 1735. Die göttlichen Schriften vor den Zeiten des Messie Jesus, Der erste Theil worinnen die Gesetze der Israelen enthalten sind nach einer freyen Übersetzung welche durch und durch mit Anmerkungen erläutert und bestätiget wird. Wertheim: Nehr. (Repr. Christian Wolff, Gesammelte Werke, III, 128, Hildesheim-New York: Olms, 2011, with an introduction by Ursula Goldenbaum.) Schmidt, S. (ed.). 1983. Alma Mater Jenensis. Geschichte der Universität Jena. Weimar: Böhlau. Schmucker, J. 1976. “Was entzündet in Kant das große Licht von 1769?” Archiv für Geschichte der Philosophie 58: 393–434. Schneewind, J.  B. 1996. “Voluntarism and the Foundations of Ethics.” Proceedings and Addresses of the American Philosophical Association 70: 25–41. Schneewind, J. B. 1998. The Invention of Autonomy: A History of Modern Moral Philosophy. Cambridge: Cambridge University Press. Schneewind, J.  B. 2010. Essays on the History of Moral Philosophy. Oxford: Oxford University Press. Schneider, W. 2004. “Monaden und Dinge and sich—Überlegungen zu Leibniz und Kant.” Studia Leibnitiana 36: 70–80. Schönecker, D. 1999. Kant: Grundlegung III. Die Deduktion des kategorischen Imperativs. Freiburg: Karl Alber. Schönecker, D. and A.  Wood. 2002. Kants 'Grundlegung zur Metaphysik der Sitten’: Ein einführender Kommentar. Paderborn: Schöningh. Schönfeld, M. 2000. The Philosophy of the Young Kant: The Pre-Critical Project. Oxford: Oxford University Press. Schrecker, P. and A. M. Schrecker (eds.). 1965 Leibniz: Monadology and other Philosophical Essays, Indianapolis: Bobbs-Merrill. Sider, T. 2007. “Parthood.” The Philosophical Review 116: 51–91. Simmons, A. 2001 “Changing the Cartesian Mind: Leibniz on Sensation, Representation and Consciousness.” Philosophical Review 110: 31–75. Sleigh, R.1990. Leibniz & Arnauld: A Commentary on their Correspondence. New Haven: Yale University Press. Smit, H. 2009. “Kant on Apriority and the Spontaneity of Cognition.” In: S. Newlands and L. M. Jorgensen (eds.), Metaphysics and the Good: Themes From the Philosophy of Robert Merrihew Adams. Oxford University Press. Spalding, P. S. 1998. Seize the Book, Jail the Author. Johann Lorenz Schmidt and Censorship in Eighteenth-Century Germany. West Lafayette, IN: Purdue University Press.

370 Bibliography Spiess, O. 1929. Leonhard Euler. Ein Beitrag zur Geistesgeschichte des 18. Jahrhunderts. Leipzig: Huber. Spinoza, B. 1982. The Ethics and Selected Letters. Trans. S. Shirley. Indianapolis: Hackett. Stang, N. 2010. “Kant’s Possibility Proof.” History of Philosophy Quarterly 27: 275–299. Stang, N. 2016. Kant’s Modal Metaphysics. Oxford University Press. Steinmetz, M. 1958. Geschichte der Universität Jena 1548/58–1958: Darstellung. Jena: Fischer. Storrie, S. 2013. “Kant’s 1768 Attack on Leibniz’ Conception of Space.” Kant-Studien 104: 145–166. Suarez, F. (S.J.), 2002 [1597]. On Creation, Conservation, and Concurrence: Metaphysical Disputations 20–22. Trans. A. Freddoso. South Bend, IN: St. Augustine’s Press. Swinburne, R., 1970. The Concept of Miracle. London: Macmillan. Swoyer, C. 1995. “Leibnizian Expression.” Journal of the History of Philosophy 33: 65–99. Talbott, T. B. 1988. “On the Divine Nature and the Nature of Divine Freedom.” Faith and Philosophy 5: 3–24. Thöle, B. 1991. Kant und das Problem der Gesetzmäßigkeit der Natur. Berlin: Walter de Gruyter. Thomasius, C. 2007. Essays on Church, State, and Politics. Ed. I.  Hunter, T.  Ahnert, and F. Grunert. Natural Law and Enlightenment Classics. Liberty Fund: Indianapolis. Timmermann, J. 2003. Sittengesetz und Freiheit. Berlin, Walter de Gruyter. Tonelli, G. 1959. Elementi metodologici e metafisi in Kant dal 1745 al 1768. Turin: Edizioni di Filosofia. Tonelli, G. 1967. “Kant Und Die Antiken Skeptiker.” In: H. Heimsoeth, D. Henrich, and G.  Tonelli (eds.), Studien zu Kants Philosophischer Entwicklung. Hildesheim: Georg Olms, pp. 93–123. Tonelli, G. 1974. “Leibniz on Innate Ideas and the Early Reactions to the Publication of the Nouveaux Essais (1765).” Journal of the History of Philosophy 12: 437–454. Trillmich, R. 1914. Christlob Mylius. Ein Beitrag zum Verständnis seines Lebens und seiner Schriften. Dissertation (Leipzig). Twelftree, G. 2011. The Cambridge Companion to Miracles. New York: Cambridge University Press. Vaihinger, H. 1881. Commentar zu Kants Kritik der reinen Vernunft. Stuttgart: Spemann. Vaihinger, H. 1927a. Die Philosophie des Als Ob. Berlin: Reuther & Reichard. Vaihinger, H. 1927b. Kant und die Als-Ob Philosophie. Stuttgart-Bad. Cannstatt: Frommann. Vailati, E. 1997. Leibniz and Clarke: A Study of their Correspondence. New York: Oxford University Press. Van Cleve, J. 1988. “Inner States and Outer Relations: Kant and the Case for Monadism.” In: P. Hare (ed.), Doing Philosophy Historically. Buffalo: Prometheus, pp. 231–247. Vilhauer, B. 2004. “Can We Interpret Kant as a Compatibilist about Determinism and Moral Responsibility?” British Journal for the History of Philosophy 12: 719–730. Van Cleve, J. 1999. Problems from Kant. New York: Oxford University Press. Van Inwagen, P. 1983. An Essay on Free Will. Oxford: Oxford University Press. Visser, S. and T.  Williams. 2001. “Anselm’s Account of Freedom.” Canadian Journal of Philosophy 31: 221–244. Voltaire. 1820–1822. Oeuvres completes de Voltaire. 60 vols. Paris: Carez, Thomine and Fortic. Walker, R. 2010. “Kant on the Number of Worlds.” British Journal of the History of Philosophy 18: 821–843.

Bibliography  371 Ward, A. 1974. Book Production, Fiction, and the German Reading Public, 1740–1800. Oxford: Clarendon Press. Warda, A. 1922. Immanuel Kants Bücher. Berlin: Martin Breslaner. Waschkies, H.-J. 1987. Physik und Physikotheologie des jungen Kant. Die Vorgeschichte seiner Allgemeinen Naturgeschichte und Theorie des Himmels, Amsterdam: De Grüner. Watkins, E. 1995. “The Development of Physicus-Influxus in 18th Century Germany. Gottsched, Crusius, and Knutzen.” The Review of Metaphysics 49: 295–339. Watkins, E. 1997. “The Laws of Motion from Newton to Kant.” Perspectives on Science 5: 311–348. Watkins, E. 1998. “From Pre-established Harmony to Physical Influx: Leibniz’s Reception in Eighteenth Century Germany.” Perspectives on Science 6: 136–203. Watkins, E. 2005. Kant and the Metaphysics of Causality. Cambridge: Cambridge University Press. Watkins, E. 2006. “On the Necessity and Nature of Simples: Leibniz, Wolff, Baumgarten and the Pre-Critical Kant.” Oxford Studies in Early Modern Philosophy 3: 261–314. Watkins, E. 2008. “Die Antinomie der teleologischen Urteilskraft und Kants Ablehnung alternativer Teleologien.” In: O.  Höffe (ed.), Immanuel Kant: Kritik der Urteilskraft. Berlin: Akademi Verlag, pp. 241–258. Watkins, E. (ed.). 2009. Kant’s Critique of Pure Reason Background Source Materials. Cambridge: Cambridge University Press. Watkins, E. 2010. “Kant and the Hiddenness of God.” In: K. James and L. Benjamin (eds.), Kant’s Moral Metaphysics. Berlin: DeGruyter, pp. 255–290. Watkins, E. 2013. “The Early Kant’s (Anti-) Newtonianism.” Studies in History and Philosophy of Science 44: 429–437. Watkins, E. 2017. “The Foundations of Transcendental Idealism: The Origin of Kant’s Distinction between Sensibility and Understanding.” In: J. O’Shea(ed.), Kant’s Critique of Pure Reason: A Critical Guide. Cambridge: Cambridge University Press, pp. 9–27. Watkins, E. and M. Stan. 2014. “Kant’s Philosophy of Science.” In: E. V. Zalta (ed.), Stanford Encylopedia of Philosophy. . Watson, J. 2012a. “Form, Harmony, and Mechanism in Leibniz’s Philosophy of Laws.” Doctoral Dissertation (Purdue). Watson, J.  L. 2012b. “Leibniz on the Laws of Nature and the Best Deductive System.” Studies in History and Philosophy of Science Part A 43: 577–584. Wierenga, E. 1989. The Nature of God: An Inquiry into Divine Attributes. Ithaca, NY: Cornell University Press. Wierenga, E. 2002. “The Freedom of God.” Faith and Philosophy 19: 425–436. Wierenga, E. 2007. “Perfect Goodness and Divine Freedom.” Philosophical Books 48: 207–216. Willaschek, M. 1998. “Phaenomena/Noumena und die Amphibolie der Reflexionsbegriffe.” In: G.  Mohr and M.  Willaschek (eds.), Immanuel Kant: Kritik der reinen Vernunft. Berlin: Akademie Verlag, 325–351. Wilson, C. 1989. Leibniz’s Metaphysics: A Historical and Comparative Study. Princeton, NJ: Princeton University Press. Wilson, C. 1990. “Confused Perceptions, Darkened Concepts: Some Features of Kant’s Leibniz-Critique.” In: G.  M.  Ross and A.  McWalter (eds.), Kant and His Influence. Bristol: Thoemmes Antiquarian Press, pp. 73–103. Wilson, C. 1995. “The Reception of Leibniz in the Eighteenth Century.” In: N. Jolley (ed.), Cambridge Companion to Leibniz. Cambridge: Cambridge University Press, pp. 442–474.

372 Bibliography Wilson, M. 1999a. “Confused Ideas.” In: Ideas and Mechanism. Princeton: Princeton University Press, pp. 322–335. Wilson, M. 1999b. “Confused vs. Distinct Perception in Leibniz: Consciousness, Representation, and God’s Mind.” In: Ideas and Mechanism. Princeton: Princeton University Press, pp. 336–352. Wilson, M. 1999c. Ideas and Mechanism. Princeton: Princeton University Press. Wilson, C. 2005. “Confused Perceptions, Darkened Concepts: Some Features of Kant’s Leibniz-Critique.” In: G. M. Ross (ed.), Kant and His Influence, New York: Continuum, pp. 73–103. Winter, E. 1957. Die Registres der Berliner Akademie der Wissenschaften 1746–1766. Dokumente für das Wirken Leonhard Eulers in Berlin. Berlin: Akademie Verlag. Wolff, C. 1713–1715. Elementa matheseos universae. 2 vols. Halle. [Ed. 1742 = GW II 29–30] Wolff, C. 1720. Vernunfftige Gedancken von Gott, der Welt und der Seele des Menschen, auch allen Dingen überhaupt. Halle. [= GW I 3] Wolff, C. 1728a. Philosophia prima, sive Ontologia . . . Frankfurt. [Ed. 1736 = GW II 3] Wolff, C. 1728b. “Principia Dynamica.” In: Commentarii Academiae Scientarum Imperialis Petropolitanae. Vol. 1. St. Petersburg, pp. 217–238. Wolff, C. 1730. Elementa Matheseos Universeae. Halle: J. E. Hoffmann. Wolff, C. 1736. Theologia Naturalis, methodo scientifica pertractata. Frankfurt and Leipzig: Renger. Wolff, C. 1737a. Theologia Naturalis, Pars II. Leipzig: Renger. Wolff, C. 1737b. De Differentia Nexus Rerum Sapientis et Fatalis Necessitatis. 2nd ed. Halle: Renger. Wolff, C. 1737c. Cosmologia Generalis. Frankfurt and Leipzig: Renger. Wolff, C. 1841. Christian Wolffs eigene Lebensbeschreibung. Ed. H.  Wuttke. Leipzig: Weidmann. Wolff, C. 1860. Briefe von Christian Wolff aus den Jahren 1719–1753. Ein Beitrag zur Geschichte der Kaiserlichen Academie der Wissenschaften zu St. Petersburg. St. Petersburg. (Repr. 1971: Hildesheim-New York: Olms.) Wolff, C. 1962–. Gesammelte Werke. Ed. J.  École et al. Hildesheim-New York: Olms. [= GW, followed by roman number of Abt., Arabic number of volume, and page number] Wolff, C. 1985. Oratio de sinarum philosophia practica/Rede über die praktische Philosophie der Chinesen. Ed. M. Albrecht. Meiner: Hamburg. Wolff, C. 1996. Discursus Praeliminaris. Ed. G. Gawlick and L. Kreimendahl. Stuttgart-Bad Cannstatt: Fromann-Holzboog. Wood, A. W. 1978. Kant’s Rational Theology. Ithaca: Cornell University Press. Wood, A. W. 1984. “Kant’s Compatibilism.” In: A. W. Wood (ed.), Self and Nature in Kant’s Philosophy. Ithaca, Cornell University Press, pp. 73–101. Wundt, M. 1924. Kant als Metaphysiker. Stuttgart: Ferdinand Enke. Wundt, M. 1945. Die deutsche Schulphilosophie im Zeitalter der Aufklärung. Tübingen: J.C.B. Mohr. (Repr. Hildesheim: Olms, 1964.) Yandell, K. E. 1988. “Divine Necessity and Divine Goodness.” In: T. E. Morris (ed.), Divine and Human Action: Essays in the Metaphysics of Freedom. Ithaca: Cornell University Press, pp. 313–344. Yandell, K.  E. 1999. Philosophy of Religion: A Contemporary Introduction. London: Routledge. Zedlers Universal-Lexicon. 1739–1764. 64 vols. Leipzig–Frankfurt: Zedler.

Bibliography  373 Zeller, E. 1862. “Wolff ’s Vertreibung aus Halle: der Kampf des Pietismus mit der Philosophie.” Preußische Jahrbücher 10: 47–72. Zeller, E. 1875. Geschichte der deutschen Philosophie seit Leibniz. Munich: Oldenbourg. Zuckert, R. 2007. Kant on Beauty and Biology: An Interpretation of the Critique of Judgment. Cambridge: Cambridge University Press.

Index Abaci, Uygar  296n.5 absolute being 79 completeness 121 determinations 226n.52 extension  90n.32, 93, 97n.52, see also pure absolute extension necessity  71, 304n.26, 305, 316 perfection 76 positing  60, 72–4, 78 real possibility  302n.18 space  18, 79, 89, 90n.32, 92, 96, 100–2 absoluteness 79n.3 of creaturely dependence  267 of divine sovereignty  273, 287 abstract/abstraction  41, 44, 61, 63, 81, 83, 88–9, 92n.39, 94–5, 98, 101, 102n.63, 105n.72, 106–8, 117, 129, 147n.14, 173, 191, 206n.71, 211, 219–21, 225–6, 230n.64, 232, 239, 257, 282, 299, 303n.22 accident argument  82, 84–5 accidental affections 271 conformity  300, 314 representations 301 accidentality 248 accidents  62, 81–2, 86, 95, 97n.52, 270n.12, 272, 288 active disposition 330 faculty  182, 186 force  47–9, 54, 157, 268, 274, 333–5, 338 grounds 62 participation 300 power  326, 330, 332–4 principles  45–6, 51 states 109 substance 225 actuality  64–6, 70, 217, 313, 343, 351 Adams, Marilyn M.  328n.20 Adams, Robert  77n.27, 102n.62, 142n.1, 143n.2, 144n.5, 148n.18, 149n.20, 152–3, 155–8, 160, 163n.40, 206n.71, 217n.14, 274, 278n.30, 282n.38, 286n.46, 288, 291n.63, 302n.17, 303n.19, 325, 326n.17, 333nn.30–1, 337

Adickes, E.  27, 38n.33, 50–2, 282n.38, 283n.40, 286 aesthetics  20, 44, 178, 242n.20 affection  182n.12, 186, 271, 288, 299, 318n.58, see also causal affection, noumenal affection agency  173, 267, 279, 283, 288, 292–3, 296, 300–1, 309, 314, 317–19, 325 agnostic  30n.11, 36, 169, 174, 354 agriculture  34, 43 Albrecht, Michael  32n.19 Allais, Lucy  143 nn.2–3,146n.11 Allison, Henry  143n.2, 246n.29, 304 all-sufficiency  285, 302–3, 306, 308, 310–11, 313–14, 316 alternate possibilities  300, 315, 316n.52, 317–18 Ameriks, Karl  143n.2, 145n.9, 276n.25, 277n.27, 286n.47, 297, 304, 322n.4, 344n.47 amphiboly  108–9, 189, 191–2, 213–14, 216, 222, 224–5, 227, 229–30, 232 analysis situs  24, 94n.44, 96, 99n.56, 100–1, see also order of situations, situs analyticity thesis  299, 306 Anderson, R. L.  58n.2 annihilation 324 Anselm  303n.20, 318n.58 anti-Cartesian conception of matter  174 anti-Cartesian insistence  157 anti-intellectualism 17n.43 anti-Leibnizian claim 235 sentiment 19 spirit 18 anti-monadist arguments  18 anti-occasionalist commitment 326 doctrine 350 model 284n.43 principle 330 anti-Spinozist principle  330 anti-Wolffian literature  33, 49 phase 20 philosophers  52, 70 polemical attacks  15n.37 sentiment 19 winners 37

376 Index anti-Wolffians  10, 37, 39–40, 44 Antognazza, M. R.  22 Aquila, R.  143n.2 Aquinas, Thomas  269, 285, 323, 328, 330, 331n.25, 342 De Potentia 329 Summa Theologica 327 Aristotelian doctrines 29 essentialism 266 philosophy  7, 31, 46–7 terms 303 Aristotelians 44 Aristotle  31, 47, 246 Arnauld, Antoine  6, 11, 26, 144n.5, 151n.26, 200n.59, 202n.62, 208n.75, 209n.76, 212n.2, 222n.36, 223n.39, 249n.43, 287n.51, 307n.38, 328–9, 338 Arthur, R.  79n.2, 80n.6, 87n.24, 90n.34, 94n.44, 97n.52, 101n.60, 142n.1, 220n.22 atheism  39, 278, 282 atheists  32, 37, 39, 72 author of sin problem  273–4 Basnage, Jacques  5 Baumeister, Friedrich Institutiones metaphysicae 16 Institutiones philosophiae rationalis 16 Baumgarten, Alexander  8n.11, 10n.16, 22–4, 29–30, 35–6, 44, 47, 49, 59n.8, 68, 71, 144, 169n.46, 270, 283, 285, 288–9, 310n.45, 340 Aesthetica 16 Metaphysica  16, 268n.8, 270n.13 Bayle, Pierre  4–5, 32, 107n.75, 202n.62, 207n.72, 307n.38 Dictionary 8n.12 Beck, Lewis White  2n.1, 7, 12n.27, 15n.39, 18n.45, 20, 27n.1, 295 Beiser, Frederick  20n.53, 29n.9 Bell, J. L.  91n.37 Belot, G.  79n.2, 93n.43, 101n.59 Bergmann, M.  299n.10 Berkeley, George  1, 22, 142, 187, 282, 286 Bernoulli family  28, 32, 34, 38 Bernoulli, Jacob  4 Bernoulli, Jean  46 Bernoulli, Johann  30, 45, 52n.61 Bernoulli, Nicolaus  46n.50 Bianco, B.  32n.20 Bierling, Friedrich  224n.42 Bilfinger, Georg Bernhard  10, 23n.63, 30, 33, 36, 46n.50 De harmonia animi et corporis humani 14 Dilucidationes philosphicae 13

biology 44 blind mechanism objection  299–302, 309, 314–15, 319 bodies  12–14, 18, 40–2, 45, 47–8, 50–1, 54, 56, 79–80, 81n.8, 84n.15, 88, 91, 93n.41, 95, 100–2, 104, 109–11, 126–7, 129, 142–5, 147–9, 151, 154, 161n.35, 162–4, 168, 174–6, 186, 195, 199, 202n.63, 203–4, 206n.71, 207–9, 219–20, 222–3, 227–8, 231, 233, 238, 244, 248–51, 255, 269, 278n.28, 324, 327–8, 335n.34, 343, see also essence of bodies, idealism Bolton, Martha  26n.68, 220n.25, 221n.28, 224n.44 Bowman, Curtis  296n.5 Boyle, Robert  22, 35, 250–2 Brandom, R.  177n.2, 196n.40, 197n.47, 198n.50, 300n.12 Brecht, M.  28n.5 Geschichte des Pietismus 28n.5 Brewer, K.  293n.67, 341n.43 Broad, C. D.  79n.2 Brucker, Johann  2n.1 Budde, Franz  10 Buffon, Comte de  246 Burnet, Thomas  352n.57 Burnett, James  204n.65 Buschmann, C.  35n.26 calculus  3, 7, 38, 50 Cartesian arguments 50 audience 212n.3 conception of matter  174 discussion 51 measure of force  56 physics 3 scholars 46 school 127 theories 283 thought 353 see also Descartes Cartesianism 7 Cartesians  30, 47, 50–2, 56 Cassirer, Ernst  5–6, 9n.14, 18n.45, 59n.9 categorical imperative  298–9 categories  73, 114n.2, 118–19, 121, 123, 136, 140, 145, 167, 176, 282, 327, 349, 351 category of actual things  82 of being  81 of causation  282 of cause and effect  121 of existence  145n.7

Index  377 of possibility  145n.9 of substance  81n.8, 171n.47 of the miraculous  323 Catholic areas/lands  8, 28n.3 causal activity  265, 284n.42 affection 166–8 concepts 145n.9 connection 65 definitions 42–3 dependence 176 ground  166, 340 inference  168, 322 influence 11 influx 285 interaction  17, 58, 272n.18, 275n.22 laws  164, 281, 343, 354n.61 nexus  245, 352 order 320 powers  176, 328 principles  280, 342–3, 345–6, 348, 352n.58 production of objects  162 realism  268n.7, 271 relations  65, 108–9, 194, 209, 265, 275–6, 282, 284n.43, 288, 324–5, 342 role  265, 300 sense 67 sequences  114n.2, 122 tendencies  222, 224 causality  42, 59, 62, 65, 67n.17, 75n.26, 253, 258–60, 265, 281, 284n.42, 285, 289, 293, 294n.68, 303n.20, 304n.23, 310–12, 316n.53, 317, 343, 354 causation  4, 14, 108, 144n.4, 145, 147n.15, 174, 202n.63, 245–6, 254, 265–8, 271–4, 276, 278, 282–4, 286–7, 289n.58, 290–2 cause and effect  114n.2, 118n.6, 121, 137n.23, 246, 328n.22 CCC principle  269–71 Châtelet, Émilie du  8, 9n.14, 50 Institutions de Physique  23n.64, 38–9, 46 chemistry 45 Chignell, Andrew  58n.3, 281n.37, 302n.17, 304n.23, 306n.34, 307n.38, 312n.48, 322n.4, 340n.40, 341n.43, 343n.45, 347n.52 Chinese culture 10–11 lecture 33 people  11, 32 philosophy 10n.21 religion 11 rites 11 Christian community 40

faith  32, 35, 37n.31 love 31 orthodoxy  10, 267, 292 religion  11, 29, 32, 39, 41 savior 33 view of the universe  268 Clairaut, Alexis  53 Clarke, Samuel  4, 6, 23–4, 79–80, 87, 89, 94–5, 107, 112, 125–33, 208n.74, 216n.10, 220n.23, 231, 323–4, 326, 331, 342 Clemm, H. W.  28n.6 coexistence  65–6, 75, 80n.6, 81, 86–9, 91–6, 100–2, 131–2, 231 cognition  13, 25, 58, 59n.9, 65–6, 80, 86n.21, 97, 102, 115, 118, 145n.9, 161–2, 166, 177n.1, 179, 181, 184, 186, 191, 211, 213, 215, 219, 230–1, 235, 241, 245, 285, 293, 305, 306n.30, 310–13 cognitive abilities 185 access 190 acts 103n.68 apparatus 188 duality 135 faculties  165, 182, 187, 211, 213–14 finitude 134 functions 213 limitations 185 limits 234 objects 232 powers  104n.70, 114n.2, 133, 252 priority 229 psychology 310n.45 purposes  227, 243 receptivity 214 state 66 subjects  148, 160 system 224 compatibilism  11, 293n.67, 354 compatibilist account of freedom  71 approach 294n.68 notion 74 picture 354 conceptual arguments  284, 292 connections 300 containment 68 discernability 108 event-space 93 impossibility 91 knowledge 99 mediation 66 objections 289–91

378 Index conceptual (cont.) obstacles 325 plurality of parts  87 representations  100, 188, 212, 224–5, 228–31 requirement 168 scheme 340 terms  187–8, 193 concurrence  266–7, 272–4, 276, 278, 283–92, 294, 318n.59, 322, 325–6, 327n.18, 329–31, 332n.27, 333, 339, 341–2, 350 concurrentism  266–7, 274, 285, 290n.60, 294 Confucianism  11, 32 confused perceptions  195–6, 202–9, 220 consciousness  133–4, 140, 163, 177n.1, 179, 186, 194–5, 197, 199–201, 274, 306, 307n.38, 310, 312–13, 316 Consentius, E.  53 conservation  268, 270, 272, 275–6, 283, 285–7, 289–91, 293, 326n.17, 341, 349 by God  270n.12 debate/dispute 266–7 law  40, 51 of causes  294 of creatures  269 of force  40, 50, 128n.18, 239, 244, 288, 328n.22 of motion  239 of power  174 conservationism  266–7, 272, 273, 274, 276–8, 277, 281, 284, 287, 288, 289n.59, 290n.60, 291–3, 325, 350 continuum  92, 93n.41, 101n.60, 220, 227, 328n.21, see also labyrinth of the continuum contradiction  11n.25, 37, 41–2, 72–3, 136, 219, 221, 270, 300, 302, 345, 354, see also principle of contradiction Copernican revolution  242 corporeal forces 158 nature  327, 342–3 realm  14, 227 souls 125 substances  207, 222, 251 things  225, 228 universe 175 world  223, 279 Corr, C. A.  42–3 cosmological connection 275–7 dialectic 146n.11 ideas  120–3, 130 implications  112, 130–2 lessons  112, 133–5, 138, 141

order 278 overtones  116n.4, 134 position 128 relations 275–6 cosmology  44, 113, 124–5, 127, 129, 136, 265 cosmology of experience  112, 119n.7, 123, 133 Coste, Pierre  4 Counter-Enlightenment thinkers/thought  15, 23 Couturat, Louis  6 Cover, Jan  80n.5, 83n.13, 90n.32, 93n.41, 206n.71, 221n.28, 232n.66, 299n.10, 337, 338n.38 creation  43, 83, 105, 128, 138–9, 236, 245, 259, 261, 267–73, 278, 283, 285–6, 288n.57, 293, 303n.20, 305n.28, 308–9, 311, 314–15, 324, 325n.12, 326, 327n.18, 328n.20, 330, 341 creation of nature  238, 253, 259–60 creator  224n.44, 250, 265, 268, 273, 293, 307, 312n.48, 313n.49, 317–18, 328, 341, 350 creator of nature  244, 259 Crockett, T.  80n.5, 92n.38 Crusius, Christian August  15n.37, 22, 29, 30n.10, 31, 43, 49, 57, 67–8, 70–1, 73–6, 78, 270, 279n.33, 283n.41, 304n.23, 316n.53, 318n.58 Dissertatio de usu et limitibus principii rationis determinatis, vulgo sufficientis 16 Sketch of the Necessary Truths of Reason 16, 59–63, 64n.14, 65–6 Cudworth, Ralph  22 D’Alembert, Jean  22, 51, 52n.61, 53, 56 Traité de dynamique  38, 51n.60 dead force  49–50 degrees of power  334–5, 338–9, 350 deism  161, 267, 274n.20, 288n.57, 294 De La Forge, Louis  269n.10 De Risi, V.  24n.65, 94n.44, 95n.47, 96n.50, 103n.68, 220n.26, 223n.38 Des Bosses, Bartholomew  5, 13n.33, 14, 81n.9, 102n.65, 150n.22, 208n.74, 227n.57 Descartes, René  7, 22, 40, 46, 64n.14, 68, 110n.83, 212n.3, 268–9, 282, 301n.14, 303n.20, 309n.42, 347 Meditations on First Philosophy 1 see also Cartesian determining ground  60, 66–7, 69–71, 73–5, 279, 315, 317 determinism  11, 17n.43, 18, 39, 41, 45, 277, 280–1, 323, 329, 339, 341n.43 De Volder, Burcher  6, 12, 13n.33, 26, 86n.22, 102n.65, 146n.10, 148–50, 154, 174, 199n.57, 206n.70, 207n.72, 227n.57, 230n.64, 232n.66

Index  379 Diderot, Denis  32n.20 Encyclopédie 3 disposition  86, 87, 220n.24, 223, 235, 253, 258, 284–5, 287, 329–30, 334, 343 divine action  266–7, 270n.14, 272, 277–8, 286–7, 288n.57, 289, 309, 315, 333 causation  265–7, 278 freedom  45, 295, 297, 299n.10, 301–2, 315–16, 318 intellect 84 will  11, 217, 295–9, 301–2, 303n.20, 305–6, 308–10, 313–17, 332, see also holy will wisdom  107–8, 306, 308 divinity  306n.30, 326–7, 345 dogmatic claims 139 controversies 31 metaphysics  1, 10, 26, 237, 282, 286, 298 philosophy  26, 29 theism 285 thinkers  13, 29 Döring, D.  8n.12, 31n.16, 33n.22, 36n.30 dualism  11, 14, 249n.41 dualists 13–14 Duchesneau, F.  32, 44 Durandus of St. Pourçain  266, 272, 288–9, 290n.60, 291 Dutens, Louis  2, 5, 20–1, 23n.63, 24–5 Leibnitii opera omnia 6 Dyck, C.  30nn.10–11 dynamics  18, 174, 208 Earman, J.  79nn.2–3, 80n.5, 321n.2, 322n.3 Eberhard, Johann August  2, 6, 9, 25n.67, 166, 186, 233, 236–7, 264 Neue Apologie des Sokrates  5n.3, 21n.58 Eberstein, Baron von  9 École, J.  12n.27, 16n.41 Edelmann, Johann Christian Moses mit aufgedeckten Angesichte 7 Eliot, T. S.  22n.62 emanation 303nn.20–1,304 empirical knowledge  44, 81, 99, 101, 103, 106, 110, 138, 140, 234 empirical miracles  353 empiricism  10, 15, 30–1, 44 empiricists  9, 45n.47, 57 engineering  34, 43, 292 Enlightenment thinkers/thought  15, 23 entities  41, 44, 81, 85, 88–9, 92n.39, 93n.41, 94–5, 97, 101n.60, 143n.2, 144–6, 148, 152, 160, 187–8, 206, 218, 224–5, 230–1, 234, 265, 268–9, 287–8, 290

Epicurus 248 epistemological arguments 284 grounds 186 hierarchy 349 pain 336 positions 20 principle 43 puzzle 339 requirement 58 sense 65 strictures 283 epistemology  4, 21, 35, 46, 167, 173, 175–6, 178, 298–9 Erdmann, B.  12n.31, 17, 27nn.1–2, 38n.33, 282n.38 Leibnitii opera philosopica 6 essence  44, 66, 73–4, 76, 88n.28, 148, 149n.21, 248, 277–8, 304n.26, 305, 307, 326, 331–5, 337–8 of bodies  150, 151n.24, 152–3, 155–60, 169–73 of things  42–3, 60–4, 71, 78, 111, 240 essence idealism  151, 153, 156–8 Euclidean geometry  93, 96n.50, 99–102, 103n.68, 105 postulates 138 space  89, 97, 99–100, 102–3 Eugene of Savoy, Prince  4 Euler, Leonhard  17, 22, 28, 30n.10, 34, 37–9, 46, 53, 55 Gedancken von den Elementen der Cörper 18 Letters to a German Princess 7n.8 evil  4, 176, 261, 274 existence idealism  150–3, 156, 158, see also weak existence idealism experience  29, 42, 44–6, 51, 100, 107n.76, 111–23, 125, 128, 132–5, 138, 140, 146–7, 160, 162–73, 176, 185n.16, 186–7, 194–6, 202–5, 209–10, 214–15, 219, 228, 233–4, 237, 243–4, 246, 259–60, 281n.37, 282, 310, 313n.49, 324, 349, 351 experiment  43, 45–6, 185n.16, 203 extension  23, 30, 47, 90–1, 97, 110n.83, 117, 128, 130, 133, 139, 190, 217, 220, 251, 270, 324, 329, see also absolute extension, pure absolute extension faith  10–11, 15, 16n.41, 32, 35, 37n.31, 49n.58, 298 fatalism  11, 39, 41, 70, 282, 305, 309, 313–14, 316n.53 Finster, R.  178n.4

380 Index Firnhaber, F. J. Defense 55 Fischer, Christian Gabriel  17 Fischer, Harald-Paul  54–6 Fischer, J. M.  300n.13 Flottwell, Eduard  27n.1, 28n.5 Fogelin, R.  322n.3 Formey, J. L. S. Histoire Abrégée de la Philosophie 12 Förster, E.  249n.40 Francke, August  29 Freddoso, A. J.  266, 272n.17, 273, 287n.52, 288n.55, 290, 329, 330n.24 Frederick II  16–17, 20n.54, 35, 37 Frederick William I  15, 17, 31, 35 free will  18, 40–2, 45, 59, 282, 297–8, 305, 316nn.52–3, 354 freedom  10–11, 26, 31, 70–1, 74–5, 235, 237, 243, 253–4, 257, 260–3, 276, 279–80, 281n.37, 282, 291, 292n.65, 293n.67, 294, 296, 300, 317, 319–20, 341n.43, 342, 354, see also divine freedom, transcendental freedom Friedman, Michael  249n.40 Futch, M. J.  79n.2, 80n.6, 86n.23 Galilei, Galileo  22, 40, 43 Garber, Daniel  23n.63, 81n.8, 108n.78, 142n.1, 144n.5, 206n.71, 249n.41, 269n.10, 324n.9, 328n.22 Gassendi, Pierre  22 geometrical concepts  41–2, 108 constructions 138 demonstrations 48 figures  97–8, 100, 222 form 106 knowledge  96, 99–100, 103 method  32, 35 objects  95–7, 105 properties  96, 98, 110 space  89, 96, 97n.52, 99, 102 structures 92 things 43n.42 truths  100, 239 geometry  8, 43, 57, 80, 93–4, 101, 137, 186, 251, 281 George I, King  7 Gerhardt, C. I.  6, 95n.47 ghosts 45 Goethe, Johann von  20, 56 Goldenbaum, Ursula  8n.10, 11n.23, 12n.27, 23n.64, 26n.70, Ursula 31n.16, 33nn.21–2, 34n.24, 35, 36n.27, 37–8, 39n.37, 41n.38, 42, 43n.41, 44nn.45–6, 45n.47, 51n.60, 54n.67, 55n.72

goodness  128, 239, 253, 256, 264, 297, 305, 309n.42, 311–14, 318, 336, 344 Gottsched, Johann Christoph  6–7, 9n.13, 17, 22–3, 28n.5, 30n.12, 32–3, 34n.23, 35–6 De optimismi macula 19 Erste Gründe der gesamten Weltweisheit 16 Gottsched, Victoria  8 Goy, I.  244n.26 gravity  45, 115n.3, 329 Grotefend, Georg  6 grounding  69, 75, 84, 149n.21, 229, 302n.18, 304n.23 of relational facts  83, 109 relations  73, 83n.12, 147n.15, 151n.25, 156, 230 Gundling, Jacob Paul von  17n.43 Guyer, Paul  26n.69, 78n.29, 242n.20, 244n.26, 249n.40, 256nn.64–5, 259n.75, 261n.84, 280n.35, 296n.5, 340n.40, 343n.46 Haller, Albrecht von  53, 55 Hansch, Michael Gottlieb  23n.63 Godefridi Guilielmi Leibnitii Principia Philosophiae  6, 24 Harig, G.  43n.42 harmony  27, 85n.19, 103n.67, 107, 146, 148, 190–3, 236, 238, 240, 245, 253–5, 263–4, 275–6, 278–9, 307n.38, 308, 311n.47, 313, 325n.12, see also pre-established harmony Harnack, A. von  17n.43, 18, 53n.63 Hartmann, G. V.  10n.20, 15n.37 Hartz, G.  80n.5, 90n.32, 93n.41, 206n.71, 221n.28, 232n.66 Hegel, G. W. F.  9, 19, 29–30 Hegelian historians  30, 34 narrative  29n.8, 38 school  29, 49 Heimsoeth, Heinz  59n.9, 286n.46 Henrich, Dieter  119n.8, 295–6, 308, 310n.45 Herder, Johann Gottfried  6, 21, 340 Gott: Einige Gespräche 21n.59 Hermann, Jakob  32–3, 36, 45, 46n.50 Hilgers, Thomas  296n.5 Hinrichs, C.  10n.18, 16n.40, 17n.43, 28nn.5–6, 31n.14, 34–5, 39 Hißman, Michael  25 Hobbes, Thomas  42–3, 323 De Corpore 87n.24 Hoffmann, Adolph Friedrich  31, 43, 310n.45 Hoffmann, Paul  142n.1, 151n.24, 152n.28 Hogan, Des  276n.25, 279n.32, 282n.38, 294n.68, 341n.43, 350n.54 holy will  295–6, 298–9, 301, 306n.31, see also divine will Humean conception of miracles  322 Humean sarcasm  344n.47

Index  381 Hume, D.  1, 20, 22, 26, 73, 322n.3 Hume’s problem  26 Hunter, I.  15n.39 Hutcheson, Francis  22 Huxford, G.  343n.46 hylozoism 248 idealism  12n.30, 13, 29, 80, 143, 147–9, 186–7, 248 about the essence of bodies  150, 151n.26, 152, 155, 159, 169–72 about the essences of individual bodies  151n.24, 158–9, 170–3 about the existence of bodies  150, 152–3, 157–9, 162, 168, 173 see also essence idealism, existence idealism, matter idealism, substance idealism, transcendental idealism, weak existence idealism, weak idealism about bodies, weak matter idealism ideality of relations  82–3, 85, 93 of space  79–81, 90–2, 101–2, 106, 161, 280 of time  161, 271, 280 identity  66, 72–3, 143, 155, 190, 193, 221, 253, 264, see also principle of the identity of indiscernibles immortality  173, 257–8, 282, 298 incarnation  324, 326–7, 341 inertia  269, 273, see also law of inertia, principle of inertia influxus physicus  27, 30n.11, 36, 39, 44n.44, 47–9, 52, 56 Insole, C. J.  266n.3, 276n.25, 293n.67, 294n.68, 341n.43 intellectual representations  58, 104, 177–8, 180–9, 191–2, 195, 210 intension  50, 51, 58n.4 intuition  58, 103–4, 119, 132n.21, 135, 137n.23, 141, 161, 166, 175n.48, 177n.1, 178–9, 182, 184–5, 188–9, 191, 194–5, 214, 218–19, 225, 227–8, 230–1, 233–4, 271, 281 Jachmann, R. B.  22n.60 Jacobi, Friedrich  6, 40–1, 167 Jauernig, Anja  12n.27, 26n.68, 84n.15, 186n.21, 187n.22, 195n.39, 335n.33 Jesuits  10–11, 294 Johann II  46n.50 Jolley, Nicholas  142n.1, 152n.28, 205n.68, 328, 329 journals  5, 28, 56 Acta Eruditorum  3–4, 6, 238 Danziger Rationaltheologie  268, 270, 278, 285n.45, 286, 287n.50, 289, 291, 292nn.65–6

Frankfurtische Gelehrte Zeitungen 54 Freie Urtheile von gelehrten Sachen 44 Frühaufgelesene Früchte der Theologischen Sammlung 35 Gelehrte Zeitungen  28n.4, 44n.45 Göttingische Gelehrte Anzeigen 54 Göttingische Gelehrte Zeitungen 55 Göttingischen Zeitungen von gelehrten Sachen 44 Hamburgische Magazin 44 Journal des Sçavans 4 Neue Gelehrte Zeitungen 38n.35 Neuen Urtheile von gelehrten Sachen 44 Nova Acta Eruditorum  38, 53–4 Teutsche Merkur 25 jurisprudence  34, 43 Justi, Johann  18 Kain, Patrick  295n.2, 298 Kant, Immanuel ‘Amphiboly’  9, 25, 26n.68, 57–8, 77, 81, 104, 105n.72, 111, 187, 190, 194, 211 ‘Antinomies’  57, 120–3, 139 ‘Canon of Pure Reason’  236, 253 Critique of Pure Reason  9–10, 20, 25, 26n.68, 57, 76, 78, 104, 112, 115, 120, 124n.13, 130, 132n.21, 160, 162, 165, 170, 187, 211, 233–5, 240, 256, 340n.40 Critique of the Power of Judgment  242, 256, 258, 261n.85, 263–4 ‘Critique of the Teleological Power of Judgment’  242, 244–5, 249 Dreams of a Spirit-Seer 24 ‘Duisburg Nachlaß’  132n.21, 137 Estimation of Forces  38, 46, 52 Groundwork for the Metaphysics of Morals  136, 257 ‘Ideal of Pure Reason’  57 Inaugural Dissertation  25, 78, 175n.48, 277, 280 Inquiry Concerning the Distinctness of the Principles of Natural Theology and Morality 20 ‘Metaphysical Exposition’  58n.4 Metaphysical Foundations of Natural Science  25, 59n.6, 160, 170, 342, 351 Negative Magnitudes 72 Nova Dilucidatio 60 On a Discovery  25, 166, 236 On Progress in Metaphysics 194 ‘On the Common Saying’  256 On the Miscarriage of all Philosophical Trials in Theodicy 293 Only Possible Argument  60, 67, 69, 71, 74, 239, 302, 311n.47 Only Possible Proof  268, 271, 278–9, 280n.34

382 Index Kant, Immanuel (cont.) Opus postumum  249, 351n.55 ‘Paralogisms’  10n.16, 20, 22n.61, 57, 145n.9 Perpetual Peace  284–6, 289, 294n.69 Physical Monadology  24, 272, 275n.21 Prolegomena to Any Future Metaphysics  2, 26, 58n.5, 112, 114n.2, 115–18, 120–5, 130, 132–3, 135, 145n.7, 175 ‘Reflections on Optimism’  306, 313n.49 Regions of Space 24 Religionslehre Pölitz  161n.36, 268, 280, 285–6, 289, 291, 292n.65, 303n.22, 310n.43 ‘System of Cosmological Ideas’  112, 120, 130 The New Elucidation (Nova Dilucidatio) 24, 60, 66–7, 70–1, 73n.22, 75, 78n.29, 124–5, 131, 133, 268, 270n.11, 272, 274–5, 279n.31, 294, 297, 310n.45, 315–16 ‘Theory and Practice’  256, 258, 263–4, 305, 309, 316 Thoughts on the True Estimation of Living Forces  23, 268n.7, 271–2, 275nn.21–3, 276 ‘Transcendental Aesthetic’  58, 115, 118n.6, 124n.13, 132n.21, 133, 230 ‘Transcendental Analytic’  116n.4, 165 ‘Transcendental Deduction’  112–19, 123, 125, 132–6, 138–41 ‘Transcendental Dialectic’  57, 241 Universal Natural History and Theory of the Heavens  59n.7, 139–40, 271, 277 What Progress? 25 Kästner, Abraham Gotthelf  18, 28n.6, 53, 54n.67 Khamara, E.  79n.2 Knutzen, Martin  27, 29–30, 34, 36, 44, 47, 48n.56 Systema Causarum Efficientium 17 Köhler, Heinrich  6 König, Samuel  18, 30, 32–4, 36, 38, 45, 51, 53, 54n.67 Korsgaard, C. M.  300, 302 Kuehn, M.  23, 27nn.1–2, 28n.3, 29n.7, 38n.33, 44n.44, 46n.49, 48n.56, 49nn.57–8, 52n.62 Kulstad, M.  222n.37 Kypke, Georg David  27n.1, 29n.7, 44 labyrinth of the continuum  80, 90–1, 94, 97 Lambert, Johann Heinrich  22, 130n.19, 137, 141 Lange, Joachim  10–11, 29, 32–5, 36n.27, 39–41, 43, 59 Modesta disquisitio 11n.24 Lange, Samuel Gotthold  35 Langton, R.  143n.2, 178n.4, 227n.56, 275–6 Lavin, D.  299–300, 301n.15

laws of action  281n.37, 347, 349–50, 353 causality 260 community  124–5, 131–3, 135, 138 connection 113 conservation of force  40, 50–1, 128n.18, 244, 328n.22 continuity  27, 50 effect  342, 345–6 gravitation 115n.3 identity 72 inertia  50, 238, 248–9, 251–2 mechanics 47 morality  262, 317 motion  18, 47, 139–40, 239, 281n.37 nature  107n.75, 113, 138, 140, 207, 234–5, 240–5, 248, 253, 275n.23, 278n.28, 279, 280n.34, 287, 301n.14, 322, 327–9, 332, 337–40, 342–3, 345, 348–9, 353, 354n.61 the series  86, 202n.63, 329, 338, 348–9 Laywine, Alison  30n.10, 55n.68, 116n.4, 119n.7, 123n.10, 132n.21, 137n.23, 138n.24, 140n.26 Le Clerc, Jean  7 Lee, S.  326 Leibniz, Gottfried Wilhelm Characteristica geometrica (CG)  90n.32, 98, 99n.55 Confessio philosophi 41n.39 De arte combinatoria 86–7 Discourse on Metaphysics  3, 11, 144n.5, 216n.10, 273, 278n.30, 324, 328 Dissertation on the Combinatorial Art 3 General Inquiries about the Analysis of Concepts and of Truths 3 Letter to Magnus Wedderkopf 41n.39 Meditations on Knowledge, Truth and Ideas  3, 23n.63 Metaphysical Disputation on the Principle of Individuation 3 Monadology  4, 6, 23–4, 197–8, 252–3, 303n.21, 324, 335 New Essays on Human Understanding  6, 20, 21n.59, 24–5, 79, 83n.11, 86n.20, 88n.29, 92n.39, 101n.60, 198, 206, 338 New Physical Hypothesis 3 New System of Nature  4, 23–4 Nouveaux Essais  20, 130n.19, 228 Novissima Sinica  10n.21, 11 On Nature Itself  4, 23, 238, 250–2, 284n.42 On the Correction of First Philosophy and the Notion of Substance 3 Primary Truths  3, 216n.10, 333

Index  383 Principles of Nature and Grace  4, 6, 23–4, 146, 196, 200, 238, 244, 251, 253, 255 ‘Recovery of Confused Qualities to Distinct Ones’ 203 Specimen Dynamicum  6, 23–4, 47, 109n.81, 334, 337 ‘Specimen of Discoveries of Marvelous Secrets’  196, 332n.28, 336 Theodicy  1, 4–7, 11, 14n.34, 19, 23–4, 83n.11, 100n.59, 269, 271, 273, 288, 291, 305, 306n.34, 307n.38, 324, 325n.12, 326–7, 332, 334, 338 Theory of Abstract Motion 3 Leibniz–Wolffian epistemology 20 living force  38, 50 metaphysics 38n.36 philosophy  6, 10, 12–13, 15, 29, 35, 46, 282 terminology,  31, 47 Lessing, Gotthold Ephraim  25, 27, 52–3, 340 Leibniz von den ewigen Straften  5n.3, 21 Pope ein Metaphysiker! 19 Levey, S.  90n.34, 108n.78, 207n.73 living forces  9n.14, 23n.64, 27, 38–9, 46n.49, 49–51, 54, 56, 128n.18 Locke, John  7, 9, 15, 20, 22, 25, 31, 32n.20, 42, 126, 187, 201, 211–12, 214 A Discourse on Miracles 44 Essay Concerning Human Understanding  1, 4, 44 Some Thoughts on the Conduct of Understanding in the Search of Truth 44 Lodge, P.  142n.1, 149n.20, 151n.26 logic  2–3, 6, 26, 31, 33, 44, 83n.11, 169n.46, 239, 301n.14 Lomonossow, Mikhail  34 Longuenesse, Béatrice  119n.7, 171n.47 Look, Brandon C.  8n.11, 12n.30, 13n.32, 15n.36, 16n.41, 144n.5, 335n.33 Löscher, Valentin Ernst  33, 35, 40, 43, 44n.43, 45, 47n.53, 48–9 Löw, R.  246n.30 Luck, M.  321n.2, 348n.53 Ludovici, Carl  10n.18 Ausführlicher Entwurff einer vollständigen Historie der Leibnitzischen Philosophie 12 Lutheran theologians  28n.5, 30–1, 35, 39, 42, 44–5, 47n.53, 48 Maaß, Johann  6 Mackie, J. L.  321–3 Mackie/McGrew Miracle (MMM)  322–7, 335n.34, 339–41, 343, 353–4 Mairan, J. J. Dortous de  38n.36

Malebranche, Nicolas  1, 40, 127, 265–72, 282, 284n.42, 287, 290–1, 349 Manteuffel, Ernst Christoph von  13 Marshall, Colin  143n.3 Martin, G.  178n.3 materialism  125–6, 128, 277, 282 materialists 13 Mates, B.  83n.13, 84n.14, 147n.14, 224n.43 mathematicians  3, 17, 33, 37, 46, 53, 233 mathematics  8, 18, 22, 24, 28, 34, 38, 40, 42–6, 48–53, 56, 58, 80, 89, 91n.37, 92n.39, 99, 101, 115n.3, 179, 233, 300, 381 matter  48, 63, 89–91, 102n.63, 106–8, 110, 115–16, 120, 122, 124, 138–9, 142, 144–7, 154, 166, 169n.46, 173–4, 176, 190, 194, 199, 209–10, 214, 225–30, 238–9, 247–52, 271, 277–8, 281n.37, 301, 331, 342, 345, 346n.50, 347–8, 351n.55 matter idealism  148–9, 152, see also weak matter idealism matter realism  153, 155, 157–8, 160 Maupertuis, Pierre Louis  9n.14, 17–18, 22, 37–9, 51, 53, 54n.67 Mavrodes, G. I.  321n.2 McGrew, T.  322–3, 325, 327, 339, 341, 343, 353–4 McRae, R.  177n.2, 196n.40, 197n.45, 200n.59, 212n.4 mechanical absolutism 40 bodies 50–1 causation  245, 254 causes 208 explanations  35, 44–5, 237–8, 244–5, 246n.29, 248–50, 301 forces 49–51 laws  48, 238, 277, 342–3, 345, 347 motion  48, 50, 56 philosophy  35, 40–2, 45, 47 processes 300n.11 properties  202n.63, 205–9 science 41 structure 254–5 system  247, 252 theory  43, 45 thinking 40 mechanics  29–30, 38, 46, 48, 52, 222, 281, 287 Meier, Georg Friedrich  22–3, 29–30, 35–6, 44n.46, 47, 270, 272n.17, 285, 288–9 Vernunftlehre 16 Meiran, M. de  9 Mencke, Friedrich Otto  54 Mencke, Otto  54 Mendelssohn, Moses  16n.41, 21–2, 40–1 Abhandlung über die Evidenz in metaphysischen Wissenschaften 20

384 Index Mendelssohn, Moses (cont.) Phaedon oder über die Unsterblichkeit der Seele  20, 22n.61 Pope ein Metaphysiker! 19 mental activity 300–1 being 347 content  26, 165n.41 faculties  188, 200 operation 220 powers  235, 241 reality 12 representations  13, 148–9 states  144, 169 Mercer, Christia  249n.41 metaphysical analysis 80 arguments 54 consequences 40 considerations 75 constraints 290 dilemma 14 doctrines  212, 232 economy 350 essence 62–4 foundations 79n.2 ground  29, 147n.15, 156, 340 harmony 307n.38 laws 351n.55 modality 314 necessity  217, 300, 302 principles  66, 73n.22, 190, 211, 213, 215, 224, 226, 232, 342, 349 project 49n.58 reasoning 239 source  238, 251 speculation 51 symmetry 287 systems  11, 19, 41, 61, 213 theory  30, 219, 267, 281 theses  5, 25, 280–1 tradition 272 views  3, 26, 47 mind–body  129, 284n.43 causation  4, 14 dualism 14 influence 127 problem  30n.10, 39–40 relation 11 union 128–9 minds  9, 12–14, 89n.30, 102–3, 144, 161n.35, 253–4, 277, 286, 304n.23, 321, 329–30, 332, 336–7, 340, 350, 353 miracles  266, 278–9, 280n.34, 281n.37, 283, 287, 292, 320–54

misconception  177, 181, 183–6, 188–9, 191–3, 204, 210 misconception objection  166, 178, 182, 185n.16, 195 Molina, L. de  294, 330 monadic perceptions  95n.49, 102, 103n.68, 146–54, 156–60, 228 monadology  9, 12–14, 23n.63, 25, 173, 249n.41 monads  4, 15, 18, 21, 27, 41, 81n.8, 84n.15, 89n.30, 95, 103, 109–11, 129, 142–5, 155, 169, 175–6, 186, 195–200, 202, 204–10, 223, 225–7, 229, 233–4, 249, 251–3, 255, 324, 333, 335n.34 monism  249n.41, 304–5, 310n.45, 314, 316n.53 monists 13 monotheism/monotheists  320, 338 moral agency 292 agents  255–7, 260, 286 arguments  282–3, 303 being  32, 261 condition 258 dictates 11 disposition  253, 258, 284, 343 evil 274 goal  238, 256 grounds  245, 260, 312, 313n.49 law  176, 235–6, 253, 257, 261–2, 282n.39, 297–8, 299n.8, 301, 309, 318, 351n.56, 354 lives  10, 32 miracles 322 necessity  305, 314, 316nn.52–3 order  11, 254 psychology 295 realm of grace  253–4 religion 343 responsibility 39 theology  282, 284, 286 truths 300 world  254–5, 260, 317 morality  32, 70, 259, 263–4, 294, 299, 315, 343, 351 More, Henry  22 Morriston, W.  299n.10 Morris, T. V.  299n.10, 301n.14, 303n.20 Mugnai, M.  82n.10, 83nn.12–14, 95n.49, 217n.14, 220n.23 Münden, Christian  55 Mylius, Christlob  53 Nadler, S.  266n.2, 307n.39 Nance, Michael  296n.5 nature and grace  235–8, 253, 255, 257, 263–4, 328 necessitarianism  11, 267, 305, 313, 314n.50, 315–16

Index  385 neo-Platonism 303n.21 Newlands, Sam  335n.34 Newtonian absolutes 233 account of mind–body union  129 account of space  101 cosmology 44 physics  3, 59 Newtonians  9n.14, 18, 37, 80, 342, 352n.57 Newton, Isaac  4, 6, 9, 15, 17, 22, 27, 35, 37, 42, 46–7, 59n.7, 79, 112, 115n.3, 125n.14, 127–30, 132–3, 139–40, 278 Opticks  51, 126, 275n.23 Principia Mathematica  18, 38, 45, 79n.3 Nisbet, H. B.  53 Noël, François  10n.21 noumena  77, 104, 106, 108–9, 175n.48, 187–94, 214, 227, 231, 260–1, 285–6, 292, 315, 354n.61 noumenal affection  166–70, 173, 282, 351, 353 character 176 choices 354 commitment 262 determination 262n.87 forces 176 grounds 106 ignorance  173–4, 340 indeterminacy 315 indeterminism 316n.53 model of causation  283 objects 176 reality 283 realm 194 self 143n.2 substance 193 Nuyen, A. T.  353n.59 occasionalism  36, 39, 127, 128n.18, 265–9, 270n.11, 271, 273–4, 283–4, 287, 288n.57, 289n.58, 290, 292, 326, 332 Oelrich, J. C. C.  30n.12 ontological argument  64n.14, 68, 76, 303n.20 dependence  152, 268–9, 278, 293–4 distinction  316n.53, 326 facts 336 framework 81 ground 268 levels  206n.71, 209 scheme 205 sense  65, 85, 90, 93n.41, 95, 142, 147, 173–4 substrates 354n.61 terms 259 ontology  24, 26, 61, 64, 206, 303n.19, 323–4, 330

of abstract objects  147n.14 of bodies  143–4 of matter  161 of phenomena  170 of relations  80 of space  94 of the self  144n.6 order of situations  87, 94–5, 97n.52, 101–3, see also situs organisms  144n.6, 237, 244–52, 259–60, 263–4, 277, 283n.41 Parkinson, G. H. R.  177n.2, 185n.16, 197nn.45–6, 212n.1, 226n.53 passive cognitive faculty  187 dispositions 330 faculty 182 forces  268, 274, 333–5, 337, 339, 350 sensibility 186 states 109 Paull, R.  322n.4 perceptions, see confused perceptions, monadic perceptions, sense perceptions Pereboom, D.  104n.70, 212n.1, 224n.47, 299n.10, 300, 302n.16, 305n.27 Perler, D.  266n.2 persecution  28n.5, 32n.20, 33–4 Pettit, P.  300 phenomenalism/phenomenalist  142–3, 145, 146nn.10–11, 147–8, 173, 176 physical influx  20, 276n.25 physics  3, 18, 48, 59, 64, 80, 89, 110n.83, 203, 207–8, 222–3 Pietism/Pietists  7n.8, 8, 10–11, 15–18, 22, 27n.1, 28–37, 39, 44, 46n.49, 47–52, 55–7, 59 Placcius, Vincent  4 Plato  2n.1, 3, 186n.19 Parmenides 82n.10 pluralism  309, 314 pluralistic dynamism  272 pluralistic substance  268 plurality  88, 102n.65 of bodies  101 of entities  94 of individuals  101 of monads  150 of object-like points  97n.52 of parts  87 of states  88n.26 of substances  304 of things  86, 88, 100 polytheism 161 Pope, Alexander Essay on Man 19

386 Index Popularphilosophen 20 possible worlds  11n.25, 100n.59, 108, 129, 237, 239, 305n.28, 312–13, 316n.53, 321, 325n.12, 327, 331, 350 pre-established harmony  4, 8, 10–15, 17–18, 20, 23n.63, 25, 30, 35–7, 39, 41–2, 45, 46n.49, 58, 75, 108, 127–31, 176, 234–5, 237, 265, 272, 324 principle of contradiction  8, 66, 73, 75 principle of inertia  37, 40–2, 49–52 principle of least action  18, 38–9 principle of sufficient reason  8, 11, 15, 17, 25, 35, 41–2, 59–60, 80, 84, 105, 107, 211, 233–4, 239, 244, 281n.37 principle of the identity of indiscernibles (PII)  24–5, 71, 73n.22, 75, 104–6, 108, 110, 197, 215–18, 222–3 priority of existence  57, 60 Protestant areas/lands  8, 28n.3, 30, 35 Pufendorf, Samuel  7 punishment  2n.1, 13, 40, 253–8, 264, 328n.21 pure absolute extension  89, 93, 96, 100, 102 pure reason  57, 60, 120–3, 125, 233, 282n.39, 343 Puryear, S. M.  206n.69 Pyra, Immanuel  35 qualities  77, 79, 81–2, 83n.11, 84, 104–9, 152n.27, 196, 199, 202–6, 209, 223–4, 302, 307n.38 Radner, M.  144n.5 Rappolt, Karl  27n.1 Raspe, Rudolf Erich  2, 4, 6, 20, 24 rationalism  15, 20, 31–2, 41, 44, 57, 59–61, 66, 73, 74n.25, 75–8, 147n.15, 320–1, 327, 329, 340, 348 Rauscher, Frederick  296n.5 Ravizza, M.  300n.13 realism/realist  142–3, 146–8, 152, 156, 173, 248, 268n.7, 271, 284n.43, 286, 293n.67, see also matter realism Reinbeck, Johann Gustav  35–6 Reinhard, Adolf Friedrich von Le Système de Mr. Pope sur la perfection du monde 19 Reinhold, Karl Leonhard  169n.46, 241–2 relationalism  18n.44, 25, 79–80, 85 Remond, N.  4–5 Robinet, André  325n.14 Rogall, Georg Friedrich  17, 28 Romantics 21 Rosefeldt, Tobias  143n.3 Rosen, G.  158n.33 Rowe, W. L.  299n.10

Rüdiger, Andreas  10, 31, 43, 310n.45 Rudolph, U.  266n.2 Russell, Bertrand  5–6 Rutherford, Donald  10n.15, 14n.34, 81n.8, 85n.19, 91n.35, 102nn.61–2, 110n.83, 142n.1, 147–58, 160, 337, 338n.38 Schepers, Heinrich  31n.15 Schmidt, Johann Lorenz  33–4 Schmidt, S.  34 Schmucker, J.  25n.66 Schneewind, J. B.  295–6, 306n.30, 308–9, 313 Schneider, W.  178n.4, 190n.28 Scholasticism  8, 9n.14, 17 Schönfeld, M.  30n.13, 277n.27, 280n.34 Schumacher, Johann  53n.66 sensation  12, 65–6, 74, 78, 126, 163, 166, 179–80, 186–7, 196–8, 200–2, 205–6, 214, 330 sense perceptions  39, 44, 179, 185, 204, 208–9, 211–12, 218, 223n.39, 228 sensibility  26, 58n.5, 61, 77–8, 103–4, 132n.21, 133–5, 138, 144n.4, 165–7, 177n.1, 178–9, 182–9, 191, 211, 213–17, 219, 226–7, 230–1, 234–5 sensible representations  136, 177, 180–1, 183–9, 191–2 Sider, T.  155n.30 Simmons, Allison  196n.40, 197n.48, 198–9, 201n.61 simple substances  12–13, 15, 90–1, 102n.64, 103n.69, 146, 148, 174, 206, 208, 249, 252, 334 situs  81, 86–7, 94–102, 106, 107n.75, see also analysis situs, order of situations Sleigh, Robert  273–4, 325n.12, 333n.30 Smit, H.  147n.14 Smith, M.  300 Socrates 81 Sophie Charlotte of Prussia  4, 99n.57, 105, 108n.78, 228n.60 space and time  18, 25, 63–4, 77–8, 79n.8, 85, 93, 101n.60, 104n.71, 112, 115, 118, 128n.17, 132n.21, 133, 135, 137n.23, 145n.7, 161, 164, 174, 179, 185, 186n.19, 188, 194, 207, 214–15, 229–31, 233, 249, 280–1, see also Euclidean space, ideality of space Spalding, P. S.  34 Spener, Jakob  31 Spiess, O.  37n.31 Spinoza, Baruch  11, 22, 32n.20, 33, 40, 145n.9, 296, 302, 304, 307–9 Ethics  1, 304n.24, 307n.38, 308n.40 Spinozastreit 40 Spinozism  11, 39–41, 161, 248, 268n.8, 284n.42, 319, 326

Index  387 spirits  13, 103n.69, 218, 341 spiritual being 281n.37 causation 284n.43 determinism 277 realm 14 substance 250 Stahl, Friedrich Julius  44 Stang, Nicholas F.  147n.15, 302n.17, 304n.23, 305n.28 Steinmetz, M.  34n.24 Steinwehr, Wolf Balthasar Adolph von  33, 38 Storrie, S.  99n.56 Suarez, F.  266, 269, 272–3, 285, 287–92, 294, 330 Metaphysical Disputations 267n.4 subordinate maxims  278n.30, 328–9, 332, 338–9, 342, 348 substance dualism  11, 14 substance idealism  148–50, 153, 156–7, 161–2 supernatural action 277 cooperation 291 events  280n.34, 281, 344n.48, 353 explanations  279–81, 285 influences 280n.34 intervention  279, 292 revelation 32 signs 323n.5 tradition 321 truth 334 supervenience  83–5, 88, 109–10, 165, 169, 175 Swinburne, R.  321n.2 Swoyer, C.  222n.37, 223n.38 Talbott, T. B.  301n.14 teleology/teleologies  233, 236–8, 241–5, 247–9, 252–4, 258–60, 263–4, 353n.59 Temmik, Aloys  83n.12 Teske, Johann  27n.1 Tetens, Johannes  22 theism/theistic  71, 248, 259, 265, 267–8, 271, 273, 278, 282–3, 285, 287n.52, 303n.20, 320, 322 things in themselves  77, 103–5, 111, 142–8, 160–3, 166–76, 181, 186–90, 194–6, 202–5, 208–10, 215, 227n.55, 285, 315, 340–1 Thöle, B.  114n.2 Thomasius, Christian  7, 15n.39, 19n.50, 31, 49n.58 Timmermann, J.  299, 306n.31 Tonelli, Giorgio  8, 12n.27, 25n.66, 59n.9 Tournemine, René-Joseph de  14 transcendental aesthetic  58, 115, 118n.6, 124n13, 132n.21, 133, 230

amphiboly  191, 213 analytic  116n.4, 165 arguments 348 concepts 120–1 deduction  112–13, 115–16, 118–19, 123, 125, 132–6, 138–41 dialectic  57, 190, 241 freedom  298, 321, 340, 353–4 idealism  58, 142, 146n.11, 190n.28, 211, 213, 268, 280, 293n.67 laws 348 philosophy  184, 282n.39 questions 111 reflection  25, 191–2, 211 Tschirnhaus, Ehrenfried von  22 Twelftree, G.  321n.2 Vaihinger, H.  25n.66, 283n.40 Vailati, E.  79n.2 Van Cleve, J.  146n.11, 225n.50, 282n.38 Van Inwagen  301n.14 Varignon, Pierre  32 Vilhauer, Benjamin  354n.61 Voltaire  9, 17, 22, 37, 38n.36 Ward, A.  20n.52 Warda, A.  22 Waschkies, H.-J.  22n.60, 28n.6, 32n.18, 38, 44, 46n.48 Watkins, Eric  11n.23, 16n.42, 26n.70, 27n.2, 30nn.10–11, 31n.15, 32n.20, 36n.28, 39, 57n.1, 59nn.7–8, 61n.11, 62, 63n.13, 64–6, 75n.26, 77n.28, 144n.5, 248n.36, 268n.7, 275n.21, 276n.25, 293n.67, 302n.17, 341n.43, 351nn.55–6 From Pre-established Harmony to Physical Influx 30n.11 Watson, Joshua  327n.19, 332n.27, 338n.38 weak existence idealism  156 weak idealism about bodies  173 weak matter idealism  149–50, 158, 162 Wertheim Bible  28n.5, 33–5, 38n.35, 41, 55 wholly passive power  333–5, 339, 350 Wierenga, E.  299n.10 Wilkins, John Essay Towards a Real Character and a Philosophical Language 90n.32 Wilson, C.  5n.3, 142n.1, 177n.2, 227n.56 Wilson, M.  196n.40, 197n.47, 198n.52, 205n.68, 212n.1 Winter, E.  37n.31 Wolff, Christian Elements of Universal Mathematics  8, 24 German Logic  23n.63, 35, 42

388 Index Wolff, Christian (cont.) German Metaphysics  13n.33, 24, 306n.34 Philosophia Prima sive Ontologia 64 Theologia Naturalis 10n.20 Wolffianism  16–17, 29–31, 33–5, 37, 39, 46, 49 Wolffians  11, 24, 28n.5, 29n.9, 30–1, 33, 35–7, 39–40, 44–7, 49–51, 54–6

Wright, Thomas  44 Wundt, Max  8n.9, 10nn.17–18, 11n.24, 12n.27, 15n.39, 21, 25n.66, 59n.9 Yandell, K. E.  299n.10, 301n.14 Zeller, E.  17n.43, 27n.1 Zuckert, R.  242n.20