Metaphysics as a Science in Classical German Philosophy (Routledge Studies in Eighteenth-Century Philosophy) 9781032030005, 9781032034898, 9781003187561, 1032030003

Citation preview

Routledge Studies in Eighteenth-Century Philosophy

METAPHYSICS AS A SCIENCE IN CLASSICAL GERMAN PHILOSOPHY Edited by Robb Dunphy and Toby Lovat

Metaphysics as a Science in Classical German Philosophy

This volume is dedicated to questions about the nature and method of metaphysics in Classical German Philosophy. Its chapters offer original investigations into the metaphysical projects of many of the major figures in German philosophy between Wolff and Hegel. The period of Classical German Philosophy was an extraordinarily rich one in the history of philosophy, especially for metaphysics. It includes some of the highest achievements of early modern rationalism, Kant’s critical revolution, and the various significant works of German Idealism that followed in Kant’s wake. The contributions to this volume critically examine certain common themes among metaphysical projects across this period, for example, the demand that metaphysics amount to a science, that it should be presented in the form of a system, or that it should proceed by means of demonstration from certain key first principles. This volume also includes material on influential criticisms of metaphysical projects of this kind. Metaphysics as a Science in Classical German Philosophy is a useful resource for contemporary metaphysicians and historians of philosophy interested in engaging with the history of the methodology and epistemology of metaphysics. Robb Dunphy is an Alexander von Humboldt Postdoctoral Research Fellow at the Albert-Ludwigs-University Freiburg, Germany. He has previously held research fellowships at the Goethe University Frankfurt, University College Dublin, and the University of Hamburg, and has taught philosophy at Northeastern University London, the University of Winchester, and the University of Sussex. His primary research interests are in the theoretical philosophy of Kant and the German Idealists and in the history of scepticism. He is the author of Hegel and the Problem of Beginning: Scepticism and Presuppositionlessness and has published research articles in journals, including The Review of Metaphysics, the Hegel Bulletin, and Apeiron. Toby Lovat teaches in the school of humanities and social science at the University of Brighton, UK. His PhD (2018) and most recent publications develop a Kantian critique of Quentin Meillassoux’s ambitious argument in After Finitude, largely on the basis that Meillassoux, and many others, fundamentally misunderstand Kant’s theoretical philosophy. In other teaching and research, Toby’s work ranges widely over broader issues in metaphysics, epistemology, ethics and political and social theory, taking in German Idealism, Frankfurt school critical theory, Marxist political economy and social theory, and the histories and ideologies of liberalism and conservatism.

Routledge Studies in Eighteenth-Century Philosophy

Beyond Autonomy in Eighteenth-Century British and German Aesthetics Edited by Karl Axelsson, Camilla Flodin, and Mattias Pirholt Kant’s Critical Epistemology Why Epistemology Must Consider Judgment First Kenneth R. Westphal The Experiential Turn in Eighteenth-Century German Philosophy Edited by Karin de Boer and Tinca Prunea-Brettonet Human Dignity and the Kingdom of Ends Kantian Perspectives and Practical Applications Edited by Jan-Willem van der Rijt and Adam Cureton System and Freedom in Kant and Fichte Edited by Giovanni Pietro Basile and Ansgar Lyssy Perspectives on Kant’s Opus postumum Edited by Giovanni Pietro Basile and Ansgar Lyssy Adam Smith and Modernity 1723–2023 Edited by Alberto Burgio Metaphysics as a Science in Classical German Philosophy Edited by Robb Dunphy and Toby Lovat

For more information about this series, please visit: https://www.routledge.com/ Routledge-Studies-in-Eighteenth-Century-Philosophy/book-series/SE0391

Metaphysics as a Science in Classical German Philosophy Edited by Robb Dunphy and Toby Lovat

First published 2024 by Routledge 605 Third Avenue, New York, NY 10158 and by Routledge 4 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN Routledge is an imprint of the Taylor & Francis Group, an informa business © 2024 selection and editorial matter, Robb Dunphy and Toby Lovat; individual chapters, the contributors The right of Robb Dunphy and Toby Lovat to be identified as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. ISBN: 978-1-032-03000-5 (hbk) ISBN: 978-1-032-03489-8 (pbk) ISBN: 978-1-003-18756-1 (ebk) DOI: 10.4324/9781003187561 Typeset in Sabon by KnowledgeWorks Global Ltd.

Contents

List of Contributors Introduction: Metaphysics as a Science in Classical German Philosophy

vii

1

ROBB DUNPHY AND TOBY LOVAT

1 Wolff on Ontology as Primary Philosophy

19

DINO JAKUŠIĆ

2 Baumgarten on the Nature and Role of Metaphysics

50

COURTNEY D. FUGATE

3 Lambert on the Certainty and Generality of Metaphysics and Geometry

73

KATHERINE DUNLOP

4 The Methodological Role of Intellectual Intuition in Kant’s Critique

105

TOBY LOVAT

5 Kant’s Promise of a Scientific Metaphysics

132

CATHERINE WILSON

6 Can Metaphysics Become a Science for Kant? GABRIELE GAVA

150

vi Contents 7 Scientific Metaphysics and Metaphysical Science: The Demand for Systematicity in Kant’s Transition Project167 MICHAEL J. OLSON

8 Kant, Reinhold, and the Problem of Philosophical Scientificity184 KARIN DE BOER AND GESA WELLMANN

9 Reinhold on the Deduction of the Categories206 ELISE FRKETICH

10 Schulze’s Scepticism and the Rise and Rise of German Idealism226 ROBB DUNPHY

11 The I and I: The Pure and the Empirical Subject in Fichte’s Science of Science251 KIENHOW GOH

12 The Science of All Science and the Unity of the Faculties: Schelling on the Nature of Philosophy270 BENJAMIN BERGER

13 Two Models of Critique of Metaphysics: Kant and Hegel294 DIETMAR H. HEIDEMANN

14 Quietism, Dialetheism, and the Three Moments of Hegel’s Logic315 G. ANTHONY BRUNO

15 Metaphysics on the Model of Natural Science?: A Kantian Critique of Abductivism339 NICHOLAS STANG

Index367

Contributors

Benjamin Berger is Visiting Assistant Professor of Philosophy at Trinity College in Hartford, Connecticut. He is the author of a number of articles on Schelling, Hegel, and the philosophy of nature. With Daniel Whistler, he authored  The Schelling-Eschenmayer Controversy, 1801 (2020) and edited The Schelling Reader (2021). Karin de Boer is Professor of Philosophy at the University of Leuven. She published numerous articles on Kant, classical German philosophy, and Heidegger. She is further the author of Thinking in the Light of Time: Heidegger’s Encounter with Hegel (2000), On Hegel: The Sway of the Negative (2010), and Kant’s Reform of Metaphysics: The Critique of Pure Reason Reconsidered (2020). She also co-edited The Experiential Turn in Eighteenth-Century German Philosophy (Routledge 2021, with Tinca Prunea Bretonnet). G. Anthony Bruno is Assistant Professor of Philosophy at Royal Holloway University of London and Co-director of the London Post-Kantian Seminar. He is the author of Facticity and the Fate of Reason After Kant, under contract with Oxford University Press. He has edited volumes on scepticism, transformation, and Schelling for Oxford University Press, Cambridge University Press, and Routledge. He has published and forthcoming articles on Kant, German idealism, neo-Kantianism, and phenomenology in British Journal for the History of Philosophy, European Journal of Philosophy, History of Philosophy Quarterly, Philosophy Compass, Fichte-Studien, and elsewhere. Katherine Dunlop is Associate Professor of Philosophy at the University of Texas at Austin. She has also taught at Brown and Stanford, and received her PhD from UCLA. She specialises in Kant, Early Modern Philosophy, and the Philosophy of Mathematics and Science. She is the author of numerous articles on these topics in journals including Synthese, Noûs, Philosophy and Phenomenological Research, and the Canadian Journal of Philosophy.

viii Contributors Elise Frketich is Visiting Instructor at Purdue University. She specialises in Classical German Philosophy, especially Kant and Reinhold. She wrote her doctoral dissertation on the philosophy of cognition and language of Karl Leonhard Reinhold, which she defended in September 2019. Some of her recent publications include “Wolff and Kant on the Mathematical Method” (2019) and “The First Principle of Philosophy in Fichte’s 1794 Aenesidemus Review” (2021). Some of her forthcoming publications include “Concepts of Truth in Schelling, Reinhold, and Hegel” (Reinholdiana),  “Kant and Hegel on Individuating Organisms”  (Idealistic Studies), and “Bardili and Reinhold on Kant’s Psychologism” (The Palgrave Handbook of Transcendental and Psychological Idealism). She is currently working on one paper on Reinhold on intellectual intuition and another on Kant on the ideas of pure water, pure earth, and pure air. Courtney D. Fugate is Associate Professor of Philosophy at Florida State University, Tallahassee. He is the author of The Teleology of Reason: A Study of the Structure of Kant’s Critical Philosophy (2013) and editor of the Cambridge Critical Guide to Kant’s Lectures on Metaphysics (2019) and Kant and Baumgarten on Metaphysics (2018). His two most recent books are Baumgarten’s “Elements of First Practical Philosophy”: With Kant’s “Reflections on Moral Philosophy” (with John Hymers, 2020) and  The Philosophical Writings of Johann Nicolaus Tetens. Volume 1: Tetens’s Writings on Method, Language, and Anthropology (with Curtis Sommerlatte and Scott Stapleford, 2022). Gabriele Gava is Associate Professor of Theoretical Philosophy at the University of Turin. He has published articles in leading philosophical journals on Kant, Peirce, pragmatism, and epistemology. His first book, Peirce’s Account of Purposefulness: A Kantian Perspective, was published in 2014 by Routledge. His second book, Kant’s Critique of Pure Reason and the Method of Metaphysics, was published in 2023 by Cambridge University Press. He is Assistant Editor of the journal Studi Kantiani.  Kienhow Goh is currently Visiting Fellow at the National University of Singapore. His primary research interest is in German idealist philosophy, especially questions related to freedom of the will and action in Fichte’s practical philosophy. He is the author of several journal articles, book chapters, and book reviews on Classical German Philosophy. Dietmar H. Heidemann is Professor of Philosophy at the University of Luxembourg. His area of specialisation is Kant and German Idealism, especially Hegel. He is the first Chairman of the Kant-Gesellschaft and Member of the Kant-Kommission of the Berlin-Brandenburgischen Akademie der Wissenschaften.  Among his publications are Der Begriff des Skeptizismus.  Seine systematischen Formen, die pyrrhonische

Contributors ix Skepsis und Hegels Herausforderung (2007) and Kant und das Problem des metaphysischen Idealismus (1998). He is also the Editor-in-Chief of the Kant Yearbook. His latest book publication is Das Einzelne und das Allgemeine. Bd 1: Kants Begründung des Dualismus der Erkenntnisstämme (2024). Dino Jakušić is Associate Tutor in Philosophy at the University of Warwick in the UK. His PhD investigated how ontology as a philosophical discipline was developed and used in the metaphysical systems of Christian Wolff, Immanuel Kant, and G.W.F. Hegel. In 2021, he held a DAADfunded postdoctoral position at the University of Tübingen. His work focuses mainly on how early modern rationalist philosophy influenced German Idealism, with an emphasis on metaphysics and natural or rational theology. Michael J. Olson teaches at Marquette University in Milwaukee, Wisconsin. His research focuses on the relationship between philosophy and natural science in the German Enlightenment. He has published numerous articles on Kant, Classical German Philosophy, and recent French philosophy. Nicholas Stang is Associate Professor of Philosophy and Canada Research Chair in Metaphysics and Its History at the University of Toronto. He is the author of Kant’s Modal Metaphysics (2016), as well as numerous articles on the intersection of Kant, metaphysics, and German idealism. He recently edited, with Karl Schafer, a collection of essays on Kant’s metaphysics and epistemology, The Sensible and Intelligible Worlds (2022). He is currently working on his second monograph, tentatively titled How Is Metaphysics Possible? A Critique of Analytic Reason, which brings Kant’s critique of metaphysics and German idealist responses to Kant into dialogue with contemporary analytic metaphysics. Gesa Wellmann is a Junior Professor at the Carl von Ossietzky Universität Oldenburg. She obtained her PhD from the KU Leuven in 2018 with a thesis entitled “The Idea of a Metaphysical System in Lambert, Kant, Reinhold, and Fichte”. The main focus of her research is investigating the problem of systematicity and method. She has published several articles in renowned journals, including the Archiv für Begriffsgeschichte and Fichte-Studien. Her current scholarly work draws upon Early German Idealism, the philosophy of Enlightenment, and post-colonial theory. Catherine Wilson is Emeritus Professor at the University of York. She has published widely on 17th- and 18th-century philosophy, the history and philosophy of science, and the reception of Epicurus and Lucretius in the early modern period. Her most recent book is Kant and the Naturalistic Turn of 18th Century Philosophy (2022). She is based in Berlin.

Introduction Metaphysics as a Science in Classical German Philosophy Robb Dunphy and Toby Lovat

0.1  Metaphysics as a Science The unifying theme of this volume is not so much any of the various first-order metaphysical theses defended in the historical context of Classical German Philosophy as it is a particular metametaphysical position which remains significant throughout the period: that metaphysics ought to be developed as a systematic, demonstrative science (Wissenschaft). This claim, variations of which are defended by, among others, Wolff, Baumgarten, Kant, Reinhold, Schelling, and Hegel, could well be taken to characterise the dominant metametaphysical stance of Classical German Philosophy. It can look like a peculiar claim to the modern reader, perhaps either because they do not suppose that metaphysics should amount to any kind of science, but rather to a creative, humanistic discipline oriented towards the practical activity of “making sense of things”,1 or because they suppose indeed that metaphysics ought to amount to a science, but in the sense that it should fall in line with the method of contemporary natural sciences.2 As one encounters it in Classical German Philosophy, the claim that metaphysics ought to amount to a science is obviously rather different to contemporary approaches, but that alone is perhaps a good reason to consider the claim anew and see what can be learned. The claim that metaphysics ought to be developed as a systematic, demonstrative science, frequently defended by German philosophers from early in the 18th century until well into the 19th century, has important ancestors outside of Germany; most importantly Descartes, who broke with the scholastic orthodoxy of his time and argued that fundamental metaphysical principles were immediately available to pure reason, that they could be treated as axioms, and that from them further principles could be strictly derived in the form of a complete system.3 The paradigm of such an approach is, of course, Spinoza’s Ethics, but one finds it also in the work of Christian Wolff, one of the first to develop such a systematic metaphysics according to this geometric method and present it in German.4 After Wolff, the ideal of DOI: 10.4324/9781003187561-1

2  Robb Dunphy and Toby Lovat metaphysics as a science that, to some extent, approximates the structure, precision, and certainty of geometry is fairly dominant in German rationalism. This ideal appears to be upheld by Kant as ‘the dogmatic procedure’ according to which metaphysics ‘must prove its conclusions strictly a priori from secure principles’ (Kant 1998: Bxxxv). And for all of the various modifications which are made to this model, it runs through much of the history of post-Kantian idealism from Reinhold to Hegel, with the latter claiming that the argument of his Science of Logic, in which his core metaphysical positions are established, is intended to have the certainty of the making of mathematical inferences (Hegel 2010: 19–20).5 Of course, such claims to scientific certainty tended to be disputed, whether from a position broadly committed to the same model – as in the case of Lambert, who argued that Wolff had not properly demonstrated the validity of his fundamental principles – or from a position that, to some extent or other, holds that the model in question might not be so successfully deployed in the discipline of metaphysics after all, as in the case of certain sceptics, Romantics, and various others. Some discussion of these kinds of criticism can be found in this volume.6 Its primary subject matter, however, is the variety of approaches that lies underneath the apparent agreement on the methodology of metaphysics common to so much of Classical German Philosophy. There is far less agreement on how fundamental metaphysical principles are to be identified, on the identity of such principles, on the nature of the demonstrative method appropriate to metaphysics, and on the criteria for systematicity. In short, while many of the most significant philosophers from Wolff to Hegel agreed that metaphysics ought to amount to a “science”, they often had very different ideas about what this actually involved. Coming to understand these differences and disputes is crucial when considering the lessons and insights that might be derived from German philosophy in the 18th and 19th centuries and applied to contemporary debates in metaphysics and metametaphysics. Accordingly, the 15 chapters collected in this volume discuss various aspects of the idea that metaphysics ought to amount to a systematic and demonstrative science, as treated by some of the most significant figures in Classical German Philosophy. It should be noted that the role played by Kant in the narrative of the development of metaphysics in Classical German Philosophy poses additional difficulties, however, beyond those of understanding and evaluating what is at stake in treating metaphysics as a science. While there is a broad agreement among pre-Kantian German rationalists that metaphysics ought to approximate the model of the mathematical sciences, and while something similar emerges in post-Kantian idealist treatments of metaphysics which follow Reinhold’s attempt to rigorously derive a philosophical system from a single, indubitable principle, Kant’s own commitment to developing a science of metaphysics continues to be a matter of dispute, from its original

Introduction 3 reception right up until the present day. At times he is read as the great “all-destroyer” of metaphysics, at times as the rescuer of the science, putting it on the firm footing it requires. The chapters in this book which focus on Kant adopt no uniform stance on this topic, but rather range across several different possible positions on Kant’s attitude towards a scientific metaphysics. The reader can decide which best reflects Kant’s own remarks. Before introducing the contents of the volume, it seems appropriate to provide a word about its limitations. The chapters assembled here address a significantly large historical expanse and a wide variety of thinkers, but their coverage is far from comprehensive. There is, for example, little to no engagement with Thomasius, Crusius. or various other contemporaries of Wolff who played important early roles in the development of philosophy in Germany in the 18th century. Nor is there any engagement with Tetens, whose significance and influence among pre-Kantian German philosophers is ever more frequently acknowledged, nor with the important works of Mendelssohn. Both Jacobi and Maimon, who in different ways played crucial roles in the reception of Kant’s philosophy and the emergence of German Idealism, are not the subjects of dedicated chapters here. The Romantic philosophy of Novalis, or Schlegel, for example, is also passed over in regretful silence. Many further examples of unfortunate omissions could be provided and much more could be said about the metametaphysical positions developed by the philosophers who are the focus of chapters here. There are, however, two important points to register in light of this. The first is that such omissions are not the result of a judgement about the importance of these philosophers and positions, nor, more specifically, about what they have to say, critically or positively, about the idea of developing metaphysics as a science. This volume simply could not be excessively long, and the chapters included simply reflect the interests of our various contributors. Secondly, this volume makes no claim to provide a comprehensive treatment of the topic of metaphysics as a science in Classical German Philosophy. It is a collection of what we believe to be high-quality contributions to the burgeoning literature on this topic. Other stand-alone articles and volumes already exist which complement it, and others still will appear in the future.7 The lacunae in this volume’s coverage, significant though they are, can therefore be filled easily enough by the industrious reader. 0.2  Overview of Chapters 0.2.1  Chapter 1: Wolff on Ontology as Primary Philosophy

In Chapter 1, Dino Jakušić provides an account of the role played by Christian Wolff’s identification of ontology as primary philosophy in his conception of a systematic metaphysical science. Following a survey of the

4  Robb Dunphy and Toby Lovat development of the terminology of “metaphysics” and “primary philosophy” and the question of the proper object of this science from Aristotle, through the Mediaeval period, to early modern European philosophy, Jakušić suggests that Wolff’s identification of ontology as the science that investigates entity qua entity (or the most general predicates and principles concerning what it is for an entity to be at all) amounts to an attempted rehabilitation of ontology, against early modern rationalist predecessors like Descartes and Leibniz who associated it with what they took to be scholastic obscurities. As Jakušić explains, against this prejudice Wolff argues that the metaphysical claims defended by earlier modern rationalists presuppose the validity of properly ontological principles and concepts which have not been properly clarified and legitimised. Jakušić goes on to examine Wolff’s defence of the claim that ontology, as the science of entity qua entity, is best placed to ground the other metaphysical disciplines, and thus all of the other sciences. Having ontology play this crucial, primary role, he suggests, allows metaphysics to be developed as a system of interconnected disciplines, with the principles of lower sciences demonstratively established in higher ones. This model amounts to the Wolffian picture of a scientific metaphysics which would orient so much of subsequent Classical German Philosophy. Jakušić proceeds to explicate four distinct, but related ways in which ontology can play the primary, fundamental role which Wolff assigns it: primacy in cognition, in demonstration, in architectonics, and in the order of being. 0.2.2  Chapter 2: Baumgarten on the Nature and Role of Metaphysics

In Chapter 2, Courtney D. Fugate defends a distinctive interpretation of Baumgarten’s conception of the science of metaphysics. He rejects the common assumption that Baumgarten’s conception of metaphysics is not importantly different from that of Wolff, turning instead to a suggestion occasionally made but not satisfactorily developed, to the effect that Baumgarten’s account of metaphysics represents an important “subjective turn” in conceptions of the discipline. Fugate’s chapter is dedicated to fleshing out this idea. He begins by explicating Baumgarten’s distinction between “archetypal” and “ectypal” philosophy, or between the perfect philosophical knowledge of God and the discipline of philosophy as it can be carried out by essentially finite human beings, striving to emulate the archetypical philosophical knowledge of God to the greatest extent possible, but in a manner distinctively shaped by the limitations of the human intellect. Fugate goes on to argue that while Baumgarten conceives of God as the archetypal theologian, cosmologist, and ontologist, he reserves the term “metaphysics” for the science of the first principles of specifically human knowledge. Fugate points out that the subjective nature of Baumgarten’s

Introduction 5 conception of metaphysics is importantly different from Kant’s, since it does not involve restricting the application of fundamental concepts and principles to appearances. Instead, its subjective character follows from the fact that humans, due to the finitude of their intellect, are not able to perceive the manner in which all things reciprocally ground one another, so there are in reality no fundamentally first principles. The human intellect, in contrast to that of God, requires first principles from which other items of knowledge can be derived, even though this does not truly reflect the nature of reality. Baumgarten’s conception of metaphysics, claims Fugate, is precisely the science of identifying and building upon those first principles which can best enable the human intellect to emulate the archetypal philosophical knowledge of God to the greatest possible extent. In so doing, it must be guided by the mathematics of intensive quantities, as that science which, according to Baumgarten, allows one to consider the extent to which any thing approaches what is best. Accordingly, Fugate claims, although Baumgarten affirms the principle of contradiction as the first principle of metaphysics and derives further principles from it, he also believes that this principle itself could be derived from other principles; its selection as the first principle of metaphysics is guided by philosophical considerations concerning its clarity, distinctness, and how effectively further principles can be derived from it, ultimately in the form of a system, by the human intellect. Fugate then illustrates this character of Baumgarten’s approach to the selection of first principles by appeal to Baumgarten’s Elements of First Practical Philosophy, which similarly acknowledges that various candidates could be put forward as a first principle, and might ultimately all be derivable from one another, and argues that the selection must therefore be guided by considerations concerning which principle would serve best as the foundation for a systematic practical philosophy, given the essential limitations of human practical philosophers. 0.2.3 Chapter 3: Lambert on the Certainty and Generality of Metaphysics and Geometry

Chapter 3, by Katherine Dunlop, is occupied with Lambert’s work on the methodology required of a scientific metaphysics. Dunlop makes a case for the importance of Lambert’s reading of Euclid’s use of postulates in his Elements for the former’s work on both the certainty of metaphysics’ first principles and the guarantee of its generality. In the first half of the chapter, Dunlop charts how Lambert, in his earlier works, attempts to overcome the limitations he finds in Wolff’s deployment of the geometrical method in metaphysics. The primary deficiency Lambert finds in Wolff, explains Dunlop, is that although he successfully structures metaphysics as a

6  Robb Dunphy and Toby Lovat demonstrative science based on first principles, he provides no proof of the certainty and fundamentality of the basic concepts with which metaphysics must begin. Lambert’s proposed model for rectifying this deficiency is to adopt Euclid’s use of postulates in order to derive the fundamental concepts of metaphysics irrefutably from inner experience. In the latter half of her chapter, Dunlop turns to Lambert’s later work, where she finds that Lambert’s appeal to the model of geometrical postulates plays a different role in his work on the methodology of metaphysics. Here the issue at hand is how to guarantee the generality that metaphysics, as the fundamental discipline, demands. In particular, Lambert raises the question of the applicability of concepts derived from experiences of corporeal things to the intellectual world. Dunlop explains that Lambert appeals again to the use of postulates in geometry in order to develop an account of kind of generality that a scientific metaphysics requires. She argues that Lambert’s appeal to postulates here cannot be taken, as in the case of Kant, as an appeal to the method of construction, but rather that it should be taken as providing an alternative account of the “general and unconditioned” possibilities of things, quite distinct from the account of generality that one can derive by abstracting from particulars. This account of general and unconditioned possibility, Dunlop explains, follows from treating postulates as practical principles for things which provide rules for the things in general and for all of their possible particulars. Because this kind of generality is established by postulates “all at once”, Dunlop explains, it is better able to guarantee the general applicability of concepts than an attempt to examine individuals and abstract to their general similarities. Dunlop closes her chapter by examining some significant moments from Lambert’s discussion of how this kind of generality is already operative in geometry. This further illustrates the kind of method to which metaphysics must aspire, if it is to amount to a genuine science, according to Lambert. 0.2.4 Chapter 4: The Methodological Role of Intellectual Intuition in Kant’s Critique

With Chapter 4, by Toby Lovat, this volume begins to zero in on Kant, the most significant figure in the development of metaphysics in Classical German Philosophy. By way of an examination of Kant’s remarks concerning the hypothetical faculty of “intellectual intuition” throughout the Critique of Pure Reason, Lovat provides a recapitulation of the limits that Kant places on theoretical metaphysical speculation by means of his antecedent project of critique, which primarily considers our cognitive faculties, the way they can be understood to give rise to genuine cognition, and the kinds of subject matter which fall beyond the boundaries of their legitimate application.

Introduction 7 The insistence upon such an antecedent project of critique, of course, is the most distinctively Kantian contribution to debates about what is required in order to render metaphysics a genuine science. Lovat affirms the merits of a broadly “deflationary” interpretation of Kant’s first Critique, the most notable advocate of which is probably Henry Allison. Along these lines, Lovat argues that Kant’s discussions of intellectual intuition, as a model of cognition that would indeed cognise things in themselves, serve a primarily contrastive and negative purpose in the argument of the Critique, shedding light on the discursive nature of human cognition and Kant’s case for limiting the latter to knowledge of appearances. Any future science of metaphysics developed with the resources of theoretical reason, according to Kant, must observe this limitation. Indeed, as Lovat notes, this limitation is, for Kant, an essential condition of even the possibility of such a science, since in the absence of its restriction to the limits and conditions of discursive cognition, metaphysics is at best mere sophistry and at worse a calamitous dead end for reason. For Lovat, Kant’s metaphysical humility not only remains an important philosophical bulwark against dogmatism and scepticism, but also adds considerable weight to a “non-metaphysical”, deflationary, and methodological interpretation of Kant’s conception of intellectual intuition and things in themselves. 0.2.5  Chapter 5: Kant’s Promise of a Scientific Metaphysics

Continuing the focus on Kant, Chapter 5, by Catherine Wilson, offers a critical perspective on Kant’s own metaphysical aspirations. Wilson focuses on Kant’s projected metaphysics of nature. Against readings which emphasise the extent to which Kant understood himself to be reforming metaphysics to enable its development as a genuine science, she argues that the prospects for a Kantian metaphysics of nature are rather dim. Wilson first sets out the background against which Kant composed his critical works, providing detailed evidence of perceived threats of materialism, atheism, and fatalism from the spread of empiricism and scepticism in European thought, as well as from certain strands of dogmatic rationalist metaphysics. Kant’s concern in the face of these developments, Wilson suggests, prompted him to attempt to establish the possibility of a respectable metaphysics of nature, compatible with religious faith, freedom, the existence of the soul, and a secure system of morals. However, she argues that, despite the urgency of this project for Kant, his own commitment to its viability can in fact be seen to waver somewhat. In addition to pointing out instances where Kant appears to retreat from his ambition in favour of leaving the project of developing a metaphysics of nature to others, Wilson argues that the sketches Kant provides of the contents of such a metaphysics are left problematically underdeveloped.

8  Robb Dunphy and Toby Lovat In the final part of her chapter, Wilson goes further, suggesting that Kant’s projected metaphysics of nature is not only left underdeveloped, but also that he could not have successfully developed it at all. Wilson contends that limitations in Kant’s understanding of the natural science of his day, as well as the difficulties he encountered in connecting his Metaphysical Foundations of Natural Science to the science of physics, meant that he was simply not in a position to develop a metaphysics of nature that could meaningfully compete with notable treatments of nature emerging in Europe at the time, along with their perceived atheistic and materialistic implications. Wilson concludes that Kant’s remarks concerning the metaphysics of nature that he would develop on the basis of his critical work express little more than the hope that such a project is feasible and might be accomplished someday, by someone. 0.2.6  Chapter 6: Can Metaphysics Become a Science for Kant?

In Chapter 6, Gabriele Gava takes Kant at his word when he states that he intends to develop a scientific metaphysics after carrying out his critical, propaedeutic work, but identifies a problem for the viability of such a project within Kant’s own Critique of Pure Reason. Gava explains that Kant’s position is that developing a discipline as a genuine science requires understanding that all of its various cognitions are ordered by an idea of reason which lends unity to that science. This applies, clearly, to the idea of a science of metaphysics. However, Gava argues, metaphysics belongs to philosophy, but Kant seems to suggest that no one can truly realise the idea of philosophy, thereby appearing to put the goal of a properly scientific metaphysics out of reach. Considering how to respond to this apparent problem, Gava suggests that it is helpful to distinguish between two senses of “philosophy”, for Kant: one according to which philosophy is simply a body of theoretical and practical a priori cognitions, and another according to which philosophy is intended to provide practical guidance for attaining virtue. Gava elucidates this distinction by examining Kant’s remarks about the “school concept” and the “worldly concept” of philosophy, the distinction between a doctrine of skills and a doctrine of wisdom, and his characterisations of the “philodox” and the “misologist”. Although the elaboration of these distinctions might be taken to suggest that the philosophical idea Kant takes to be unrealisable concerns only that of the philosopher as a paragon of morality, and thus presents no obstacle to conceiving metaphysics as a unified body, or science, of a priori cognitions, Gava does not think that the matter is so straightforward. Instead, he suggests, Kant seems to think that a philosophical discipline can only become a science according to its worldly concept, that is, taken in relation to its essential, practical ends. This means that the problem in

Introduction 9 question is not so easily dissolved. Gava responds by tentatively suggesting that there might be two different ways of conceiving of philosophy, and metaphysics in particular, according to its ends. In the first case, this involves acknowledging that metaphysics must take the highest good to be the fundamental end of human reason, while in the second it involves the additional task of providing practical guidance concerning how the good is to be achieved. He suggests that conceiving metaphysics according to the first model might allow the discipline to take into account its essential, practical ends, without thereby automatically committing the metaphysician to the attempted approximation of a moral perfection that Kant takes to be impossible. Gava concludes with some reflections on the implications of this line of thought for the purposes of clarifying the relation between the respective doctrines of method in Kant’s first and second Critiques. 0.2.7 Chapter 7: Scientific Metaphysics and Metaphysical Science: The Demand for Systematicity in Kant’s Transition Project

In Chapter 7, Michael J. Olson turns his attention to the development that takes place in Kant’s views on the relation between the metaphysics of nature and empirical science in the 1790s. After characterising Kant’s views on the matter in the First Critique and the Metaphysical Foundations of Natural Science (in which the possibility of metaphysics as a systematic science, the fundamental principles of which are known a priori is defended, while the principles that can provide unity and thus systematicity to the empirical sciences are thought only ever to be assumed hypothetically, according to a purely regulative use of reason), Olson turns to Kant’s developing views on this topic, in the direction of his Opus postumum. Here, as Olson explains, it appears that Kant comes to think that a constitutive use of reason can provide a fundamental principle for physics a priori, thus apparently blurring the distinction he so carefully drew between metaphysics and physics in the first Critique. Olson argues that this peculiar change in Kant’s position is at least partly prompted by the lack of anything like a significant consensus on several of the fundamental issues in the physics of his day, which he illustrates by considering the wildly varying views circulating at the time on the topic of the ether. Olson argues that the disturbing lack of anything like an emerging unified picture of physics might have prompted Kant to attempt to extend the reach of his systematic metaphysics of nature so that it could provide constitutive principles for physics, thus setting the empirical sciences on surer foundations and better allowing for the emergence of them in systematic form. Olson closes by reflecting on the significance of the systematicity of the sciences for Kant’s views of both metaphysical and empirical investigations of nature, but suggests that Kant’s later tendency to blur the lines between metaphysics and physics risks

10  Robb Dunphy and Toby Lovat undermining his case for the apodictic certainty of his metaphysical framework, especially if he were to tie the table of the categories too closely to the contents of the questionable empirical sciences of his day. 0.2.8 Chapter 8: Kant, Reinhold, and the Problem of Philosophical Scientificity

In Chapter 8, Karin de Boer and Gesa Wellmann provide a critical appraisal of Reinhold’s attempt, during the period of his “Elementary Philosophy”, to reformulate the project of Kant’s Critique of Pure Reason according to what he believed to be a properly rigorous, scientific model, in order to better accomplish Kant’s stated task of bringing metaphysics “onto the secure path of a science”. Wellmann and de Boer argue that although Reinhold derived his understanding of the model of a genuine philosophical science from his reading of Kant, this reading, particularly of parts of the Transcendental Analytic, proves to be deeply questionable. They argue that Reinhold is not justified in claiming that Kant’s project acknowledges the need for a foundational principle from which the results of the Transcendental Doctrine of Elements would be derived. Having set out what they take to be Reinhold’s misreading of Kant, de Boer and Wellmann show how this interpretation leads Reinhold to formulate his “principle of consciousness” as a putative foundation for a supposedly improved, axiomatic-deductive version of the arguments of Kant’s first Critique. Despite Reinhold’s modifications to Kant’s approach – which extend even to the claim that, upon recognising the principle of consciousness as the fundamental, self-evident principle, the enterprise ceases to be critical – de Boer and Wellmann argue that Reinhold does not abandon Kant’s distinction between a necessary propaedeutic discipline and a subsequent science of metaphysics. In Reinhold’s view, they argue, the science of metaphysics must presuppose a scientific treatment of representation in general: the primary purpose of the Elementary Philosophy. Despite this degree of continuity with Kant’s critical project, they conclude that the popularity of Reinhold’s metaphilosophical innovations is largely responsible for the “disappearance” of the Kantian idea of a critique of pure reason from German Idealism. 0.2.9  Chapter 9: Reinhold on the Deduction of the Categories

In Chapter 9, Elise Frketich also takes up Reinhold’s appropriation and revision of Kant’s critical philosophy during the period of his Elementary Philosophy. Rather than focusing heavily on Reinhold’s potential misunderstandings of the criteria that Kant takes to apply to a properly scientific philosophical project, Frketich attends specifically to the limitations

Introduction 11 Reinhold perceives in Kant’s metaphysical and transcendental deductions of the categories of the understanding. Reinhold’s position, as she reconstructs it, is that Kant has failed to successfully demonstrate the completeness of his table of categories, and, relatedly, that his justification for his metaphysical deduction fails because it relies on the results of his transcendental deduction, which is itself viciously circular. On the basis of these criticisms, Frketich sympathetically explicates Reinhold’s attempts to provide improved versions of the two deductions, in the opposite order to Kant: firstly a transcendental deduction, and on its basis, a metaphysical deduction. In this manner, Frketich exhibits Reinhold’s attempt to revise a central element of Kant’s metaphysics according to what he conceives as a properly scientific model. Frketich explains that Reinhold takes himself to prove the results of an improved transcendental deduction by deriving the claim that the categories are the conditions of the possibility of cognition of objects and thus originate in the understanding from the concept of representation itself, which in turn he derives from what he takes to be the self-evident principle of consciousness. In this way, she argues, Reinhold claims to avoid the vicious circularity he detects in Kant’s own transcendental deduction. Next, Frketich details how Reinhold takes himself to derive Kant’s table of categories from his own principle of consciousness in the form of a complete system, thus solving the other of the two problems he identifies in Kant’s deductions. 0.2.10 Chapter 10: Schulze’s Scepticism and the Rise and Rise of German Idealism

In Chapter 10, Robb Dunphy turns his attention to G.E. Schulze’s early sceptical response to Kant and Reinhold’s critical and metaphysical projects, and in particular to Jessica Berry’s recent case to the effect that the genuine force of Schulze’s scepticism has been largely underappreciated. After a brief introduction to Schulze’s text and some of its better known claims, Dunphy presents the essentials of Berry’s interpretation, which emphasises Schulze’s recovery of the Pyrrhonian Scepticism of Sextus Empiricus in order to argue that the attraction of Schulze’s scepticism and the force of the criticisms that it directs towards the metaphysical projects of Kant and the German Idealists has largely been overlooked. In the remainder of the chapter, Dunphy provides a critical examination of Berry’s interpretation. He argues that the attribution of a genuinely Sextan or Pyrrhonian scepticism to Schulze is not a straightforward matter: although Berry is right to direct attention to the relationship between Schulze and Sextus, elements of Schulze’s scepticism are distinctively modern and do not fit with the portrayal of Schulze as a Pyrrhonist, while other elements

12  Robb Dunphy and Toby Lovat are reminiscent of Sextus’ scepticism, but not necessarily in the ways that Berry takes them to be. In the final part of the chapter, Dunphy argues that the metaphysical projects of the German Idealists might not have been as vulnerable to Schulze’s scepticism as Berry suggests, and he emphasises some of the affinities between their approach to philosophical inquiry and that of the Pyrrhonian Sceptics. 0.2.11 Chapter 11: The I and I: The Pure and the Empirical Subject in Fichte’s Science of Science

In Chapter 11, Kienhow Goh provides a discussion of the fundamentals of Fichte’s early presentations of the Wissenschaftslehre, emphasising the need to distinguish between Fichte’s account of the “pure I”, even when it is conceived as finite and determined, and the “empirical I” of an individual’s inner experience. Maintaining this distinction, Goh argues, both prevents a reading of Fichte’s subjectivism as merely formal and arbitrary, and better allows a proper appreciation of how Fichte’s idealism answers one of Maimon’s most significant sceptical challenges to Kant’s transcendental deduction of the categories. Explicating the fundamentals of Fichte’s conception of the Wissenschaftslehre in this way, Goh argues that the primary orientation of Fichte’s project is an epistemological one, directed towards securing the possibility of scientifically certain knowledge, particularly against sceptical challenges concerning the legitimate application of a priori concepts to empirical content. Goh proceeds to analyse Fichte’s conception of a science and the function and justification of its first principles, arguing that philosophy, as the “science of science”, attempts to present the “human mental system” or the system of cognitive acts which make all other sciences and human knowledge in general possible. He argues that, from the standpoint of philosophy, no sharp division is drawn between the I and the objects of knowledge, such that a systematic account of experience is possible a priori, even though, from the standpoint of experience, one’s knowledge is of objects external to oneself, known only empirically. Maintaining these two perspectives, Goh suggests, allows Fichte to answer the Maimonian criticism that it is illegitimate to apply a priori categories to the contents of empirical experience, which appear quite heterogeneous to them. Finally, Goh reiterates Fichte’s epistemological or metaepistemological reorientation of the science of metaphysics along transcendental lines, concerning the a priori conditions of the possibility of experience. Maintaining the distinction between the pure and empirical I, Goh explores Fichte’s development of Kant’s distinction between determining and reflective judgements, in order to illustrate Fichte’s case for unifying sensible nature and the moral order in a single, scientific, philosophical system.

Introduction 13 0.2.12 Chapter 12: The Science of All Science and the Unity of the Faculties: Schelling on the Nature of Philosophy

In Chapter 12, Benjamin Berger takes Schelling’s 1803 Lectures on the Method of Academic Study, in which he presents a series of recommendations concerning the organisation of the university, as a launch pad from which to provide an account Schelling’s conception of the scientific metaphysics he identifies with “philosophy” and its relation to other disciplines. On the basis of Schelling’s remarks in those lectures, Berger identifies a distinctive account of philosophy in his thought, as a science which provides insight into the essential structure of reality by purely rational means; a task which, he suggests, requires abstracting from one’s individual perspective and taking up the universal standpoint of reason as such. Berger takes Schelling to argue that this universal standpoint lends a holistic perspective to philosophy capable of incorporating and providing full intelligibility to the knowledge developed in the other sciences. Philosophy is thus essentially concerned with organising knowledge in the form of a system. Berger argues that Schelling’s conception of the role of philosophy in lending full intelligibility to the results of the other sciences and organising them in the form of a system is crucially different from the otherwise similar sounding project of Fichte’s Wissenschaftslehre. He details Schelling’s rejection of the Fichtean project of providing a foundation for all of the other science by means of a preparatory metaphilosophical investigation into the possibility of knowledge and contrasts this with the former’s project of elaborating an account of philosophy as the science of science which maintains from the outset an interest in making claims about the nature of reality, but from a universal standpoint that allows it to identify and unify the philosophical insights developed in the specific contexts of the various natural and human sciences. For Schelling, Berger suggests, this means that all other disciplines, even if they are not to be identified with philosophy, contain philosophical insights and are to some extent inspired by the philosophical goal of gaining an understanding of the universal nature of reality. Berger then reflects on the fact that Schelling’s resistance to Fichte’s idea of a science whose object is scientific knowledge itself, rather than reality, means that he has rather narrowed the scope of Kant’s conception of “pure philosophy”. For Kant, “pure philosophy” indicates both the project of a preparatory critique of our cognitive faculties and a subsequent scientific metaphysics, but Schelling appears to want to do away with the necessity of the first of these, since he takes it to push one towards an excessively subjective conception of philosophy. Berger concludes by considering the manner in which their differences lead Kant and Schelling to alternative accounts of how the autonomy of philosophical reason is to be preserved and of its role in relation to the other sciences within the university.

14  Robb Dunphy and Toby Lovat 0.2.13 Chapter 13: Two Models of Critique of Metaphysics: Kant and Hegel

In the antepenultimate chapter of this volume, Dietmar Heidemann considers Kant’s and Hegel’s competing critical accounts of the presuppositions and method of dogmatic metaphysics and the lessons that they draw from their respective critiques. Heidemann begins with Kant’s “capacities first” model, examining Kant’s account of the cognitive faculties of sensibility and understanding, along with the capacities of receiving intuitions and generating concepts that belong to each faculty, respectively. By means of this analysis, Heidemann clarifies Kant’s case for his cognitive dualism: the idea that only the cooperation of these two fundamentally different faculties suffices for genuine cognition. Since, he argues, Kant takes dogmatic, rationalist metaphysics to rely upon the method of amplifying cognition purely on the basis of concepts, a priori, the “capacities first” model of critique and the cognitive dualism it establishes has the implication that metaphysical cognition is not possible, and thus that rationalist metaphysics, if understood as the science of super-sensible objects, is not possible. Heidemann then turns to Hegel, suggesting that he is in broad agreement with Kant’s rejection of dogmatic metaphysics. Nevertheless, he takes Hegel to propose a “thought determination” model of the critique of metaphysics, among the results of which is a positive estimation of the possibility of developing metaphysics as a genuine science. This looks curious next to Hegel’s apparent agreement with Kant’s criticisms of dogmatic metaphysics, however, since the contention that genuine metaphysical knowledge can be established by means of the analysis of thought determinations or categories alone appears to suggest that Hegel is straightforwardly returning to the rationalist presupposition that Kant criticised. To differentiate Hegel’s position from that of dogmatic rationalism, therefore, Heidemann offers a brief reconstruction of the development of Hegel’s views on metaphysics. He suggests that, according to Hegel’s mature view, the thought determinations that are the primary subject matter of the science of logic can be legitimately taken to capture the nature of things, and as such must differ fundamentally from the contents of earlier, rationalist metaphysics. Heidemann illustrates this difference by way of a discussion of the “untrue” and “true” uses of thought determinations, intended to show how Hegel takes dogmatic metaphysics to attempt to know the objects of metaphysics by means of finite, limited concepts, drawn empirically from representation, which are fundamentally inadequate for this purpose, in contrast to the “true” use of thought determinations which he himself develops in his Science of Logic, precisely by means of the critique and sublation of the limitations of the merely finite grasp of concepts characteristic of the understanding. Heidemann thus illustrates that, although

Introduction 15 Kant and Hegel both offer similar sounding critiques of dogmatic metaphysics, they are on very different grounds. The former criticises dogmatic rationalism for making metaphysical claims on the basis of an unjustified assumption concerning human understanding’s capacity to cognise things without sensible intuition, while the latter criticises it for proceeding on the basis of finite thought determinations that have themselves not been critically examined or justified. In the final part of the chapter, Heidemann supports his illustration of Kant’s and Hegel’s competing models of critique of metaphysics by presenting them as diagrammatic schema, and suggests that what should decide the validity of one model over the other is the question of Hegel’s ability to capture the role played by sensible intuitions for Kant in purely conceptual terms. The last two chapters of this volume develop its theme in slightly different directions, away from straightforward contributions to the history of Classical German Philosophy. The first takes the period’s characteristic emphasis on the nature of the demonstration appropriate to a scientific metaphysics and uses examples of inattention to its details to provide an illustration of how incautious readings in the history of philosophy can misrepresent both the thought of historical figures and our proximity and distance to them (and thus, perhaps, also what we might learn from them). The second utilises reflections from the period on the methodology and epistemology of metaphysics and the natural sciences in order to develop a critical perspective on the predominant metametaphysical orientation of contemporary philosophy. 0.2.14 Chapter 14: Quietism, Dialetheism, and the Three Moments of Hegel’s Logic

In the penultimate chapter of the volume, G. Anthony Bruno considers Hegel’s characterisation of the three “moments of every logical reality … every concept or every truth”, which, taken together, provide a rough model of the kind of demonstration that, according to Hegel, is immanent to the subject matter of logic and metaphysics. Bruno uses Hegel’s account of these three moments to critically evaluate two contemporary appropriations of Hegel’s thought: McDowell’s interpretation of Hegel’s idealism as shoring up a common sense realism and Priest’s attribution to Hegel of a logical and a metaphysical dialetheism. Bruno finds that McDowell’s recruitment of Hegel’s idealism for the task of advocating a therapeutic approach to sceptical questions and a philosophically quietistic orientation towards the deliverances of common sense problematically skips over the second, sceptical or dialectical, of Hegel’s moments, resulting in an idealist standpoint that Hegel himself would be bound to reject as unacceptably dogmatic. He then argues that Priest’s dialetheist reading of Hegel fails to

16  Robb Dunphy and Toby Lovat appreciate the importance, for the latter, of the systematic resolution of the contradictions that emerge in the course of his logical and metaphysical science. For Bruno, this misinterpretation follows from Priest’s tendency to halt at the negative results of the second, dialectical moment of Hegel’s logical procedure, without attending to the possibility of arriving at a positive result in which the contradictions in question are speculatively sublated and unified, as Hegel’s characterisation of the third logical moment suggests. In both cases, therefore, and despite their substantial differences, Bruno finds that McDowell’s and Priest’s misreadings of Hegel stem from their inattention to Hegel’s claim that the three moments constitutive of the method of logical demonstration must be grasped together, as a unity. Bruno concludes that misreadings of the kind he finds in McDowell and Priest, which spring from insufficient attention to Hegel’s characterisation of the method proper to the subject matter of logic and metaphysics, result not only in an inaccurate picture of Hegel’s own philosophy and its concerns, but also of the distance of much contemporary philosophy from Hegel’s own. Yet gaining an accurate picture of such things, he argues, is necessary for the proper evaluation of what historical philosophers like Hegel can offer us. 0.2.15 Chapter 15: Metaphysics on the Model of Natural Science? A Kantian Critique of Abductivism

In the final chapter, Nicholas Stang develops a critique of “abductivism” about metaphysics. In order to develop this critique, Stang first provides a sketch of abductivism, a set of metametaphysical positions defended in one form or another by many contemporary metaphysicians, the key component of which is the claim that the basic methodology of metaphysics is the same as that of the sciences: one centred on the use of inference to the best explanation. Stang points out that, in Kantian terms, abductivists are committed to a “critique of metaphysical reason”, not only in the sense that they are interested in identifying limits of reason’s capacities when it comes to metaphysics, but also in the sense that they are interested in explaining how it is possible for human reason to achieve knowledge in the discipline of metaphysics. The primary goal of the chapter is not to set out Kant’s alternative, non-abductivist position on methodology and epistemology of metaphysics in any detail, but to argue that abductivism cannot provide an adequate explanation of the possibility of knowledge when it comes to metaphysics. Stang develops this critical account by way of a consideration of Kant’s explanation of the possibility of abductively achieved knowledge in the natural sciences, arguing that contemporary abductivists can provide no equivalent explanation of abductively achieved knowledge in metaphysics.

Introduction 17 Stang shows, primarily by appealing to the Appendix to the Transcendental Dialectic, that Kant has “at least the beginnings” of explanations of three theses that are required in order to explain the possibility of abductively achieved knowledge in the natural sciences: that the (phenomenal) world has enough in the way of ontic explanatory structure for our natural scientific explanations to correspond to, that abductive reasoning in natural science tracks both the truth and the underlying ontic explanatory structure in a manner that is not merely lucky, and that the terms or concepts of our best natural scientific theories carve the ontic explanatory structure of the world at its joints, so as to provide genuine explanations. By contrast, Stang claims that contemporary abductivists about metaphysics tend either to make no real attempt to explain why the world should have sufficient ontic explanatory structure for our metaphysical explanations to correspond to, why abductive reasoning in metaphysics should track the truth and the ontic explanatory structure of the world in a way that is not merely lucky, or why the terms or concepts of our best metaphysical theories carve the ontic explanatory structure of the world at its joints, or, when such an attempt is made, it tends not to succeed. The respective theses about the possibility of abductively achieved knowledge in metaphysics turn out to be more or less arbitrarily asserted. In Kantian terms, Stang points out, this makes contemporary abductivists into “dogmatists”: uncritically assuming that (abductive) reasoning can make genuine progress when it comes to metaphysics. Stang closes by pointing out that even if Kant’s own explanation of the possibility of metaphysics, on the basis of his transcendental idealism, is found wanting, this should not automatically direct us back in the direction of abductivism. He suggests that one might instead give serious consideration to the views of those post-Kantian idealists who manage to combine Kant’s critical rejection of dogmatism with an alternative explanation of the possibility of metaphysics, one which does not depend upon transcendental idealism. Notes 1 See Moore 2012: xviii, 1–22, 595–606 2 Williamson 2016, for example, represents a clear instance of this tendency. A more extreme version of the claim that metaphysics should follow the methodology of the natural sciences is provided in Ladyman and Ross 2007. 3 Descartes’ methodological innovations were at least in part prompted by the increasing prevalence of sceptical philosophy in Europe. This, coupled with his not unusual view that metaphysics deals with the most fundamental matters, upon the results of which other disciplines depend, lent some urgency to the project of reformatting the methodology of metaphysics on more secure grounds. 4 See, for example, Wolff 1983: §383, and Chapter 1 in this volume. 5 See de Boer 2011 for a good discussion of points of continuity in the conceptions of metaphysics operative in Wolff, Kant, and Hegel.

18  Robb Dunphy and Toby Lovat 6 See Chapter 3 in this volume for some discussion of Lambert’s criticisms of Wolff in the service of an improved science of metaphysics. See Chapter 10 in this volume for some discussion of one of the more notable sceptical criticisms of aspirations towards developing a scientific metaphysics, from Schulze, although Schulze notably does not claim outright that the model of such a science itself should be abandoned. 7 For some other helpful collections related to the topic of this volume, see, for example, Gerhard, Sell, and de Vos 2012; Schleich 2021; Fugate and Hymers 2018 (on Baumgarten and Kant specifically); Altman 2014 (on German Idealism); or Bondeli, Chotaš and Vieweg 2016 (on scepticism and metaphysics in post-Kantian philosophy).

Bibliography Altman, M. (Ed.) (2014) Palgrave Handbook of German Idealism (Basingstoke: Palgrave Macmillan) Bondeli, M., Chotaš, J., and Vieweg, K. (2016) Krankheit des Zeitalters oder Heilsame Provokation? Skeptizismus in der Nachkantischen Philosophie (Paderborn: Wilhelm Fink Verlag) de Boer, K. (2011) “Transformations of Transcendental Philosophy: Wolff, Kant, and Hegel” Hegel Bulletin Volume 32 Issue 1–2, pp. 50–79 Fugate, C. and Hymers, J. (2018) Baumgarten and Kant on Metaphysics (Oxford: Oxford University Press) Gerhard, M., Sell, A, and de Vos, L. (2012) Metaphysik and Metaphysikkritik in der Klassischen Deutschen Philosophie (Hamburg: Felix Meiner Verlag) Hegel, G.W.F. (2010) Science of Logic. Translated by G. di Giovanni (Cambridge: Cambridge University Press) Kant, I. (1998) Critique of Pure Reason. Translated by P. Guyer and A. Wood (Cambridge: Cambridge University Press) Ladyman, J. and Ross, D. (2007) Every Thing Must Go: Metaphysics Naturalized (Oxford: Oxford University Press) Moore, A. (2012) The Evolution of Modern Metaphysics: Making Sense of Things (Cambridge: Cambridge University Press) Schleich, N. (Ed.) (2021) Philosophie als Wissenschaft: Wissenschaftsbegriffe in den philosophischen Systemen des Deutschen Idealismus (Hildesheim: Georg Olms Verlag) Williamson, T. (2016) “Abductive Philosophy” The Philosophical Forum Volume 47 Issue 3–4, pp. 263–280 Wolff, C. (1983) Vernünftige Gedancken von Gott, der Welt und der Seele des Menschen, auch allen Dingen Überhaupt (Hildesheim: Georg Olms Verlag)

1

Wolff on Ontology as Primary Philosophy Dino Jakušić

1.1 Introduction Christian Wolff (1679–1754) is arguably the most influential German philosopher in the period between Leibniz and Kant. His work can be read as connecting the wider world of Early Modern thought with what will become known as Classical German Philosophy. Given this significance, it is both unfortunate and remarkable that Anglophone philosophers have largely neglected Wolff’s work. Indeed, if he is known at all, he is likely to be recognised as a mere systematiser of Leibnizian thought,1 or else as a punching bag for Kant and Hegel. Recently, however, things have begun to change; today we are witnessing a surge in translations and scholarly interest in Wolff’s work.2 Wolff’s influence lies more in the form in which he presented his ideas than in the ideas themselves, many of which can be traced to earlier thinkers. Together with Christian Thomasius, Wolff is considered as the inventor of the German philosophical language, due to his choice to write his logical and philosophical work in German.3 After his exile from Halle in 1723, which earned him the reputation of an ‘Enlightenment martyr’, he produced an extensive series of publications in Latin in order to reach a wider European audience.4 This resulted in the publication of the Latin Logic in 1728, Primary Philosophy or Ontology in 1730, General Cosmology in 1731, etc.5 Although Wolff’s philosophy never received the same acclaim in the rest of Europe as it did in Germany, his influence on philosophy still persists. A strong argument can be made for the claim that it is due to Wolff that the terms such as “ontology”, “cosmology”, and “teleology” entered mainstream philosophy. Moreover, in the preface to the second edition of his German Metaphysics, he seems to have been the first to use the term “idealism” as a classificatory term. Finally, Wolff is often referred to as an originator of the “traditional” division of metaphysics into Metaphysica Generalis and Metaphysica Specialis – even though such attempts predate him, and, while this kind of division is central to his philosophy, these exact terms do not appear anywhere in his work.6 DOI: 10.4324/9781003187561-2

20  Dino Jakušić In what way does Wolff attempt to establish metaphysics as a science? Like most classical German philosophers, Wolff argues that a properly scientific metaphysics must be formulated as a system. And almost 60 years before the Critique of Pure Reason, Wolff anticipates Kant’s complaint, stating that ‘much darkness has come over metaphysics, thereby also obscuring other sciences’.7 Wolff envisions a general philosophical system that aims to ground all sciences, be they natural, metaphysical, or mathematical. This project will succeed, he thinks, if philosophy is (re-)constructed according to a sufficiently rigorous demonstrative method, and its disciplines organised in a proper architectonic order. At the pinnacle of this system of sciences, Wolff places the discipline of primary philosophy, which he identifies with the then fringe term ontology. In this chapter, I will provide an account of Wolff’s conception of ontology and of the significance of its identification with primary philosophy, which was regarded as controversial at the time. I believe that it is largely due to Wolff that the term ‘ontology’ survives its Early Modern birth, and therefore that Wolff enabled the term to enter into mainstream philosophical discourse, a legacy which lasts to this day.8 I will not try to demonstrate this claim here, but in showing the importance of Wolff’s identification of primary philosophy with ontology, and in explicating the role of ontology in his overall philosophical system, I hope to make Wolff’s influence on the subsequent attempts to formulate metaphysics as a science more apparent. 1.2  What Ontology Was and What It Was Not Today, ontology is far from a peripheral discipline. The term finds its home across the sciences, be they natural, social, or humanistic, where it is often understood in distinct ways. Even within philosophy, the use of the term varies widely across traditions and approaches. For example, one notable way of understanding ontology, inspired by W.V.O. Quine, sees it as an investigation into ‘what there is’, or into which (kinds of) entities exist. We also encounter discourses on ‘regional’ ontologies, such as the ontology of mind, mathematics, social ontology, etc. In a different tradition, inspired by Martin Heidegger, ontology is best regarded as an investigation into Being, rather than into existence.9 None of these senses, however, fit the way in which ontology was understood by 17th- and 18th-century philosophers. For them, ontology was the science of an entity qua entity. The term ‘ontology’ enters the history of philosophy relatively late, with its first known appearance in a 1606 text by Jacob Lorhard, who for a brief time was Professor of Theology at Marburg, entitled Ogdoas Scholastica. In this book, probably written as a textbook, the term appears three times. It is used as a synonym for metaphysics and is never defined or explained.10 Another significant appearance comes from Rudolph

Wolff on Ontology as Primary Philosophy 21 Goclenius’ 1613 Lexicon Philosophicum, which is still sometimes erroneously cited as a first appearance of the term. Unfortunately, the term is a hapax legomenon, appearing in a margin to the article on Abstraction, simply stating ‘ὀντολογία, philosophia de ente’.11 Another pertinent appearance is in works of Johannes Clauberg, the first rector of the University of Duisburg, who is considered the author of the first proper treatise on ontology.12 Of especial significance is his Elementa philosophiae sive ontosophia of 1647, which was republished in 1660 as Ontosophia nova, quae vulgo metaphysica, and again in 1664 as Metaphysica de ente, quae rectius Ontosophia. Wolff cites the last of these editions, from which we can assume that it exerted direct influence on his own understanding of what ontology should be. The term ‘ontology’ therefore predates Wolff’s work on the subject by more than a century. However, the philosophers who are regarded as having defined the Early Modern period use it very sparingly, or not at all. There is no mention of “ontology” in Locke, Descartes, Spinoza, or Leibniz’s published works.13 Instead, it tends to appear as an entry in philosophical dictionaries and in works by authors now mostly forgotten, where it is used either as a synonym or as a proper term for metaphysics. While there are some variations regarding what ontology was understood to be, it was always defined as a science of an entity qua entity.14 But what does that mean? Let us look at one of Wolff’s definitions of ontology: There are some things which are common to all entities [enti omnia communia] and which are predicated both of souls and of natural and artificial bodies. That part of philosophy which treats of an entity in general [de ente in genere] and of the general affections of entities is called Ontology, or Primary philosophy. Thus, ontology, or primary philosophy is defined as the science of an entity in general, or insofar as it is an entity [scientia entis in genere, seu quatenus ens est]. (DP §73) Wolff’s conception of “ontology” is strikingly different to current uses of the term. Since ontology was conceived as a science investigating what is common to all entities, there could be no regional formulations of it (e.g., the ontology of mathematics, or of mind). Ontology was also not concerned with existing entities: it neither draws up the inventory of ‘things that are’, nor of kinds of actually existing entities.15 The aim of Wolff’s ontology was to provide us with a set of concepts that constitute the necessary and essential predicates of any possible entity. At the same time, this science cannot, and does not attempt to demonstrate that the entities to which these predicates belong actually exist. For example, ontology shows that all entities can either be simple or complex, and it further

22  Dino Jakušić explicates the properties of simplicity and complexity as such. It does not ask whether any simple or complex entity actually exists.16 An existing entity (called an actual entity [ens actuale])17 belongs only to a specific class of possible entities. Hence, ontology cannot privilege existing entities. The concern with existence only reaches as far as this relates to the investigation of the universal predicates that can be ascribed to any possible entity. Hence, Wolff defines existence as a ‘complement of possibility’ [complementum possibilitatis].18 Instead of concentrating on “existence”, Wolffian ontology consists in the investigation of concepts such as essence, attribute, mode, necessity, contingency, place, time, perfection, etc. According to him, these concepts are not explained in other parts of philosophy, since all philosophy relies on them and the principles derived from them.19 Taken together, these “ontological” predicates constitute the most general predicates that can be ascribed to any possible entity, regardless of its particular conditions. To summarise: to understand Wolff’s attempt at developing a scientific metaphysics, we need to understand, I argue, his identification of primary philosophy with ontology. Ontology, as we have seen, is the science of an entity in general. It investigates what is common to all entities qua entities and is to be identified with primary philosophy. In that case, the questions to be addressed concern what primary philosophy is and how the concepts of ontology relate to it. 1.3  Primary Philosophy and Metaphysics While the term “ontology” predates Wolff by more than a century, the terms “primary philosophy” and “metaphysics” predate his work by almost two millennia. “Primary philosophy” [prōtē philosophia] was both the title of and the science formulated in the collection of Aristotle’s texts today known as Metaphysics. The term “metaphysics”, which was unknown to Aristotle, is traditionally considered to originate in the 1st century BCE when Andronicus of Rhodes placed the treaties on Primary Philosophy behind the treaties on Physics – meta ta physika.20 We might use the term primary philosophy merely to point out the original title for the text we know as Metaphysics, and leave the issue at that. However, the fact that the name metaphysics appears as the title for a work that never once employs this term, unbeknownst to its author and for unclear reasons, has resulted in attempts to clarify the relation between the two. And importantly, these attempts were not motivated by merely historical or philological reasons. Instead, the difficulty of distinguishing between primary philosophy and metaphysics has been linked to an ambiguity in Aristotle’s text, and has contributed to the problem that is sometimes referred to as the problem of the object of metaphysics.

Wolff on Ontology as Primary Philosophy 23 The problem arises from apparently conflicting accounts in books 4 and 6 of Metaphysics, regarding what primary philosophy is supposed to investigate. In Metaphysics 4, we are told that primary philosophy will ‘investigate an entity as an entity [theōrei to on hē(i) on] and the attributes which belong to it in itself’ (Metaphysics 4: 1.1003a20–1. Translation modified).21 In contrast to other sciences, which focus on a particular genus of entities, primary philosophy will focus on an entity qua entity which, according to Aristotle, is not a genus. However, in Metaphysics 6, we are told that primary philosophy deals with divine objects, i.e., those substances that are immaterial and unchangeable, and with god as the first principle of everything.22 According to this second account, primary philosophy is an investigation, not into an entity qua entity but into a genus of “supernatural” or “super-sensible” substances. Aristotle, there fore, names this science theologikē – theology (or theologics, theological art).23 The need to resolve this unclarity regarding the object of primary philosophy is heightened once Aristotle’s texts are taken up by subsequent traditions. According to Jan Aertsen (2005), among the Neoplatonic commentators, the theological interpretation regarding the object of primary philosophy was dominant, since it aligned with their approach of investigating the One beyond ousía. This changes in Mediaeval philosophy, where the problem of the object of metaphysics seems to arrive into Catholic Scholasticism by way of the Arabic tradition. As Duns Scotus suggests in his Questions on the Metaphysics: ‘the first question is whether the proper subject of metaphysics is an entity qua entity, as Avicenna claimed, or God and the intelligences, as Averroes assumed’.24 The problem is later picked up in the German Protestant Scholasticism of the early 17th century, through the renewal of interest in Aristotle’s philosophy.25 As Ernst Vollrath pointed out, every commentator and interpreter of Aristotle attempted to resolve the ambiguity in Aristotle’s text.26 But the problem of the object of metaphysics was not (merely) the problem of how to resolve an apparent tension found within a text by Aristotle. While the issue might have originated from a difficulty in understanding Aristotle’s work, it also expresses an independent and important question regarding what the object of metaphysics is supposed to be: whether metaphysics should be a study of what is common to all (possible) entities (“primary philosophy”), or whether it should be a study of a particular genus of entities, e.g., of those entities which are in some way “beyond” physics (“meta-physics” or “theology”). According to J.F. Mora, once Scholastic philosophers (such as Francisco Suárez and Pedro Fonseca) moved away from primarily writing commentaries on Aristotle, and started to develop general systems of metaphysics, the question of the object of metaphysics gained importance independently of the questions regarding the correct interpretation of Aristotle. It is in this more general form that the question

24  Dino Jakušić of the object of metaphysics, and hence of the relation between “metaphysics” and “primary philosophy”, moves from the Mediaeval period into Early Modern philosophy, and after 1635, the attempts to resolve this issue take the form of trying to classify philosophy into different branches and subbranches. It is from these attempts that ontology as a discipline emerges.27 Since ontology as a philosophical discipline arises from the Aristotelian– Scholastic problem regarding the object of metaphysics, the discipline was mostly associated with the philosophers connected to the Scholastic tradition.28 In his 1726 reflections on his own writings, Wolff reports that as a student he was taught to treat ontology with contempt due to its association with Scholasticism.29 This could explain why the term does not feature in the works of philosophers who saw themselves as representing the “new”, post-Scholastic philosophy, as Descartes, for instance, did. These “new” philosophers, however, still contributed, if only indirectly, to the question of the object of metaphysics. For example, in a letter to Mersenne dated 11 November 1640, Descartes expresses his views on the relation between metaphysics and primary philosophy: ‘I have sent my Metaphysics to M. de Zuylichem…. I have not put any title on it, but it seems to me that the most suitable would be René Descartes’ Meditations on First Philosophy because I do not confine my discussion to God and the soul, but deal in general with all the first things to be discovered by philosophizing’ (Descartes 1991: 157/AT III 235. Emphasis mine).30 Further comments on the relation between the two can be found in his Principles of Philosophy: The first part of philosophy is metaphysics, which contains the principles of knowledge, including the explanation of the principal attributes of God, the non-material nature of our souls and all the clear and distinct notions which are in us. The second part is physics, where, after discovering the true principles of material things, we examine the general composition of the entire universe…. The first [part of the Principles] contains the principles of knowledge, i.e., what may be called ‘first philosophy’ or ‘metaphysics’; so in order to gain sound understanding of this part it is appropriate to read first of all the Meditations which I wrote on the same subject. (Descartes 1985: 186–187/AT IXB 13-17) It seems that, for Descartes, primary philosophy and metaphysics are used more or less synonymously, with the former term emphasising, rather than restricting, the focus on the first principles of cognition. There also seems to be no space here for the science of ontology à la Clauberg or Wolff, i.e., a discipline expressly formulated as an investigation into an entity qua entity. Instead, Descartes’ primary philosophy more resembles meta-physics,

Wolff on Ontology as Primary Philosophy 25 i.e., a science of that which is prior and non-reducible to physics, e.g., God, human souls, and our clear and distinct ideas.31 Leibniz also conceives of primary philosophy as closely related to metaphysics, though he puts more emphasis on its architectonic function. In On the Correction of Primary Philosophy, and the Concept of Substance from 1694, he writes: I notice that most people who delight in the doctrines of the mathematicians abhor those of the metaphysicians; in the ones they see light, in the other darkness…. This problem has spread to other disciplines, which are subordinate to this primary and architectonic science…. That is why we should not be surprised that this principal science [scientiam illam principem], which rises to the name of Primary Philosophy [quæ Primæ Philosophiæ nomine venit], and which Aristotle called the ‘sought after’, is still to be found. (Leibniz 1998/1694: 140/110. Translation modified) Leibniz thus also identifies metaphysics with primary philosophy, though he claims that a metaphysics worthy of the name “primary philosophy” has not yet been developed. However, Leibniz’ conception of primary philosophy does not resemble the shape of Wolff’s ontology. For Leibniz, the darkness of metaphysics arose due to its deficient conceptual clarity, brought about by the lack of a sufficiently rigorous method (a sentiment that Wolff will share). But instead of solving the problem by focusing primary philosophy on an entity qua entity as its proper object, Leibniz suggests developing this new metaphysics through the investigation of the concept of substance, from which ‘the most important truths about God, the soul, and the nature of body’ will follow (Leibniz 1998/1694: 141/111). 1.4  Primary Philosophy as Ontology The examples above show that the terms “primary philosophy” and “metaphysics” are ubiquitous in the “new” philosophy that was inaugurated by Descartes, but that the term “ontology” and the investigation into an entity qua entity are not. According to Wolff, this is one of the shortcomings of (post-)Cartesian philosophy. While he acknowledges that Descartes and Leibniz have significantly improved the primary philosophy of the past, their efforts failed to develop a properly scientific metaphysics.32 In line with earlier philosophers such as Clauberg and Micraelius, Wolff holds that one needs to divide metaphysics (and philosophy in general) into several disciplines on the basis of the object treated by each discipline. The goal behind this approach was to avoid the ambiguities and conflations that allegedly plagued philosophy during the Scholastic period. If one

26  Dino Jakušić could break philosophy up into disciplines on the basis of the object they treat, and nothing else, this would result in a more precise and rationally organised system of science.33 But, in addition to establishing specialised disciplines within philosophy on the basis of their objects – as Descartes did by presenting philosophy as a tree34 – Wolff, and the ontologists preceding him, held that we need to establish a discipline whose object is that which is common and presupposed by any other mode of inquiry. And since every scientific discipline focuses on a certain, specific kind of entity, what will unite them all is a science looking into an entity qua entity.35 With this in mind, Wolff divides metaphysics in the following way: “primary philosophy” designates the most abstract and universal branch of metaphysics. What makes this part of metaphysics the most abstract and universal is its object: an entity in general, i.e., what is common to all entities. Since it is the science of an entity in general, it is named “ontology”. Ontology is followed by more specialised metaphysical disciplines, also named according to their objects. Specifically, pneumatics (encompassing both natural theology and psychology) is the branch of metaphysics that deals with God and souls independently of Revelation, while cosmology deals with the world in general.36 Without all of these, Wolff claims, metaphysics is incomplete.37 Wolff was aware of the controversies to which his identification of primary philosophy with ontology might give rise. He complains that ‘hardly any other name is more despised today than that of ontology’ (WO §1*), and that students of philosophy are still taught to avoid it.38 While he does not reference writers who explicitly disparaged ontology, it is clear that he expects his readers to associate “ontology”, and the investigation into an entity qua entity, with Scholasticism. Wolff tells us that while ‘ontology has been pursued with the greatest zeal and very highly exalted by the scholastics; it has been despised by Descartes and the moderns’ (Aus. Nach.: §69. Cf. WO Præfatio: 7; §1*),39 even gaining the reputation of being merely a lexicon of philosophical, or even of “barbaric”, terms – technical terms from foreign languages that are probably best abandoned.40 Since Scholastic philosophy, in its investigation of an entity qua entity, failed to provide clear and distinct terms, Wolff’s contemporaries, according to him, associated the obscurity of Scholastic terminology with the ontological investigations themselves. Hence, after Descartes’ attempt to expel confused and obscure concepts from philosophy, the investigation into an entity qua entity was itself abandoned.41 In fact, in the first two editions of his German Logic, Wolff himself avoids the term “ontology”. The term is also absent from every edition of his German Metaphysics. It appears for the first time in the third edition of the German Logic in 1722, when Wolff had already started working on his Latin opus. To the originally conceived four parts of philosophy (logic,

Wolff on Ontology as Primary Philosophy 27 natural theology, physics, and ethics), he adds a fifth part called ontology or Grundwissenschaft (“fundamental science”). He later explains that he chose to call it Grundwissenschaft because the more literal Dingen-Lehre (“Doctrine of things”) sounded silly, and also because he became sensitive to the need for a science that explains the first principles of cognition of all things in general.42 He states, as I have already acknowledged, that the attempts of his predecessors to formulate a metaphysics without ontology were motivated by impatience and prejudice against the Scholastics. As a result, the moderns threw out the baby with the bathwater: their resistance to Scholastic obscurantism blinded them to the necessity of the ontological project. Therefore, while the Scholastics, according to Wolff, included ontology, psychology, and theology as parts of metaphysics, Descartes retained only the latter two, out of his mistaken belief that the ontology of the Scholastics contains nothing but obscure and indistinct concepts.43 But even if Wolff is correct in his claim that Scholastic ontology developed many philosophically useful and important concepts,44 which Descartes, according to him, unjustly omitted, this does not explain what exactly is lacking in the Cartesian system of metaphysics. Unfortunately, Wolff does not provide a clear and detailed explanation of what goes wrong in the Cartesian system due to its exclusion of ontology. He simply states that it is clear that theology and psychology cannot be ‘thoroughly elaborated’ without ontology, and accepts as a given that Descartes and his followers did not succeed in their attempts to improve metaphysics.45 However, it might be possible to glean what Wolff had in mind from some of his other references to Descartes. For example, Wolff seems to agree with Descartes’ claim that the Cogito proposition is certain in the face of radical doubt, but argues that this certainty presupposes the principle of contradiction.46 Descartes’ system, therefore, relies on the principle of contradiction, without clearly and distinctly establishing the validity of the principle, thereby violating its own method. Similarly, Wolff argues that the first axiom of Descartes’ argument for the existence of God from the second set of Replies to the objections to the Meditations presupposes the principle of sufficient reason without clearly and distinctly formulating this principle beforehand.47 According to Wolff, every truth one discovers is ultimately grounded on ontological principles,48 and thus the attempt to ground the system of philosophy on anything else, such as certainty of one’s own existence or on God’s benevolence, will result in some fundamental philosophical principles being assumed, rather than firmly established. And this, in short, is the overall goal of Wolff’s entire project: to show that primary philosophy should properly be conceived as ontology, as the science of an entity qua entity, i.e., the science of the most general principles and predicates, which are presupposed by every metaphysical and scientific discipline. Hence, if one attempts to start their system of

28  Dino Jakušić metaphysics by investigating objects less general than an entity qua entity (such as souls or God), they will inevitably end up presupposing certain ontological principles. In summary, for Wolff, the term metaphysics becomes primarily a classificatory term for the sciences that deal with entities, as well as with the principles of their cognition, in the most general ways possible. In metaphysics, one can investigate what is common to all entities as such (ontology or primary philosophy), what is common to all immaterial entities (pneumatics), and what is common to all material entities (cosmology). This indirectly solves the problem of the object of metaphysics.49 Aristotle’s definition of primary philosophy from Metaphysics 4 is identified with ontology, while the definition from Metaphysics 6, or Aristotle’s theology, does not apply to primary philosophy, but is subsumed under more specific philosophical disciplines. Furthermore, with the addition of general cosmology within metaphysics, Wolff moves away from the idea that metaphysics should be understood as meta-physics, i.e., an investigation into non-physical objects and entities, as Descartes and Aquinas understood it.50 Now that we know that it is ontology, the science of an entity qua entity, that takes on the mantle of primary philosophy, the question remains of what it is that makes ontology “primary”. 1.5  Ontology as Primary Philosophy We have already seen three ways in which metaphysics was considered to be primary philosophy. For Descartes, it was because it investigated the first principles of cognition. For Leibniz, it provides a systematic unity to all other sciences – a true scientia architectonica. For Aristotle (in Metaphysics 4), the primacy of philosophy lies in its universal application, its treatment of what is common to all entities. All of these senses of primacy can be found in Wolff’s conception of ontology. For the purposes of clarity, I will differentiate four interrelated ways in which ontology is the primary scientific discipline for Wolff: primacy in cognition, demonstrative primacy, architectonic primacy, and primacy in the order of being. 1.5.1  Primacy in Cognition

By primacy in cognition I mean that ontology contains the fundamental principles of cognition, not that it is a discipline that should be studied first.51 Wolff defines ontology in this way in several places, stating that this is the reason why its German name should be Grundwissenschaft – since the ground or reason [Grund/ratio] is that from which we understand [verstehen/intellegere] why something is.52 Similarly to Descartes’ understanding of primary philosophy, ontology provides the most fundamental

Wolff on Ontology as Primary Philosophy 29 concepts and principles involved in cognition. These fundamental concepts correspond to the ontological concepts that were established in Wolff’s Ontologia, such as finitude, extension, space, time, quantity, quality, identity, similarity, possibility, impossibility, cause, and effect. We cannot know what something is, for example, if we do not know whether it is possible or not. We cannot know whether the soul is distinct from the body, unless we know whether the soul is extended or non-extended. We cannot know how any two things are related if we do not know whether they are identical or not, or whether they stand in a causal relation or not. Since these concepts are, according to Wolff, derived a priori from the notion of an entity in general, and since an entity is whatever is possible, they are supposed to feature in all of our cognitions, regardless of how particular our cognition might be.53 Beyond these, the fundamental principles of cognition that are presented in the Ontologia are the principle of contradiction and the principle of sufficient reason. The principle of contradiction establishes the very field of possibility, since for something to be possible, and hence to be an entity, it is sufficient that it is not self-contradictory.54 The role of the principle of sufficient reason as a fundamental principle of cognition becomes clearer once we understand Wolff’s general theory of cognition. In short, Wolff divides cognition into historical, philosophical, and mathematical kinds.55 The first of these is either autonomous or achieved through testimony, in which case Wolff calls it belief or faith.56 Autonomous historical cognition provides knowledge of facts, either through sensory experience or through introspection and is, as such, indubitable.57 We speak of philosophical cognition in regard to the reasons why a fact obtains. Mathematical cognition occurs when philosophical cognition is accompanied by the explication of the quantitative relation between the explanans and the explanandum. For example, by observation, I might learn that water boils when sufficiently heated. This amounts to historical cognition. If I discover that this is due to water molecules being transferred enough energy to evaporate, I undertake a philosophical cognition of the previously known fact: I know why the water boils. If I further establish that this happens at 100 °C under 1014 hPa, I complement my philosophical cognition with the mathematical one, since I know the quantitative relations inherent in my explanation of why the water boils. From this we can see that Wolff’s very conception of philosophical cognition already refers to sufficient reason.58 It is not controversial to say that we cannot cognise anything self-contradictory, since nothing can be self-contradictory,59 or that if we were to unknowingly affirm two mutually contradictory judgements, we could not say that we fully know either of them.60 What is perhaps more controversial is that Wolff thinks that it is necessary that we rely on the principle of sufficient reason to expand

30  Dino Jakušić our cognition beyond what is merely not impossible, especially if we want to discover what is actual. He points out that many claims that we accept as true depend on it, such as Euclid’s first and fifth axiom,61 Newton’s concept of inertia, and ‘innumerable philosophical and theological conclusions’ put forward by Leibniz. Moreover, the real world would itself collapse into a dream-world [mundus fabulosus] in which we would not be able to cognise anything.62 There are two more principles of cognition that Wolff establishes in the first part of his Ontologia, specifically the law of excluded middle (WO §52) and the ‘principle of certainty’.63 According to Wolff, these principles can be a priori derived from the principle of contradiction, showing that the principle of contradiction is “the source of all certainty”. For this reason, however, they should not be considered first principles. They are merely corollaries of the principle of contradiction.64 How do we establish the two fundamental principles? Perhaps anticlimactically, Wolff claims that we ‘experience them as the nature of our mind’ (WO §§27, 74). By this he seems to simply mean that the truth of these principles is self-evident on reflection. To deny them would be like denying that we see the sun when apprehending it.65 The fact that they obtain is presented as a direct, indubitable intuition of a fact – an example of historical cognition.66 Having established these two (or four) principles, the first part of the Ontologia ends, and the remaining concepts and principles are, according to Wolff, obtained a priori through the fundamental ones.67 In summary, for Wolff, ontology exemplifies primacy in cognition in the sense that the principles and concepts developed there will need to be presupposed by every other discipline, philosophical or otherwise, in order for it to be scientific. If a discipline wants to be properly scientific, its doctrines cannot be mutually or internally contradictory and its explanations need to express sufficient reason for why what it is concerned with is or occurs. Moreover, there could be no scientific explanations regarding relations between two things, if we could not say whether these things are similar or dissimilar, identical or different, or causes or effects of each other. Ontology has primacy in cognition since it will ground and explain these concepts and principles without which, according to Wolff, nothing can be known or explained in any scientific discipline. 1.5.2  Demonstrative Primacy

For Wolff, “science” provides absolute certainty regarding the truth or falsity of propositions.68 There are three ways in which such certainty can be reached. One way is through direct experience in historical cognition. This is the way in which the principles of contradiction and sufficient reason

Wolff on Ontology as Primary Philosophy 31 were established. Another way is through definitions, which can be understood analogously to Kant’s analytic a priori judgements, i.e., by showing that the predicate of a judgement is contained in its subject. Finally, certainty can be obtained through demonstration. In fact, Wolff defines science as such as the habit or capacity to demonstrate propositions.69 In order to understand Wolff’s notion of demonstration, it will be useful to first explicate his distinction between two inferential processes, namely proof and demonstration.70 Simply put, a proof is a process in which the truth of a proposition is established by means of a valid syllogistic inference. In case the proposition we want to affirm is reached through a single syllogism, we speak of a simple proof [probatio simplex]. In case of a composite proof [probatio composita], we need to construct a chain of syllogistic inferences, ultimately resulting in the proposition that we want to affirm. Since this is all that is required for something to count as a proof, there is nothing preventing us from developing “dummy” proofs from false premises (e.g., for the purposes of logical instruction), or from drawing inferences from mere hypotheses.71 Demonstration is, first and foremost, a species of proof, meaning it is a process of establishing propositions through valid syllogistic inference. However, unlike mere proof, a demonstration cannot use false or even hypothetical propositions for its premises. If a proof is to count as a demonstration, its premises must consist of either propositions the truth of which is indubitable or conclusions of previous demonstrations. Demonstration, therefore, is a chain of syllogistically valid inferences (ultimately) grounded upon definitions, indubitable experiences, or axioms. The model that Wolff has in mind when developing the concepts of “demonstration” and “demonstrative certainty” is the one found in Euclid’s Elements. Euclid demonstrates a vast number of propositions from only a small number of definitions, postulates, and axioms. Wolff is very explicit in this attempt to develop philosophy more geometrico in the style of Euclid, often comparing his own approach to Euclid’s.72 According to Wolff, the benefits of doing so should be obvious to anyone who realises that there is no difference between the philosophical and the mathematical method abstractly conceived: they both are derived from the concept of certainty.73 Mathematicians were just more successful in utilising this ‘true’ method than anyone else in other disciplines, or at least by anyone else prior to him (DL Vorrede: 4).74 Hence, in a Euclidian manner, Wolff will sometimes refer to ontological principles as “axioms” [axiomata/Grundsätze], i.e., indemonstrable theoretical propositions, and “theorems” [theoremata/Lehrsätze], or demonstrable theoretical propositions.75 Coupled with the demonstrative method, these principles allow us to establish other ontological concepts and principles that are not immediate and self-evident.76 As for Wolff the

32  Dino Jakušić result of a demonstration is as certain as the premises constituting it, the remaining content of his ontology, since it can be demonstrated from the principles of contradiction and sufficient reason, is equally certain.77 This shows us how and why it is that Wolffian ontology uses the demonstrative method in order to expand beyond the principles of contradiction and sufficient reason. However, it does not show why ontology can be ascribed “demonstrative primacy” in relation to all other scientific disciplines. And Wolff is quite clear that it can, since ‘without ontology, philosophy cannot be developed according to the demonstrative method’ (DP §73*; WO §§4, 9). Why is that so? The answer is twofold. Firstly, in the previous section, I have explained how for Wolff the principle of contradiction is the most general philosophical principle. It is the first, indemonstrable, and universal principle of cognition. Whatever violates it is impossible and hence cannot be cognised or even be an entity. Moreover, Wolff argues that any logical proof depends on the principle of contradiction,78 and since a demonstration is a species of proof, we cannot develop the notion of demonstration without the principle of contradiction. Furthermore, since the principle of contradiction is, according to Wolff, an ontological principle, i.e., one that is applicable to any possible entity, it makes sense that it will be established and discussed within ontology and applied to all other sciences from there. Therefore, one way in which ontology exemplifies demonstrative primacy is based upon the claim that since all demonstration is grounded upon the principle of contradiction, which is itself an ontological principle, we need to develop an ontology in order to make it possible for any science to use demonstrative inference.79 The other way in which demonstrative primacy is attributed to ontology can be gleaned from Wolff’s claim that without ontology, the results of no other science can reach demonstrative certainty.80 As we can see from Wolff’s definition of demonstration, while proofs can always be formulated by way of valid syllogistic inference, demonstrations require that the premises of proofs are grounded in axiomatic propositions. According to Wolff, it is the ontological concepts and principles that will in this way ground the concepts and principles of all other sciences. According to the primacy in cognition of ontology, discussed in the previous section, to know that one billiard ball caused the other one to move upon impact, we need to know what it means for something to be a cause. Demonstrative primacy, however, spells out a more specific grounding relationship between ontological concepts and their application in other sciences. In terms of primacy in cognition, we must understand ontological concepts in their abstract form in order to properly understand more concrete propositions that utilise them. According to demonstrative primacy, ontological concepts will serve as axioms or theorems for demonstrations in other

Wolff on Ontology as Primary Philosophy 33 sciences. This means that, at least in principle, we should be able to formulate a polysyllogistic chain starting with the ontological definition of ‘cause’ and ending in a proposition that “billiard ball A caused billiard ball B to move”. As Wolff puts it: ‘if we continue proving the principles [of sciences] until they become manifest a priori, we always end up with the principles of primary philosophy’ (WO §§8–9*). In summary, ontology does not possess demonstrative primacy in the sense that ontological investigations have to be undertaken before other scientific disciplines can begin their investigation. Rather, ontology possesses demonstrative primacy because (a) in ontology the principle of contradiction is established, without which no demonstration would be possible; (b) ontology attempts to formulate a system of principles and concepts that serve as fundamental axioms and theorems to which demonstrations of more concrete scientific disciplines should be ‘brought back’ [reducere] (Not. Dir. §3). While without ontology other sciences would be able to undertake empirical investigations, formulate hypotheses, and develop proofs (assuming they presuppose the principle of contradiction), it would be impossible for them to bring their proofs to the level of demonstrative certainty that Wolff requires from any science. 1.5.3  Architectonic Primacy

The idea, characteristic of Classical German Philosophy more broadly, that any science worthy of the name must take the form of a system is already present in Wolff. Hence, if philosophy is to acquire the status of a science, it must be presented in a systematic order.81 In fact, this is obvious from the example of mathematics since, according to Wolff, all great mathematicians have always approached their discipline systematically, which is the reason for its unparalleled success.82 As mentioned earlier, in his text On the Correction of Primary Philosophy, which Wolff references on several occasion,83 Leibniz stated that the properly architectonic science, to which all other sciences are subordinated, has yet to be developed. Wolff believed himself to be the first to have developed it, stating that it is his ontology that ‘deserves the name of the architectonic science’, since he is the first one to present philosophy in a truly systematic way (Not. Dir. §§1, 11).84 What, then, is a system? Similarly to how Kant will later see it, a system for Wolff cannot be a mere aggregate of disciplines, or a collection of ‘scattered truths’ (Int. Sys. §11).85 This inadequate conception of a “system” was prevalent in Scholasticism: ‘everything concerning the same subject is put together in one place, without any reason being found how the cognition of one should depend on the cognition of another’ (LL §829).86 The real system, on the other hand, is ‘a collection of truths connected to each

34  Dino Jakušić other, and to their principles’(LL §889). What does this mean? Wolff explains that two propositions are “connected” when ‘one is demonstrated through the other as through a principle’ (Int. Sys. §2). This means that a proposition A and B are connected, for example, if A can serve as a premise in a syllogism which infers B as a conclusion.87 This presents us with one kind of a system, specifically a ‘system of doctrines’ (Int. Sys. §3; LL §889), examples of which are Euclid’s Elements, as well as Wolff’s Latin Logic and Ontology. Science as a whole, then, cannot simply be a collection of disparate systems of doctrines. According to Wolff, ‘the supreme law of the philosophical method is that those things must come first through which later things are understood and established’ (DP §133). This law is to be observed on every level of scientific inquiry. In the very notion of “demonstration”, this means that axiomatic propositions, as well as propositions logically derived from them, serve as premises for further demonstrations. Within a system of doctrines, this means that conclusions of preceding syllogisms will serve as premises in the following syllogisms. This same principle of ordering is also to be observed between scientific disciplines.88 The system of sciences must be so arranged that those disciplines that provide principles for other disciplines ‘come first’ (DP §87). Wolff explains how metaphysics exemplifies this hierarchy of disciplines. In metaphysics, he states, ontology comes first, general cosmology second, psychology third, and natural theology last. There are two ways in which we should not understand this order. Firstly, the architectonic primacy of ontology does not imply that it is necessary that we learn or undertake ontology prior to other disciplines. For example, Wolff tells us that in studying and doing philosophy, the study of logic should take first place, since logic guides our thinking in any field of rational engagement. However, architectonically, according to the demonstrative order, the science of logic is established later in the system and not as a part of metaphysics, since, according to Wolff, it requires principles taken from psychology in order for its propositions to be fully grounded.89 Hence, the architectonic primacy of ontology, as in the case of its primacy in cognition, does not imply primacy in the order of learning, or in the order of scientific research. Another thing that the architectonic primacy of ontology does not imply is the claim that all propositions of subsequent disciplines need to be syllogistically derived from ontological principles. Wolff does not advocate the idea that all the content of subsequent disciplines can ultimately be analytically inferred from the principle of contradiction, or from the totality of ontological predicates. Instead, the relation between disciplines is such that more specific disciplines must “borrow principles” from the more general ones. One example Wolff gives is between cosmology and

Wolff on Ontology as Primary Philosophy 35 ontology.90 General cosmology attempts to give the most general account of the world, bodies, and the elements of bodies. In order to differentiate between bodies and their elements, cosmology understands the former as complex entities and the latter as simple entities. But while the concept of a “body” is not an ontological concept, the division of entities into simple and complex is an ontological one. We are free to use concepts of simplicity and complexity in cosmology before we have demonstratively established these concepts in ontology. We might even be very successful in our investigations by doing so. However, since bodies are complex entities, our discoveries regarding the nature and behaviour of bodies cannot be demonstratively certain, until ontology has established a clear and distinct definition of simplicity and complexity. Similar situation obtains in natural theology, which needs to “borrow” principles not just from ontology (since God is the most perfect entity), but also from cosmology (since God is supposed to be the author of the world), and psychology (since God is not a body, but a spirit lacking limitations).91 The system of metaphysics, consisting of ontology, cosmology, psychology, and theology, due to the generality of its subject matter, “lends principles” to any other science that deals with bodies, agents, or divine things. This effectively means that ontology architectonically grounds any possible science, since every science treats of some region falling under these genera. Since ontology is an investigation into an entity qua entity, all metaphysical disciplines need to “borrow” some principles from it, and all other scientific disciplines need to do the same, either indirectly, through borrowing principles from other metaphysical disciplines, or directly, by utilising ontological principles (e.g., if we ask whether the moon causes the tide to rise, our question depends on our understanding of the concept “cause”, which is an ontological concept). This is even the case in geometry since, as Wolff claims, what Euclid treated as axioms in his Elements are in fact ontological theorems that he simply adopted without proof, and the truth of which Wolff claims to have demonstrated in his Ontologia from the notion of quantity. Hence even Euclidean geometry, which serves as a model for Wolff’s conception of ontology, ultimately ‘owes all its certainty to Primary Philosophy, from which it takes its first principles’ (Not. Dir. §6; Int. Sys. §11; WO [2005], Præfatio 8–9, 15). 1.5.4  Primacy in the Order of Being

According to Wolff, ontology deserves the name “primary philosophy” since it ‘teaches the first principles and first concepts used in reasoning [ratiocinando]’, but primary philosophy equally needs to be called ontology, since its ability to provide us with the first principles of cognition arises from its being the investigation into the concept of an entity in general

36  Dino Jakušić (WO §1*). I have so far explained Wolff’s ontology in its role as primary philosophy, i.e., as an attempt to ground all cognition on propositions we know with certainty to be true, to formulate a method for expanding our knowledge from these cognitions, and to harmonise the collection of true propositions into a system. But this role, for Wolff, overlaps with the role of ontology to discover the most general properties of all possible entities. The distinction between the two functions is nominal, since the investigation into the first principle of our cognition is, at the same time, the investigation into the most general nature of things: ‘Primary philosophy, as one is wont to call it, or fundamental science [Grundwissenschaft], as I am wont to call it, deals with the first general concepts that belong to all things’ (Anm. §56).92 Therefore, if ontology, i.e., an investigation into properties of all possible entities, is itself possible, then true ontological propositions enable us to understand the real relations that hold between entities as such and particular species of entities.93 For example, earlier I explicated how Wolff sees the principles of contradiction and of sufficient reason as grounds of our cognition. But since these are also supposed to be ontological principles, this means that any possible entity will not contain contradictory properties or lack a sufficient reason for why it is. The same is supposed to hold for other ontological predicates and theorems. If in ontology we can demonstrate that any entity is either simple or complex, this does not only mean that this is how we must understand entities discussed in cosmology or theology, but also that every entity will be either simple or complex.94 Hence, the primacy in the order of being refers to the claim that ontological principles, definitions, and theorems provide us with the fundamental structure of any possible entity.95 So is ontology itself possible? Wolff believes that it is, although he gives no straightforward demonstration of its possibility, i.e., of the capacity of human cognition to access the structure of things in themselves. Instead, one can find a series of separate arguments in favour of its possibility. One kind of argument draws an analogy with Euclidean geometry. Firstly, Euclidean geometry shows us that abstract a priori demonstrative systems, applicable to empirical reality, are possible.96 Euclid’s system is built entirely a priori from definitions, axioms, and postulates, yet also tracks the ways in which existing things are related. One example Wolff gives is of builders building a door. They will use a piece of string to measure an existing door in order to make a new, matching door. They know that if the new door is the same height as the length of a string, it will also be the same height as the old door.97 What they might not know is that in doing so, they are applying Euclid’s first axiom: two things equal to a third are equal to each other. Wolff states that we can confidently assume that mathematical propositions tell us something about the way existing

Wolff on Ontology as Primary Philosophy 37 things are, since this is attested to by millennia of practical application of abstract mathematical propositions. The truth of Euclidian geometry has been ‘verified’, i.e., a priori demonstrations have been confirmed by observation or experimentation (Int. Sys. §12).98 Therefore, since Wolff believes that Euclid’s axioms are simply ontological theorems that are held to be true without demonstration, an abstract a priori system of ontology is likely to be possible. The truth and practical utility of mathematics, then, provides Wolff with one reason to believe that ontology is possible. In addition, since ontology grounds mathematics, all those ontological theorems that are, according to Wolff, presupposed by Euclid’s axioms, and which in ontology are derived a priori, can also be verified through mathematics. Since ontology is a science of any possible entity, its propositions can additionally be verified, either directly through “experimental ontology” or indirectly through experimental physics, theology, or any other discipline.99 The possibility of ontology, if it is not demonstrable a priori, will eventually be verified a posteriori. The most direct argument for the possibility of ontology, however, can be extracted from Wolff’s distinction between natural and artificial ontology.100 While the ontological theorems are ‘derived entirely a priori [prorsus a priori deducantur]’ (Int. Sys. §11), the content that is supposed to be ordered in this demonstrative, systematic manner, is taken from “natural ontology” – our ordinary everyday use of mental faculties. Just as, according to Wolff, “artificial logic” is an a priori formalised and systematised explication of our natural powers of reasoning,101 artificial ontology, the discipline I have been discussing here, is also a formalised and systematised explication of the most abstract terms we use in everyday life when expressing universal judgements about entities. The same is the case for Euclidean geometry, since its axioms are simply abstractions from everyday experience. When we compare objects by their size, we arrive at abstract concepts of “the same”, “bigger”, “smaller”, etc. allowing Euclid to formulate axioms such as “two things equal to a third are equal to each other” or “the whole is greater than its part”. Since the building blocks of ontology and geometry are universalisations of the particulars we observe, it should be possible to develop an a priori demonstrative science regarding the nature of merely possible, as well as actual things.102 Possible, that is, provided there is a rigorous and systematic enough treatment of the universals and the relations between them. Hence, unlike the empiricists, who hold that the abstraction from everyday experiences allows us access only to general ideas of observed particulars that need not necessarily correspond to the structure of things themselves, Wolff, guided by his understanding of Euclidean geometry, holds that abstraction allows us to “see” the universal in particular, ordinary experiences (WO §19).

38  Dino Jakušić 1.6  Conclusion: Ontology and Properly Scientific Metaphysics Wolff’s attempt to formulate a complete, scientific system of philosophy brings together various ideas from the Scholastic and Early Modern period and points towards a variety of issues that will characterise Classical German Philosophy. Metaphysics can, and should, amount to a science, but for that it needs to be formulated in a new way. Primary philosophy, a discipline that was sought after since Aristotle, is identified with ontology, a discipline praised by the Scholastics, but despised by the Moderns. Metaphysics is not understood as a science of those things that are “beyond” physics. Instead, it designates the most abstract and general investigation into entities either absolutely (ontology) or under a certain specific condition (pneumatics and cosmology). As an investigation into an entity qua entity, ontology is worthy of the name “primary philosophy”, since it demonstrates fundamental principles required to ground cognition. It also develops a method of demonstration and harmonises all sciences into a system. It can do so since it is an investigation into the fundamental structure of all that can possibly be. As such, it is able to formulate a unified, scientific procedure for treating any possible scientific subject matter, and also to bring any possible discipline in a higher systematic unity with all other disciplines. Scientific metaphysics, since it will allow us to reach certainty in any field of discourse, is useful for any mode of human cognitive and practical engagement. As such, it benefits the ‘higher faculties’ of medicine, law, and theology; it can guide our personal ethical decisions; and it is useful even to ‘the lowest manual arts’ such as wood-cutting (DP §§39, 109, 139*). Today, we mostly remember Wolff through Kant’s criticism of “dogmatic” philosophy – of the overreach Wolff allegedly committed by claiming to be able to provide us with a metaphysics that is unbounded by any cognitive limits. Kantian or Lockean humility, i.e., the restriction of metaphysical ambitions to the scope of the experiential, might seems to be a more sober and rational approach. Interestingly, already in his German Logic, Wolff considers this type of objection. Some may suggest, he says, that it is advisable not to understand philosophy as the science of all that is possible, but to consider ‘a more modest definition’. He replies, however, that it is better to attempt to define philosophy from the standpoints of ‘its highest possible perfection’, rather than to restrict it from the onset to some assumed limits of human cognition (DL Vorbericht §§8–9). We have tried to restrict philosophy in a such a way before, and it led us to believe that Aristotle gave the final word on what constitutes the boundaries of human understanding. We would be much better off if we assumed much greater or even no limits to our cognition. This is, for Wolff, what will lead to true intellectual humility. Under the assumption that our cognition has no bounds, we will always feel that there is more to know, and never be under

Wolff on Ontology as Primary Philosophy 39 the illusion that we have learned everything there is to learn. The result of limiting cognition is not humility, but complacency, making us believe that the answers to everything we can know are just around the corner. Wolff believed that he presented all of his proofs with the highest degree of demonstrative rigour, and that he achieved the highest degree of certainty that his abilities allowed him. While he acknowledged gaps and potential weaknesses in his demonstrations, he believed that his system would ultimately be justified through the vast utility and improvement it would bring to philosophy, the disciplines of the higher faculties, the sciences in general, as well as to the moral and practical education of ordinary people. While this practical function of the system was never achieved, Wolff’s demand for rigour in demonstration, insistence on relating philosophy and mathematics, and the idea of science as a system of interrelated disciplines crowned by ontology provided an influential model of scientific metaphysics for his contemporaries and for subsequent philosophers. By the time of Wolff’s death in 1754, his German Logic had been published in 14 editions, and his German Metaphysics in 12 editions. His views on metaphysics were adopted and developed by subsequent philosophers such as Alexander Gottlieb Baumgarten, Martin Knutzen, and Johann August Eberhard – all important influences on Kant – as well as J.F. Flatt, who taught Schelling, Hegel, and Hölderlin at the Tübinger Stift.103 Ultimately, even Kant and Hegel felt the need to refer to Wolff in their own attempts at developing scientific metaphysics, despite the fact that their approaches at grounding metaphysics as a science involved rejecting Wolffianism. Kant presents Wolffian philosophy as the prime example of dogmatism which is to be replaced by his own criticism in order for it to be possible to develop scientific metaphysics, but also claims that Wolff was the first to provide an example of how metaphysics can be developed as a science.104 Hegel, similarly, considers Wolff’s system as the paragon case of the historically “outdated” philosophy of the understanding, but also presents his own Objective Logic as a discipline that fulfils the role of Wolff’s ontology, i.e., of the science of the nature of entity in general.105 In conclusion, while Wolff’s philosophy today remains largely unappreciated, it had an immense influence on Classical German Philosophy, especially with regard to the project of developing scientific metaphysics. As Étienne Gilson observed: ‘throughout the schools of Europe, and especially in Germany … [t]o innumerable professors and students of philosophy, metaphysics was Wolff and what Wolff had said was metaphysics’ (Gilson 1952: 119). Notes 1 This accusation was already present in 1723. Some suggest it was formulated by Wolff’s Pietist opponent Joachim Lange. See Int. Sys. [2019]: 10.

40  Dino Jakušić The abbreviations below will be used in referencing Wolff’s texts. When possible, I only specify the section of the work from which the reference is taken. If a number is followed by an asterisk (*), it designates Wolff’s notes to the section. If a page number is provided in square brackets, this means that the page lacks official pagination. Beyond Wolff’s text, I provide a reference to a translation (in German or English) when one was consulted. All translations into English are mine, except when an English version is listed in the Bibliography and I have not modified it. I have mostly tried to work with the first editions of texts, but that was not always possible. When necessary, I have specified which edition the abbreviation refers to by designating the year in square brackets. All the references to 18th-century editions of Wolff’s text are to the versions digitalised by Münchener Digitalisierungszentrum. Anm. Met. (Anmerkungen to German Metaphysics): Wolff 1724. Aus. Nach. (Ausführliche Nachricht): Wolff 1733. Int. Sys. (Intellectus systematicus): Wolff 1729a & 2019b. Not. Dir. (Notionibus directribus): Wolff 1729b. DL (Deutsche Logik): Wolff 1722. DM (Deutsche Metaphysik): Wolff 1720 & 2019a. DP (Discursus Præliminaris): Wolff 1735a & 1963. LL (Latin Logic): Wolff 1735b. WO (Wolff’s Ontologia): Wolff 1730 & 2005. 2 In 2017, Thomas Ahnert published a revised version of the J.H. Drake’s 1934 edition of Wolff’s Jus Gentium, and in 2019 Corey W. Dyck published selections from Wolff’s German Metaphysics and The Refutation of Spinoza. Most new research on Wolff investigates the affinity between his philosophy and Kant’s (de Boer 2011, 2020; Dyck 2011a, 2011b, 2014; Gava 2018; Lu-Adler 2019; Frketich 2019; McQuillan 2019). Debates regarding the nature of Wolff’s metaphysics itself have also started to take root (Anderson 2015 and Dunlop 2018). 3 Starting from his Vernünfftige Gedancken von den Kräfften des menschlichen Verstandes (referred to today as the German Logic) in 1712, and Vernünfftige Gedancken von Gott, der Welt und der Seele des Menschen (German Metaphysics) in 1720. 4 Although Wolff states that this project of self-Latinisation already started prior to the exile, motivated by the popularity of his German texts. See DM [1722], Vorrede zu der anderen Auflage, 5. 5 Due to this, Wolff’s work tends to be separated into the ‘German’ and ‘Latin’ opus. For Wolff’s bibliography, as well as a short biographical account, see Biller 2018. For more details regarding Wolff’s exile and what became known as Causa Wolffiana, see Marschke 2015 and Schönfeld 2002. Regarding Wolff and Thomasius’ influence on philosophical German, see Beck 1969: 243–275. 6 For Wolff’s claim to have invented cosmology and teleology, see DP §§78*, 85. For Wolff and “idealism”, see DM[1722] and Guyer and Horstmann 2022. For an account of the history of the division of metaphysics into Generalis and Specialis, see Vollrath 1962. 7  DL, Ch. 1: §23. This echoes Leibniz 1998: 140. 8 As Josep Olesti puts it, while the term is not invented by Wolff, it is only with him that ‘the concept was given full citizenship in the universe of philosophy’ (Olesti and Zimmer 2020: 81). 9 ‘Ontological inquiry is indeed more primordial, as over against the ontical inquiry of the positive sciences…. Basically, all ontology … remains blind

Wolff on Ontology as Primary Philosophy 41 and perverted from its own most aim, if it has not first adequately clarified the meaning of Being [den Sinn von Sein]’ (Heidegger 2001/1976: 31/11). For Heidegger, any questions concerning existence in the Quinean sense, or investigation into (kinds of) entities would count as an ontic, rather than ontological investigation: ‘The task of ontology is the explication of Being itself [die Explikation des Seins selbst] and the differentiation of Being from entities [Die Abhebung des Seins vom Seienden]’ (Ibid: 49/27. Translation modified). 10 From this we cannot see what ontology was supposed to be, or why Lorhard felt the need to coin this neologism, but see Øhrstrøm et al. 2005, 2007 for some discussion on this topic. 11 For discussions regarding Goclenius’ use of ontology, and its subsequent impact, see Mora 1963 and Bardout 2002. 12 See Bardout 2002. 13 The term does appear several times in Leibniz’ unpublished notes, but there is no developed treatment of it. The most detailed use of it can be found in his Introductio ad Encyclopaediam Arcanam where he states: ‘Ontology, or the science of Something and Nothing, Entity and Non-entity, Thing and the mode of a thing, Substance and Accident’ (Leibniz 1999: 527). 14 See, for example, Clauberg 1664: ch. 1, §§1–2: ‘There is a certain science which investigates an entity insomuch as it is an entity…. It is commonly [vulgὸ] called Metaphysics, but more properly Ontology’. 15 See de Boer 2001: 78–79, for further discussion of this point. 16 See DP §129*. 17 See WO §175. 18 In the Beweisgrund, Kant (1992b: 121 / BDG, AA 02: 76.05–14) complains about this definition, stating that it is ‘very indeterminate’. He prefers Baumgarten’s concept of ‘thoroughgoing internal determination’ [die durchgänginge innere Bestimmung]. 19 See DP §73*. 20 Werner Jaeger, however, argued that the term must predate Andronicus and that it could not have originated from a mere arbitrary ordering of texts. For this, see Owens 1963: 74. 21 All references to Aristotle are to the W.D. Ross collection (Aristotle 1995). The Greek is taken from the Loeb edition (Aristotle 1933). The phrase τὸ ὂν ᾗ ὂν will be more familiar to most readers as “being qua being”. However, I translate it, along with the equivalent Latin phrase ens quatenus ens est, as “an entity qua entity”. The decision to translate τὸ ὂν and ens (as well as German Seiendes) as “an entity” rather than as “being” is to avoid ambiguity between these terms and the terms είναι and esse (and Sein), which, if differentiated at all, tend to be translated into English as “Being”. 22 See Metaphysics 6: 1.1026a10. 23 See Metaphysics 6: 1.1026a19. One might notice that the definition of the object of primary philosophy from Metaphysics 4 sounds similar to earlier given definitions of ontology. This is not accidental, since the 17th- and 18thcentury ontological tradition supports taking this definition as postulating the proper object of metaphysics. Due to this, the two conceptions of the object of Aristotle’s primary philosophy are sometimes referred to as the “ontological” and the “theological” conception. I avoid this terminology, since the term “ontology” postdates Aristotle significantly. 24 Scotus, Quaestiones in Metaphysicam, apud. Aertsen 2005: 384. 25 See Bardout 2002: 131; Beck 1969: 73, 135ff.

42  Dino Jakušić 6 See Vollrath 1962: 267. 2 27 See Mora 1963: 39. 28 See Mora 1963: 38. 29 See Aus. Nach. §69. 30 I cite Descartes by providing both the English edition used and the reference to Oeuvres de Descartes edited by Charles Adam and Paul Tannery (AT). Roman numerals indicate the volume of AT referenced, and are followed by the Arabic numerals indicating the page. 31 This is more in line with Aristotle’s conception from Metaphysics 6. 32 See Aus. Nach. §§69, 71; WO §7. 33 See Mora 1963: 39–40. 34 ‘The whole of philosophy is like a tree. The roots are metaphysics, the trunk is physics, and the branches emerging from the trunk are all the other sciences, which may be reduced to three principal ones, namely medicine, mechanics and morals’ (Descartes 1985: 186/AT IXB 14). 35 As mentioned earlier, this idea is not unique to Wolff, and can already be found in Aristotle’s Metaphysics: ‘There is a science which investigates an entity qua entity and the attributes which belong to this in virtue of its own nature. Now this is not the same as any of the so-called special sciences; for none of these others deals generally with an entity as an entity’ (4: 1.1003a20–25, translation modified). 36 See DP ch. 3 and cf. DL Vorbericht §§10–4. 37 See Aus. Nach. §4. 38 See Anm. Met. 102; Aus Nach. §69. 39 Descartes is the only example Wolff provides of “those who disparaged ontology”, but Descartes never used this term. 40 See WO §§25–6. 41 See WO §7. 42 See Aus. Nach. §69. In his early writings, Wolff also avoids the term “system”, for which see Int. Sys. [2019]: 14. He likely adopts the term “ontology” from Clauberg, but it is unclear whether Wolff actively avoided using the term “Ontologie” – or his own “Grundwissenschaft” – in his early works, or whether he only came across it in 1720s. His reluctance to use the term “system” suggests active avoidance. 43 See Anm. 61; Aus. Nach. §68; WO §§ 7, 12, 15*. 44 See WO §§11–2. 45 See Aus. Nach. §§68, 69, 71. 46 See WO §55*. 47 See WO §71* and Descartes 1984: 116/AT VII 164-5. 48 See WO Praefatio: [3]/9. 49 I say indirectly, since, as far as I am aware, Wolff himself did not have this problem explicitly in mind. 50 I am not aware of any point at which Wolff defines metaphysics as a science that treats of the things that are ‘beyond’ physics. Cosmology, which studies the most abstract characteristics of physical things, is a part of metaphysics and also of physics (DP §77). Physics is a part of philosophy that treats of bodies (§59), and cosmology ‘treats of the total bodies of the world’ (§77). Furthermore, since ontology is the first part of metaphysics, and treats of what is ‘common to all entities’ and ‘predicated both of souls, as well as of natural and artificial bodies’ (§73), it follows that metaphysics cannot be understood as merely a science of the non-physical. In fact, Wolff claims that

Wolff on Ontology as Primary Philosophy 43 the three metaphysical objects (God, souls, world) are the only ones that metaphysics (special) treats of (and they can only be treated by metaphysics), because we cannot conceive of any other species of an entity that is more general, or even as general, as these three (§§55–6). 51 Hence it is first in the ‘order of knowing’, but not in the ‘order of learning’ which should begin with logic and Euclidean geometry (DP §91). For Descartes one should start with ethics, proceed to logic, basic mathematics, and then metaphysics (Descartes 1985: 186, AT IXB 13-14). For Aristotle (Metaphysics 7: 3.1029b1–12), primary philosophy comes last in the order of learning. See also Kant 1998: A52/B76. 52 See Aus. Nach. §69 and WO §1, for the connection of ontology with first principles of cognition, as well as DL §14. For Grund/ratio, see DM §29; §DP 31, and also WO §65. 53 See WO §8; Int. Sys. §11. 54 See WO §135. 55 See DP ch. 1. Cf. Kant 1998: A713/B741; A836/B864. 56 Glaube/fides; see DL ch. 7, §3; LL §611. Cf. Baumgarten 1973: §35, as well as Fugate & Hymers’ introduction to Baumgarten 2014: 21. 57 Wolff is not troubled by (global) perceptual scepticism or possibility of perceptual error for several reasons. Firstly, for Wolff, what is indubitable in the context of sensory experience is that something appears to me to be suchand-such. He points out that neither the ancient sceptics nor Descartes in his Meditations ‘denied the phenomena’, i.e., denied that the world appeared to them in a certain way, or that they imagined it in a certain way (WO §27*, Descartes 1984: 19/AT VII 29). Regarding error, Wolff holds that most cases of error arise due to sensory misrepresentation (‘when our senses represent different things as identical’), but sensory illusion is insufficient for something to count as an error – an error requires a mistake in judgement, and any judgement that goes beyond our immediate experience needs to be systematically harmonised with other judgements, which only reason can do (DM §§371, 381-2, 396-7; cf. Kant 1998: A293/B350). Ultimately, Wolff is not that interested in how we can ground the veracity of particular judgements based on sensory experience, but in discovering reasons why we experience things in a certain way. 58 In fact, in DP §4, Wolff explicitly appeals to the principle of sufficient reason as the ‘foundation of philosophical cognition’. Cf. WO §56. 59 See WO §§59, 60, 62. 60 See WO §31. 61 Respectively: that two things equal to a third are equal to each other, and that the whole is greater than its part. 62 See WO §§75*, 77. Corey Dyck (2011b: 481; 495) sees Wolff’s invocation of the dream-world as an attempt to reintroduce and “refight” Descartes’ battle against global scepticism, but it seems to me, at least at this point, that Wolff uses it as a reductio ad absurdum, rather than as a reintroduction of sceptical challenges within primary philosophy. 63 The latter is more commonly referred to as the principle of identity: ‘if A is, it is true in any case that A is’ (WO §55). 64 In his German Metaphysics, Wolff does not begin with the explication of the principle of contradiction, elaborating on it in the second chapter, but with a Cartesian intuition that we cannot doubt we are conscious, and hence that we exist. In the Ontologia this order is reversed, and we are told (§55*) that even

44  Dino Jakušić Descartes depends on the principle of contradiction to establish the Cogito. It is not widely known that Kant’s Nova Dilucidatio begins by problematising this very relation between the principle of contradiction and the principle of identity. 65 See WO §27*. 66 Hettche and Dyck (2019: §5.1) explain the appeal to the self-evident nature of the principle of contradiction as recognition of ‘the psychological impossibility of denying it’: we cannot think contrary to it, and the denial of it leads to the denial of the certitude we have of our own existence. I am uncertain of this interpretation since it suggests that the principle of contradiction is established inferentially. This is in tension with Wolff’s idea that this principle cannot be inferentially demonstrated, which the authors correctly note. 67 I presented the principle of sufficient reason as established by Wolff only through the appeal to direct experience of its truthfulness. He indeed does so (WO §74), but this is preceded by his controversial attempt to prove it from the principle of contradiction (WO §70). I omit discussion of this proof since it is not straightforward, seems to subreptively utilize the concept of ‘nothing’, and is not relevant for the topic presented here. For an account and a discussion of this issue, see Hettche and Dyck 2019: §5.1. For Kant’s critique of Wolff’s attempts to establish the principle, see Kant 1992a: 19 / PND AA 01: 397. 68 See DP §§33, 87. 69 See DL ch. 1, §2; DM §383; DP §30. While a case could be made for translating scientia and Wissenschaft as “knowledge”, I will translate them as “science”. 70 In Latin probatio and demonstratio, in German Erweis and Beweis. See LL §§496, 498; DL c.10, §11. 71 It might seem odd to use the term “proof” for what would today amount only to a valid argument. However, while Beweis is usually translated as “proof”, this mostly happens in texts which do not contrast Beweis and Erweis/probatio. Since all three terms can equally be translated as “proof”, I have decided to translate them in the way explained earlier. While I am not aware of a generally accepted practice for translating these terms, the translators of Georg Friedrich Meier’s 1752 Excerpt from the Doctrine of Reason translate Erweis and Beweis in the same way (see Meier 2016: 44). 72 See WO [2005] Præfatio, 7; §§7, 9*; DP §9; Int. Sys. §11. 73 See DP §139; LL §792; Int. Sys. §11. 74 Wolff, however, is not claiming that philosophy and mathematics are identical, only that the method of obtaining certainty through axiomatisation, definition, and demonstration should be used in both. The failure of philosophers to do so is explained by the claim that ‘the scientific treatment of primary philosophy is more difficult than of mathematics, since metaphysical concepts do not fall under the senses and the imagination in the same way as the mathematical’ (WO §7). 75 See LL §§275–6; DL c. 3, §14. See also Aus. Nach. §70. It might be worth pointing out that Wolff is much more permissive than Descartes regarding the use of experience for grounding metaphysics and other sciences, claiming that there is a “holy marriage” of reason and experience throughout the sciences, and that fundamental concepts of physics, primary philosophy, ethics, and even pure mathematics “must be derived from experience” (DP §12) – even though it is not always clear how this marriage and derivation from experience is supposed to proceed. See Rumore 2018, Dyck 2021, Leduc 2021 for discussion. 76 See Not. Dir. §1 also for an example of how Wolff derives concepts of mode, modification, and accident.

Wolff on Ontology as Primary Philosophy 45 7 See Int. Sys. §10. 7 78 See WO §29*. 79 It might be worth emphasising that for Wolff, the principle of contradiction is established in ontology, and this version grounds its application in all other sciences, including logic and mathematics. After defining what a contradiction is (‘simultaneous affirmation and negation’ (WO §30)), Wolff states that the logical definition of contradiction (understood as contradiction between propositions) ‘flows from [fluit]’ the definition established in his ontology (WO §30*, cf. LL §288). This framing of the principle of contradiction as primarily a metaphysical principle which as such applies not only to logic, but to everything, already takes place in Wolff’s earlier German Metaphysics (DM ch. 2, §§10-11). 80 See DP §§29, 73*, 89, 94, 96; Aus. Nach. §73; WO §4, 9. 81 See Int. Sys. §11. 82 See Int. Sys. §§6, 9. 83 See, for example, Int. Sys. §12, or WO §§7, 25*. 84 See also WO [2005], 9–11; Aus. Nach. §78. 85 Cf. Kant 1998: A832/B860. 86 See also Int. Sys. §3. 87 See Int. Sys. §14. 88 See DP §134. 89 See DP §§89–91*. 90 See DP §97. 91 See DP §96. 92 See also DP §73, WO §25. ‘Thing’ [Ding] is Wolff’s translation of the Latin ens, hence I am here using the terms ‘thing’ and ‘entity’ interchangeably, but with preference for the latter (although see WO §243). 93 Despite the clear Aristotelian influence, Wolff considers ‘entity’ to be the ‘highest genus’ [genus summus] (Not. Dir. §6) which is an anathema for Aristotle or Aquinas. 94 As a reminder, ontology cannot tell us whether there are any simple or complex entities. 95 This claim is complicated by certain religious ideas, such as the Trinity, but Wolff holds that ultimately we can demonstrate that there is a ‘miraculous consensus’ between propositions demonstrated to be true and those revealed by God. See Theis 2018 for further discussion. 96 See Int. Sys. §11. 97 See Not. Dir. §6. 98 Cf. Kant 2002: 316n / ÜE AA 8: 226n. 99 Wolff announces the project of developing experimental ontology, and of showing how experimentation can be integrated in every part of philosophy in his De Notionibus Directribus of 1729 (§§7, 9), but does not take it up again. See also Int. Sys. §12; DP §26. 100 See WO §§19–24. 101 For Wolff’s conception of logic, including his distinction between “natural” and “artificial” logic, see Corr 1970: 136; Gómez-Tutor 2018: 86; Cataldi Madonna 2018. 102 See LL §57, WO §19. 103 For a discussion of Flatt, and other (post-)Kantian Wolffians, see Beiser 1987. 104 See Kant 1998: Bxxxv-xxxvii; A855/B883. See also Gava 2018. 105 See Hegel 1989: 93–105; 383, and Hegel 1986: 13; 61.

46  Dino Jakušić Bibliography Aertsen, J. (2005) “Metaphysics as a Transcendental Science” Questio: Yearbook of the History of Metaphysics Volume 5, pp. 377–389 Anderson, R. L. (2015) The Poverty of Conceptual Truth: Kant’s Analytic/Synthetic Distinction and the Limits of Metaphysics (New York, NY: Oxford University Press) Aristotle (1933) Metaphysics, Volume I: Books 1–9. Translated by H. Tredennick (Cambridge, MA: Harvard University Press) Aristotle (1995) “Metaphysics’. In J. Barnes (Ed.) The Complete Works of Aristotle. Translated by W. D. Ross (Princeton, NJ: Princeton University Press) Bardout, J. (2002) “Johannes Clauberg”. In S. Nadler (Ed.) A Companion to Early Modern Philosophy (Oxford: Blackwell), pp. 140–151 Baumgarten, A. (1973) Acroasis Logica in Christianum L. B. De Wolff (Hildesheim: Georg Olms Verlag) Baumgarten, A. (2014) Metaphysics: A Critical Translation with Kant’s Elucidations, Selected Notes, and Related Materials (London: Bloomsbury) Beck, L. (1969) Early German Philosophy: Kant and His Predecessors (Cambridge, MA: The Belknap Press of Harvard University Press) Beiser, F. (1987) The Fate of Reason: German Philosophy from Kant to Fichte (Cambridge, MA: Harvard University Press) Biller, G. (2018) “Biographie und Bibliographie”. In R. Theis and A. Aichele (Eds.) Handbuch Christian Wolff (Weisbaden: Springer), pp. 5–31 Cataldi Madonna, L. (2018) “Logik”. In R. Theis and A. Aichele (Eds.) Handbuch Christian Wolff (Weisbaden: Springer), pp. 93–114 Clauberg, J. (1664) Metaphysica de Ente, Quae rectius Ontosophia, Aliarum Disciplinarum, ipsius quoque Iurisprudentiæ & Literarum, studiosis accomodata, 3rd ed. (Amsterdam: Daniel Elzevir) Corr, C. A. (1970) “Certitude and Utility in the Philosophy of Christian Wolff” The Southwestern Journal of Philosophy Volume 1, pp. 133–142 de Boer, K. (2011) “Transformations of Transcendental Philosophy: Wolff, Kant, and Hegel” Bulletin of the Hegel Society of Great Britain Volume 63, pp. 50–79 de Boer, K. (2020) Kant’s Reform of Metaphysics: The Critique of Pure Reason Reconsidered (Cambridge: Cambridge University Press) Descartes, R. (1964) Oeuvres de Descartes. Edited by C. Adam and P. Tannery. Revised edition (Paris: Vrin/C.N.R.S) Descartes, R. (1984) The Philosophical Writings of Descartes. Translated by J. Cottingham, R. Stoothoff, and D. Murdoch. Volume 2 (Cambridge: Cambridge University Press) Descartes, R. (1985) The Philosophical Writings of Descartes. Translated by J. Cottingham, R. Stoothoff, and D. Murdoch. Volume 1 (Cambridge: Cambridge University Press) Descartes, R. (1991) The Philosophical Writings of Descartes: Correspondence. Translated by J. Cottingham, Robert S., D. Murdoch, and A. Kenny. Volume 3 (Cambridge: Cambridge University Press) Dunlop, K. (2018) “Definitions and Empirical Justification in Christian Wolff’s Theory of Science” Logical Analysis and History of Philosophy Volume 21, pp. 149–176

Wolff on Ontology as Primary Philosophy 47 Dyck, C. (2011a) “A Wolff in Kant’s Clothing: Christian Wolff’s Influence on Kant’s Accounts of Consciousness, Self-Consciousness, and Psychology” Philosophy Compass Volume 6, pp. 44–53 Dyck, C. (2011b) “Kant’s Transcendental Deduction and the Ghosts of Descartes and Hume” British Journal for the History of Philosophy Volume 19, pp. 473–496 Dyck, C. (2014) Kant and Rational Psychology (Oxford: Oxford University Press) Dyck, C. (2021) “Before and Beyond Leibniz: Tschirnhaus and Wolff on Experience and Method”. In K. de Boer and T. Prunea-Bretonnet (Eds.) The Experiential Turn in Eighteenth-Century German Philosophy (New York, NY: Routledge), pp. 17–36 Frketich, E. (2019) “Wolff and Kant on the Mathematical Method” Kant-Studien Volume 110, pp. 333–356 Gava, G. (2018) “Kant, Wolff, and the Method of Philosophy” Oxford Studies in Early Modern Philosophy Volume 3, pp. 271–304 Gilson, É. (1952) Being and Some Philosophers (Toronto: PIMS) Gómez-Tutor, J. I. (2018) “Philosophiebegriff und Methode’. In R. Theis and A. Aichele (Eds.) Handbuch Christian Wolff (Weisbaden: Springer), pp. 73–92 Guyer, P. and Horstmann, R-P. (2022) “Idealism’. In E. Zalta (Ed.) The Stanford Encyclopedia of Philosophy https://plato.stanford.edu/archives/spr2022/entries/ idealism/ Hegel, G. W. F. (1986) Wissenschaft der Logik I (Frankfurt am Main: Suhrkamp Verlag) Hegel, G. W. F. (1989) Enzyklopädie der philosophischen Wissenschaften im Grundrisse; Erster Teil: Die Wissenschaft der Logik mit den mündlichen Zusätzen (Frankfurt am Main: Suhrkamp Verlag) Heidegger, M. (1976) Sein und Zeit (Tübingen: Max Niemeyer Verlag) Heidegger, M. (2001) Being and Time (Oxford: Blackwell) Hettche, M. and Dyck, C. (2019) “Christian Wolff’. In E. Zalta (Ed.) The Stanford Encyclopedia of Philosophy https://plato.stanford.edu/archives/win2019/ entries/wolff-christian/ Kant, I. (1992a) “A New Elucidation of the First Principles of Metaphysical Cognition” In D. Walford and R. Meerbote (Eds.) Kant – Theoretical Philosophy, 1755–1770 (Cambridge: Cambridge University Press), pp. 1–46 Kant, I. (1992b) “The Only Possible Argument in Support of a Demonstration of the Existence of God (1763)”. In D. Walford and R. Meerbote (Eds.) Kant – Theoretical Philosophy, 1755–1770 (Cambridge: Cambridge University Press), pp. 107–202 Kant, I. (1998) Critique of Pure Reason. Edited and translated by P. Guyer and A. W. Wood (Cambridge: Cambridge University Press) Kant, I. (2002) “On a Discovery Whereby Any New Critique of Pure Reason Is to Be Made Superfluous by an Older One” In H. Allison and P. Heath (Eds.) Immanuel Kant: Theoretical Philosophy after 1781 (Cambridge: Cambridge University Press), pp. 271–336 Leduc, C. (2021) “The Role of Experience in Wolff’s General Cosmology”. In K. de Boer and T. Prunea-Bretonnet (Eds.) The Experiential Turn in EighteenthCentury German Philosophy (New York, NY: Routledge), pp. 37–54 Leibniz, G. W. (1694) “De Primæ Philosophiæ Emendatione, & de Notione Substantiæ” Acta Eruditorum, pp. 110–112.

48  Dino Jakušić Leibniz, G. W. (1998) “Reflections on the Advancement of True Metaphysics and Particularly on the Nature of Substance Explained by Force”. In G.W. Leibniz: Philosophical Texts. Translated by R. Francks and R. S. Woolhouse (Oxford: Oxford University Press), pp. 139–142 Leibniz, G. W. (1999) “Introductio ad Encyclopaediam arcanam”. In Sämtliche Schriften und Briefe (Berlin: Akademie Verlag), Part A, Series VI, Volume IV, pp. 527–531. Lu-Adler, H. (2019) “Ontology as Transcendental Philosophy”. In C. Fugate (Ed.) Kant’s Lectures on Metaphysics (Cambridge: Cambridge University Press), pp. 53–73 Marschke, B. (2015) “Pietism and Politics in Prussia and Beyond”. In D. Shantz (Ed.) A Companion to German Pietism, 1660–1800 (Leiden: Brill), pp. 472–526 McQuillan, J. C. (2019) “The Remarriage of Reason and Experience in Kant’s Critique of Pure Reason” Epoché: A Journal for the History of Philosophy Volume 24, pp. 53–69 Meier, G. F. (2016) Excerpt from the Doctrine of Reason (London: Bloomsbury Academic) Mora, J. F. (1963) ‘On the Early History of “Ontology”’ Philosophy and Phenomenological Research Volume 24, pp. 36–47 Øhrstrøm, P., Andersen, J., and Schärfe, H. (2005) “What Has Happened to Ontology” In F. Dau, M.-L. Mugnier, and G. Stumme (Eds.) Conceptual Structures: Common Semantics for Sharing Knowledge (Berlin: Springer-Verlag), pp. 425–438 Øhrstrøm, P., Uckelman, S., and Schärfe, H. (2007) “Historical and Conceptual Foundation of Diagrammatical Ontology”. In U. Priss, S. Polovina, and R. Hill (Eds.) Conceptual Structures: Knowledge Architectures for Smart Applications (Berlin: Springer-Verlag), pp. 374–386 Olesti, J., and Zimmer, J. (2020) “Aufklärung: Christian Wolff und Alexander Gottlieb Baumgarten”. In J. Urbich and J. Zimmer (Eds.) Handbuch Ontologie (Stuttgart: J.B. Metzler), pp. 81–88 Owens, J. (1963) The Doctrine of Being in the Aristotelian Metaphysics (Toronto: Pontifical Institute of Mediaeval Studies) Rumore, P. (2018) “Empirical Psychology”. In R. Theis and A. Aichele (Eds.) Handbuch Christian Wolff (Weisbaden: Springer), pp. 175–196 Schönfeld, M. (2002) “German Philosophy after Leibniz”. In S. Nadler (Ed.) A Companion to Early Modern Philosophy (Oxford: Blackwell), pp. 545–561 Theis, R. (2018) “Theologie”. In R. Theis and A. Aichele (Eds.) Handbuch Christian Wolff (Weisbaden: Springer), pp. 219–250 Vollrath, E. (1962) “Die Gliederung der Metaphysik in eine metaphysica generalis und eine metaphysica specialis” Zeitschrift für philosophische Forschung Volume 16, pp. 258–284 Wolff, C. (1720) Vernünfftige Gedancken von Gott, der Welt und der Seele des Menschen, auch allen Dingen Überhaupt. 1st ed. (Halle: Rengerischen Buchhandlung) Wolff, C. (1722) Vernünfftige Gedancken von den Kräfften des menschlichen Verstandes & ihrem richtigen Gebrauche in Erkäntnis der Wahrheit. 3rd ed. (Halle: Rengerischen Buchhandlung)

Wolff on Ontology as Primary Philosophy 49 Wolff, C. (1724) Anmerckungen über die Vernünfftige Gedancken von Gott, der Welt und der Seele des Menschen, auch allen Dingen Überhaupt, zu besserem Verstande und bequemerem Gebrauch derselben. 1st ed. (Franckfurt am Mayn: Andreäischen Buchhandlung) Wolff, C. (1729a) “De differentia intellectus systematici & non systematici”. In Horæ subsecivæ Marburgenses anni MDCCXXIX, quibus philosophia ad publicam privatamque utilitatem aptatur (Francofurti & Lipsiæ: Bibliop. Rengeriano) Wolff, C. (1729b) “De notionibus directribus & genuino usu philosophiæ primæ”. In Horæ Subsecivæ Marburgenses Anni MDCCXXIX, quibus philosophia ad publicam privatamque utilitatem aptatur (Francofurti & Lipsiæ: Bibliop. Rengeriano) Wolff, C. (1730) Philosophia Prima, sive Ontologia, methodo scientifica pertractata, qua omnis cognitionis humanæ principia continentur. 1st ed. (Francofurti & Lipsiæ: Officina Libraria Rengeriana) Wolff, C. (1733) Christian Wolffens Ausführliche Nachricht von seinen eigenen Schriftten, die er in Deutscher Sprache von den verschiedenen Theilen der WeltWeißheit heraus gegeben, auf Verlangen ans Licht gestellet. 2nd ed. (Franckfurt am Mayn: Andreäischen Buchhandlung) Wolff, C. (1735a) “Discursus præliminaris de philosophia in genere”. In Philosophia rationalis sive Logica methodo scientifica pertractata, et ad usum scientiarum atque vitæ aptata. 3rd ed. (Verona: Dionysius Ramanzini) Wolff, C. (1735b) Philosophia rationalis sive Logica methodo scientifica pertractata, et ad usum scientiarum atque vitæ aptata. 3rd ed. (Verona: Dionysius Ramanzini) Wolff, C. (1963) Preliminary Discourse on Philosophy in General. Translated by R. J. Blackwell (Indianapolis: The Bobbs-Merrill Company, Inc.) Wolff, C. (2005) Erste Philosophie oder Ontologie, nach wissenschaftlicher Methode behandelt, in der die Prinzipien der gesamten menschlichen Erkenntnis enthalten sind §§ 1–78. Translated by D. Effertz (Hamburg: Felix Meiner Verlag) Wolff, C. 2019a. “Rational Thoughts Concerning God, the World, and the Human Soul, and Also All Things in General” In C. Dyck (Ed.) Early Modern German Philosophy (1690–1750) (Oxford: Oxford University Press), pp. 95–134 Wolff, C. (2019b) Über den Unterschied zwischen einem systematischen und einem nicht-systematischen Verstand: Lateinisch – Deutsch. Translated by M. Albrecht (Hamburg: Felix Meiner Verlag)

2

Baumgarten on the Nature and Role of Metaphysics Courtney D. Fugate

2.1 Introduction The aim of this chapter is to reach a deeper understanding of Alexander Gottlieb Baumgarten’s views on the nature and role of metaphysics. By the “nature” of metaphysics, I mean just what it is – its definition – and whatever might be the foundation of this definition, whereas by its “role” I mean both the general function that metaphysics is supposed to have with respect to other sciences and the role it is supposed to have in our lives more generally. Neither topic has received much attention in the literature, although they are clearly central to understanding the reception of his philosophy among Kant and his contemporaries. Indeed, with only one exception, most commentators seem to assume that in respect to these key issues, Baumgarten is not notably different from his famous predecessor, Christian Wolff.1 In this chapter, I will demonstrate that there are several fundamental and original aspects to Baumgarten’s conception of metaphysics that have been overlooked or at least insufficiently investigated. If I am correct, then Baumgarten departs from his predecessors, and from many of his contemporaries, by regarding metaphysics as a uniquely human science whose essential purpose is to provide the best instruments for knowing and realising perfection in human life, given that we are subject to essential limitations. I will argue further that this instrumental view of metaphysics leads him to develop several Leibnizian ideas regarding perfection and the best possible world into highly articulated metaphysical doctrines that are able to serve as universal guiding principles for all other sciences. In the final part of this chapter, I will illustrate three ways in which metaphysics plays this role through an examination of the foundations of Baumgarten’s own practical philosophy.

DOI: 10.4324/9781003187561-3

Baumgarten on the Nature and Role of Metaphysics 51 2.2  The Nature of Metaphysics 2.2.1  Baumgarten’s Definition of Metaphysics

To understand the nature of metaphysics as Baumgarten saw it, we must begin by considering what he says about philosophy, since metaphysics would seem to be one, if not the main part of that discipline. His statements on this topic are found distributed in three separate works, namely his Acroasis logica (hereafter simply Logic) published in 1761 and reissued posthumously in a revised and arguably less authoritative form in 1765 (and 1773),2 his Philosophia generalis, posthumously published in 1770, and his Metaphysica (hereafter Metaphysics), the first edition of which appeared in 1739. It is in the Logic that we discover Baumgarten’s initial definition of philosophy: PHILOSOPHY is the science of the qualities in things that are to be known without faith. (PHILOSOPHIA est scientia qualitatum in rebus sine fide cognoscendarum.) (Baumgarten 1761: §1) This definition has three main parts. Firstly, philosophy is ‘science’, meaning it is ‘certain knowledge based on what is certain’ (Baumgarten 1761: §2). Hence, the first grounds of philosophy, also called its ‘principles’, must be certain and all further philosophical propositions must be demonstrated based upon them. Now, since demonstration is a process realised by the intellect and, in particular, by reason, and is one through which concepts are rendered distinct, philosophy is also a product of reason and itself consists of distinct knowledge. Secondly, philosophy is the science ‘of the qualities in things’. In the Metaphysics, Baumgarten defines qualities to be internal distinguishing marks of a thing that can be known distinctly ‘without assuming or relating them to anything else (without the presence of anything else)’ (Baumgarten 2014: §69). Hence, although both philosophy and mathematics are scientific in the formal sense just stated, the two disciplines differ with respect to their objects or materially; the former studies the qualities or non-relational realities in things, while the latter studies quantities, which cannot be understood unless a thing is related to something else, which is taken as its measure. Thirdly and finally, the definition tells us that this science of qualities must be known ‘sine fide’. Translating this as ‘without faith’ is correct but also apt to mislead, since by ‘fides’ Baumgarten does not exclusively, nor indeed primarily, intend us to think of religious faith. Rather, the Logic informs us that ‘faith’ is one of three principal sources of knowledge, the

52  Courtney D. Fugate others two being sensation or experience and reason. Faith is assent based upon testimony (Baumgarten 1761: §357). Knowledge based on ‘faith’ thus broadly encompasses anything that requires the reliance on the assertions of others and so includes history proper, but also jurisprudence insofar as the latter relies on witnesses or historical records, and finally also revealed religious doctrines. For this reason, a more transparent but also looser rendering might be that philosophy is the science of qualities that does not require belief in anything told to us by someone else.3 This qualification would already be implied by the definition of science as ‘certain knowledge based on what is certain’, if not for the fact that Baumgarten admits not only rational certainties, but also sense-based and historical ones, which are clear but not yet distinct. Thus, philosophy excludes from its first principles only historical certainties. And yet while Baumgarten does admit that there exist certainties of sense, he also argues that we must always strive to reduce these to certainties of reason, that is, to axioms or postulates that are not only certain and clear, but also certain and clear in a distinct or demonstrated manner. In sum, then, philosophy is formally considered as a science that excludes the evidence of non-first-person testimony; materially considered, it concerns the qualities, or non-relative internal realities in things. In his Logic, which is mainly just a commentary on Wolff’s German textbook, this definition is presented without much discussion or context. But in Philosophia generalis, which is the ‘science of the many general predicates of philosophy that are common to its parts’ (Baumgarten 1770: §1), Baumgarten provides a fuller theory of this definition’s origins and consequences. As he now explains, to understand the above definition and its correctness, we must carefully attend to the logical canons for the formation of definitions, which are covered in the Logic (§§64–95). Among other things, these canons instruct us to begin by establishing a nominal definition, that is to say, a clear statement of the attributes of a thing signified by a word when the latter is understood properly in accordance with conventional usage. Establishing a nominal definition, on this account, requires primarily a historical investigation into the etymology and public use of a word, combined with skilled judgement in determining its specific predicates. Only once the signification of term is precisely fixed, is it then possible to begin developing the real definition of what it signifies. With this in mind, it is clear that the definition of philosophy offered in the Logic is purely nominal (Baumgarten 1770: §21) and as such allows us only to pick out and name something indeed referred to by common usage, but whose reality has yet to be established and whose genuine essence, if it has one, has yet to be fully determined. Now, how is a philosopher to go about establishing and determining the real definition of philosophy? As

Baumgarten on the Nature and Role of Metaphysics 53 a species of science, which involves representing the rational connections between things, it is something that can be realised only by an intellect and exists only insofar as it is found in one. So what philosophy really is, or, in other words, the real definition of philosophy, will express the essence of what the intellect able to realise philosophy would have in mind, and philosophy itself will be possible only if such an intellect is possible. This individual, or better, this individual’s intellect, will then exhibit what philosophy, understood in terms of its essential realities, consists in, and will have deduced its real definition from its nominal definition through their very own activity (Baumgarten 1770: §29). Having established this much, the next natural question Baumgarten addresses is what form this philosophy will in fact take. In pursuing an answer to this question in the Philosophia generalis, Baumgarten immediately brings to bear a set of principles which are distinctive to his thought, and which, I believe, are also key to understanding its reception and development in Kant, namely that portion of his ontology found under the heading ‘The first principles of the mathematics of intensive quantities’. Due to the significance of these principles and the fact that little attention has been directed to them in the literature, I want to take a moment here to briefly describe them and then explain below how they are used by Baumgarten to reach a better understanding of what philosophy, according to its real definition, must be like. As indicated in the preface to the second edition of his Metaphysics, Baumgarten believes that ‘a new sphere of meditation, as it were, is often opened up’ by a special discipline he has invented under the title of the ‘mathematics of intensive quantities’. This new part of metaphysics, which is to be the foundation for a ‘mathematics of non-extended things’ (Baumgarten 2014: §249), ‘will in the end grant you, a thinker, an indisputably singular joy alongside a million other amusements if you were to conceive of what is the greatest of that which is real and positive and thus discover God and the divine’ (Baumgarten 2014: 90). At its core, the mathematics of intensive quantities is to be nothing less than a higher mathematics that formalises the ways in which the relative degrees of reality and perfection in anything are determined or measured, and how these are related to both the minimum and the maximum within their kinds, the maximum being the ultimate measure and model with respect to all lower degrees.4 Kant would later reformulate and limit this mathematics to the single principle governing the reality in phenomena (the socalled Anticipations of Perception, A165–176/B207-218), but Baumgarten takes it to be the single, universal means for providing insight into the many degrees of realities or perfections in finite beings, the concept of the most perfect being, and the relation between the two. And since a precise knowledge of the relative perfection of a thing is also essential to judging

54  Courtney D. Fugate whether something is better or worse, more or less perfect, complete or incomplete, it is evident why the first principles of the mathematics of intensive quantities must play a fundamental guiding role in the construction of all sciences, especially in those that require wise judgement. If one further examines this set of principles in detail, a few features stand out. The first group of the principles (§§165–79) runs serially through all the basic ontological concepts and subjects them to a common formula for defining the lowest degree of that concept, the manner in which it may be increased, and the maximal degree it can reach. The list of concepts covered includes, among others, possibility, ground, nexus, sufficient ground, essence, unity, identity, and defect. Here, notably, Baumgarten introduces the key perfections of grounds, namely their fecundity and weight (gravity or nobility), which are closely tied to their perfections as motives for willing. A second group then follows (§§180–9), which deals with rules of perfection, the various degrees of strength in a law, how to determine the strength of a law relative to another law, how to measure the conformity of things to rules and laws, the connection between the magnitude of order and its conformity to rules, and metaphysical perfection, or conformity of things to the first principles of human knowledge. Finally, the principles culminate in definitions of the evidently central concept of the most real being, which is the basis of the ontological proof, as well as the related concepts of the supreme good, the metaphysical best, and the contingent supreme good (§190). Now, as I indicated above, Baumgarten draws on this set of principles to further determine the real definition of philosophy. He does this by first using the definition of philosophy as science to determine its various perfections, and then subjecting each of these latter to the appropriate formula from the mathematics of intensive quantities. Since science is a species of cognition, for example, ‘that which is more, greater, truer, clearer, more certain, more ardent, arises from more and more certain things, is the greater science, until it is the very most, greatest, truest, clearest, most certain, most ardent, flowing from the most and most certain things’ (Baumgarten 1770: §31). Over the next sixteen paragraphs (§§31–46), this strategy is employed to reach a definition of the highest forms of these perfections, which then are understood to be the determinations of philosophy, and so also of the philosopher, in their highest forms.5 This provides the ground for finally presenting the real definition of philosophy in the first two sections of a chapter entitled ‘The Philosopher’. The two sections in question bear the headings, ‘The Archetype’ and ‘The Ectype’, the former referring to philosophy according to its real definition and the latter to philosophy as it is developed among finite beings and, in particular, among human beings. As we saw above, philosophy according to its real definition will be the actual body of knowledge known and

Baumgarten on the Nature and Role of Metaphysics 55 deduced by the philosopher based upon the nominal definition found in the Logic. Consequently, The philosopher is the one in whom philosophy exits. Hence, the supreme philosopher is the one in whom there exits (1) all, therefore the maximum, (2) the truest, (3) the clearest, (4) the most certain, (5) the most ardent, (6) the most based on reason, hence (7) the most certain, knowledge of qualities without the need for faith. (Baumgarten 1770: §260) Referring back to what was demonstrated in the Metaphysics, Baumgarten recalls that we ‘venerate’ these same determinations in God, conceived of as the supreme being, and ‘[t]herefore God is the supreme philosopher’ (‘Ergo deus est sumus philosophus’) (Baumgarten 1770: §261).6 Hence, for Baumgarten, there is ultimately only one philosophy that corresponds to and exhibits the real definition of this idea, namely the singular archetype found solely in the divine intellect, which realises in the maximum degree all possible perfections belonging to philosophical knowledge. This single, divine philosophy is also properly called an ‘archetype’, according to Baumgarten, because the philosophy of finite beings is bound by duty to its imitation (Baumgarten 1751: §92), making the former into the latter’s proper ectype.7 This language of archetype–ectype, which will be familiar to most from Kant’s later adoption of it, is based on a passage in the metaphysics, which reads as follows: An EXEMPLAR is that to which the similar is intended [to be similar], and since it is an impelling cause (§342), it is a cause (§307), which is called exemplary, and what is caused by this is called the replica (ectype, copy). An EXEMPLAR which does not have another exemplar is an ARCHETYPE (original). (Baumgarten 2014: §346) Philosophy, according to its real definition, is thus to be understood as the exemplar of exemplars, or the unconditioned exemplar of the most perfect philosophy, and since we are bound to its imitation, it is also the archetype of our philosophy. Moreover, because of its singularity, philosophy is not in truth a general or class concept, but rather an idea or a representation of a unique individual in relation to which all particular philosophies are to be measured. In this respect, philosophy shares in the nature of the divine being itself insofar as the latter is completely determined as an individual, and therefore in its existence, through its very concept, and at the same time provides the highest instantiation and model of perfection.

56  Courtney D. Fugate I will mention only two further, general features of this archetype here, although Baumgarten himself goes into quite a bit more detail in the text itself. Firstly, the archetype differs in structure in clear and essential ways from philosophy as developed among finite beings, i.e., from its ectype. According to Baumgarten, this is a necessary consequence of the essential difference between the finite and the infinite. Every finite being possesses less than the maximum of any reality, and hence is contingent, can be increased or diminished, and ‘can be defined as a being that is not actually everything that it can be with respect to its internal determinations’ (Baumgarten 2014: §260). Hence, the philosophy of finite beings is, at best, always in a state of development towards greater similarity with the archetype. This feature of human philosophy is essential to finitude as such and so must be denied of the infinite being. What is more, the concept of progression, which is essential to the nature of human philosophy, itself involves that of an order and a priority that must likewise be denied of the divine. And yet, if we must imitate the divine, then we must represent it, and since we can only represent matters through an order, we must attribute an order even to divine philosophy. But in doing so it will be necessary to remember that this representation is only one constructed by means of an analogy with what is finite, and thus one which attributes to the divine – though in an otherwise unknown way – the realities found in the finite (through what is called ‘eminence’), while separating off and denying as much as possible, any associated negations.8 With this in mind, Baumgarten argues that the primary part of philosophy is not ontology, but rather archetypal theology, since it is most natural for the divine being to know best and with the greatest pleasure what is the greatest, namely itself. In a word, the primary object of the most perfect philosophy should also be the most perfect being. After this, the archetype of philosophy also contains an archetypal cosmology, psychology, and ontology, insofar as the divine being also knows completely and in the clearest way (hence not abstractly or generally, but most philosophically) every individual monad of every possible world, hence also all minds and all the representations within all minds, hence all possible things from every possible vantage point, and so all the eternal essences of all possible things.9 One quirk of this ordering seems to be that archetypal ontology, which knows the full essences of all things, must be thought to precede the others, not because divine knowledge, like human knowledge, must proceed from the abstract to the concrete, but rather because, unlike in the human case, the concrete eternal essences are within God himself and so must be represented as having priority in terms of perfection (Baumgarten 1770: §264). A second, notable feature of the archetype is that within it, these things must all be represented as known without any of the more notable

Baumgarten on the Nature and Role of Metaphysics 57 negations belonging to human knowledge, hence in the most perfect way or ‘maximus philosophicus’. As Baumgarten explains, Divine philosophy knows no external experiences, abstractions, acts of attention, reflections, comparisons, combinations or our successive reasoning, insofar as these are imperfections, but is immutable, without increase or decrease, and hence most holy. (Baumgarten 1770: §269) Nevertheless, apart from these negations, human knowledge also contains many realities or perfections, among which is the ability to know one thing based upon another, i.e., connectedness or groundedness. More perfect knowledge thus knows more things from more things, while the most perfect knowledge must be represented as knowing all things from all other things. By this use of analogy, the reality contained in the idea of connectedness found in human knowledge is retained while its associated negation, namely priority, is denied. To know all things in the most philosophical way is therefore to be able to see how each thing is as much the ground as the consequence of all other things, and hence ‘the supreme philosopher sees all things flowing from all things’ (Baumgarten 1770: §45). For simplicity, I will refer to this core principle of archetypal philosophy as the “equipollence” of grounds. Remembering what was said above regarding the analogical ordering we nevertheless ascribe to divine knowledge, it is clear that Baumgarten embraces another core strategy of traditional theology, which lies in balancing a positive theology based upon analogy with a negative theology that denies that this same analogy, or really any, is able to adequately represent the divine; for Baumgarten, as we have seen, recognises that we must represent the divine knowledge and so we must ascribe some ordering to it by analogy with human finitude (though one reordered to better reflect its perfection and nature) and yet also, from another point of view, deny that this very ordering can adequately represent the divine. So in one sense we ascribe an order to archetypal philosophy, while in another we must deny any priority within it. This runs parallel to the fact that we must ascribe some primary essence to God in any scientific theology while recognising that no concept of the divine is truly prior to any other. These essential differences between divine, archetypal philosophy and our human, ectypal philosophy is consistent with an important expansion found in the second edition of the Logic, where the nominal definition of philosophy is followed by a more determinate formulation of what here in the Philosophia generalis is called the ectype of philosophy. It reads: ‘OUR PHILOSOPHY * (*philosophy insofar as it is constructed among human beings, or considered as a collection of disciplines) is the

58  Courtney D. Fugate science of several qualities in things known without faith: distributed into disciplines’ (Baumgarten 1773: §38). Clearly, God’s knowledge can be described in several ways, as theology, cosmology, psychology, and ontology, but it cannot properly be regarded as divided into disciplines, since a discipline, by its very nature, describes a practice for improving our imperfect capacities for knowledge and action. Hence, no matter how much the division of archetypal philosophy seems to resemble our human division of philosophy, it is crucial to recognise that the latter essentially reflects a division into disciplines, which in turn reflects the order most conforming to human imperfections and the means to their improvement. 2.2.2  Metaphysics an Essentially Human Science

Some, including myself,10 have casually suggested that Baumgarten perhaps anticipates Kant and some of the latter’s contemporaries in taking a subjective turn with regard to metaphysics. This turn, if indeed genuine, lies somehow in that portion of the definition of metaphysics that qualifies it as pertaining specifically to ‘human knowledge’. As Baumgarten writes, METAPHYSICS is the science of the first principles in human knowledge. (Baumgarten 2014: § 1) Yet, while it is true and demonstrable from an examination of the lecture notes that Kant developed his own definition of metaphysics by applying pressure to this exact point, i.e., to the question of how metaphysics is possible as a human science, I have always remained mildly sceptical about the extent to which this was what Baumgarten had in mind. I would now like to make the case for the thesis that Baumgarten intended to institute a certain sort of subjective turn not found in other philosophers, at least not in his fellow Wolffians, and that this is reflected in his definition of metaphysics. To anticipate my conclusion somewhat, I will argue that Baumgarten regards metaphysics as something belonging exclusively to human beings due to their finitude, and so not to God, who is infinite. In this respect, he is creatively drawing on critical elements and strategies already found in the traditions of positive and negative theology. And although this does represent a subjective turn generally absent in Wolff and his other followers, it remains fundamentally distinct from that of Kant, who instead restricts the validity of our concepts, in their theoretical application, to objects of experience and so denies any sense to them in a supposed transcendent application (although he concedes sense to certain theoretically transcendent concepts in their practical application).11

Baumgarten on the Nature and Role of Metaphysics 59 The first piece of evidence for this claim is indirect. In all of his works, Baumgarten fiercely applies the canons of definition, including that according to which ‘[a]ny note which is sufficiently determined by another does not enter into a definition’ (Baumgarten 2014: 88). He believed that he was permitted and should forgo the usual practice of including derivative notes or marks,12 which are otherwise useful in avoiding misinterpretation, since his writings were intended to be fleshed out in the lecture hall. Hence, if “human” is included as a mark in the definition of metaphysics, then undoubtedly Baumgarten regarded it as both an essential and a primitive mark of that concept. Note also that Baumgarten, as we have seen, does not qualify philosophy itself as human, except in its ectypal form. So whereas he does present a definition of philosophy that in principle is applicable to the divine being, he presents no such definition of metaphysics. Presumably then, if we take Baumgarten’s definitions as seriously as he asks us to take them, then we must conclude that for him God can be the supreme philosopher, but not the supreme metaphysician. But why would this be? Why would Baumgarten exclude ‘human’ from his definition of philosophy, only to include it as an essential mark of metaphysics? I believe a clue to an answer can be found in the preface to the third edition of the Metaphysics, where he responds to an unnamed critic who proposes that there is a contradiction in assuming the principle of sufficient reason (PSR) to be universal, since one thereby must posit both a first ground of all things and yet also that this ground itself has a further one. As Baumgarten explains, [W]e must always distinguish the ground from the ground that we specifically know by proceeding through every single thing. The notes of things that are relatively first, or first with respect to human knowledge (also to clear human knowledge), certainly still have a ground, although we do not know this ground in respect to the rest of its predicates in a way that would allow us to separate it sufficiently from the others. And hence there is always something prior to the first in clear and distinct human knowledge, which can only ever be known somewhat obscurely by humans. …But the absolutely first being as such neither abstracts, nor reflects, and hence, since it knows all things most distinctly from everything else, it is never led back to absolutely first notes. (Baumgarten 2014: 80) Here Baumgarten directly contrasts what he will later call ectypal and archetypal philosophy and, just as in the Philosophia generalis, denies the appropriate negations of the divine being while asserting the equipollence of grounds. But in addition, he now uses this contrast to explain how the

60  Courtney D. Fugate PSR can be true, although human metaphysics generally, and his metaphysics in particular, must always start from a first or absolutely primary principle. To put the argument in the language of the Philosophia generalis, Baumgarten here contends that within archetypal philosophy, since all things flow from all things, everything has a sufficient reason and there is never a true first in the chain. But in human philosophy there must always be a first, and indeed a first beyond which – given the essential limit attaching to human nature – our knowledge can never progress. Now, if there is no first concept or principle in archetypal knowledge, then it follows that there can be no first principles of such knowledge, and, consequently, there can be no basis for any conception of a divine metaphysics. On this understanding, the way of analogy requires the denial of a divine metaphysics, for the same reason it requires us to deny an absolute priority within archetypal philosophy. First principles are only needed by finite beings and so a science of such principles, i.e., metaphysics, is essential only to ectypal philosophy. We reach the same conclusion by observing that metaphysics contains essentially abstract knowledge and, as we saw in the Philosophia generalis, and again in the third preface to the Metaphysics, Baumgarten affirms that the supreme philosopher knows nothing of abstractions. From this, we must conclude that Baumgarten takes the very possibility of a human metaphysics to be rooted in a kind of need that arises only because all finite beings, and hence also human beings, have an essential limit beyond which they cannot pass. The basis for such a claim is explained in the following paragraphs of the Metaphysics: To be a real being is a quality (§69) belonging to every being (§136). And since there is a certain number of realities in every being (§136, 159), every being has a certain degree of reality (§246, 159). Hence, this will either be the greatest, or not (§10, 247). And since that degree of reality in comparison with which a greater is possible, or that which is not the greatest (§247), is called a LIMIT (a boundary , cf.§350, end, cf. §341), that which has a limit will be FINITE (cf. §341, limited). … Finite beings have a limit, hence a degree (§248), and therefore a quantity (§246). Hence philosophical and mathematical knowledge of all finite beings is possible (§93, 22). The ESSENTIAL LIMIT of a given finite being is that limit beyond which it is impossible for there to be anything more in this being due to its own essence. The mathematics of non-extended things is the MATHEMATICS OF INTENSIVE QUANTITIES (§247). (Baumgarten 2014: §248–249) By means of this passage, we can see how Baumgarten links the essentially human character of metaphysics, i.e., its basis in the finitude of human

Baumgarten on the Nature and Role of Metaphysics 61 nature, with the possibility of a mathematics of intensive magnitudes that articulates from our own finite point of view the relation of ourselves to the infinite. Hence, we can see that the very limit that makes metaphysics necessary is also what makes it possible for us to know our own limitedness, and thereby also the relation of ourselves to God and of our ectypal philosophy to the single, supreme archetypal philosophy. In this manner, Baumgarten holds that God can be represented, though inadequately and with qualifications, as the supreme theologian, the supreme cosmologist, and the supreme ontologist, because these name classes of real things known by God, namely God himself, contingent beings, and essences, which in turn means that the knowledge of them is also a reality that must be ascribed to God analogically through eminence. However, because metaphysics itself is rooted in a negation and limit, the same analogical attribution of the perfections of knowledge to God requires the denial of it as a division of archetypal philosophy. Hence, the divine being cannot likewise be represented as the supreme metaphysician without detriment to his infinitude. Metaphysics therefore must be understood as referring essentially to a special group of principles that is necessarily first in our knowledge of things only because we are unable to know them fully adequately and immediately. On Baumgarten’s account, then, the primacy of metaphysics should be seen as inseparable from its abstractness; for we must know things by means of intermediate representations, namely abstract concepts and principles, only because we are unable to know individual things fully adequately and immediately. It is a simple corollary to this, that if our knowledge is nevertheless to amount to science, then we must be able to begin from some set of concepts and principles that are (fortunately) immediately certain to us, although in truth they are also demonstrable in principle, or by God, from other things that, again for us, are forever furthest from certainty. Two other pieces of evidence strongly support the essentially human character of Baumgarten’s metaphysics. The first stems from his comments on the PSR in the third preface to the Metaphysics. As I have just argued, Baumgarten regards our need to comprehend things by reference to prior and indeed first principles, as based not in the nature of things themselves, but rather in the discursivity13 and finitude of the human intellect. This view is directly reflected in his seeming indifference in this passage about whether the PSR be regarded as a first, indemonstrable truth, or instead as a derivative and demonstrable one. In response to criticisms of his proof of this principle, he notes that if one is not willing to accept the argument in §20 of the Metaphysics (which derives the PSR from the principle of contradiction), then one could just as well regard the principle as indemonstrable and self-evident. Not without exasperation, he retorts: ‘If only these sorts of claims [i.e., to self-certainty] were never made about things

62  Courtney D. Fugate that are less evident!’ (Baumgarten 2014: 78). Alternatively, if one wished, one could regard the principle as proven by, and so as following from, certain things found later in the system, which, as matters stand, are now presented as following from it. The exact way this principle is treated in his system, it seems, is based solely on a choice about how best to present truths and their connections in a way that conforms to the needs of the human intellect. Such flexibility in the manner of presentation can be seen as a direct consequence of the principle of the equipollence of grounds; for all truths in themselves are equally interconnected and reciprocally grounding and so every part of the system – including the PSR itself – can be regarded as a ground for all the other parts. It is for this same reason that Baumgarten is able to make the otherwise startling claim that if God did not exist, then the principle of contradiction would be false (Baumgarten 2014, §824),14 although this last principle obviously forms the absolute bedrock, or ‘absolutely primary’ principle, of his metaphysics (Baumgarten 2014, §7). The second piece of evidence lies in Baumgarten’s analogous statements regarding human theology in the following passages from the Metaphysics: §816. The first concept of God is the internal perfection of God from which all the remaining internal perfections can thus be eventually deduced such that this very perfection cannot be afterwards deduced from another internal perfection of God by those desiring to avoid circular logic (§40, 39). Now, such a deduction of the remaining internal perfections is possible from the infinite perfections of God (§24, 49), since any given ground is the greatest (§812, 166), most sufficient (§169), and hence the unqualifiedly final (§170) and greatest essence (§171). Therefore, the first concepts of God are infinite, and any of these, when chosen to be the essence, is nevertheless the unique essence of: God (§40, 77). §817. Although there is in God the greatest (§808, 167) and maximally universal nexus (§172), and thus everything in God in the truest sense originates from everything else (§876), nevertheless it is easier for us to know the rest of his perfections from one perfection of God than it is from another (§527). Therefore, it is preferable to choose as the essence that perfection from which we hope to deduce the rest most easily (§816). (Baumgarten 2014) Again, the specific character of our metaphysics, in this case of our natural theology, is based not on an order of priority found in things themselves, but instead on the way of thinking most suited to our human understanding. Furthermore, it is clear from these passages that the very fact that we

Baumgarten on the Nature and Role of Metaphysics 63 must choose any particular essence of God to be first, and so construct a metaphysics of God at all, is due only to this same finitude. But while we must construct just such a metaphysics to fulfil the need for a positive theology, we must also seek as much as possible to both recognise the equipollence of grounds and extend our knowledge of things so that we can increasingly (though always inadequately) understand the reciprocal ground of all things in one another. In this respect, the relationship between archetypal and human theology provides another example confirming that the need for metaphysics generally belongs only to human, ectypal philosophy. Finally, the purely human character of our metaphysics follows logically from two doctrines already mentioned, which are exclusive to Baumgarten, at least with respect to the Wolffian tradition. Firstly, as we just saw, he purports to demonstrate that every one of the infinite concepts of God is his unique essence, from which all others can be demonstrated. This implies that God’s actuality is also his unique essence. Secondly, Baumgarten also holds that if God were not actual, then the principle of contradiction, which is the absolutely first principle of metaphysics (Baumgarten 2014, §7), would be false.15 Hence, the truth of the principle of contradiction, the ground of our entire metaphysics, could be seen as depending and following from God’s actuality. Now, if we add to these two unique claims the doctrine that God also knows all things, including all contingent things, entirely from and through his own self-knowledge, then it follows not only that the absolutely first principle of our metaphysics is not the absolutely first principle of knowledge as such or in itself (since there is no such thing), but indeed also that infinitely many alternative forms of metaphysics are possible, although only a few may be possible for us and only one may be the best suited to the specific limitations of our nature. 2.3  The Role of Metaphysics 2.3.1  How Metaphysics Guides the Selection of a First Principle

As the first and fundamental human science, metaphysics evidently plays a number of essential roles in all other sciences and so within the scope of human life. Most of these are not unique to Baumgarten, nor important for understanding the reception of his philosophy. However, two in particular are such: the role that it plays with regard to the beginnings or principles of the sciences and with respect to their internal ends or perfections. If the argument of the last section represents Baumgarten’s considered view, then the selection of the principle of contradiction as the absolutely first principle of metaphysics is grounded solely in its unique suitability to play this role for our form of knowing. In this respect, its selection rests

64  Courtney D. Fugate on the philosopher’s wise judgement, which itself must be based upon an application of the first principles of the mathematics of intensive quantities, including the metaphysical principles of the best.16 Yet this is a feature attaching to all human knowledge, and hence should not be exclusive to metaphysics or to its own first principle. Indeed, due to the equipollence of grounds, all first principles of whatever human science should in some sense be a matter of such wise selection. As surprising as this consequence might seem, it is readily confirmed and illustrated in Baumgarten’s Elements of First Practical Philosophy. The purpose of the Elements is to provide a scientific treatment of the first principles of all practical philosophy, under the title of ‘first practical philosophy’, in a way analogous to how metaphysics provides the same for all human sciences as such (Baumgarten 2020: §7). Accordingly, practical philosophy is defined by Baumgarten as ‘the science of the obligations of a person that are to be known without faith’ (Baumgarten 2020: §1) and first practical philosophy as ‘the science containing the first principles that are proper but also common to the rest of the practical disciplines’ (Baumgarten 2020: §6). Now, just like metaphysics, first practical philosophy must have its own internal or ‘domestic’ first principle, something analogous to the former’s principle of contradiction. But what should this first principle be? Baumgarten extensively treats this very question, but two passages will suffice to capture his main line of thought. In §89, we read regarding the first principle of the right of nature: And even if (8) it may not be denied that the institution or selection of this sort of first principle, as in other disciplines, thus also in the right of nature in this sense, is something CHOSEN, … still, just as nominal definitions, despite being chosen, are not left simply to a BLIND CHOICE, which would be to strive against reason on who knows what grounds, but rather follow a PRUDENT CHOICE, i.e., the best cognition of the best that there can be: thus also if there are many principles belonging to any given discipline or right of nature that you could institute as the first according to preference, the one to be preferred is that which is better among the rest according to a wise preference (§70). (Baumgarten 2020: §89) Three things are notable here. Firstly, the seemingly open question in the first line about whether there are other possible first principles for this discipline was answered affirmatively in the previous paragraph.17 Secondly, Baumgarten here explicitly regards this first principle as chosen, or as a matter of convention, in a manner perfectly analogous to how nominal

Baumgarten on the Nature and Role of Metaphysics 65 definitions are chosen, which was discussed above. Finally, he stresses that although the choice of a principle is indeed a matter of convention, it must be anything but arbitrary in the sense of blind or without good reason; indeed, it must be deliberately selected with the aim of providing the best possible principle, i.e., the one most conducive to realising the necessary perfections of this sort of knowledge. Although Baumgarten’s reference to the ‘best’ and to ‘wise preference’ here should be enough to indicate connection of the criteria for such a choice and the first principles of intensive quantities, §90 goes on to explain in detail that ‘prudent choice’ would settle on the principle that is most conducive not only to the achievement of those perfections specific to practical knowledge as such, but also to their achievement among human beings like ourselves, with all our flaws and mediocrity. In this respect, wise or prudent choice indicates selecting the principle that is easiest to understand and employ in demonstrations by those only ‘moderately prepared’ to undertake practical philosophy. The first principle is thus uniquely human in at least three respects: insofar as any principle must serve as first at all, insofar as this first principle should be adequate for allowing us to deduce everything essential to our practical philosophy, and finally insofar as it should be selected with an eye to the actual teaching and learning of the discipline among people of ordinary talent.18 Most interestingly, Baumgarten cites the latter as a reason why the ‘first principle of practical philosophy, and of the right of nature, which is a part of philosophy, ought to be demonstrable without faith’, thereby indicating that the need for philosophy itself, and of first practical philosophy as a part of it, lies in the need to ground human life in that form of thinking most suited to influencing the most human beings in the strongest possible way. Presumably, then, it is based upon these very criteria that Baumgarten himself ‘wisely’ selects as the first principle of practical philosophy ‘furnish the good to seek perfection as much as you are able’, which he equates with ‘furnish the good that is to be known with certainty and without faith as much as you are able’, warning however that ‘it should not for that reason be denied that any of these [i.e., other derivative principles in practical philosophy] can ever be considered primitive’ (Baumgarten 2020: §91). The virtuous circularity of Baumgarten’s thinking here should not be overlooked. To furnish the good and to seek perfection, which is our duty, we must first know what goodness and perfection are for us, what degree of these we possess, and what increasing this degree as much as we are able to looks like. Consequently, this first principle makes it an obligation to develop, know, and apply the mathematics of intensive quantities in our own lives, while that same mathematics provides the ground for choosing this principle as the best, or as the one most suitable to our own finite nature.

66  Courtney D. Fugate From this examination of the rules for selecting a primary principle for first practical philosophy, which Baumgarten discusses far more extensively than any other, we are able not only to confirm and illustrate what we saw earlier regarding the selection of the principle of contradiction as the primary principle of metaphysics, but also to better understand how the selection of all such principles – insofar as it aims at perfecting our knowledge and so ourselves, and so at imitating the divine in a way suited to human finitude – constitutes a genuine ethical duty. 2.3.2 How Metaphysics Guides the Formation of Any Complete System of Science

The choice of a first principle thus rests on a judgement with regard to what is ‘best’, and so must depend upon a scientific assessment that the principle, in comparison with others, leads best to a certain end, namely to the greatest realisation among human beings of the perfections of a given science. Now, although the ideal might be a complete system of perfectly demonstrated knowledge, human beings, like all finite things, are subject to limits, one of which is that the vast majority of our cognitions must remain undemonstrated, confused, and sensory. What is more, the specific laws of nature, and so also of practical philosophy – a knowledge of which is required for us to know and pursue what is best – are themselves dependent upon God’s own inscrutable but nevertheless ‘wise’ choice. As I have explained elsewhere,19 the formal features of this divine choice of laws can however be grasped through a specific set of principles, found in the Metaphysics, which define what is natural and show that God gives priority to natural over supernatural order, but also spell out the conditions under which the latter would in fact obtain. What this means for our construction of any particular science is that we should reasonably expect its laws to conform to the most harmonious order that can be known from the natures of individual things and from the natural whole to which they belong; for every natural order – individual and natural – has its common norms (Baumgarten 2014: §472) and, all else being equal, God prefers what conforms to that order. Now, as Baumgarten explains, ‘the collection of the laws of the order of nature is the RIGHT OF NATURE, IN THE BROADEST SENSE, whose parts are the laws and rules of motion, and the laws of the nature of spirits’ (Baumgarten 2014: §472). The metaphysical priority of natural over supernatural orders, combined with the mathematics of intensive quantities, thus leads to something like what Kant would call a regulative principle for investigating the right of nature, broadly considered, and so for determining its further particular laws. In constructing any system of the right of nature (which includes both physics and practical philosophy),

Baumgarten on the Nature and Role of Metaphysics 67 we should be guided by an expectation that the law harmonising most fully with those laws of nature already known and with the most perfect possible whole of natural law will also be a true law of the right of nature, broadly considered. This is because – given that God has prioritised natural order, which is knowable by human reason – he has established this law in order to realise that very harmony and thus has done so with the kind of foresight (or with what Baumgarten will also call ‘vision’) that human reason can have insight into, at least in principle. This shows yet another way in which metaphysics provides a guide for wise judgement, this time not with respect to first principles, but instead with respect to determining the laws of order governing any systematic body of natural knowledge. Baumgarten’s Elements again provides a perfect illustration of metaphysics playing this unique role. That specific part of the right of nature, taken in the broadest sense, which comprises ‘the collection of natural laws that obligate human beings is NATURAL RIGHT, BROADLY CONSIDERED’ and includes ‘both internal and external moral laws’ (Baumgarten 2020: §65). This subsumption of moral laws, or of our moral nature, directly under the right of nature, broadly considered, has three fundamental consequences for the form of any proper system of practical philosophy. Firstly, laws obligate through rewards and punishments and natural laws through natural rewards and punishments, which, though chosen by God, follow and can be known from the natures of mundane things. Since, all else being equal, God prefers natural over supernatural order, he also prefers to obligate through natural rewards and punishments. This means that ‘wise’ judgement will reliably conclude from the natural consequences of a type of action to a specific law, i.e., a duty, of its commission or omission. Whence, there arise two formal laws from which many other particular laws can be inferred through wise judgement: ‘The law of nature is: commit whatever promises the most and greatest rewards, and omit its opposite’ (Baumgarten 2020: §111); ‘The law of nature is: omit what threatens the most and greatest punishments, and commit its opposite’ (§120).20 Likewise, since human law has vision – i.e., is founded in conformity with a knowledge of what would be best – to the same extent that it agrees with divine laws, and, furthermore, since the above regulative principle informs us that divine laws have natural rewards and punishments, it follows that the correctness of existing or proposed human laws can be confirmed by the existence of natural rewards and punishments for the very same acts (§111, §121). A second consequence of this subsumption of moral nature under the form or right of nature as such is that it allows us to bring to bear several other metaphysical principles for placing any actual system of natural right within the broader context of the archetypal system of moral laws,

68  Courtney D. Fugate and so for gaining an insight into the former’s limitations as well as into the goal towards which we ought to strive in its development through imitation of the latter. The true system is of course also the ‘best’, where the precise sense of this word is explicated by the mathematics of intensive quantities. Now, due to the same finitude that prevents us from ever possessing the archetypal philosophy, we also can never fully possess the best system of moral philosophy at any actual moment, nor can we ever know clearly and distinctly all things that pertain to practical philosophy (Baumgarten 1741a: §5). Hence, the whole system can be divided into the territory known by ‘a given human being, or even the whole human species’ and the unknown territory (Baumgarten 2020: §95). Now, of the known territory, part is known by reason; and this is possible precisely because the right of nature, itself based upon the divine preference for a natural order, is what is knowable by reason. But the same divine preference leads us to expect that what is not known to reason is yet such as could not only be known to it in principle, but is also supremely knowable in itself and the most suitable object for maximally philosophical knowledge. Now, in addition to reason, we also possess various faculties for representing the laws connecting things, i.e., their nexus, indistinctly or sensitively, but still clearly, which faculties constitute what Baumgarten terms the ‘analogue of reason’, or in legal theory, is called ‘experience’. It follows, then, that the known territory can be further divided into that of science, known by reason, and that of sensory experience, known by means of the analogue of reason. This knowledge of the terrain of practical philosophy, combined with the duty to ‘furnish the good and to seek perfection’, leads Baumgarten finally to affirm a principle demanding of us that we perfect our own practical knowledge by extending, as far as possible, the territory of the known into that of the unknown and the territory of science into that of experience (Baumgarten 2020: §97). A third consequence concerns how we ought to interpret law, given that the whole of genuine law should form such a perfectly harmonious whole, all of which can be known to the supreme philosopher, and only some of which can be known naturally by human reason. As we saw above, the laws governing free beings should form a system containing in the greatest possible degree those perfections proper to practical philosophy. This leads to the expectation that every law will stand in the greatest possible nexus of ground and consequence relations with all other laws and things in nature. Now, in the Elements, Baumgarten explains that ‘the RIGHT and PRUDENCE of a legislator as such are NOMOTHETIC’ (Baumgarten 2020: §105), meaning it is these things – right and prudence – that qualify any promulgation of a norm to be a genuine act of legislation. For this reason, ‘LAWS strictly considered according to nomothetic rules POSSESS VISION broadly considered, and those contrary to the same are BLIND,

Baumgarten on the Nature and Role of Metaphysics 69 broadly considered’ (§105). God has these things in the supreme degree and hence he is the supreme legislator (§100). The true laws of natural right, then, are never blind, but instead have the greatest vision. Thus arises the special rule for interpreting law: ‘take as true and genuine the literal sense of positive laws that, if it were true, would harmonise best with the nomothetic rules, until the contrary be understood, or: do not interpret any positive law as blind without necessity’ (§179). According to this rule, any judge interpreting positive human law must again possess wise judgement, which is a faculty guided by a knowledge of the rules of perfection (and so informed by the mathematics of intensive quantities), of the distinction between the natural and the supernatural, and of the rules according to which God prioritises natural over supernatural means – in a word, must be’ guided by metaphysics. This rule in particular directs us to use such metaphysical knowledge to insure that human law is always interpreted in a way that aligns as much as possible with what is truly legislative. 2.4 Conclusion Previous scholarship has established that part of Baumgarten’s originality lies in the way he connects certain core doctrines regarding perfection and goodness, which are more typical of Leibniz’s thought, with specific parts of the Wolffian systematic metaphysics popular in his own day. In this chapter, I have shown that these connections are not isolated and unmotivated, but instead stem from the very heart of Baumgarten’s well-considered conception of the nature and role of metaphysics as a uniquely human science. According to him, this discipline is supreme in human knowledge, not because it corresponds to what is first in divine knowledge, but instead because it provides the most basic, most indispensable, and most appropriate instrument for perfecting human knowledge and thereby also the human being. Metaphysics is able to do this for two reasons: (1) Its own first principle has been selected for this very purpose and (2) that chapter of it concerning the mathematics of intensive quantities provides the criteria for wise or prudent judgement, i.e., for judgement regarding the degree to which anything approaches what is truly best. If this knowledge is then combined with another part of metaphysics, namely that treating the laws governing divine choice, then there follow at least three further formal principles for the development of all other sciences, as illustrated in our examination of the Elements. These principles direct us in discovering and confirming laws of nature, in conceiving and extending the boundaries of a given science, and in interpreting the existing body of scientific knowledge. Finally, we have seen that practical philosophy demonstrates that developing and teaching this sort of metaphysics – one not based on faith – is itself a duty for those who are able, since, like all human beings,

70  Courtney D. Fugate they are determined to ‘furnish the good or seek perfection as much as [they] are able’ (91). In this respect, Baumgarten argues that some of us have a natural as well as divine duty to become philosophers and to teach the type of practical philosophy that is well-grounded in his metaphysics and so appropriate to the moderate talents of human nature. Notes 1 Aichele 2010 defends a sharp distinction between Wolff’s and Baumgarten’s conceptions of philosophy based upon differing views regarding its proper object. 2 See Schwaiger 2017 for an account of the origins of the first edition of the Logic and its relation to subsequent editions. 3 Compare Baumgarten 1741b: 12. 4 A typical example reads: ‘§185 The smallest perfection is only one smallest agreement of the fewest and smallest beings in one smallest being (§94, 161). Hence, the more and greater beings, the more beings in which they agree and the greater they are, the more often and the more closely they agree, the greater is the perfection (§160), until it is the greatest perfection, which is the greatest agreement of the most and the greatest beings in one (§161, 169). Moreover, since the supreme perfection is thus maximally composite (§183, 96), simple perfection, however great it may be, is nevertheless not the greatest (§96)’. 5 Compare Baumgarten 1741b: 12, where a similar list of determinations is developed. 6 Cf. Baumgarten 1741b: 12–13. 7 As Baumgarten explains in the passage cited, this is properly an ethical duty based upon the natural law which bids us to increase our perfection. Striving to imitate the divine archetype involves increasing the perfection of one’s own intellect and hence falls under the duty to perfect ourselves. 8 The well-known ‘way of analogy’ is endorsed and explained in Baumgarten 2014: §§826 and 827. For a recent discussion covering Baumgarten and Kant, see Chance and Pasternack 2018. 9 There is obviously an order here. What is the basis of that order, if there is no order in God? It must be the order in which we are alone able to represent God most properly to ourselves. 10 See Fugate 2015. 11 In fact, the matter is still more complicated; for Kant does admit a limited sense to theoretically transcendent principles in a theoretical respect, and indeed does so using the device of analogy (which he employs in his practical philosophy for the same purpose), but this sense is restricted to the regulative function that metaphysical concepts can play in empirical research and in no way (not even inadequately) represents the supposed metaphysical objects of these concepts. 12 For the definition of a mark or note as ‘the ground of distinction in a thing’, see Baumgarten 2014: §67. 13 That is, the essential reliance on chains of reasons in which one member is prior to another. 14 For if every truth both implies and is implied by every other truth, then they are all equivalent and so the denial of any one implies the denial of the others. The force of this claim, however, is to emphasise that the truth ‘God exists’, which is generally held to be less certain than the principle of contradiction, should

Baumgarten on the Nature and Role of Metaphysics 71 be regarded just as primary as the principle of contradiction, since there is no priority among truths. 15 This is the second principle unique to Baumgarten. He explains his acceptance of it in Baumgarten 1741b: 65–68. 16 How this precisely works remains unclear and deserves further study. At the least, it seems that the selection of a first principle such as that of contradiction would rest on a survey and comparison of its perfections (its self-evidence, clarity, distinctness, its ‘fecundity’ or adequacy for deriving all other principles from it, etc.) with that of all other potential principles, informed by the mathematics of intensive quantities. 17 §88 reads in part: ‘If some specific principle in our system of demonstrating the right of nature is the first of the domestic principles, then do not let it be inferred that no other principle could also be correctly instituted as first in any other series for demonstrating the right of nature. For, what is first in a certain series of demonstrations of domestic principles can perhaps be deduced in a slightly modified chain of conclusions from another domestic principle, and this, then, only from propaedeutic principles’. 18 In the first preface to the Metaphysics, Baumgarten himself claims to have adopted this as a law for his own teaching: ‘[T]here is not reason why I should regret the very rule that I have resolved I must follow, since my task has been not solely to learn, but also to promote learning: To regulate whatever I say so that a person of average intelligence is able to know clearly and perspicuously what I mean when mildly familiar with the doctrines that I offer and that must be taught, and when only mildly interested in that which I treat and that must be learned. Weighing whatever I am about to expound to YOU according to this law, I will neither seize upon novelties because they are recent, nor spurn the old because it seems obsolete’ (Baumgarten 2014: 92). Also in the Philosophische Brieffe, after explaining the perfections of the highest philosophy, Baumgarten warns that although we should pursue this ideal, we should not presume to ever match it. Rather, we should measure our philosophy to the fact that ‘the best philosophers among those that are mortal are the mean between fools and the all-too-wise’ (Baumgarten 1741b: 12). If my interpretation is correct, then the basis for this law actually lies in the duty to employ wise choice, informed by the finitude of human nature, of the best and most suitable form of education. 19 Fugate 2018, especially 150–155. 20 For this to hold, the consequence must be naturally knowable from the action, not just follow upon it. Baumgarten warns in particular of subreptive inferences possible here as well as of the fallacy of post hoc, propter hoc. See Baumgarten 2014: §548; 2020: §§113 and 122.

Bibliography Aichele, A. (2010) “Allzuständigkeit oder Beschränkung? Alexander Gottlieb Baumgartens Kritik an Christian Wolffs Begriff der Philosophie” Studia Leibnitiana Volume 42, pp. 162–185 Baumgarten, A. G. (1741a) De vi et efficacia ethices philosophicae. Praeside Alexandro Gottlieb Baumgarten. Auctor Samuel Wilhelm Spalding Traicti ad Viadrum. Latin edition and German translation with an introduction by A. Emmel: http://www.ruhr-uni-bochum.de/aesth/Emmel/Spalding.pdf

72  Courtney D. Fugate Baumgarten, A. G. (1741b) Philosophische Brieffe von Aletheophilus (Frankfurth und Leipzig) Baumgarten, A. G. (1751) Ethica philosophica scripsit acroamatice (Halle: Hemmerde) Baumgarten, A. G. (1757) Metaphysica 4th ed. (Halle: Hemmerde) Baumgarten, A. G. (1760) Initia philosophiae practicae primae acroamatice (Halle: Hemmerde) Baumgarten, A. G. (1761) Acroasis logica in Christianum L. B. de Wolff (Halle: Hemmerde) Baumgarten, A. G. (1770) Philosophia generalis. Edited with an introductory dissertation by J. C. Foerster (Halle: Hemmerde) Baumgarten, A. G. (1773) Acroasis logica, aucta et in systema redacta, Ioanne Gottlieb Toellnero (Halle: Hemmerde) Baumgarten, A. G. (2011) Metaphysica/Metaphysik Eds. G. Gawlick and L. Kreimendahl (Stuttgart-Bad Cannstatt: frommann-holzboog) Baumgarten, A. G. (2014) Metaphysics: A Critical Translation with Kant’s Elucidations, Selected Notes, and Related Materials. Translated and edited by C. D. Fugate and J. Hymers (London: Bloomsbury) Baumgarten, A. G. (2020) Baumgarten’s “Elements of First Practical Philosophy”: A Critical Translation with Kant’s Reflections on Moral Philosophy. Translated and edited with an introduction by C. D. Fugate and J. Hymers (London: Bloomsbury) Chance, B. and L. Pasternack (2018) “Baumgarten and Kant on Rational Theology: Deism, Theism and the Role of Analogy”. In C. D. Fugate (Ed.) Kant’s Lectures on Metaphysics: A Critical Guide (Cambridge: Cambridge University Press), pp. 214–232 Fugate, C. D. (2015) “The Unity of Metaphysics in Kant’s Lectures”. In R. Clewis (Ed.) Reading Kant’s Lectures (Berlin: Walter de Gruyter), pp. 64–87 Fugate, C. D. (2018) “Kant’s Cosmology, Miracles and the Autonomy of Reason”. In C. D. Fugate (Ed.) Kant’s Lectures on Metaphysics: A Critical Guide (Cambridge: Cambridge University Press), pp. 122–155 Kant, I. (1998) Critique of Pure Reason. Translated by P. Guyer and A. Wood (Cambridge: Cambridge University Press) Schwaiger, C. (2011) Alexander Gottlieb Baumgarten – Ein Intellectuelles Porträt. (Stuttgart-Bad Cannstatt: frommann-holzboog) Schwaiger, C. (2017) “Die Rezeption von Wolffs Deutscher Logik bei Alexander Gottlieb Baumgarten”. In A. Pelletier (Ed.) Christian Wolff’s German Logic: Sources, Significance and Reception (Hildesheim: Olms), pp. 175–196 Wolff, C. (1738) Philosophica practica universalis method scientific pertracta, 2 vols (Frankfurt and Leipzig: Renger) Wolff, C. (1739) Theologia naturalis 2nd ed., 2 vols (Frankfurt and Leipzig: Renger)

3

Lambert on the Certainty and Generality of Metaphysics and Geometry Katherine Dunlop

3.1 Introduction The topic proposed for the 1763 Berlin Academy of Sciences essay competition was a comparison between metaphysics and geometry. Competitors were specifically asked to pronounce whether metaphysical truths ‘admit of distinct proofs to the same degree as geometrical truths’, and, if they answered no, to explicate the nature and degree of metaphysics’ certainty, and say ‘whether this degree is sufficient for complete conviction’ (Walford and Meerbote 1992: lxii). The Academy’s brief did not state or imply that such highly distinct proofs, and certainty of the same nature and degree as geometry’s, are necessary to qualify as a science. Yet three of the most illustrious essayists – Immanuel Kant, Johann Heinrich Lambert, and the eventual winner, Moses Mendelssohn – began their essays by remarking on the faddishness of metaphysical theories, and suggesting that metaphysics will never achieve the status of science as long as it lacks the evidence of geometry.1 Mendelssohn explicitly considers that philosophy might be said not to ‘fulfill the requirements of a science’ (1764/1997); his own view, however, is that metaphysics is a science and indeed as certain as geometry, just not as perspicuous in its reasoning. In his essay, Kant refers to metaphysics as a science, but there are hints of the questions he will go on to raise as to how and whether metaphysics, as traditionally understood, qualifies as science.2 In contrast to both, Lambert appears to take metaphysics’ status as a science for granted; and in sharp contrast to Kant, Lambert holds that philosophy should emulate the method of mathematics. In this respect, Lambert’s view diverges less than Kant’s from that of Christian Wolff, whose claim to institute mathematical method in philosophy prompted the Academy to pose their question. But Lambert faults Wolff’s understanding of mathematical method, finding it incomplete. According to Lambert, Wolff grasps how the deductive ordering of mathematics’ propositions conduces to its certainty, but DOI: 10.4324/9781003187561-4

74  Katherine Dunlop fails to recognise how mathematics (in particular, geometry) proves the correctness of its concepts: through its use of postulates. We owe to Alison Laywine the important observation that ‘Lambert gave Kant the interesting idea that postulates in Euclid’s sense were somehow essential to the reform of metaphysics they both sought’ (Laywine 2010: 114). I propose to build on Laywine’s treatment by distinguishing different roles that Lambert finds for postulates in geometry, and considering how they can have those roles in metaphysics as Lambert conceives it. After a brief introduction (§2) to the texts in which Lambert proposes a method for metaphysics, I turn (in §3) to Lambert’s (never-completed) essay for the prize competition and a contemporaneous, thematically related manuscript. I show in §§3.1–2 that Lambert proposes to fill the gap left by Wolff, that of identifying and ratifying the fundamental concepts of metaphysics, by deriving these concepts from inner experience. This view initially seems at odds with Lambert’s conviction that metaphysics should follow geometry’s method. But Lambert takes care to argue, as I show in §3.3, that the manner in which geometry establishes its fundamental concepts is relevantly similar. On Lambert’s view, geometry establishes its fundamental concepts by means of postulates, which prescribe certain practices or activities; just like the experience of one’s own thoughts, these practices or activities can be renewed or repeated on every occasion that doubt arises. In §4, I argue that Lambert’s comparison between geometry’s method and the method he prescribes for metaphysics serves a further purpose in his later, published philosophical work. Lambert continues to make use of this comparison to explain the certainty of metaphysics’ starting point. But he now locates this starting point in simple concepts which we designate using words for physical things, rather than in specifically inner experience. As I explain in §4.1, this makes it a problem to establish the general applicability of metaphysical propositions (across what Lambert calls the “corporeal world” and the “intellectual world”). Lambert holds that metaphysics can achieve this general applicability by following the method of geometry, and in particular by having a basis in postulates. But it is difficult to see how basing metaphysics on postulates can give it the requisite generality, particularly when we approach Lambert already familiar with Kant’s view of geometrical construction, as the introduction of particular mathematical objects (in pure intuition). §4.2 argues that Lambert instead understands construction, or more generally the practice enjoined by postulates, in terms of algorithms or procedural rules. §4.3 considers Lambert’s contrast between the “general” or “unconditioned” possibility conferred by postulates and the indeterminate generality attained by abstraction, and argues that metaphysics requires the former for its general applicability. In §4.4, I show how Lambert finds examples

Lambert on Certainty and Generality of Metaphysics and Geometry 75 of such maximal generality in mathematics, in order to further establish the relevance of mathematics (and in particular geometry) to his reform of metaphysics. 3.2 Is the Method Lambert Prescribes Intended for Metaphysics? The science whose methodology is of most concern to Lambert is metaphysics. He writes to Kant that ‘whenever a science needs methodical reconstruction and cleaning, it is always metaphysics’ (10: 62). But the methodology Lambert sets out in his published writings is for other sciences, reconceived or invented by him, whose relationship to metaphysics is somewhat obscure. So the features of mathematical thought that Lambert seeks to replicate must be shown to have application in metaphysics, in particular. The main textual sources for Lambert’s philosophical views are his correspondence; the draft he composed for the 1763 essay competition, titled Über die Methode, die Metaphysik, Theologie und Moral richtiger zu beweisen [ÜM], and a related manuscript titled Abhandlung vom criterium veritatis [CV]; his Neues Organon, oder Gedanken über die Erforschung und Bezeichnung des Wahren und dessen Unterscheidung vom Irrthum und Schein [NO], published in 1764; and his Anlage zur Architectonic, oder Theorie des Ersten und des Einfachen in der philosophischen und mathematischen Erkenntnis [Arch.], published in 1771. There are significant doctrinal shifts between the unpublished (ÜM, CV) and the published (NO, Arch.) writings, which I will designate as “earlier” and “later”, respectively. Of these works, only ÜM designates its topic as metaphysics. CV concerns scientific cognition in general. Each of NO’s four parts presents a science as a ‘tool’ for ‘cogniz[ing], present[ing], and distinguish[ing] from error and appearance what is true’ (NO “Vorrede” n.p., trans. Watkins 2009: 259). The sciences address questions which Lambert says are ‘very naturally’ prompted by the clashes between schools and between the ‘main doctrines’ of ‘Aristotle, Gassendi, Descartes, Newton, Leibniz, Wolff, etc.’ (NO “Vorrede” n.p., trans. Watkins 2009: 258).3 But the “organon” which these sciences together compose seems intended, not specifically for metaphysics, but for every science that aims at truth. Lambert describes Arch. as an ‘investigation undertaken entirely anew into the fundamental disciplines [Grundlehren] of metaphysics’ (Arch. i). Lambert’s titles for the book’s divisions indicate that the book itself constitutes a single fundamental discipline. The first part is a general plan [Anlage] of such a discipline, while the second and third concern, respectively, what is “ideal” and “real” in the discipline (the fourth treats magnitude).

76  Katherine Dunlop In the Preface, Lambert informs us that the word “Architectonic” in the title is taken from Baumgarten’s treatise of metaphysics. It is an abstraction from architecture and has a wholly similar meaning in regards to the structure [Gebäude] of human cognition, at least if it is related to the first foundations, the first plans, the materials and their preparation and overall organization, in such a way that one advances toward making from them a whole fit for purpose [ein zweckmässiges Ganzes]. (Arch. xxviii–xxix) This description of “architectonic” indicates that it applies broadly to the entire “structure of human cognition” (or at least scientific cognition), just as the “organon” of the earlier work applies more broadly than only to metaphysics. Lambert’s introduction of the topic in the main text confirms its broad application. Having catalogued the simple concepts which he takes to be ‘the first foundations of our cognition’, and the particular sciences founded on them, Lambert maintains that these concepts ‘singly and combined with one another, taken together constitute a system, which necessarily contains every first ground of our cognition’ (Arch. §74). This system affords scientific cognition for which the names ‘fundamental discipline [Grundlehre], fundamental science, architectonic, basic discipline [Urlehre]’ already exist (Arch. §74). Lambert writes in a letter to Kant that ‘architectonic’ is ‘to include all that is simple and primary in every part of human cognition’ and that he does not ‘count under “architectonic” all of what one treated in metaphysics until now’ (10: 51–52). Given architectonic’s broad scope, the only obvious way that matters previously treated in metaphysics could be excluded from it is if they are insufficiently “simple and primary”, or not a “first ground” of cognition. In the passage of Arch. just quoted, Lambert is at pains to make clear that ‘ontology on a literal understanding’ is not identical with the fundamental discipline, but rather constitutes just a part of it, as do all other sciences founded on particular simple concepts or combinations thereof. He refers back to an earlier passage that identifies ontology with ‘the theory of a thing in general’, thus giving insight into how at least this part of metaphysics fails to be wholly simple and primary. There Lambert argues that the notion of thing, ‘which indeed is applied more generally, considered in itself’, is nonetheless grounded on the concept of solidity, insofar as ‘it is to signify something real and not some mere figment of the brain’ (Arch. §57). So for Lambert ontology is no longer, as in the tradition that includes Wolff’s metaphysics, the most general and fundamental discipline.4 For him, the concept of thing’s dependence on the concept of solidity makes

Lambert on Certainty and Generality of Metaphysics and Geometry 77 it – and the science dedicated to it – derivative, complex, and less than maximally general. Since the traditional understanding of ontology also equated it with “general metaphysics”, the question arises whether Lambert would also accept this identification, or whether he thinks there is a discipline more general than ontology which is still a kind of metaphysics. This question is important because the method Lambert sets out and follows in Arch. is, in the first instance, for a maximally general “fundamental” discipline.5 If this discipline cannot be conceived as a kind of metaphysics, then it will not be clear whether its method is appropriate for metaphysics. The question can perhaps be sharpened by a comparison with Kant’s “critique of pure reason”. Kant conceives critique as a ‘propaedeutic’ (A11/B25) and as ‘preparation’ (A12/B26; Bxxxvi) for cognition from pure reason, in much the same way that Lambert’s architectonic deals with the “first plans” and the “preparation of the materials” for scientific cognition.6 To the extent that these disciplines are conceived as merely preparatory, it is in question whether they belong to the reformed scientific cognition that is to follow.7 Thus, in Kant’s case, it is disputed whether the contents of the Critique count as metaphysics. Fortunately for our purposes, it is clear that Lambert conceives the fundamental discipline as a kind of metaphysics. Early in Arch. he describes metaphysics as a theory [Lehrgebäude] of the concepts and propositions that are found in all parts of human cognition, both common and scientific. He claims this theory is ‘to contain the first grounds of the whole of human cognition [der gesammten menschlichen Erkenntnis]’ (Arch. §3), which gives it the same scope as the fundamental discipline (Arch. §74). Lambert also speaks of the ‘advantages’ which the ‘universal applicability of such a theory seems to promise’ and which are already possessed by ‘algebra and geometry’ (Arch. §3); elsewhere, he elaborates the advantages ‘geometry and scientific cognition in general’ have, ‘which metaphysics, and especially the Grundlehre, should have as well’ (Arch. §14). Our concern is with the way in which geometry serves as a model for metaphysics, and I will suggest that it plays a broader role in Lambert’s later than in his earlier philosophical writings. But Lambert’s argument that the fundamental discipline must emulate the method Euclid follows in geometry, in order to preclude doubt, carries over from his earlier discussions of philosophical methodology. In Arch., more explicitly than the earlier writings, this point appears as a criticism of Wolff’s attempt to introduce mathematical method in philosophy. Here Lambert writes ‘one cannot say that Wolff employed the mathematical method thoroughly’, because ‘in his metaphysics the postulates and exercises are almost entirely lacking, and the question of what one is defining is therefore not fully decided’. He emphasises that in Euclid’s work, in contrast to Wolff’s, ‘one

78  Katherine Dunlop could not doubt the possibility of’ concepts, thanks mainly to Euclid’s use of postulates to secure their ‘general possibility’ (Arch. §12). 3.3  Postulates and the Certainty of Metaphysics’ Starting Point 3.3.1  Fundamental Concepts in Philosophy

My first objective is to show how, in Lambert’s early unpublished writings, the comparison with Euclid serves to explain the certainty of philosophy’s starting point. Towards the beginning of ÜM and CV, Lambert points out some respects in which philosophy has already been raised, in part, to the certainty he finds in geometry. As I have observed elsewhere,8 Lambert finds two main factors responsible for geometry’s certainty (which he lists in the first of the numbered notes at the start of ÜM, “Notanda” 1). The first is the tightness of the deductive linkages by which geometry’s theorems are securely grounded on the first principles. The second factor is ‘that one cannot easily err in the concepts of the figures, because they are simple [and] lie before the eyes’. Lambert observes that logic [Vernunftlehre] is already secure from error in the first of the two ways (ÜM, “Notanda” 2). Even in Arch. (where his criticism of Wolff is most explicit), Lambert allows that Wolff sought to deduce every truth (of philosophy) from axioms and definitions;9 since fulfilling this aspiration would give philosophy as a whole a deductive structure comparable to geometry’s, Wolff can be credited with having partially introduced “mathematical method” to philosophy (CV §§2–3). Where philosophy still falls short of geometry’s certainty is at the level of concepts. In ÜM, Lambert writes that the second factor responsible for geometry’s certainty ‘seems peculiar to it, at least [insofar as] it concerns the simplest and most recognizable [kenntlichst] among all sensible concepts’ (ÜM, “Notanda” 3). His commendation of Wolff is accordingly qualified: Wolff’s introduction of the mathematical method has ‘very much decreased the difficulties, to which the discovery of the criterion of truth is subject’, but only ‘so long as in addition, people can distinguish the most correct of their concepts from the more confused and doubtful ones’ (CV §22). In his correspondence, Lambert includes Wolff among the thinkers who grasp only ‘barely the half’ of mathematical method, since the method is not restricted to ‘how one should connect inferences’, but extends also to ‘how one must seek out the simplest and first in the concepts … in order to begin correctly’ (letter to Holland, 18 March 1765, IX.7).10 Accordingly, Lambert devotes the bulk of his effort to explaining how philosophy can have certainty at the conceptual level. On his view, what is needed is to identify the basic concepts [Grundbegriffe] and derive

Lambert on Certainty and Generality of Metaphysics and Geometry 79 the others [Lehrbegriffe] from them in accordance with certain rules, on analogy with the derivation of theorems [Lehrsätze] from first principles [Grundsätze] (ÜM “Notanda” 19, §44; CV §25). Throughout his philosophical writings, Lambert emphasises that we must seek “simple” concepts on which to found scientific cognition (and metaphysics in particular). Lambert shares Kant’s and Wolff’s understanding of concepts as structured in terms of “marks”. Kant counts these as concepts, calling them ‘partial concepts’ (9: 95); likewise, Lambert speaks of them as parts of concepts, while allowing that a mark can itself be a concept (Arch. §7), presumably as an improper part of that concept. By the same token, for Lambert a simple concept can be a mark, but it cannot be decomposed into distinct marks. (Lambert counts as foundational concepts those which ‘either have no distinguishable parts, and are consequently simple in themselves, or … their parts are so similar to the whole that one must necessarily cover them with the same name’ [CV §43].) One obvious way in which simple concepts are secure against error is that they cannot have any components that contradict one another. In CV, Lambert explains how the insufficiency of Wolff’s approach is rectified by finding concepts that cannot harbour contradiction. He argues that ‘Wolff’s criterion of truth’, namely the deducibility of a proposition from first principles (axioms and definitions), ‘is in the most proper understanding applicable and necessary only with respect to theorems [Lehrsätzen]’ (CV §3). It remains to find how the first principles can have the certainty that they are supposed to transmit to the theorems; Lambert claims this is by the “Cartesian” criterion of correctness of concepts.11 Lambert then argues that ‘as the Wolffian mark of truth is analyzed [auflöst sich] into the Cartesian one, the latter is further analyzed [auflöst sich] into the question, whereby [woran] can one cognize, that a concept does not contradict either itself or other concepts’ (CV §7). Having thus traced the certainty of theorems to the logical consistency of the concepts that comprise first principles, Lambert elaborates that ‘it is not enough that one sees nothing contradictory in the concept, rather one must prove that there neither is nor could be anything contradictory in it’ (CV §7). He contends that ‘the way one has so far taken’ in regard to this problem is not sufficient. That is to analyse a given concept into its components and ‘more exactly develop’ these. If it is found that ‘one of them has such a mark that it negates just that which the mark of another affirms in the same sense’ (CV §8), the question of consistency is settled; but if not, ‘the question of whether anything more that is contradictory remains’ (CV §10). In Arch. §7, Lambert attributes this way of proceeding to Leibniz in particular, and distinguishes between the case in which ‘the analysis continues into the infinite’ (leaving the latter question undecided) and the case in which one arrives at simple marks. While Lambert is less explicit in his

80  Katherine Dunlop early manuscripts, he clearly holds that the only way to ensure avoidance of contradiction is to begin with simple concepts. Lambert also holds that beginning with simple concepts avoids the regress that ensues in trying to justify any non-self-evident starting point. In Arch. §22, he contends that such a regress arises on Wolff’s approach, and in CV §74 he describes a similar regress as the ‘difficulty’, concerning ‘the long-sought criterion of truth’.12 3.3.2  Inner Experience and Certainty

In Arch. §8, Lambert explicitly formulates the problem of ‘seeking out every single concept that is simple’, as distinct from that of stating the criteria [Kennzeichen] of conceptual simplicity (which is a problem for ‘the general theory of concepts’). While the early manuscripts do not separate these problems so sharply, they do specify where simple concepts can be found. Lambert repeatedly invokes ‘the Cartesian Cogito, ergo sum’ as a starting point that is invulnerable to doubt, as when he argues that it avoids the just-mentioned regress, because the concepts occurring in it are of the sort that require no proof (CV §76); he also characterises them more positively, as such that ‘their possibility and correctness are immediately evident as soon as one represents them’ (CV §36). Generalising from the example of the cogito, Lambert holds that every concept ‘the soul has of its own thoughts, actions, and sensations’ is simple (CV §43), and that this is the origin of all fundamental concepts (ÜM “Notanda” 14, §20). It would be open to Lambert to hold that we acquire these concepts through a rational or intellectual process of reflection, of the sort that Leibniz seems to mean by his term “apperception”. But Lambert makes very clear that he takes them to originate in experience of a particular kind; on his view, their origin in inner experience explains their certainty. In ÜM, Lambert gives an account of the fundamental concepts’ origin in the course of his argument that metaphysics can share the certainty of geometry and logic. His first objective is to show that logic ‘deserves just as much glory as’ geometry, specifically that it shares geometry’s ‘distinctness’ and compels assent as least as forcefully (ÜM §18). Lambert explains how logic’s fundamental concepts give it these advantages: §20. [T]he fundamental concepts, on which [logic] bases its propositions and on which its first principles rest, can have a verification [Prüfung] just as immediate as we first ascribed to the propositions. For each of these concepts, one may merely think back on oneself, and each time one either sees it originate in one’s soul, or one finds it in the consideration of one’s own thoughts.

Lambert on Certainty and Generality of Metaphysics and Geometry 81 Lambert then concedes that ‘all these concepts are in this manner empirical, insofar as they are not composed from others’ (ÜM §21); I think this just serves to restrict his claim to the fundamental concepts of logic, rather than to leave open a non-empirical origin for logical concepts. Lambert clarifies that ‘occur[ing] as empirical concepts’ suffices ‘for certainty, because the experiences are such that one repeats them also [when] in doubt and therefore must admit them in the rejection [of them]’ (ÜM §26).13 The penultimate step of Lambert’s argument is his observation that ‘the fundamental concepts of logic are already in themselves metaphysical concepts’ (ÜM §21), and the final step is showing that fundamental metaphysical concepts which do not occur in logic (such as order, completeness, and degree) also originate in one’s own thoughts (ÜM §§30–31). In CV §45, Lambert states explicitly that ‘all fundamental concepts in general are actually [eigentlich] empirical concepts’. Here he gives a fuller account of the certainty of metaphysics’ basic concepts, still in terms of the repeatability of inner experience. He is specifically concerned to show how metaphysics can withstand the doubts of ‘an egoist’, for which purpose it must begin with the distinction between simple and composite thoughts. I take the concepts “simple” and “composite” and posit that, for the sake of introducing them into metaphysics right at the beginning, one must not abstract them from outer things. For then an egoist would not admit them. Since however [they] must be proved not only for a little while, but rather rigorously, thus I have indicated how these concepts are derived from inner sense [aus dem sensu interno], since they occur in every case with thoughts, as there are simple and composite thoughts. (CV §80) Shortly afterwards, Lambert contends that since metaphysics must resist ‘even the greatest doubters’, it ‘is always more advisable and necessary to take its fundamental concepts from our inner sensations, such as we again have immediately at hand in every case of doubt’ (CV §85). Lambert’s references to “inner sense” and “inner sensations” make clear that the source of fundamental concepts is specifically a kind of experience that can always be ‘again renewed’ and lies within one’s own thoughts (see, for instance, CV §92, item 2). In ÜM and especially in CV, where he is more concerned with previous approaches to the ‘criterion of truth’, Lambert understands the incorrigibility of ‘the Cartesian Cogito, ergo sum’ in terms of such repeatable inner experience, ‘for whoever would put it in doubt or nullify it, always comes back to it’ (CV §76; ÜM §38). Lambert gives Wolff credit for basing his own metaphysics on such a secure starting point (as in the opening

82  Katherine Dunlop paragraph of Wolff’s German Metaphysics), which incidentally shows that Lambert does not think Wolff’s attempt to derive all truths from axioms and definitions forecloses the possibility of empirical justification.14 What Wolff misses, on Lambert’s view, is how this experience’s justificatory power is due to the simplicity of the concepts occurring in it; accordingly, Wolff misses the need for an ordering of concepts, from simple to complex, parallel to the deductive ordering of propositions. 3.3.3  Certainty and Practical Principles in Geometry

Given Lambert’s ultimate aim of showing that metaphysics can have the same certainty as geometry, his emphasis on the empirical origin of metaphysics’ fundamental concepts may seem out of place. But in the passages just cited, Lambert takes care to argue that the fundamental concepts of geometry are shown possible by similar means, though without saying that these concepts are empirical. In comparing logic with geometry in ÜM, just after conceding that the fundamental concepts of logic are empirical, Lambert adds: ‘Only here too geometry has no comparative advantage, and Euclid himself sets the possibility of a straight line among the postulates’ (ÜM §21; see also §CV 18). The implication is that we become certain of the straight line’s possibility through experience. In his account of how metaphysics can be made secure from even an egoist’s doubts, Lambert anticipates the objection that ‘the determination or definition of these concepts [“simple” and “composite”] becomes too narrow, because they are not abstracted from all particular things that are simple or composite, rather only from thoughts’ (CV §80). In reply, Lambert argues:15 one does better if, from the beginning, the concept extends no further than it is proven. Just here lies that which I above called the mode of generation of a concept. The egoist sees [the concept] arise, as it were, in his own soul, and must contradict himself if he refuses to admit it. This has much in common with Euclid’s proof of the possibility of an equilateral triangle and is of the same strength. To better understand how the method of metaphysics can be similar to that of geometry, we can turn to Lambert’s account of how Euclid makes geometry invulnerable to doubt. On Lambert’s telling, Euclid aimed specifically to ward off the doubts of ‘the Sophists’, who were to him ‘approximately’ what an egoist is (to philosophers of Lambert’s time), and ‘according to dialectic would grant nothing to him’ (CV §80). Lambert recounts being surprised that the first proposition of the Elements (I.1) is a problem, which presents the reader with something to make or do, rather than a theorem, which is a claim needing proof: ‘What is this, I

Lambert on Certainty and Generality of Metaphysics and Geometry 83 thought, should not the theory precede, before one puts it into practice? Only Euclid presumably thought much further’ (CV §79).16 The problem is, specifically, to construct an equilateral triangle having a given segment as its base. As Lambert explains, for this Euclid ‘needs only one line, and since this can vary only in respect of length, so in this way he puts the possibility of the equilateral triangle before the eyes of those who would put it in doubt, that they can themselves refute [the doubt], in that he shows them how such a triangle can be made from [or “of”, von] any given magnitude’. The ‘exercise’ thus shows that in the theory ‘no merely imaginary figures are considered, rather only possible ones’. In contrast, ‘if for proof [Euclid] would have already drawn some figures’, then one who would deny the figures’ possibility could still ‘put their generality in doubt’: ‘since for a triangle three lines are required, as its sides, nothing would have been easier than to put before him three lines, of which one is greater than the other two taken together’ (CV §79). Such lines do not satisfy the condition under which, as Euclid shows in Elements I.22, a triangle can be constructed from any three lines. Lambert concludes that ‘the trick of it lay in … that the possibility of the equilateral triangle manifested so to speak of itself. One cannot better refute someone who believes something to be impossible than if one shows him how he himself could effect it [ins Werk setzen]’ (CV §79). He urges that this ‘means’ should be employed for the proof of all concepts that are to be assumed as basic [Grundbegriffe], claiming that ‘Vernunftlehre at least seems to attain its certainty in this way’ (CV §79). Lambert refers back to CV §43, which gives Descartes’s cogito as an example of knowledge attained through simple concepts (in Vernunftlehre). What the comparison with Euclid’s method shows about the distinctive certainty of that which one must “come back to”, even in doubting – such as the cogito – is that the doubter must “effect” it in the course of, or for the sake of, some activity or practice. Lambert generalises this point by insisting that “practical” elements must have equal standing with “theoretical” ones in the deductive structure of any science (insofar as it is to be certain), as they do in Euclid’s Elements. As briefly noted above, the propositions proved in the Elements are of two kinds: theorems and problems. They are distinguished by the lines that conclude their proofs: those of theorems end with ‘(Being) what it was required to prove’ (quod erat demonstrandum), and those of problems, with ‘(Being) what it was required to do’ (quod erat faciendum). This stylistic difference naturally suggests a classification of the propositions as theoretical and practical, respectively. Lambert holds that the theoretical/ practical distinction can also be drawn at the level of the first principles, between the two groups of principles listed under the headings “common notions” (or “axioms”) and “postulates”.17 For Lambert, it is crucial that the

84  Katherine Dunlop first principles must include at least some practical ones. This means that the theory must include postulates, understood as practical propositions whose solutions do not depend on any further conditions or assumptions, as well as the problems that stand to them as theorems to axioms. The postulates … are actually problems [Aufgaben], whose solutions must be admitted as soon as one understands the words. Problems are, however, a kind of questions, in which something to find or to do is put forth. The solution shows how one is to do it [wie … es machen solle], and the proof shows that it really [wirklich] can be accomplished in this way. (CV §50) The questions either require a solution, and then they are called problems, or one admits their solution in all rigor as possible and doable [thunlich], because one knows and comprehends it as soon as one understands the words, and these are called postulates. (CV §51) If no practical proposition qualifies as a first principle, then every solution to a practical proposition will require more than just “understanding the words”, and we lack a guarantee that these additional requirements can be met. Thus, to ensure that a theory contains elements that show their certainty by being put “in effect”, the theory must be based on postulates as well as axioms. We will see that throughout his philosophical writings, Lambert continues to emphasise the importance of postulates, which he still conceives as first principles and as practical. But as his conception of metaphysics’ starting point changes, it becomes harder to see how giving metaphysics a basis in practical principles can raise it to the certainty of geometry, and give it the universal applicability required of the fundamental discipline. 3.4  Postulates and the Generality of Metaphysics 3.4.1 The Problem of General Applicability in Lambert’s Later Writings

I take myself to have shown how Lambert’s comparison between metaphysics and geometry explains the certainty of their starting points, specifically in terms of an activity or a process that is always ‘immediately at hand’ (CV §85), and can thus be repeated whenever doubt arises. What remains to be investigated is whether and how his comparison between the methods of metaphysics and geometry extends further. As I will now

Lambert on Certainty and Generality of Metaphysics and Geometry 85 explain, Lambert takes metaphysics, in its role as the fundamental discipline, to face a problem of applicability. In his earlier writings, he is explicit that this problem does not arise for geometry, and the means to its solution are not geometrical. But in his later writings, where the problem takes a different form, he finds a solution to it precisely in geometry’s use of postulates. So in NO and Arch., the comparison of metaphysics’ method with geometry’s has a further purpose than accounting for the certainty of their starting points. We will be concerned to understand how it can accomplish this further goal. As we have seen, in ÜM and CV, Lambert locates philosophy’s starting point in the soul’s consideration of its own thoughts. There naturally arises the question of how knowledge grounded on such inner experience can be extended to things outside of the soul. As Christian Leduc emphasises in his study of ÜM, no such question arises for geometry, which remains a merely ‘ideal’ science (Leduc 2016: 217). In metaphysics, this problem of general applicability is solved by distinguishing appearance from “the truth”, on the model of astronomy, thus employing resources beyond the method borrowed from geometry.18 In NO and Arch., Lambert continues to emphasise the importance of simple concepts. Since Wolff adopts the ‘Leibnizian analysis of concepts’, the inconclusiveness of this method (Arch. §7) remains a problem for his philosophy (Arch. §11); it is corrected by beginning with simple concepts. Lambert now, however, credits Locke with showing how to begin (Arch. §14).19 Locke’s approach, as Lambert describes it, is to ‘take up our cognition, as [so wie] it is’; separate out the ‘abstract and even merely symbolic’ from ‘what is called clear representation and [an] actual concept’; ‘observe, for every sort of concept, which senses and impressions it arises from [zu danken haben] and which arise from mixed impressions’; and finally ‘separate the simple from the rest, and put [the simple] in certain classes’ (Arch. §9). Clearly, in taking cognition “as it is” and investigating its sources in inner and outer sense, Lambert is no longer restricting the starting point to specifically inner experience. Accordingly, he no longer seems concerned with the doubts of “egoists” (and it becomes easier to see how the concepts of colours and sounds can be fundamental [Grundbegriffe], as Lambert already claims in ÜM §§42–3). Lambert also indicates that it is no longer necessary to follow the method of astronomy, according to which ‘concepts taken from appearance are always laid at the ground, until one can find out so much from them, that the real and the true can be determined thereby’ (Arch. §43). In a letter written (to Georg von Holland) in 1765, after reading Locke, Lambert gives a fuller account of how to base metaphysics on this broader starting point. In metaphysics, ‘one cannot set before one’s eyes the thing

86  Katherine Dunlop itself, which is abstract, but must rather be content for the most part with word and concept’. The word, however, ‘is almost always taken from the corporeal world and made metaphysical. And one does well to follow this trace’ (IX.30–31). In Arch., Lambert suggests it is inevitable that words come from the ‘corporeal world’, because it is better known (than the intellectual world).20 There thus arises a new problem of applicability: of how concepts given through terms whose significance is exclusively corporeal can apply broadly to the corporeal and intellectual worlds, so that metaphysics (in the guise of Grundlehre) can have the generality required of it. In Arch., Lambert explains that this problem is solved by formulating the fundamental discipline in terms of “transcendent” concepts. The terms for things of the intellectual world are derived from things of the corporeal world, insofar as they share a resemblance with them according to our manner of representation, and if we designate both by the same name, then the abstract concept, which is combined with that name, is transcendent. The word “power” [Kraft] may serve as an example. Originally it is derived from motive powers in the corporeal world, insofar as something can be brought about by them. On account of the similarity of the manner of representation, however, we also count the understanding and the will as [eignen … zu] powers … [T]he concept of power thereby becomes not just more general, but rather entirely transcendent, because it occurs with things that have nothing at all in common with one another. (Arch. §29) While Lambert held in CV and ÜM that metaphysics must at first restrict itself to what he here calls “things of the intellectual world”, he now contrasts two ways in which the fundamental discipline can proceed: either by considering each species of power separately or by making the theory of powers ‘wholly transcendent, so that it remains applicable to every species’. It is best, Lambert claims, if one ‘combines the transcendent theory right at the beginning with the special one, division by division [eintheilungsweise]’, so that the theory’s applicability is apparent from the beginning (Arch. §29). Lambert’s choice of “power” as the example of a transcendent concept is presumably not arbitrary, for this concept has a special status in his metaphysical system. Whereas in CV Lambert claimed that postulates establish their solutions as “doable” [thunlich], in Arch. he claims that they

Lambert on Certainty and Generality of Metaphysics and Geometry 87 establish only possibility of a more general kind. To establish “doability”, the “theory of powers” is also required:21 And whoever grants to Euclid his postulates and especially their generality, must necessarily grant him … still more distinct possibilities. Since with every composite concept the generality, and with arbitrarily composed ones the possibility, must be discussed, thus the postulates actually [eigentlich] occur only with simple concepts, and hence they must be brought forward with every simple concept. … For actual doabilities [wirklich Thulichkeiten], the theory of powers must specify the foundation. (Arch. §20) The concept of power is important, in particular, for the transition from the “ideal” to the “real” part of the fundamental discipline. Lambert begins his treatment of this part by remarking that he has so far considered ‘powers and the solid’ only so far as was necessary ‘to indicate that ideal relations were not a mere dream, but rather the rudiments [Anlage] for them occur in things themselves’. Now that the rudiments come into consideration, ‘we begin with powers, for all of the connections, real relations, compositions, positive possibilities, etc. occurring in the things themselves depend on powers’ (Arch. §372). But what is most important for us is the general methodological point that concepts taken “transcendently” should be considered in conjunction with their species, “division by division”, at every stage of the theory. This turns out to be crucial for metaphysics to attain the generality it is supposed to have (in its role as fundamental discipline). As we will see in detail, Lambert maintains that metaphysics can achieve this generality by following the method of geometry, and in particular, by having a basis in postulates. This part of his view presents a puzzle, because Lambert continues to regard geometry as an “ideal” science (Arch. §80). So in metaphysics, postulates somehow fulfil a need (guaranteeing applicability to real, not only to ideal, things) that does not even arise in geometry. 3.4.2  Construction as Geometry’s Characteristic Activity

In order to understand how postulates play this role, we must take care not to assimilate Lambert’s conception of postulates to Kant’s view. The example Kant chooses to illustrate the constructive methodology that he considers distinctive of mathematics is Elements I.32 (A716/B744), which suggests that Euclid’s method of proof informs his view in the same way it informs Lambert’s. The postulates of the Elements, as they were known

88  Katherine Dunlop to Kant and Lambert,22 uniformly enjoin the practice or activity of constructing particulars (lines that extend given lines or join given points, and circles with given centre and radius). Interpreters such as Jaakko Hintikka understand Kant’s notion of construction in intuition in terms of certain steps in the proofs of Euclid’s Elements that apply these constructive procedures to produce particular lines, points, and so forth.23 Seen against this background, Lambert’s account of geometrical method as based on repeatable activity or practice appears to centrally involve the introduction of particular objects. Our understanding of “constructive” proof may also lead us to emphasise the introduction of particulars, since a proof qualifies as constructive by supplying either an object that verifies a general claim or at least an algorithm for finding such an object. Furthermore, there is direct evidence that Lambert identifies the practice or activity characteristic of geometrical reasoning with construction. He states in the Preface to NO that ‘he has used the expression, typical of geometrical demonstrations, of per constructionem, in order to point out the similarity’ of his method to that of geometry (n.p., trans. Watkins 2009: 260). This understanding of the practice or activity characteristic of geometry lends itself to Lambert’s project, in CV and ÜM, of showing that metaphysics’ fundamental concepts are “proved” in just the same way geometry’s fundamental concepts are shown to be possible, specifically by means of postulates. Indeed, in the course of arguing that geometry has no ‘comparative advantage’ over logic and metaphysics, despite the empirical basis of their concepts, Lambert claims that Euclid ‘sets the possibility of a straight line among the postulates’ (ÜM §21). For the analogy between metaphysics’ method and geometry’s is very close on the supposition that metaphysical concepts are proved by supplying ourselves with particular thoughts and sensations, just as the possibility of geometry’s concepts is shown by introducing particular lines and figures. But on this understanding of geometry’s characteristic activity or practice, it is difficult to see how Lambert can sustain the analogy in NO and Arch. One way to bring out the difficulty is to compare Lambert’s position with Kant’s. Though Kant refrains from describing geometry as ‘ideal’, he contrasts the ‘pure’ intuition (of space) that presents its objects to us with ‘empirical’ intuition, in which ‘the real’ is represented by means of sensation (B146–7). The fact that pure intuition also gives us the ‘form’ of sensibility appears crucial for explaining how general solutions are achieved through the use of particular constructed figures.24 But since determination of pure intuition gives us cognition of only the form (space) in which objects could be presented, mathematical concepts count as cognitions of objects only insofar as ‘there are things that must be intuited in this form’ (B147). Kant thus faces a question concerning how far geometry can be known to apply to real things, and in some passages at least, his answer

Lambert on Certainty and Generality of Metaphysics and Geometry 89 seems to be that experience of real things involves the very same constructive activity as mathematical cognition.25 Kant typically puts this point in terms of the “synthesis” by which real things are apprehended, and at least once he explicitly claims this synthesis involves construction (A224/B272). Since Kant relies on the association between construction and the synthesis through which real things are apprehended to show that the categories of quantity apply to objects of experience,26 it would appear that on his view, at least this much of metaphysics (insofar as it is genuine science) is proven applicable in the same way as geometry. We thus see how, in the context of Kant’s idealism, the problem of applicability can be solved by making real things subject to construction. But Lambert does not suggest that the real things to which metaphysics and other sciences apply also somehow belong to the ideal realm to which geometry is restricted (as Kant’s “empirically real” objects are “transcendentally real”), so it seems he cannot solve the problem this way. We are not forced to conclude that Lambert first understood geometry’s characteristic activity or practice in terms of the construction of particulars, however. For even in CV and ÜM, we find a broader understanding of the practice enjoined by postulates. This broader conception is in evidence in Lambert’s account of problems as ‘a kind of questions, in which something to find or to do is put forth’ (CV §50). Lambert is concerned to argue that problems are not genuinely practical unless they are solved by means of commands or rules: [O]ne does well to distinguish what is called a proposition [Satz] in the most proper sense from questions, rules, commands [Befehlen], prescriptions [Vorschriften] and so forth. If one asks what a thing is, does, has, etc. then the answer takes the form of a proposition. If one asks what to do [was zu thun seye] then the answer is a rule, prescription, command etc. This constitutes the difference between theoretical and practical problems. (CV §50) Lambert gives two examples to illustrate this distinction. If one is tasked with finding the relation of the side of a square to its diagonal, ‘then the solution is, it is as 1 to the square root of 2’. This problem ‘is absolutely theoretical, because the solution contains no rules, but rather consists of inferences’. But if one is tasked with finding the square root of any given number, then ‘the solution consists of rules’ and shows ‘what to do [was zu thun seye] to find the sought root in each case’ (CV §50). This shows at least that Lambert’s understanding of the “practical” in mathematics coincides with our understanding of “constructive” proof in that both require a procedure or an algorithm, not necessarily a particular instance.

90  Katherine Dunlop In fact, Lambert’s “rules, prescriptions, commands, etc.” may fail to introduce (fully determinate) particulars. The procedure for extracting square roots typically does not terminate, in contrast to the construction of lines and circles, which terminates in particulars that are a basis for further constructions (as in Elements I.1).27 Lambert indicates that for him, this difference is irrelevant. I do not doubt that Euclid also had a method for extracting the root, as far as he wished. But he must also have known that one never thereby comes to the end and also never fully attains the root. The complete acuity which he sought therefore remained in the realm of possibility, and this made it so that he set the extraction of roots among the postulates. One must admit that the root and its discovery, considered as such, are possible, as one also admits in respect of the geometrical line. (CV §49).28 Hence for Lambert, the “rules” that comprise the solution to the problem of finding the square root of any given number are just as exemplary of geometrical activity or practice as the construction of lines and circles. This strongly suggests that he does not require or assume that geometrical postulates enjoin the production of particulars. There is also explicit acknowledgement in Lambert’s writings that the familiar understanding of construction (in terms of introducing particulars) is not entirely apt for his purposes. In a letter (to Sulzer, quoted and translated in Basso 2012: 67) he writes: ‘for lack of another technical term [term d’art], I felt induced [entrainé] to use’ the expression per constructionem in NO. 3.4.3  Lambert’s Account of “General and Unconditioned” Possibility

We may observe that in the passages where the expression per constructionem occurs in NO (Aleth. §173 and §200), no particulars are produced. Rather, certain very abstractly described cases are stipulated at the outset to satisfy certain requirements. Instead of going into those examples in detail, in this section I will try to give a more general account of the use of postulates that Lambert has in mind. We may begin by considering Lambert’s account of postulates as expressing “general and unconditioned possibilities”. Martin Hammer has recently drawn attention to Lambert’s claim that postulates contain such possibilities ‘in the realm of truth’ (NO Aleth. §246), where metaphysical concepts are so to speak located (Arch. §511). I will draw on Hammer’s account of this distinctive kind of

Lambert on Certainty and Generality of Metaphysics and Geometry 91 generality in order to explain Lambert’s view that giving metaphysics a basis in postulates, understood as practical principles, can make it maximally general. Lambert is at pains to distinguish the required kind of generality, which Hammer aptly calls ‘possibility generality’ [Möglichkeitsallgemeinheit], from generality attained through abstraction [Abstraktionsallgemeinheit]. As Hammer puts it, for Lambert the ‘cardinal error’ of metaphysics up to his time is the belief that one can find the simple, or at least approach it more closely, through abstraction (Hammer 2022: 164, referring to Arch. §515). Lambert observes (ironically, on Hammer’s reading) that ‘a concept from which many determinations are left aside is, precisely through [this abstraction], no longer as composite, but rather simpler than before the abstraction’ (Arch. §515); but this simplicity is of the wrong kind. According to Lambert, the aim of abstraction is to separate out, from particular concepts, what is general ‘in accordance with similarities’ (Arch. §516), so as ultimately to reach the highest concept in a genus–species hierarchy. He takes it that ‘in metaphysics there is complete agreement that the highest genus and consequently the most general concept is something and nothing, thing and non-thing’ (Arch. §517). Lambert first charges that this procedure ‘abstracts away from that which one actually [eigentlich] seeks in the complete analysis of a concept into its simple marks’. For, ‘since the simple is opposed to the composite, but both occur in things, thus one moves higher toward the general concept of a thing and accordingly represents it such that neither the simple nor the composite, rather only the possibility of one or the other remains within it, as it were as a Fundamentum divisionis’ (Arch. §517). By the same token, ‘the concept of a thing in general, or in the generality with which it is understood in metaphysics, is strictly speaking not simple’, since one must take all ‘Fundamenta divisionis’ and also ‘Fundamenta subdivisionis’ along with it (Arch. §521).29 As Hammer notes, Lambert sets forth his alternative to abstraction in the same terms in which he characterises postulates. According to Lambert, ‘instead of general similarities, by which things are stepwise distinguished and divided into species and higher genera’, we are to seek out ‘general and unconditioned possibilities’ (Arch. §523). “General and unconditioned” possibility is precisely what Lambert says the postulates guarantee. He maintains that for Euclid, concepts are represented ‘completely and purely’ by figures (diagrams), but their ‘general possibility’ is secured by Euclid’s ‘postulates, which represent general, unconditioned, and thinkable-in-themselves, or simple possibilities, or doabilities [Thulichkeiten], and which he brings forth in the form of problems’ (Arch. §12).

92  Katherine Dunlop Lambert gives an example to illustrate the difference between “general similarities” and “general and unconditioned possibilities”. This latter generality is now markedly distinguished from the former, because one takes the former such that it extends to [gehe … auf] all things; on the other hand, the latter has its proper subject, and it is unrestricted with respect to this [subject]. For example, that a body set in motion has a direction and velocity is a proposition which is general in the first regard, because therein all moving bodies are similar to one another. On the other hand, that a body can be set into motion along any direction and with any velocity, is [sic] a generality of the other kind, or an unrestricted possibility. (Arch. §523) Lambert elaborates that generality of the former kind (“general similarity”) ‘applies to [geht … auf] the subject, such that one says: All A are B’, whereas generality of the latter kind (“unrestricted possibility”) applies ‘to the predicate, so that one says: A can be B, according to every modification of B’ (Arch. §523). Although Lambert does not here bring in Euclid’s postulates as his model, he must have been impressed by the fact that they are general in just this respect; the clearest example is the third postulate, which reads “To describe a circle with any centre and distance”. Lambert says of ‘general and unconditioned’ or ‘unrestricted’ possibilities not only that they are general with respect to their predicates, but that each ‘has its proper subject’, with respect to which it is unrestricted (Arch. §523). This marks a contrast with “general similarities”, the subjects of which must be considered common to other (possible) propositions, which express the relations (of similarity and difference) that locate things within a genus–species hierarchy.30 Taking a step further, we may conjecture that for Lambert, the subjects of “general and unconditioned” possibilities are not represented through the “special [special] concepts” within which similarities are found through abstraction.31 Something along these lines is suggested by Lambert’s contrast between how we would proceed in geometry if we followed the procedure that is usual in metaphysics, and how geometry does proceed. According to the local metaphysical order, one would have to begin in geometry by seeking out that which is common to all figures, e.g., every figure is extended, has a magnitude, boundaries, etc. But in geometry one does not even let this come to mind; rather, one begins with points, lines, and angles, as with the simplest elements,

Lambert on Certainty and Generality of Metaphysics and Geometry 93 lays down their axioms and postulates, and then investigates which figures can be generated from them, how far their possibility extends etc. (Arch. §524) Although Lambert thus contrasts the way that metaphysics has heretofore “sought what is common” with geometry’s procedure of beginning with simple elements, by this point it is difficult to see the relevance of geometry – and of postulates, in particular – to his reform of metaphysics. Once we distance ourselves from the familiar conception of postulates in terms of construction, we can indeed discern a connection between the practical character of postulates, as pertaining to practice or activity, and the sort of generality they express, in terms of predicates. For predicates such as “being set into motion” express actions we can take, or processes we can make things undergo. But it is harder to understand how exactly generality of this sort is less conditioned or restricted than generality with respect to the subject. In this chapter, I will not try to explain this difference, but merely point out that generality in respect of determinations that distinguish different objects (e.g., being in motion or at rest, or moving with this or that velocity or along this or that path) is presupposed by the notion of a thing in general. For this notion must be thought as containing all the “Fundamenta divisionis” and also “Fundamenta subdivisionis”, but only in a hazy and indeterminate, or, at best, an anticipatory manner. (I think this is at least part of what Lambert intends when he describes the ‘general concept of a thing’ as ‘as it were a skeleton, general image, impression, silhouette’ of an individual [Arch. §522], while also holding that this concept inevitably reaches back to the notion of an individual [Arch. §520]).32 This raises an acute problem for metaphysics because it must begin, according to Lambert, with simple concepts designated by names for corporeal things; if it then ascends by abstraction to the notion of a thing in general, it will lack the specifications or determinations by which it can also apply to things in the intellectual world. But generality with respect to determinations that distinguish different objects is precisely what postulates secure, on Lambert’s view. So it can at least be said that the use of postulates puts in place, all at once, generality of the kind that metaphysics seeks but has so far lacked the means to secure. 3.4.4  The Completeness of Division in Geometry

Understood in this light, the overall point of Lambert’s contrast between “general similarities” and “general and unconditioned” (or “unrestricted”) possibilities is that the former are at best a fallible means to the

94  Katherine Dunlop “lawlike” and maximally general order that the latter already involve, or put in place. [I]n scientific endeavours, one will have a beginning, and therein, as in the realm of truth, the composite concepts will occur as predicates, before they occur as subjects. One does not proceed thus in metaphysics, however; rather, right at the beginning one takes as subject the concept of a thing, which, as we have previously seen, is so highly composite that one hardly finds an end in the analysis. … The division [Eintheilung] of things into species and genera is as it were a merely local order; on the other hand, the lawlike order begins with the simple and unconditioned possibilities, and precisely thereby goes in a completely different way. (Arch. §523) This gives us a way to understand how geometry, in particular, can be Lambert’s model for a reformed metaphysics (as based on postulates that make it maximally general). For Lambert frequently turns to geometry for illustrations of what he calls “lawlike”, as opposed to “merely local”, order.33 My final objective is to clarify this use of geometry as a model by showing in more detail how geometry exhibits this maximally general order. To this end, it will be helpful to keep in mind that geometry is not limited to basic results about circles and triangles. As we saw above (in §4.2), Lambert takes Euclid’s procedure for extracting square roots as exemplary of geometrical practice or activity. In the continuation of the passage just quoted, he uses the example of the numerical expression of √2 to illustrate lawlike, in contrast to local, order. It is, accordingly, highly possible that one can begin with the latter and proceed step by step, whereas with the former the order [is] only piecemeal; on the whole, however, there is simply an absolute disorder. For this would be just the same [eben so viel] as if one were to determine the places of the numbers in the square root of 2 according to their similarity. Every place has something particular, which no general rule will allow [zulassen], notwithstanding that the entire decimal series 1.414213562373095048 etc. is found and formed according to one, and indeed very simple, rule. (Arch. §523) This example illustrates, in particular, the power of a rule to determine a complete run of what we can call special cases (the digits appearing at each place in the decimal expansion). Lambert also uses trigonometry as

Lambert on Certainty and Generality of Metaphysics and Geometry 95 an example of such global or complete determination,34 and in the opening passages of Arch., he uses this example to illustrate the advantages that ‘geometry and scientific cognition in general’ have, ‘which metaphysics, and especially the Grundlehre, should have as well’ (Arch. §14). His account of these advantages provides additional evidence that geometry already realises the sort of generality that metaphysics strives towards. Lambert elaborates the advantages that already belong to geometry just after he characterises Locke’s simple concepts and Wolff’s deductive method as needed complements to each other’s approaches (Arch. §14). He takes the opportunity to further criticise Wolff for conceiving the advantage of scientific cognition too narrowly, merely as the ‘conviction and certainty to which it gives rise’ (Arch. §15). One example of a desirable feature overlooked by Wolff is that ‘all that can be expressed concisely [in die Kürze gezogen] and generally is actually expressed concisely, so that one need not begin anew in each particular case’, and Lambert claims mathematics again supplies many examples (Arch. §16).35 Lambert finds specific examples of these features of scientific cognition in the study of spatial figures, and especially in trigonometry, in which ‘all the cases are enumerated whereby [wie] from three elements [Stücke] of a triangle one can find the remaining three’ (Arch. §15). In this context, the six “elements” of a triangle are its three sides and three angles. The completeness with which trigonometry enumerates the various cases turns out to be crucial for fulfilling the requirements on scientific cognition, as Lambert indicates in a subsequent paragraph. It is undisputed that a correct enumeration of cases, classes, etc. contributes enormously to distinctness, order, and especially completeness and trustworthiness. The concepts of genera can, of course, be more correctly determined if one has before oneself an enumeration of the species. … What is applicable to divided [eingetheilten] concepts can also be immediately applied to the taxa [bei den Gliedern] of the division. The specific [specialere] determinations thereby obtained can at once be specified in detail, and the taxa [Gliedern] [into which] the division [is divided] in one respect can be compared with the taxa [into which] the division [is divided] in another respect, [to find] to what extent they can occur together in individual cases. In this manner are many of the above-mentioned (Arch. §§15–17) requirements on Grundlehre satisfied. (Arch. §34) Lambert also links the “correct enumeration of cases, classes, etc.” to the application of scientific cognition to real things. I do not think he is making the trivial point that a maximally general classification or ordering

96  Katherine Dunlop will capture everything, real things included. Rather, Lambert holds that a “correct” classification into species and genera shows the applicability of “ideal” elements to “real” things by making it possible to determine how far the properties articulated in the theory ‘can occur together in individual cases’ (Arch. §34).36 We can see this by drawing on NO’s chapter on division (which Lambert frequently cites in Arch.), since Lambert indicates that the desiderata of concise expression and minimisation of assumptions are largely or wholly realised by the correct “division” [Eintheilung] of concepts into species and genera (Arch. §34). Lambert claims, with respect to the “determinations” on which divisions can be based, that it is required that only those determinations be taken together as are capable of existing together [beysammen seyn können]. The variation [Abwechslung] which remains possible therein is the ground of the division, since one thereby attains as many species of this characteristic [Merkmaal] as there are possible variations. (NO, Dian. §83) The issue of how much variation (in a given respect) can occur in the objects falling under a given concept – which we now see to be the “ground” of genus/species groupings – can be addressed through the methods of geometry, as Lambert indicates much later in this part of NO. We have already remarked that one can often be content, to begin, with the mere possibility of a concept, still not knowing in advance how far it extends, and to what extent the characteristics [Merkmaale] that accompany it in a given case can [do so] in every case, and which other determinations they accompany. For example, a single triangle is enough to give us the concept, that a triangle is possible; but how far it is possible can be made out [only] by closer consideration, whereby one seeks [to find] how far its sides and angles allow of variation. In Euclid’s Elements of Geometry one finds plenty of propositions which are actually there only to fix [festzusetzen] the possibility of figures, which remains wholly undetermined in the definitions. (NO, Dian. §695) An obvious example of such a proposition is Elements I.22, which sets a limit on the variation that a triangle’s sides can undergo (in respect of length), namely that the triangle is possible only as long as no side is greater than the other two taken together. The circumstance that the range of possible variation is not settled by the construction that first proves the possibility of the (equilateral) triangle

Lambert on Certainty and Generality of Metaphysics and Geometry 97 (in Elements I.1), but only by additional propositions, can make it seem that Euclid’s method addresses this issue only indirectly. But, again, we should keep in mind that Lambert conceives of geometry more broadly than as just the first books of Euclid’s Elements. The theory of curves in the plane, for instance, settles the range of variation more directly: Mathematicians seek … to render their concepts, propositions, and problems more general, but this does not take place in such a way that, in abstracting, one omits nearly everything; rather, one includes still ever more circumstances, and thereby their general formulae appear much more composite than the special [ones], because in [the general formulae] they retain all the varieties that occur in particular cases and are partially left out of many of these [cases]. One can take as examples the general equations for curved lines of the second, third, fourth degree etc. and the Newtonian binomial formula. (Arch. §193) Lambert uses the theory of curves in the plane as his example of how mathematics achieves completeness with respect to division [Eintheilung]. In mathematics, One leaves all circumstances and magnitudes undetermined, or one does not abstract from them but rather carries them along in the calculation, and in this way the general formulas become so wideranging. On the other hand this procedure serves so that one not only determines each particular case and species more easily, but can also make sure that one has them all. One can take for an example the general equations for lines of the third degree, which very concisely express [ungemein abkürzt] many single cases. (NO, Dian. §110) Lambert gives this example in the course of his comparison between the two basic kinds of “determinations” with respect to which concepts can be divided into species: degree and quality. His main point, as he summarises in a later passage, is that mathematics’ treatment of generality is ‘different and incomparably more advantageous than that which occurs in respect of qualities’ (NO, Dian. §452). But qualities are representative of the similarities in terms of which genus–species hierarchies were customarily organised, at least in philosophy, so Lambert can be understood to make a more general comparison between the generality of mathematics and that of traditional philosophy. The comparison shows, as in Arch. §§523–4, that geometry already realises the sort of generality that philosophy strives towards, using less apt means.

98  Katherine Dunlop 3.5 Conclusion In this chapter, I have sought to make sense of Lambert’s view of geometry, specifically geometry’s use of postulates, as the model that metaphysics must follow in order to fully qualify as science. As natural as it has become for us to think of the postulates as enjoining the production of particular mathematical objects, I think we can better understand Lambert by thinking of postulates as making possible mathematics’ supreme generality. They thus afford what Ernst Mach was to call “economy of thought”, which Lambert already singles out (though not by that name) as distinctive of scientific cognition. One aspect of this economisation is that conclusions are derived from the fewest premises or assumptions. Another, perhaps the most obvious, feature is that it spares the trouble of working out for individual cases that which it establishes in full generality.37 The examples discussed in the last section make clear that Lambert takes geometry to have these features, and the opening paragraphs of Arch. signal their importance for his project: The general applicability of such a theory [Lehrgebaüde] [as metaphysics] appears to promise many advantages. It is to contain the first grounds of the whole of human cognition [der gesammten menschlichen Erkenntnis], and what is therein established once for all does not need to be established anew in each case that comes up. … Of this kind are the merits of algebra and geometry, and such should also be the merit of metaphysics. (Arch. §3) Notes 1 According to Mendelssohn, ‘those who regard metaphysical conceptions as convincing and irrefutable must ultimately concede that they still have not been provided with the evidence of mathematical sciences; otherwise it would not have been possible for them to meet with so much contradiction’ (Mendelssohn 1764/1997: 255). For similar statements by Kant and Lambert, see Kant: 2: 275; Lambert: ÜM §2, §§10–15; CV §§19–20). 2 Kant will go on to claim that metaphysics has not yet ‘been so favored by fate as to have been able to enter upon the secure course [Gang] of a science’ (Bxiv. See also 4: 271–75). In the second edition of the first Critique’s Preface, Kant elaborates that in metaphysics ‘we have to retrace our path [Weg] countless times, because we find that it does not lead where we want to go, and it is so far from reaching unanimity in the assertions of its adherents that it is rather a battlefield’ (Bxiv–xv). Similarly, in the Prize Essay, Kant claims that the posing of the prize question ‘shows that there is good reason to ask about the path [Weg] in which one proposes to search for metaphysical understanding in the first place’ (2: 283). 3 The first question concerns the human understanding’s power to attain truth, and is addressed in “dianoiology”. The second concerns “the truth itself”

Lambert on Certainty and Generality of Metaphysics and Geometry 99 and its discernibility from error (NO “Vorrede” n.p., trans. Watkins 2009, 258), and is addressed in “alethiology”. The third concerns language insofar as it bears on the cognition of truth, and is addressed in “semiotics”, while the fourth concerns the differentiation of truth from “appearance” and is addressed in “phenomenology”. 4 Karin de Boer helpfully explains that in this tradition, ‘ontology’ was the ‘comprehensive treatment of the concepts and principles that inform any cognition of objects whatsoever and, accordingly, cannot be treated within disciplines devoted to a particular kind of object’ (De Boer 2020: 2). 5 The assumption that Lambert follows the method he articulates in the philosophical writings culminating in Arch. within those works themselves can be questioned. In a recent paper, Gesa Wellmann argues forcefully that the ‘Reflexionsniveau’ of the works ‘does not extend to the systematic construction [Aufbau] of the works themselves’ (Wellmann 2018/19: 156). I am happy to grant that the close attention Wellmann gives to Lambert’s manuscript writings on “systematology” brings into view a more general method than the one Lambert prescribes for metaphysics (which, as ‘a systematic connection of truths’, constitutes ‘only one part’ of the system he conceives there). But I do not think this difference in method is apparent within Arch. or the earlier philosophical writings. 6 I do not wish to claim that Kant’s understanding of “architectonic” is the same as Lambert’s, or that Kant conceives his critique of pure reason as an architectonic. Kant does say that critique is to proceed architectonically (A13/B27). But his explication of ‘architectonic’ as ‘the doctrine of that which is scientific in our cognition in general’ (A832/B860) makes it broader than critique, which occupies a place in the system outlined by architectonic (A841/B869). 7 Wellmann’s (2017: 352) interpretation of Arch. as ‘a kind of propaedeutic metatheory’ suggests it may not be included, but Wellmann is not explicit on this point. 8 See Dunlop 2009. 9 Lambert puts it grudgingly in Arch. §15: Wolff ‘attained no other concept of scientific cognition, than that all therein must be proved from grounds’. Yet even here, Lambert says that ‘the honor of introducing a correct and useful method into philosophy was reserved for Wolff’ (Arch. §11). 10 Compare Lambert’s letter to Holland of 27 April 1767, IX.189–190. For additional references, see Basso 2012: 17. 11 ‘One must use a criterion which is not very different from the Cartesian, and ultimately [höchstens] allows the reduction of certainty to the correctness of concepts. One calls “first principles” those propositions whose correctness one admits and must admit as soon as one understands the words in which they are expressed. To understand the words is, however, just as much as to have clear, distinct, and nice concepts of them. This, however, is the Cartesian criterion of truth’ (CV §3). 12 Lambert proposes to set out this difficulty ‘in all its severity’: ‘One has a word and indicates its meaning to me. I question whether it [the meaning] is correct. One gives me the proof. This consists of more words; and my first question, which is only about one [word], now extends to more. If I should require the meaning and proof of the correctness of each, then new words present themselves’, and their amount continues to increase ‘as far as the whole extent of language’. It thus appears that ‘the hypothetical’ never leaves off, and ‘the categorical’ never begins (CV §74).

100  Katherine Dunlop 13 Similarly, in the notes that begin ÜM, Lambert asserts that the ‘actus reflexus, [by] which the understanding casts back to [züruckdenkt auf] itself and its representations, gives immediate experiences, whose actuality one must admit even in doubting’ (“Notanda”, 4). 14 On the contrary, Lambert objects that Wolff leaves ‘the question of what one is defining’ undecided (Arch. §12), implicitly allowing that this question is answered through experience. In his essay on the theory of parallel lines, Lambert’s criticism of Wolff shows that he thinks Wolff abstracts concepts from experience – in an insufficiently rigorous manner – even in mathematics. See Dunlop 2009: 50. 15 Lambert argues, more specifically, that ‘it would be a clear offense against logic if one were to admit a concept with a greater extension [Umfang] than one can prove at the place [Ort] where one brings it forward. It would be hypothetical to the extent that the proof was still incomplete, and one could accept it only to the point that the proof can be made complete. But this is always precarious, for one can much too easily confuse the hypothetical with what is proven’ (CV §80). 16 Lambert reports that by the time he read Euclid, he was already familiar with Wolff’s work and ‘knew more or less what mathematical method and Scholastic method were’, and so expected Euclid to begin with ‘the first theorems in which angles are compared’ (CV §79). This seems to refer to Theorem 1 in Wolff’s Anfangsgründe aller mathematischen Wissenschaften, Vol. 1, which shows that the angles created by the intersection of two lines are equal to two right angles. Euclid proves the result as Proposition 13 in Book I of the Elements. 17 In Euclid’s Elements, the axioms and postulates are preceded by definitions. Lambert contends that Euclid in fact treats his definitions as hypotheses, not resting any claims on them until they themselves are proved. See Dunlop 2009. 18 ‘Geometry and logic are ideal sciences, and in metaphysics [the egoist] admits nothing other than what he can have from experience in itself. All that is outside him is to him only appearance. One must … go so far with him and let it be appearance to him, until one has grounds enough to conclude the truth from the assumed appearance. This is what astronomers do, who at first have only what is apparent in the movement of the heavens, but penetrate through this layer to the truth’ (ÜM §33). For further discussion, see Leduc 2016. 19 For instance, in his letter to Kant of February 3,1766 (10: 66). Lambert explains in the Preface to NO that he first read Locke’s work only after he had completed his own account of simple concepts and ‘was thereby led to shorten it’ (n.p., trans. Watkins 2009: 260). 20 ‘[T]he intellectual world is known [bekannt] to us no otherwise than through a kind of similarity with the corporeal world, and all the words through which we represent the former are taken from the latter and made metaphorical’ (Arch. §39). 21 It does not seem, however, that the role of the “theory of powers” separates cleanly from that of the postulates. Lambert’s broader aim in this passage is to distinguish two ways of determining possibility: a posteriori, by concluding possibility from actuality, and a priori, through simple concepts and their associated axioms and postulates. Later in Arch., he claims ‘the most immediate source, however, of positive possibilities, a priori as well as a posteriori, is

Lambert on Certainty and Generality of Metaphysics and Geometry 101 powers, for without these nothing can take place’. In ‘the actual world’ nothing is possible for which ‘the actually occurring powers’ are insufficient, and in ‘the realm of possibility’ powers pertain to all that is free of contradiction ‘absolutely or categorically in the strictest sense’, so that ‘the limitations on what is possible through powers are already determined by the simple concepts, their axioms and postulates’ (Arch. §243). 22 Throughout the various editions of Euclid’s Elements, the propositions listed as “postulates” share a stylistic peculiarity: they are infinitive constructions, preceded by a passive imperative (which can be translated “Let it be asked that”). But the earliest known editions, which are Greek, contain five postulates, of which the fourth “lets it be asked” that all right angles are equal to one another, and the fifth, a formulation of the Postulate or Axiom of parallels, “lets it be asked” that two lines have an intersection point in a certain direction under certain conditions. In contrast to the first three postulates, these principles seem not to demand any activity on the part of the reader. Accordingly, Greek thinkers such as Proclus do not distinguish between postulates and axioms as practical and theoretical. Early modern editions of the Elements, by contrast, typically list Euclid’s last two postulates as axioms and give only his first three postulates. For details, see De Risi 2016. 23 These steps are ekthesis (“setting-out”), which ‘applies the content of [a general proposition] to a figure’ that is assumed to be drawn, and kataskeue (“machinery”, or auxiliary construction), in which this figure is ‘completed by drawing certain additional lines, points, and circles’ (Hintikka 1967/1974: 168; see also Laywine 2020). Kant’s account of the exhibited triangle that ‘express[es] in the representation universal validity’ (A713/B741) corresponds to ekthesis, and his further account of mathematical method, in reference to Elements I.32, emphasises auxiliary constructions (A716/B744). 24 See Shabel (2006). 25 Since this is a view about experience, not about philosophical activity, it does not conflict with Kant’s separation between the methods of mathematics and philosophy. 26 See the “Axioms of Intuition”, A163/B204. That Kant shows the applicability of the categories of quantity in this way is also suggested by his claim that ‘we create the objects themselves in space and time through homogeneous synthesis, considering them merely as quanta’ (A723/B751). For extensive discussion, see Sutherland 2022. 27 See Friedman 1992: 61–65. 28 According to Lambert, Euclid ‘sets the extraction of square and cubic roots among the postulates’ insofar as Euclid ‘leaves out the solution of this problem for actual performance [für die wirkliche Ausübung]’ (CV §49). He explains that the sort of possibility that is guaranteed by postulates does not extend to ‘actual performance’ even in the case of lines: ‘Euclid actually required only the possibility of drawing a geometrical line, which can also take place merely in thought, because the whole of geometry is ideal, and its acuity is always missing [wegfällt] in actual performance [Ausübung]’ (CV §48). This distinction between “actual performance” and “ideal” possibility may prefigure Lambert’s later distinction between “doability” and the sort of possibility guaranteed by postulates. 29 Lambert further argues that in metaphysics, a thing is thought as having ‘a inner metaphysical unity, truth, and goodness’ (Arch. §519, citing Arch. §304

102  Katherine Dunlop and §350). He also holds, as we have seen, that the general concept of a thing presupposes the concept of solidity (Arch. §57). 30 Lambert claims that in ‘analysis’ [Auflosen], his term for his alternative to abstraction, ‘one considers the concept in itself, and leaves it as it is; with abstraction, however, one especially takes the general, and precisely thereby compares the concept with others’ (Arch. §516; see also Arch. §525). This aspect of Lambert’s contrast calls to mind Kant’s discussion of ‘the analytical unity of consciousness’ (B133–34n.), and the similarity supports Hammer’s view that Lambert’s account of postulates – and the kind of generality they guarantee – was a source for Kant’s analytic/synthetic distinction. 31 ‘One abstracts, in the most proper sense, so as to bring forth the general from the special [specialen] concepts’ (Arch. §515). 32 Lambert here cites Arch. §195, where he says that ‘if the Individua belonging under a genus are correctly known to us, the concept that we have of the genus also represents to us as it were a model, image, formula, or call it what you will, of the individual’. 33 For extensive discussion, see Basso 2012. 34 See, e.g., NO Dian. §450. 35 Lambert gives two examples. The first is that ‘in every case that comes up in which [the science] is applicable, from the least number of given elements the rest, which are determined from or stand in relation to [these], can be found’. Lambert asserts that ‘one still finds very few examples’ of this sort of minimisation in metaphysics, but ‘in the whole of Mathesis one makes it a law to assume neither too few nor too many Data, and to determine from the Datis what is given together with them or can be found from them’ (Arch. §15). In fact, Lambert places so much importance on this relationship between “given” and “sought” elements that he singles it out, together with postulates, as missing from Wolff’s version of mathematical method (Arch. §14). 36 Here, it bears noting that the concepts of genus and species (and the closely related concepts of universal and particular, as well as those of property [Eigenschaft], modification, and essence) are themselves ideal. Lambert describes them as ‘concepts of concept’ rather than ‘concepts of things’, and also as ‘merely ideal and logical’ (Arch. §123). These concepts indeed occur both in Vernunftlehre [logic] and in Grundlehre, where they not only find application to real things, but make possible the application of other concepts. Thus, Lambert describes Arch.’s chapter on “the general and particular” (which explains how concepts are to be divided [eingetheilt] into species) as the first of the ‘main elements [Hauptstücken] of the fundamental discipline which pertain much more to our manner of representing things than to the things themselves, but through which, nonetheless, the ideal must be considered in relation to the things themselves’ (Arch. §161). Lambert’s account of the concepts of genera and species as belonging both to logic, which is an “ideal” science, and to the fundamental discipline may give us a way to understand how elements of geometrical method – in particular, postulates – can remain ideal in their home science of geometry, while also serving to establish applicability to real things in the fundamental discipline. 37 In an unpublished manuscript, Lambert characterises science in terms of its concern with ‘what is related [das Zusammenhängende] in cognition, which is found between the general and the particular. It thereby puts us in position to apply [the general] to [the particular], and spares us the trouble of undertaking new efforts in each case that occurs’ (IX.2, 407).

Lambert on Certainty and Generality of Metaphysics and Geometry 103 References Primary Sources Kant’s works are cited as reprinted in the Berlin Academy of Sciences edition, except for the Critique of Pure Reason, which is cited according to page numbers of the “A” (1781) and “B” (1787) editions. Lambert’s correspondence is cited as reprinted in Band IX of Philosophische Schriften, Johann Heinrich Lambert (Hildesheim: Olds, 1965–2008). Other works by Lambert are cited according to the following scheme of abbreviation: [ÜM] “Über die Methode, die Metaphysik, Theologie und Moral richtiger zu beweisen”. In Philosophische Schriften, Entwürfe und Rezensionen aus dem Nachlaß. Band X of Philosophische Schriften, Johann Heinrich Lambert (Hildesheim: Olds, 1965–2008). Teilband 2 [CV] “Abhandlung vom criterium veritatis”. In Philosophische Schriften, Entwürfe und Rezensionen aus dem Nachlaß. Band X of Philosophische Schriften, Johann Heinrich Lambert (Hildesheim: Olds, 1965–2008). Teilband 2 [NO] Neues Organon, oder Gedanken über die Erforschung und Bezeichnung des Wahren und dessen Unterscheidung vom Irrthum und Schein (Leipzig, 1764) [Arch.] Anlage zur Architectonic, oder Theorie des Ersten und des Einfachen in der philosophischen und mathematischen Erkenntnis (Riga: Hartknoch, 1771) Additional Sources Basso, P. (2012) The Other Side of Euclid: Lambert’s Epistemology of Constructive and Diagrammatic Strategies (Milan: Ledizioni) De Risi, V. (2016) “The Development of Euclidean Axiomatics” Archive for History of Exact Sciences 70, pp. 591–676 Dunlop, K. (2009) “Why Euclid’s Geometry Brooked No Doubt: J. H. Lambert on Certainty and the Existence of Models” Synthese 167, pp. 33–65 Friedman, M. (1992) Kant and the Exact Sciences (Cambridge, MA: Harvard University Press) Hammer, M. (2022) “Lamberts Postulate als Quelle der Synthesis Kants”. In Motta, Schulting and Thiel (Eds.) Kant’s Transcendental Deduction and the Theory of Apperception: New Interpretations. Kantstudien-Ergänzungshefte vol. 218 (Berlin: de Gruyter), pp. 133–192 Hintikka, J. (1967/1974) “Kant on the Mathematical Method” Monist 51. Cited as reprinted in Knowledge and the Known (Dordrecht: Reidel), pp. 160–183 Laywine, A. (2010) “Kant and Lambert on Geometrical Postulates in the Reform of Metaphysics”. In Domski and Dickson (Eds.) Discourse on a New Method (Chicago: Open Court), pp. 113–134 Laywine, A. (2020) Kant’s Transcendental Deduction (Oxford: Oxford University Press) Leduc, C. (2016) “Métaphysique und mathématique chez Lambert: Une réponse à la Preisfrage de 1763”. In Les Métaphysiques des Lumières (Paris: Classiques Garnier), pp. 207–225

104  Katherine Dunlop Mendelssohn, M. (1764/1997) “On Evidence in Metaphysical Sciences”. In D. Dahlstrom (Trans. and Ed.) Philosophical Writings (Cambridge: Cambridge University Press), pp. 251–306 Sutherland, D. (2022) Kant’s Mathematical World: Mathematics, Cognition, and Experience (Cambridge: Cambridge University Press) Walford, D. and Meerbote, R. (Eds.) (1992) Theoretical Philosophy, 1755–1770 (Cambridge: Cambridge University Press) Watkins, E. (2009) Kant’s Critique of Pure Reason: Background Source Materials (Cambridge: Cambridge University Press) Wellmann, G. (2017) “Towards a New Conception of Metaphysics: Lambert’s Criticism on Wolff’s Mathematical Method” Revista de Estudios Kantianos 2, pp. 135–148 Wellmann, G. (2018/2019) “Lamberts Begriff eines metaphysischen Systems” Archiv für Begriffsgeschichte 60/61, pp. 143–162

4

The Methodological Role of Intellectual Intuition in Kant’s Critique Toby Lovat

4.1 Introduction ‘Criticism’, says Kant, ‘is the preparatory activity necessary for the advancement of metaphysics as a well-grounded science’ (CPR B36).1 While Kant agrees with Wolff, ‘the greatest of all dogmatic philosophers’ (CPR B36), that a scientific metaphysics ought to be developed a priori and according to the strictest inferences and systematicity, for Kant the ground of this science needs to be both cleared and established by determining the nature and scope of reason in relation to the nature and scope of our cognitive faculties. Since, for Kant, metaphysics is, by definition, an a priori science (having to do with that which is beyond or before physics), the first half of the Critique concerns an investigation of our cognitive faculties precisely to establish what can be known a priori. And although Kant affirms that a ‘dogmatic procedure’ is essential to any well-formed science, he characterises dogmatism in metaphysics as ‘the presumption of getting on solely with pure cognition from (philosophical) concepts according to principles … without first inquiring in what way and by what right it has obtained them’ (CPR B35). Unlike Descartes, for whom “clear and distinct ideas” can provide direct insight into the nature and existence of, e.g., God and the soul, Kant denies that we are cognitively equipped for such ostensibly “intellectual intuition” of metaphysical objects. Indeed, for Kant, both our a priori and empirical knowledge are fundamentally limited precisely because our intuition is constrained. While Kant sets out the fundamental concepts through which we can think anything at all, he also restricts the legitimate scope of these concepts to objects of possible sensible intuition. His critique of reason aims to show that when not so restricted, our understanding is driven to confusion, error, and contradiction. Nonetheless, Kant has much to say about intellectual intuition. While human cognition is discursive, requiring both concepts and sensible DOI: 10.4324/9781003187561-5

106  Toby Lovat intuitions, Kant characterises intellectual intuition as pertaining to a nondiscursive cognition, which is to say, to a rational being whose cognition does not entail sensibility, i.e., God. Many of Kant’s claims about the nature of intellectual intuition and the cognition of such an intellect might seem to suggest that he is setting out a metaphysical account of both God’s mode of cognition and what God would in fact cognise. In this chapter, I argue that Kant’s account of intellectual intuition should be understood as fundamentally contrastive, ultimately concerned with better understanding the nature and limitations of discursive cognition, rather than with establishing putatively true claims about God’s cognition. The central aim of this chapter is to show how Kant deploys the contrastive idea of intellectual intuition to support both positive and negative aspects of the Critique, both of which are crucial to his efforts to put metaphysics on the path to being a “well-grounded science”. On the positive side, Kant deploys the contrast to establish both the possibility and grounds of a metaphysics of empirical reality predicated on the view that human discursive cognition, in contrast to divine cognition, is fundamentally bound by a priori conditions of sensibility (i.e., the forms of intuition). On the negative side, Kant deploys the contrast between a purely intuitive intellect and discursive cognition to illustrate why efforts to extend our cognition beyond the confines of sensibility are ruinous to the prospects for metaphysics. By focusing on the contrastive approach to intellectual intuition, I aim to provide a clear understanding of how and why Kant restricts metaphysical speculation in his endeavour to establish the possibility and grounds of a genuinely scientific metaphysics. In this respect, this chapter offers a reminder of the philosophical virtues of Kant’s metaphysical humility, too often ignored in both contemporary metaphysics and Kant interpretation.2 In the first section of this chapter, on the distinction between discursive cognition and intellectual intuition in Kant’s philosophy, I set out the key differences between the former and the latter, emphasising that the latter is best understood as offering a contrast to the former, principally to clarify the nature and limits of discursive cognition. The second section concerns Kant’s distinction between appearances and things in themselves in relation to his efforts to establish a firm footing for metaphysics. I draw on Henry Allison’s interpretation and defence of Kant’s idealism, emphasising that the distinction between appearances and things in themselves concerns contrastive ways of considering objects in relation to contrastive conceptions of cognition. Again, the aim here is to show that these two contrasting ways of considering objects (as appearances and as things in themselves) in relation to their correlated epistemic-intuitive standpoints (discursive and non-discursive cognition) are fundamental to Kant’s project of developing a metaphysics whose success

Methodological Role of Intellectual Intuition in Kant’s Critique 107 depends on establishing and maintaining the boundaries of our possible cognition. The third section concerns Kant’s restriction of the legitimate use of the categories of the understanding to the conditions of sensibility, and hence to spatiotemporal/empirical reality, emphasising that when not so restricted, we overstep the discursive nature of our cognition. The fourth section concerns Kant’s account of how and why reason both leads the understanding astray and comes into conflict with itself when human cognition fails to take note of its discursive and hence sensible nature. This failure leaves reason unrestrained, compelling human cognition to treat objects of a merely sensible intuition as if they were objects of a pure or unrestricted understanding (i.e., an intellect intuition). For Kant, awareness of, and adherence to, the limitations of sensible intuition/ discursive cognition functions as a philosophical bulwark: defending the possibility of a well-grounded science of metaphysics against the illusion, error, and contradiction that follows the unrestricted use of speculative reason. 4.2  Discursive Cognition and Intellectual Intuition I take Kant’s distinction between sensible intuition and intellectual intuition to revolve around the claim that, unlike the latter, the objects given in sensible intuition are not themselves a product of our mental activity. This, of course, is a central feature of Kant’s claim that human cognition is discursive, in need of both concepts and sensible intuitions. This discursivity thesis is encapsulated in the following passage from the Transcendental Logic: Our cognition arises from two fundamental sources in the mind, the first of which is the reception of representations (the receptivity of impressions), the second the faculty for cognising an object by means of these representations (spontaneity of concepts); through the former an object is given to us, through the latter it is thought in relation to that representation (as a mere determination of the mind) … thoughts without intuitions are empty, intuitions without concepts are blind. (CPR A51/B74) Kant’s discursivity thesis entails that human experience, construed as ‘empirical cognition’ or ‘cognition that determines an object through perception’ (CPR B218) requires, and consists in, the synthesis of sensible intuitions and concepts. While, for Kant, the understanding thinks by means of concepts, human experience and knowledge of the world also requires

108  Toby Lovat an independent source of representations through which objects are given. In the absence of sensible intuitions, concepts have no reference, and are in that sense empty. Without concepts, there would be no thought, and therefore no cognition of the world. Without sensible intuition, there would be nothing to be thought about, and therefore no cognition of the world. Kant contrasts a discursive understanding, dependent on sensible intuition, with a non-discursive and, therefore, non-sensible understanding. As Marcel Quarfood (2011: 143) notes, in addition to the term intellectual intuition, Kant also refers to such an understanding as an ‘“intuitive understanding”, “infinite understanding”, “intuitus originarius”, and “archetypal understanding”. Despite the variety of terms, Quarfood identifies two central aspects of Kant’s conception of a non-sensible, cognitive capacity: (1) that an intellectual intuition would be productive rather than receptive and, therefore; (2) it would immediately (i.e., without synthesis) intuit everything in its totality. For Quarfood, these two aspects are ‘not neatly separable’ because both are set against Kant’s ‘general idea of a non-human cognitive capacity, which he uses for various contrastive purposes, sometimes stressing its divine, productive character and sometimes giving a more austere description of its power in terms merely of an immediate holistic viewing’ (Quarfood 2011: 151).3 Indeed, for Quarfood, the key to understanding Kant’s various characterisations of, and determinations about, non-discursive understanding is that they are to be contrasted with the human understanding, i.e., a ‘discursive, concept-employing understanding’ (Quarfood 2011: 152). A brief survey of relevant passages from the first and third Critiques evidences these two key aspects of intellectual intuition, and also, crucially, their contrast to discursive intuition. [O]ur kind of outer as well as inner intuition … is called sensible because it is not original, i.e., one through which the existence of the object of intuition is itself given (and that, so far as we can have insight, can only pertain to the original being); rather it is dependent on the existence of the object, thus it is possible only insofar as the representational capacity of the subject is affected through that. (CPR B72) Kant’s claim here is that our sensible intuition through which we represent the world and its objects requires that objects are given to us from a source that is independent of our minds. By contrast, an intellectual intuition is original because its representation of reality would not be derived from, or grounded in, an external source. An intellectual intuition would create reality directly through representation, and hence through intuition: ‘a divine understanding … would not represent given objects, but [would

Methodological Role of Intellectual Intuition in Kant’s Critique 109 be one] through whose representation the objects would themselves at the same time be given, or produced’ (CPR B145). And it is for this reason that, unlike a discursive intellect, an intuitive intellect would not require synthesis or combination for cognition. For Kant, our ability to grasp the manifold of sensation given in an intuition, to represent that manifold as, say, a unified object, necessarily requires a unified consciousness through which a manifold of sensation can be combined into a unity through concepts. But, says Kant, the ‘principle’ of the necessary unity of apperception is not a principle for every possible understanding, but only for one through whose pure apperception in the representation I am nothing manifold is given at all. That understanding through whose selfconsciousness the manifold of intuition would at the same time be given, an understanding through whose representation the objects of this representation would at the same time exist, would not require a special act of the synthesis of the manifold for the unity of consciousness, which the human understanding, which merely thinks, but does not intuit, does require … the human understanding cannot even form for itself the least concept of another possible understanding.4 (CPR B138–9) So, unlike a discursive intellect, an intuitive intellect would not require a synthesis or combination for cognition, precisely because such an intellect is not given a manifold from an external or independent source. And because an intuitive intellect would not require a synthesis for combination, it would not cognise through categories: [I]f I wanted to think of an understanding that itself intuited … then the categories would have no significance at all with regard to such cognition. They are only rules for an understanding whose entire capacity consists in thinking, i.e., in the action of bringing the synthesis of the manifold that is given to it in intuition from elsewhere to the unity of apperception, which therefore cognises nothing at all by itself, but only combines and orders the material for cognition, the intuition, which must be given to it through the object. (CPR B145) Since, for Kant, the categories are the most basic rules or functions through which we unify what appears in a spatiotemporal manifold of given sensible intuitions into a representation (without which we would not experience determinate objects), an understanding that does not synthesise a manifold of given sensible intuition, but instead directly produces what it

110  Toby Lovat cognises, would not require the use of concepts. Whereas sensible intuition depends upon objects (i.e., an external reality), an intellectual intuition would be one for which objects would wholly depend upon its intuition. In that sense, as A. B. Dickerson (2011: 166) puts it, ‘[t]he intuitive intellect (i.e., God) writes the word of experience; we have to read it’. And it is because an intellectual intuition would not synthesise or unify a manifold given in sensibility through concepts that such an understanding would instead immediately grasp, or “view” all that is. In the Critique of Judgment (CJ), Kant explains that ‘[o]ur understanding must go from the analytical universal (of concepts) to the particular (of the given empirical intuition) … from the subsumption of empirical intuition … under the concept’ (CJ 5: 407). As concepts are universals, discursive understanding builds up cognition of intuited content through a composite of concepts (the most basic of which are the categories). What this means is that discursive cognition grasps ‘a determinate form of the whole’ (CJ 5: 407) by way of the synthesis or unification of parts. In contrast, says Kant, ‘we can also conceive of an understanding which, since it is not discursive like ours but is intuitive, goes from the synthetically universal (of the intuitions of a whole as such) to the particular, i.e., from the whole to the parts’ (CJ 5: 407). For Kant, the contingency built into discursive cognition would not pertain to an intuitive understanding. In §76 in the third critique, Kant ties the distinction between possibility and actuality to ‘the subject and the nature of its cognitive faculties’ (CJ 5: 403). Through concepts as predicates of possible objects of sensible intuition, we can, he says, ‘always have something in our thoughts although it does not exist’ (CJ 5: 403). And through sensible intuition, we can ‘represent something as given’ or actual, even ‘if we do not have a concept of it’ (CJ 5: 403).5 By contrast, for an understanding that does not represent things through sensible intuition, and which, therefore, does not use concepts, all objects that [it] cognise[d] would be (exist), and the possibility of some that did not exist, i.e., their contingency if they did exist, as well as the necessity that is to be distinguished from that, would not enter into the representation of such a being at all. (CJ 5: 403) For such an understanding, the modalities of contingency (or possibility) and necessity would not apply. And a correlate of this, says Kant, is that an intellectual intuition would be one for which there is ‘no distinction between what should be done and what is done’ (CJ 5: 404), which is to say it would not regard moral law as a duty, but necessarily act in accordance with it.

Methodological Role of Intellectual Intuition in Kant’s Critique 111 In the above, I have identified key characteristics of Kant’s conception of an intellectual intuition, emphasising that these are invariably developed in contrast to (and as a foil for) discursive cognition in general – and human cognition in particular – rather than as positive claims about what a putatively divine intellect would in fact cognise. These themes will return as I now move to a closer examination of the distinction between sensible and intellectual intuition in the first critique. My contention throughout is that that contrastive distinction plays a fundamental role in the central doctrines and overall project of the Critique, precisely because that distinction is chiefly concerned with determining the limits of human cognition and the implications for human cognition when these limits are unnoticed. If a well-grounded metaphysical science is to be possible, the legitimate use of our reason must be limited to the constraints implied by sensibility. When not so constrained, metaphysical speculation is illegitimate. One implication of this is that Kant’s determinations about intellectual intuition and objects considered in relation to such a cognitive standpoint (things in themselves) cannot be taken as positive claims about either God’s actual cognition or things construed from such a standpoint. 4.3 Kant’s Copernican Revolution and the Distinction between Appearances and Things in Themselves At the outset of the Critique, Kant contends that traditional metaphysical enquiry is guided by an epistemological approach tacitly directed by the question of how pure concepts, or a priori thought, can conform to objects. Kant suggests that metaphysics might fare better if its enquiries instead begin from the question of how objects can conform to pure concepts. Since, for Kant, cognition of any sort requires an object of intuition (whether sensible or intellectual), he first asks after the nature of the objects in question. The positive side of Kant’s “experiment” in philosophy consists in an effort to show that we (mere rational mortals) can make some headway in metaphysics, but only once we have established that our sensible intuition has pure a priori forms (space and time), since that provides at least the possibility of demonstrating that pure concepts of the understanding apply with necessity to nature or empirical reality. For Kant, this experiment ‘succeeds as well as we could wish, and … promises to metaphysics the secure course of a science’ because we can now establish that ‘objects appropriate [to a priori concepts] … can be given in experience’, and can, therefore, ‘provide satisfactory proofs of the laws that are the a priori ground of nature, as the sum total of objects of experience’ (CPR Bxix). If, by contrast, we start out with a conception of objects either as such (i.e., in general) or as that which is

112  Toby Lovat beyond sensible intuition or experience, we have, for Kant, no means of determining whether or how our understanding might relate to such objects. Kant’s experimental response to what he sees as the failings and irresolvable battles of traditional metaphysics is premised on explicitly adopting an ‘alteration in our way of thinking’ (CPR Bxix) concerned first with determining the nature of objects in relation to the a priori conditions of their cognition. At the heart of that experiment is a bifurcation of contrasting ways of considering objects, namely as appearances and as things in themselves. For Kant, these two contrasting ways of considering objects are fundamentally conceived as correlates of contrastive conceptions of cognition. It is because our intuition is sensible and not intellectual that, for Kant, we could have no cognition of objects conceived in their relation to an intellectual intuition. And on that score, Kant’s conception of intellectual intuition is the cognitive standpoint apposite to the possible cognition of things considered as they are independent of sensible intuition, i.e., things in themselves. In this way, Kant uses the word appearances to refer to objects construed in their relation to the forms of our sensible intuition, which is to say that appearances refer to the representation of objects of a spatiotemporal manifold. But, says Kant, to avoid ‘the absurd proposition that there is an appearance without anything that appears’ (CPR Bxxvi), or a ‘constant circle’ (CPR A252) as he puts it, the notion of appearances must be contrasted with the notion of something that is not an appearance but which we nonetheless must assume grounds appearances. That assumption is required because the concept of an appearance only makes sense if what appears is grounded in something that is not itself an appearance. In the absence of such a ground (i.e., appearances all the way down), appearances would contradict their own concept. Indeed, in the absence of a ground that is not appearance, appearances would be self-grounding and, therefore, in effect, things in themselves (and not appearances). Besides being contradictory, for Kant, such a view is tantamount to Berkeleyan phenomenalism and, therefore, also to transcendental realism. I will turn to Kant’s critique of transcendental realism later, but it is here useful to register that his distinction between appearances and things in themselves is fundamental to the possibility of avoiding the final catastrophic destination of all transcendentally realist metaphysics, namely ‘the euthanasia of pure reason’ (CPR A407/B434), i.e., the untimely death of reason when it is compelled to accept mutually incompatible positions. Indeed, on that score, Kant’s distinction between appearances and things in themselves is not only central to his foray into what he takes to be the hitherto confused and fruitless battleground of metaphysics but also fundamental to the possibility of anything but Pyrrhic victory.

Methodological Role of Intellectual Intuition in Kant’s Critique 113 The contrastive approach to understanding Kant’s distinction between appearances and things in themselves follows Henry Allison’s interpretation and defence of Kant’s idealism. Against traditional, metaphysical interpretations of Kant’s distinction between appearances and things in themselves – according to which appearances and things in themselves refer to different sorts of objects in different worlds – Allison, following Gerold Prauss, argues that the distinction ought to be understood ‘as a contrast between two ways in which … objects can be considered in a philosophical reflection on the conditions of their cognition’ (Allison 2012: 67).6 On this reading, Kant’s distinction between appearances and things in themselves pertains to contrasting concepts of things construed in their relation to contrasting concepts of cognition. As we have seen, “appearances” refers to things as cognised by a discursive understanding and, by way of contrast, “things in themselves” refers to things as cognised by a non-discursive, or pure, intuitive understanding. To be clear, on this interpretation, appearances and things in themselves are not concepts of different things or different objects; rather, they are different, contrastive, ways of considering objects. Kant’s conception of things in themselves should be understood as entailed by the conception of appearances, so that, as Michelle Grier puts it, ‘the representation of the thing in itself’ can be ‘very generally characterized [as] methodologically entailed by the critical procedure of reflecting on objects in relation to our cognitive faculties’ (Grier 2001: 89). Since human cognition is discursive, that entailment is premised on the conclusion of Kant’s Transcendental Aesthetic; namely ‘that space and time are only forms of sensible intuition, and therefore only conditions of the existence of things as appearances’ (CPR Bxxvi). In this way, as Allison puts it, ‘the expression appearances is parasitic upon, or at least correlative with, the expression thing in itself’ (Allison 2004: 55). So construed, the notions of appearances and things in themselves are polar concepts. Just as the notion of “up” makes sense only in contrast to a spatial orientation that is “not up”, and “black” makes sense only in contrast to colours that are “not black”, the notion of objects as appearance makes sense only in contrast to the notion of objects considered as they are not appearances, which, as above, is to say as they are not objects of sensible intuition. And this is why, for example, Kant claims that space as a property of appearances qua the form of human sensible intuition, ‘represents no property at all of any things in themselves nor any relation of them to each other’ (CPR A26/B42). The contrastive pairs of intellectual and sensible intuition, and of things in themselves and appearances, should be understood as correlates of contrasting ways of considering objects. Insofar as both sets of correlated distinctions are contrastive, both serve to clarify the nature of discursive

114  Toby Lovat cognition and determine its limits. As we have seen, while the cognitive model of intellectual intuition is conceivable (via its contrast with discursive cognition), what such a cogniser (e.g. God) would or does cognise is unknowable, at least for discursive cognisers. Similarly, while objects considered as things in themselves are conceivable, they are nonetheless unknowable (for discursive cognition) precisely because “things in themselves” refers to objects (or things) as cognised by a non-sensible or intellectual intuition. There is, however, one crucial difference between the two sets of comparisons, correlated though they are. While the comparison between intellectual and sensible intuition does not at all commit Kant to the position that an intellectual intuition exists, his contention that appearances entail a ground which is not itself appearance means that we must suppose that things in themselves exist, i.e., we must suppose that there is a way that things (both qua objects and in general) are in themselves. It is, however, worth emphasising that this supposition is predicated on Kant’s experimental gambit of considering objects or things in relation to contrastive epistemic-intuitive standpoints: first as appearances as cognised by discursive cognition, and then, ex hypothesi, as things in themselves as cognised by a non-discursive cognition. From his day to ours, commentators have complained that Kant seems to make positive claims about putatively unknowable things in themselves (e.g. that they exist, that they are non-spatiotemporal, but somehow ground appearances and affect sensibility). While Kant’s theory of appearances does indeed entail the claim that things in themselves exist (i.e., that objects have a grounding in a super-sensible reality), this is not at all equivalent to cognition, or knowledge, of things so construed. Nonetheless, his various characterisations about the nature of things considered in themselves (principally that they are non-spatiotemporal and affect sensibility) have been taken as metaphysical claims about things as they are in themselves. So taken, however, Kant would seem to overstep the boundaries that he himself establishes as required for the possible development of a legitimate, scientific metaphysics. One of the central virtues of Allison’s epistemological interpretation and defence of Kant’s idealism is that it provides a means of dissolving such prima facie contradictions, precisely by insisting that the most philosophically tenable approach is to understand Kant’s claims about things as they are in themselves as a contrasting conception of objects, developed by abstracting from the conditions of discursive cognition. Rather than viewing Kant’s claims about things so considered as positive knowledge claims, the contrastive epistemological approach interprets these claims as fundamental features of Kant’s restriction of the legitimate scope of our cognition. Insofar as we can conceive of things considered as they are in themselves (i.e., in abstraction from, or in contrast

Methodological Role of Intellectual Intuition in Kant’s Critique 115 to, the conditions of our sensibility), cognition of things so construed entails a non-sensible (i.e., intellectual) intuition. And insofar as we can conceive of the cognition of such an intuition (again by way of abstraction from, or contrast to, discursive cognition), that conception teaches us that our cognition of things (whether a priori or empirical) is predicated on, and so limited to (or by) the sensible form of our intuition. Thus, although Kant thinks that we must suppose that empirical reality (and what appears to us therein) is grounded in a non-sensible or supersensible reality, consideration of the epistemic-intuitive model appropriate to cognition of such a form of reality serves chiefly to determine the limitations and hence prospects of our possible knowledge, metaphysical or otherwise. Allison’s interpretation has been subject to important and sustained critique. I cannot address that criticism in detail here, and I would certainly hesitate to conclude that Allison’s approach fulfils the demands of what Karl Ameriks has characterised as the ‘holy grail’ of Kant scholarship.7 Nor would I defend Allison on the basis that his interpretation is clearly best supported by the text, since, as Lucy Allais has remarked, there is ‘an abundance of apparent textual evidence … that can be appealed to in support of phenomenalist idealist interpreters [e.g., Strawson and Guyer] … deflationary … interpreters [e.g., Allison and Bird]’ and, I would add, metaphysical realist interpreters (e.g. Langton and Allais), so that interpreters ‘can’ indeed ‘always find textual evidence which challenges the other’s view’ (Allais 2015: 101). Nonetheless, as above, I take the chief virtue of Allison’s interpretation to be its philosophical strength, not least because it offers a riposte to long-standing complaints about the coherence of Kant’s philosophy, perhaps especially to various charges of phenomenalism and contradiction. As we have seen, while Kant’s theory of appearances does entail the supposition of the existence of non- or super-sensible grounds of empirical reality, that metaphysical commitment is part and parcel of his arguments against our possible cognition of things so considered, precisely because the legitimate reach of our cognition is limited to, or restricted by, our sensibility and its forms. Kant’s considerable efforts to diagnose and remedy the dangers of extending theoretical reason in metaphysics beyond the confines of objects of possible experience would seem especially puzzling if his various determinations about things considered in themselves (or about intellectual intuition) are understood as positive metaphysical claims about a super-sensible reality and its cognition. Indeed, in the second half of the Critique, the Transcendental Dialectic, Kant is chiefly concerned with demonstrating how and why reason’s efforts to extend our cognition beyond the confines of sensibility leads to a dead end, not least for the prospects of metaphysics.

116  Toby Lovat 4.4  The Scope and Restriction of the Categories Kant’s contrastive approach to discursive and purely intellectual cognition is central to the ways in which he restricts categorial cognition to objects of possible experience, and consequently also to the ways in which he develops and defends positive claims about the possibility of a scientific metaphysics within the boundaries of our possible knowledge. Pure general logic, says Kant, concerns ‘the absolutely necessary rules of thinking, without which no use of the understanding takes place … without regard to the difference of the objects to which it may be directed’ (CPR A55/B79). Pure general logic, therefore, ‘considers only the logical form in the relation [i.e., combination, unification, subordination] of cognitions to one another, i.e., the form of thinking in general’ (CPR A55/ B79). By way of contrast, while transcendental logic is also pure and a priori, it is, says Kant, ‘special’ because it ‘has to do merely with laws of the understanding and reason, but solely insofar as they relate to objects a priori’ (CPR A55/B79). In contrast to pure general logic, then, transcendental logic, as Hanna notes, is ‘ontically restricted, i.e., objectuallycommitted, and thereby presupposes the existence of certain specific categories or kinds … of objects’ (Hanna 2022: §2.1.1). Since, for Kant, cognition of objects requires that they are given in intuition, our legitimate application of transcendental logic is restricted to appearances, i.e., to the possible objects of our sensible intuition. When not so restricted, Kant warns us that the understanding is apt to overstep the limitations entailed by the nature of discursive cognition: [B]ecause it is very enticing and seductive to make use of these pure cognitions of the understanding and principles by themselves, and even beyond all bounds of experience, which however itself alone can give us the matter (objects) to which those pure concepts of the understanding can be applied, the understanding falls into the danger of making a material use of the merely formal principles of pure understanding through empty sophistries, and of judging without distinction about objects that are not given to us, which perhaps indeed could not be given to us in any way. Since it should properly be only a canon for the assessment of empirical use, it is misused if one lets it count as the organon of a general and unrestricted use, and dares to synthetically judge, assert, and decide about objects in general with the pure understanding alone. (CPR A63/B88) Indeed, for Kant, deciding ‘about objects in general with the pure understanding alone’ leads to forms illusion and error; and from there to

Methodological Role of Intellectual Intuition in Kant’s Critique 117 dogmatism, antinomy, and ultimately ‘the euthanasia of pure reason’. As we will see, for Kant, the proximate source of these problem is a failure to restrict the legitimate use of our understanding to objects of our possible sensible intuition. Another way to put this is that the unrestricted use of the categories ignores the discursivity thesis and in so doing tacitly presumes to treat objects of a merely sensible intuition as if they were objects of a pure, or intuitive understanding, i.e., things in themselves. Narrowly construed, Kant’s “metaphysical deduction” consists in the production of the table of categories from the table of judgements. Unlike the table of judgements, which concerns pure general logic and attends to the functions of thought when abstracted from all content, the table of categories sets out the a priori rules for the possible judgements about objects. The “transcendental deduction” aims to show that the a priori categories derived in the metaphysical deduction legitimately apply to objects given in sensible intuition. I can here leave aside the endlessly contested details of the transcendental deduction.8 For my purposes, in this chapter, I will accept that Kant succeeds in establishing that the categories are both rules for possible thought of objects of a sensible intuition and that they apply necessarily to whatever is given in human sensibility, which means that empirical reality necessarily stands under the categories. A crucial feature of Kant’s categories is that while their legitimate application with respect to knowledge (whether a priori or empirical) of objects is restricted to our sensible intuition (its forms and the objects that appear therein), they are nonetheless the concepts by which to ‘think objects in general without seeing to the particular manner (of sensibility) in which they might be given’ (CPR A254/B309). Indeed, insofar as ‘we cannot think any object except through the categories’ (CPR B165), they ‘have an unbounded field’, as Kant puts it (CPR B166n). Of course, for Kant, the unbounded field of thought is to be contrasted with the bounded field of cognition, since knowledge and experience of objects ‘requires intuitions that correspond to those concepts’ (CPR B166). 9 Nonetheless, because the categories are the rules by which objects can be thought, they too determine our possible conception of objects as such, or in themselves. Something construed as a thing in itself is hence ‘a thing insofar as it is not an object of our sensible intuition’ (CPR B166). On that score, Kant maintains that we have only a negative conception of things in themselves, a conception drawn from, and in contrast to, objects viewed in their relation to the conditions of our cognition. If ‘one assumes an object of a non-sensible intuition as given’, then, says Kant, one can certainly represent it through all of the predicates that already lie in the presupposition that nothing belonging to sensible intuition pertains to it: thus it is not extended, or in space, that its

118  Toby Lovat duration is not a time, that no alteration (sequence of determinations in time) is to be encountered in it, etc. But it is not yet a genuine cognition if I merely indicate what the intuition of the object is not, without being able to say what is then contained in it; for then I have not represented the possibility of an object for my pure concept of the understanding at all, since I cannot give any intuition that would correspond to it, but could only say that ours is not valid for it. (CPR B149) If instead we want to make positive claims about ‘an object of a nonsensible intuition’, to say, for example, that things in themselves, or things as such, are in fact non-spatiotemporal, or that they fall under the categories of the understanding, then, says Kant, ‘we assume a special kind of intuition, namely intellectual intuition … the possibility of which we cannot understand’ (CPR B308). Insofar as “things in themselves” refers to objects considered apart from the conditions of our sensibility, we can think of the grounds of appearances as non-spatiotemporal and not governed by causation. But insofar as things in themselves are understood as objects of a non-sensible understanding, i.e., of an intellectual intuition, we have no means of acquiring knowledge of the super-sensible ground of appearances that we must otherwise assume exists. Here we find Kant introducing intellectual intuition, as a contrast to discursive cognition, again precisely as a means of illustrating the limitations of our cognition. While we can consider objects as correlates of a non-sensible intuition as if they were given to such an intuition, deploying categorial cognition to produce determinations about objects so construed, these determinations can only be negative, i.e., only about what objects so considered (in their contrast to objects of discursive cognition) are not. But since intuitions that might “correspond” to such objects are impossible for us, these determinations are not knowledge of things so considered. Kant’s model of intellectual intuition, a hypothetical form of cognition that would indeed cognise non-sensible objects, is fundamentally alien to discursive cognition; although conceivable as a model of cognition, it is otherwise wholly mysterious, precisely because we could have no real understanding of what such cognition could be like. 4.5 The Battleground of Metaphysics: Transcendental Realism and Transcendental Illusion We have seen that Kant admits the use of the categories to generate negative determinations of things considered apart from the conditions of our sensibility. And that makes sense because the categories are the rules by which we think objects. But such determinations cannot yield knowledge,

Methodological Role of Intellectual Intuition in Kant’s Critique 119 precisely because the legitimate use of the categories with respect to possible knowledge is restricted to a form of sensible intuition and what appears therein. By showing that whatever we perceive in a spatiotemporal manifold necessarily stands under the categories Kant aims to prove that a priori knowledge of empirical reality is possible. But the argument of the Transcendental Deduction relies on establishing that it is through the a priori forms of our intuition (i.e., the conditions of our sensibility) that our minds provide a framework for, and so contributes to, what appears. And it is on that score that the legitimate use of the categories is restricted. Because discursive cognition requires – indeed entails – a form of sensible intuition (e.g. space and time), the legitimate use of the categories with respect to our possible knowledge of objects (whether a priori or empirical) is restrained by, or restricted to, that form and what appears therein. Kant’s critical or transcendental revolution entails that we are locked ‘within the boundaries of possible experience’ (CPR Bxxiii) and so denied a key aspect of traditional metaphysical speculation.10 Where we persist in our efforts to reach beyond experience, on the assumption that our task is to determine the relationship between pure concepts and a world beyond experience, we are, for Kant, tacitly asking after an unrestricted understanding of things, for which the transcendental model of cognition is a non-sensible and hence intellectual intuition. Construed in its comparison to our discursive cognition, the cognition of a non-discursive intellect would be wholly unrestricted, and its reason would indeed grasp the unconditioned ground of things (e.g., things in themselves). In the Transcendental Dialectic and its section on the antinomies Kant aims to illustrate how and why our efforts to extend our cognition as if unrestricted, result in irresolvable confusion and conflict, ‘from which’ he says reason ‘can indeed surmise that it must somewhere be proceeding on the grounds of hidden errors’ (CPR Aviii).11 One way to approach these hidden errors is to begin with Kant’s comparison between transcendental idealism and transcendental realism. Kant’s ideality thesis, the core of transcendental idealism, is the position that space and time are both ideal and real. That thesis, says Kant, ‘assert[s] the empirical reality of space (with respect to all possible outer experience), though to be sure [also] its transcendental ideality, i.e., that it is nothing as soon as we leave the condition of the possibility of all experience, and take it as something that grounds the things in themselves’ (CPR A28/B44). For Kant, transcendental idealism is equivalent to the position that space and time are not predicates of things ‘when they are considered in themselves through reason, i.e., without taking account of the constitution of our sensibility’ (CPR B308). Kant explicitly contrasts transcendental idealism with transcendental realism precisely because the

120  Toby Lovat latter treats space and time ontologically – either as things in themselves (i.e., substances) or as properties or conditions thereof: To [transcendental] idealism is opposed transcendental realism, which regards space and time as something given in themselves (independent of our sensibility). The transcendental realist therefore represents outer appearances (if their reality is conceded) as things in themselves, which would exist independently of us and our sensibility and thus would also be outside us according to pure concepts of the understanding. It is really this transcendental realist who afterwards plays the empirical idealist; and after he has falsely presupposed about objects of the senses that if they are to exist they must have their existence in themselves even apart from sense, he finds that from this point of view all our representations of sense are insufficient to make their reality certain. (CPR A369) It is because she presupposes a transcendentally realist position that the empiricist, for whom the mind can have immediate access only to its own ideas or representations, is forced to concede that knowledge of the existence of external objects is problematic. Although empiricists reason correctly when they conclude that we have no direct cognitive access to objects beyond our ideas or representation, on Kant’s view that reasoning is predicated on a failure to note that it is through the forms of space and time that sensibility itself makes an a priori contribution to the cognition of what appears. And it is on account of this error that the empiricist tradition treats empirically external objects (objects in space and time) as transcendentally external, i.e., as things in themselves. This same transcendentally realist presupposition underpins the rationalist tradition. According to Kant, whereas Locke ‘totally sensitivized the concepts of the understanding’, ‘Leibniz intellectualized the appearances’; he ‘compared the objects of senses with each other as things in general, merely in the understanding’ (CPR A271/B327). Consequently, he contends that for Leibniz space and time became the intelligible form of the connection of the things (substances and their states) in themselves … 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. (CPR A276/B332)

Methodological Role of Intellectual Intuition in Kant’s Critique 121 Thus, for Kant, the cardinal error of transcendental realism, whether it sensitivises concepts or intellectualises appearances, is that by presupposing that space and time and the objects therein are things in themselves, or pertain to things in general, it conflates appearances (that which is given) with things in themselves (that which is independent of the conditions of our sensibility). The error of transcendental realism, therefore, is that it assumes that knowledge of the universe and its furniture is, as Allison (2006: 114) puts it, ‘to be measured and evaluated in terms of its conformity (or lack thereof) to the norm of a putatively perfect divine knowledge’, i.e., a purely intellectual intuition. By contrast, Kant’s formal idealism – his Copernican revolution – shifts the ground of our possible knowledge to a conformity (or lack thereof) with the sensible and intellectual conditions of our cognition.12 In his account of transcendental illusion, Kant explains why and how this tacit assumption characteristic of transcendental realism leads to irresolvable antinomies and ultimately to the ‘euthanasia of pure reason’. In his brief preliminary discussion of transcendental illusion, Kant notes that the source of illusion and hence error in our judgements can be neither the ‘understanding by itself nor the senses by themselves’ (CPR A294/ B350). By itself (i.e., unless affected by some other force) understanding merely functions according to its laws, so that its judgements ‘must necessarily agree with these laws’ (CPR A294/B350). And since sensibility does not itself make judgements, it too cannot err. Because sensibility and the understanding are the only sources of our cognition, Kant concludes that it is only in the relation or connection between the two that illusions, and therefore errors of judgement, can arise. ‘Error’, concludes Kant, ‘is effected only through the unnoticed influence of sensibility on understanding’ (CPR A294/B350). While that claim makes good sense in relation to visual illusions, where we make erroneous judgements about what appears due to the unnoticed influence of a visual illusion (e.g., a bent stick in water), Kant does not really explain in what way errors of judgement are only an effect of ‘the unnoticed influence of sensibility on understanding’.13 In his very brief example of possible illusion and error in ‘pure judgments a priori’, Kant explains that such illusion and error can be avoided through ‘transcendental reflection’, by assigning representations to the ‘faculty of cognition proper to’ them (CPR A294/B350). Although Kant does not spell it out, in the case of pure a priori judgements, such transcendental reflection would seem to consist in noting the ‘influence of sensibility on understanding’ by determining sensibility’s potentially distorting influence on our determinations of pure a priori judgements.14 By separating out what properly belongs to the understanding and to sensibility (through transcendental reflection), we can avoid the possible illusion and error generated by falsely ascribing aspects of the latter to the former. Allison

122  Toby Lovat (2004: 322–327) has suggested that we can also regard the error of efforts to extend categorial cognition beyond the confines of sensibility as predicated on a failure to notice that the legitimate scope of the understanding, vis-à-vis knowledge, must be restricted to the a priori conditions of sensibility (i.e., the forms of intuition). In this way, Allison suggests that Kant’s critique of the transcendental deployment of the understanding (i.e., its use independently of the conditions of sensibility) in the Analytic can be understood as a treatment (through transcendental reflection) of the illusions and errors that are the effect of the failure to note that sensibility and its forms are a condition of (or ‘influence’ on) the legitimate use of the categories (with respect to a priori or empirical knowledge). If we are to take Kant at his word (that all error in judgement is due to ‘the unnoticed influence of sensibility on understanding’), then it does seem plausible to suggest, as Allison does, that the Analytic is, in part, a treatment of the illusion germane to the metaphysical errors of transcendental realism, because it (transcendental realism) fails to note the proper relationship between sensibility and understanding, which leads to ‘the illicit extension of categories to things in general’ (Allison 2004: 325).15 While the treatment of transcendental reflection serves to address the illusions and errors of judgement caused by ‘the transcendental use or misuse of categories, which’, says Kant, ‘is a mere mistake of the faculty of judgment when not properly checked by criticism’ (CPR A296/B352), such a remedy does not apply to transcendental illusion because that illusion cuts deeper as it concerns principles proper to reason itself. According to Kant, the very same demands of reason that guide our rational scientific enquiry lead us unavoidably to transcendental illusion, the cause of which is that in our reason (considered subjectively as a human faculty of cognition) there lie fundamental rules and maxims for its use, which look entirely like objective principles, and through them it comes about that the subjective necessity of a certain connection of our concepts on behalf of the understanding is taken for an objective necessity, the determination of things in themselves. (CPR A297/B353) In short, Kant’s claim is that the particular nature of reason is such that it takes its own principles and interests to be objectively necessary in relation to things in themselves. And it is this propensity, he argues, that ‘incite[s] us to tear down all those boundary posts [of the Analytic] and to lay claim to a wholly new territory that recognizes no demarcations anywhere’ (CPR A296/B352). The roots of this propensity lie in the fact that ‘the proper principle of reason in general (in its logical use) is to find

Methodological Role of Intellectual Intuition in Kant’s Critique 123 the unconditioned condition for conditioned cognitions of the understanding, with which its unity will be completed’ (CPR A307/B364). Crucially, Kant contends that this ‘logical maxim cannot become a principle of pure reason unless we assume that when the conditioned is given, then so is the whole series of conditions subordinated one to the other, which is itself unconditioned, is also given’ (CPR A308/B364). In other words, the very demand to find the unconditioned assumes, as a metaphysical principle, that it can be found. And it is this supposition, argues Kant, which forms the background assumption upon which rests metaphysical claims about God, the Soul, and the ultimate grounds of the universe. Kant is not advocating that we attempt to cure reason of this illusion by removing its dependence on the supposition that the unconditioned exists. Not only is this supposition unavoidable (since it is ‘the supreme principle of pure reason’ (CPR A308/B36), part and parcel of reason’s condition-hunting), it is also the very life-blood of rational enquiry.16 It is, nonetheless, the source of error because the unconditioned posited by the principle of reason cannot be met with in any possible experience. The problem, in short, is that reason’s injunction to explain everything by seeking out ever more conditions deceives us by assuming that the unconditioned ground of all conditions is itself given, and can, therefore, be found. In this way, reason leads us to make claims about things in themselves, and in so doing tacitly, but fallaciously, attributes to us a faculty of intellectual intuition. According to Kant, even if we cannot avoid this illusion, the means by which we avoid its deception is through the self-discipline of reason. At the heart of this self-discipline is Kant’s contention that reason’s proper use is regulative, not constitutive. In its regulative use, reason works on the material of the understanding, directing it towards unity and completeness. However, when reason is treated as constitutive, we are deceived into thinking that the objects we posit to satisfy its demands to find an unconditioned refer to real objects. Kant’s point is that the ideas reason compels the understanding to postulate do not provide us with concepts of objects that we could ever come to know. Thus, in its constitutive use, ‘[c]ontrary to all the warnings of criticism [reason] carries us away beyond the empirical use of the categories, and holds out for the semblance of extending the pure understanding’ (CPR A295/B352). In the deception of transcendental illusion, we are led by the principles and presuppositions of reason to work over the products of the understanding without recognising the restricted scope of the understanding entailed by sensibility, which is to say (following Allison), we fail to note the ‘influence of sensibility on understanding’. The Antinomies chapter is intended to demonstrate that when the world and the objects therein are treated as things in themselves, when we judge as if our understanding is unrestricted and thereby able to intellectually

124  Toby Lovat intuit the truth, reason drives the understanding into irresolvable conflict. For example, in the first Antinomy (CPR A426–427/B454–455), Kant develops apagogic proofs of the theses that the world is finite, and that it is infinite. The proof of the former is established by demonstrating the incoherence of the latter, and vice versa. If we assume that all events are preceded by an infinite series of events, then, says Kant, any given moment or event (a present) consists in a completed infinity. But since a completed infinity contradicts the concept of infinity, it cannot be the case that the world is infinite, so it must be finite, i.e., it must have had a beginning. If, however, we assume that the world had a beginning, then we must suppose that there was a time at which it was not. But since nothing can occur, or arise, in an ‘empty time’, we must conclude that the world is infinite. Thus, according to Kant, whether we begin from either thesis, that the world is finite or that it is infinite, reason compels us to deny and accept both. That contradiction arises for Kant where we assume that ‘the condition as well as its conditions are things in themselves’, so that when the first [condition] is given not only is regress to the second given as a problem, but the latter is thereby really already given along with it; and, because this holds for all members of the series, then the complete series of conditions, and hence the unconditioned is thereby simultaneously given, or rather presupposed by the fact that the conditioned, which is possible only through that series, is given. (CPR A498/B526) But, as Kant explains, the problem here is that ‘the synthesis of the conditioned with its condition is a synthesis of the mere understanding, which represents things as they are without paying attention to whether and how we might achieve acquaintance with them’ (CPR A498/B526). The untimely death of reason – its requirement to accept mutually incompatible positions – is, thus, a consequence of presupposing that what is given in experience is itself independent of the conditions of our sensibility. His point, as Allison explains it, is that from a transcendentally realist perspective – i.e., one that treats objects of experience as things in themselves, illicitly attributing to us the ‘semblance’ of an unrestricted or intuitive understanding – the way in which the understanding synthesises the conditioned with its condition ‘involves a pure or transcendental use of the categories, which, by abstracting from the sensible conditions of their application … yields merely rules for thought of “objects in general”’ (Allison 2004: 386). In other words, abstracted from sensible conditions, the mere thought of the existence of a conditioned object entails that the unconditioned condition, or totality of conditions, can in fact be cognised. Thus, the transcendentally realist outlook must assume that the conditions

Methodological Role of Intellectual Intuition in Kant’s Critique 125 we are impelled to hunt down by reason are not only ‘given as a problem’ or as a task, but also that they are given, and can, therefore, be found. In contrast, transcendental idealism – which, says Kant, is ‘the key to solving the cosmological dialectic’ (CPR A490/B518) – need not assume that sought-after conditions are given. By viewing objects of experience as appearances, transcendental idealism has a means of separating reason’s task of hunting down conditions from its assumption that the whole series (or totality) of conditions is given. As Kant explains: [I]f I am dealing with appearances, which as mere representations are not given at all if I do not achieve acquaintance with them (… for they are nothing except empirical cognitions) then I cannot say … that if the conditioned is given, then all the conditions (as appearances) for it are also given; and hence I can by no means infer the totality of the series of these conditions. (CPR A499/B527) In the case of the first antinomy, unlike the transcendental realist (who is compelled to conclude that the world is both finite and infinite in time), transcendental idealism construes the world as a series of representations or appearances, not as a thing in itself given as a totality which must be either infinite or finite in time (CPR A503/B531–A505/B533). In this way, transcendental idealism has a means by which to loosen the grip of transcendental illusion by denying reason’s principle that all conditions, and the totality of conditions, are given.17 Thus, transcendental idealism can avoid the illusion’s deception by maintaining that both theses (that the world is finite or infinite in time) are false, because, unlike transcendental realism, it is not committed to the view that the totality of conditions, qua appearances, is given. Consequently, however “natural” it remains for the transcendental idealist to think of the series of conditions as given, she is ‘equipped with a critical tool, making it possible to resist being seduced by this way of thinking’ (Allison 2004: 386). Indeed, it is this critical tool that allows transcendental idealism to dissolve the antinomies and thereby avoid reason coming into conflict with itself, i.e., “the euthanasia of pure reason”. For Kant, transcendental illusion is an unavoidable feature of transcendental realism because the latter assumes that the categories have an unlimited scope, treating mere rules of thought about objects as if they were rules for all possible reality, for things in general. In that respect, transcendental realism tacitly treats our cognition as if it were purely intellectual. Kant’s elaboration of the concept of a purely intellectual intuition, developed by way of contrast to a discursive cognition, helps us to see that the assumption of the unrestricted scope of the categories (with respect

126  Toby Lovat to knowledge) problematically presupposes that we can grasp things in themselves, as if they were a mere product of our intellect. And therein lies both the foundational problem of transcendental realism and the reason Kant thinks that transcendental idealism (for which the objective reality of the empirical is predicated on the formal ideality of space and time) is the only way to avoid perpetual wrangling and confusion in metaphysics, and ultimately reason’s dead end in self-contradiction. Furthermore, the restrictions to legitimate cognition that Kant introduces in the Aesthetics and the Analytic, and in the diagnosis of transcendental illusion in the Dialectic, suggest that his conception of intellectual intuition must be understood as contrastive (concerned with establishing both the limits of discursive cognition and the problematic presuppositions entailed by a failure to note or adhere to these limits) rather than as a positive theory of a divine cognition, since such a theory would be an exemplar of the speculative, transcendent, and ultimately dogmatic and doomed metaphysics that the Critique aims to refute and remedy. 4.6 Conclusion I will conclude with a summary of the function and import of Kant’s distinction between sensible and intellectual intuition in relation to his approach to the possibility of metaphysics. For Kant, the term appearances just means what is given to the mind in sensory experience. But his central innovation is the argument that what appears to us is mediated or framed by our mind’s forms of sensibility, i.e., space and time. And it is on account of this contribution or framing that Kant argues that things as they appear can be shown to accord necessarily with the basic concepts (i.e., categories) through which objects are amenable to cognition. But while this provides metaphysics with a means to establish a priori knowledge of reality, the cost of that advance is the restriction of our possible cognition to things as they appear, i.e., to empirical reality. As such, Kant contends that ‘the proud name of ontology … must give way to the more modest one of a mere analytic of the pure understanding’ (CPR A247/B303). While an ontology in the proud sense is the a priori science of the most general principles of things as they are in themselves, of being qua being, Kant’s transcendental idealism can be construed as offering a more modest ontology of appearances. At the same time, Kant reframes ontological questions about what exists as epistemological questions concerned with the rules of our understanding and their legitimate application. What that analysis shows is that insofar as ontology ‘presumes to offer synthetic a priori cognitions of things in general’ (CPR A247/B303), it assumes the possibility of unconstrained, or unmediated intellectual access to non-sensible objects. For Kant, the

Methodological Role of Intellectual Intuition in Kant’s Critique 127 proximate error here is a failure to conceive of objects in relation to contrasting epistemic-intuitive frameworks of their possible cognition. Since human cognition of objects is fundamentally mediated by forms of sensibility, unmediated cognition of objects entails a non-sensible intuition. Thus, for Kant, objects of sensible intuition, appearances, are objects considered in their relation to the cognitive framework of a sensible intuition; and objects of non-sensible intuition, things in themselves, are objects considered in their relation to the cognitive framework of a non-sensible intuition, i.e., an intellectual intuition. While, for Kant, the notion of appearances necessarily entails the notion of things in themselves, the latter is unknown and unknowable because cognition of things so construed would require intellectual intuition, ‘the possibility of which we cannot understand’. In the Transcendental Dialectic, Kant vindicates both his Copernican revolution in general and transcendental idealism in particular by showing how our efforts to approximate a God’s-eye view, by extending our cognition beyond the confines of sensibility, leads to either scepticism about possible knowledge (or even the existence) of empirical reality, or to an unrestricted use of reason that compels the understanding towards dogmatism and then contradiction by assuming that the ultimate, or unconditioned ground of appearances, is within its possible grasp.18 Notes 1 All quotations from Kant’s Critique of Pure Reason (CPR) are from Guyer’s and Wood’s Cambridge edition. 2 That reminder is important, because (1) much contemporary speculative metaphysics seems not to have considered Kant’s warnings about the dangers of overstepping the legitimate bounds of metaphysical reasoning, and (2) strongly metaphysical readings of Kant’s claims about things in themselves (or intellectual intuition) would seem to imply that he himself ignored these warnings. While I briefly address the latter of these claims in this chapter, a treatment of the former is beyond its scope. Nonetheless, I here endorse Robert Hanna’s judgement that ‘contemporary Analytic metaphysicians really and truly need to learn Kant’s eighteenth-century lessons’: (1) about the inherent limits of human cognition and knowledge, (2) about the unsoundness of all possible ontological arguments from logical or analytic necessity to actual or real existence, (3) about the essential cognitive-semantic difference between (3i) mere logical, analytic (a.k.a. “weak metaphysical”) possibility and (3ii) real, synthetic (a.k.a. “strong metaphysical”) possibility, and (4) about the essential ontological difference between noumena and phenomena. For without these insights, they have been, are, and forever will be inevitably led into the very same “obscurity and contradictions” that beset classical metaphysics prior to Kant (Avii). (Hanna 2017: 763)

128  Toby Lovat In Continental philosophy, perhaps the most ambitious recent effort to transcend Kant’s limitations on metaphysical reasoning can be found in Quentin Meillassoux’s After Finitude (Meillassoux 2009). I have previously argued that Meillassoux’s efforts to transcend Kantian ‘finitude’ by way of a proof of the necessary (or absolute) contingency of all reality fail, partly because the ‘obscurities and contradictions’ of that ontology are very clearly the result of what Kant calls transcendental illusion, ultimately a consequence of a failure to restrict reason to the legitimate scope of discursive cognition. See Lovat 2018. 3 Quarfood’s approach is developed against Förster’s (2010: 323) contention that, in the Critique of Judgment, ‘intellectual intuition and intuitive understanding are thematized as two distinct cognitive powers irreducible to one another’. For Förster, while these distinctive cognitive powers are ‘always identified with one another in the literature, they are not the same: in the first case the alternative is between receptivity and spontaneity, in the second case the alternative is between a discursive and intuitive understanding’ (Förster 2009: 203). Quarfood recognises these differences, but argues that their purpose is primarily contrastive, concerned with different aspects of discursive cognition, rather than with fundamentally different conceptions of a non-discursive cognition. 4 Kant’s claim that we ‘cannot … form … the least concept of’ an understanding whose mind would produce the content of reality might seem to conflict with the concept and characterisation of such an understanding he develops. Yet formulating a hypothetical conception of an intellectual intuition in its contrast with a sensible intuition is quite different to grasping what such an intuition might be like. 5 While the latter of these claims might seem to imply a contradiction with Kant’s claim that ‘intuitions without concepts are blind’, his point is that a determinate concept of every given object isn’t required. We can represent objects for which we have no determinate concept (e.g. an elephant, upon encountering one for the first time), but that is only because we have a raft of basic concepts (categories) and other empirical concepts through which we represent anything that appears. 6 According to Prauss (1977: 20), the phrase “thing in itself” is an abbreviation of “thing considered in itself” (Ding an sich selbst betrachtet). In this sense, Allison belongs to the group of Kant interpreters in the “two aspect” camp. 7 Ameriks characterises this ‘holy grail’ as follows: ‘[T]o find a meaning for the doctrine of transcendental idealism that is not only consistent, understandable in its origins, and not immediately absurd, but also does full justice to the complex fact that Kant insists on claiming both that there are “real appearances”… that is, appearances disclosing to us features of physical objects that are empirically real, and also that these features are nonetheless “mere appearances” in contrast to “things in themselves”’ (Ameriks 2011: 48). 8 My interpretation of the core argument of the deduction follows Allison (2004, 2015) and Gomes (2010) for whom Kant’s argument hinges on the claim that both the prior synthesis of space and time required for representation (and intuition) of sensible objects and the formal logic upon which the categories are based are grounded in the synthetic unity of apperception, i.e., ‘the understanding itself’ (CPR B134n). My reconstruction and interpretation of the deduction can be found in Lovat 2018. 9 It is this correspondence that Kant’s transcendental deduction aims to prove.

Methodological Role of Intellectual Intuition in Kant’s Critique 129 10 It is worth noting that this restriction pertains only to the use of theoretical reason, the subject matter of the first critique. Practical reason, says Kant, is not so restricted because in its practical use, reason can legitimately transcend possible experience: ‘[A]fter speculative reason has been denied all advance in this field of the supersensible, what still remains for us is to try whether there are not data in reason's practical data for determining that transcendent rational concept of the unconditioned, in such a way as to reach beyond the boundaries of all possible experience, in accordance with the wishes of metaphysics, cognitions a priori that are possible, but only from a practical standpoint’. (CPR Bxxiii) 11 Kant continues ‘but [reason] cannot discover them, for the principles on which it is proceeding, since they surpass the bounds of experience, no longer recognise any touchstone of experience’. 12 By tying both sensibility and the legitimate deployment of the understanding (with respect to knowledge of the world) to our form of intuition and hence to appearances, Kant’s idealism combines the strengths of both empiricist scepticism (for which we are incapable of a priori knowledge of things in themselves, because the mind can have immediate access only to its own ideas or representations) and rationalist dogmatism (for which we are capable of a priori knowledge of the world – but, for Kant, only considered as appearance). 13 Indeed, Kant excludes ‘logical illusions’ from this general characterisation of errors in judgement, since, e.g., ‘fallacious inference’… arises solely from a failure of attentiveness to … logical rule’ (CPR B353/A297). 14 Transcendental reflection, says Kant, is ‘[t]he action through which I make the comparison of representations in general with the cognitive power in which they are situated, and through which I distinguish whether they are to be compared to one another as belonging to the pure understanding or to pure intuition’ (CPR A261/B317). 15 Thus, for Kant, on Allison interpretation, Locke’s error of ‘sensitivizing the concepts of the understanding’ is predicated on the assumption that the conditions of our sensibility are also conditions of things as such, rather than merely conditions of things as objects of possible experience. This is the extension of the legitimate field of application of the categories from objects of possible experience to things in themselves. In contrast, Leibniz’s error of ‘intellectualiz[ing] the appearances’ i.e., his construction of ‘an intellectual system of the world … [by which he believed he was] able to cognise the inner constitution of things by comparing all objects to the understanding and the abstract formal concepts of its thinking’ (CPR A270/B326) is based on a failure to notice the ‘positive role of sensibility as a source of the realizing conditions of the understanding’ (Allison 2004: 325). In other words, this is the extension of categorial cognition under the assumption that sensibility is not a genuine condition of cognition. Both ontological systems thus entail the illicit extension of the categories beyond their legitimate field of application; both fail to note, albeit in opposite ways, the ‘influence of sensibility on understanding’. 16 Of reason’s role in all our cognitive activities, Kant writes, ‘the law of reason to seek unity is necessary, since without it we have no reason, and without that, no coherent use of the understanding, and, lacking that, no sufficient mark of

130  Toby Lovat empirical truth … we simply have to presuppose the systematic unity of nature as objectively valid and necessary’ (CPR A651/B679). 17 See also CPR A500/B528, where Kant suggests that advocates of these kinds of problematic claims about the world as a thing in itself are guilty of equivocating on the term “condition” – slipping between an empirical use as it applies to appearances in time, and a “transcendental” use which illegitimately applies a pure category to the world in order to make a claim about it as a thing in itself. 18 Thanks to Robb Dunphy for the many suggestions and invaluable editing support on various drafts of this chapter.

Bibliography Allais, L. (2015) Manifest Reality: Kant’s Idealism and His Realism (Oxford: Oxford University Press) Allison, H. E. (2004) Kant’s Transcendental Idealism: An Interpretation and Defense (New Haven, CT: Yale University Press) Allison, H. E. (2006) “Kant’s Transcendental Idealism”. In G. Bird (Ed.) A Companion to Kant Oxford: Blackwell Publishing), pp. 111–124 Allison, H. E. (2012) “Transcendental Realism, Empirical Realism, and Transcendental Idealism”. In Essays on Kant (Oxford: Oxford University Press), pp. 67–83 Allison, H. E. (2015) Kant’s Transcendental Deduction: An Analytic-Historical Commentary (Oxford: Oxford University Press) Ameriks, K. (2011) “Kant’s Idealism on a Moderate Interpretation”. In D. Schulting and J. Verburgt (Eds.) Kant’s Idealism: New Interpretations of a Controversial Doctrine (Dordrecht: Springer), pp. 29–53 Dickerson, A. B. (2011) Kant on Representation and Objectivity (Cambridge: Cambridge University Press) Förster, E. (2009) “The Significance of §§76 and 77 of the Critique of Judgment for the Development of Post-Kantian Philosophy (Part 1)” Graduate Faculty Philosophy Journal 30, no. 2, pp. 197–217 Förster, E. (2010) “The Significance of §§76 and 77 of the Critique of Judgment for the Development of Post-Kantian Philosophy (Part 2)” Graduate Faculty Philosophy Journal 31, no. 2, pp. 323–347 Gomes, A. (2010) “Is Kant’s Transcendental Deduction of the Categories Fit for Purpose?” Kantian Review 15, no. 2, pp. 118–137 Grier, M. (2001) Kant’s Doctrine of Transcendental Illusion (Cambridge: Cambridge University Press) Hanna, R. (2022) “Kant’s Theory of Judgment”, The Stanford Encyclopedia of Philosophy (Spring 2022 edition). Hanna, R. (2017) “Kant, the Copernican Devolution, and Real Metaphysics”. In M. Altman (Ed.) The Palgrave Kant Handbook (London: Palgrave Macmillan), pp. 761–789 Kant, I. (1998) Critique of Pure Reason. Translated by P. Guyer and A. Wood (Cambridge: Cambridge University Press) Kant, I. (2000) Critique of the Power of Judgement. Translated by P. Guyer (Cambridge: Cambridge University Press)

Methodological Role of Intellectual Intuition in Kant’s Critique 131 Lovat, T. (2018) “Back to the Great Outdoors? A Kantian Reply to Meillassoux’s Argument”. PhD thesis, The University of Brighton https://research.brighton.ac.uk/en/ studentTheses/back-to-the-great-outdoors-a-kantian-reply-to-meillassouxs-argume. Meillassoux, Q. (2009) After Finitude: An Essay on the Necessity of Contingency. Translated by R. Brassier (London: Continuum Books) Prauss, G. (1977) Kant Und Das Problem Der Dinge an Sich (Bonn: Bouvier) Quarfood, M. (2011) “Discursivity and Transcendental Idealism”. In D. Schulting and J. Verburgt (Eds.) Kant’s Idealism: New Interpretations of a Controversial Doctrine (Dordrecht: Springer), pp. 143–158

5

Kant’s Promise of a Scientific Metaphysics Catherine Wilson

5.1 Introduction In the Preface to the first edition of the Critique of Pure Reason, Kant declared his intention to some day write a speculative Metaphysics of Nature, far shorter but ‘incomparably richer in content’ than the present work (Axxi). He asserted even more strongly in the Preface to the second edition of the first Critique (hereafter CPR) that the critical philosophy was merely a ground-clearing operation for a future metaphysics to be presented ‘dogmatically, and systematically … in a way that complies with school standards (rather than in a popular way)’. This would apparently follow a Wolffian format (Bxxxvi–xxxvii). It has been noted since the late 19th century that Kant came to restore and renew German school philosophy, not to destroy it. This thesis has been revisited by more recent scholars (Freuler 1991; de Gurbert 1995; De Boer 2011, 2020), challenging the 20th-century textbook tendency to present Kant – not however without reason – as a rebellious destroyer of dogmatism.1 As Descartes had wished to supplant the academic philosophy textbooks of his time with his Principles of Philosophy, Kant apparently wished to fashion an indispensable replacement for the metaphysics of Wolff and Baumgarten. He was a particular foe of eclectic ‘popular philosophy’, as well as radical empiricism.2 ‘Naturalists’ of pure reason are contrasted later in the CPR with ‘scientists’ of pure reason. Naturalists are said to take immediate experience and common sense as constituting knowledge. They are ‘misologists’ or haters of reason. Like their forebears, the Epicureans, such people take the moon to be the size of a dinner plate, or whatever size it looks at a given moment.3 But like the mathematical physics that can reveal the actual size of the moon, metaphysics, equally intellectually demanding, can, Kant implies, become a science, pushing beyond naive sense-experience and common sense (A855/B883).

DOI: 10.4324/9781003187561-6

Kant’s Promise of a Scientific Metaphysics 133 In his second Preface, Kant claimed to have produced, in the previous edition of the CPR, a perfect account of ‘pure speculative reason’. ‘I have found nothing to alter’, he states, either in the propositions themselves or in their grounds of proof or in the form and completeness of the book’s plan; … [It] contains a truly articulated structure of members in which each thing is an organ, that is, in which everything is for the sake of each member, and each individual member is for the sake of all. … I hope this system will henceforth maintain itself in this unalterability. (Bxxviii) So that project was, at least in 1787, entirely complete. It is nevertheless controversial whether Kant was consistently committed to the further project of personally articulating a metaphysics of nature, complementary to his finished and allegedly unimprovable account of pure speculative reason. Did he already have a coherent idea of the rich content it should contain, and was he prevented only by lack of time and distractions from making greater headway on this constructive project? And even if he was consistently committed to it, did he possess the intellectual and conceptual resources to carry it out? If he did not possess them, no amount of extra time or a longer life would have made possible a metaphysics of nature parallel, as he intended it to be, to the metaphysics of morals. The thesis of this chapter is that Kant could not, and at some, though not all stages, knew that he could not write such a metaphysics of nature. To defend these claims, I begin with a short summary of metaphysics and its opponents in the 18th-century context, and with Kant’s motivation to usher in a revolution conservatrice for the sake of religion and morality. I argue that while the ‘Architectonic of Pure Reason’ supplies the ‘form’ of a metaphysics of nature, the ‘content’ is not only unspecified but unspecifiable.4 Finally, I summarise briefly some of the speculations of 18th-century theorising about nature in order to show how, although Kant strongly desired to provide an alternative, he was not in a position to do so. 5.2 18th-century Metaphysics and Its Critics Metaphysics was a topic on which Kant lectured for years. Baumgarten’s Metaphysica, which went through six editions between 1739 and 1768, served as his basic text for elucidation, commentary, and criticism. The work was divided into four parts: Ontology, Cosmology, Psychology, and Natural Theology. Ontology included, among other topics, the treatment of possibility, truth, cause and effect, necessity and contingency, singular and plural. Cosmology included the simple parts of the universe, its

134  Catherine Wilson origins, the perfection of the world, the nature of bodies, the natural and the supernatural. Psychology was divided into empirical psychology, the study of our mental faculties, and rational psychology: the nature of the soul, immortality, animal souls, and non-human spirits. Natural theology included the nature of God, Providence, the purpose of the world, divine decrees, and revelation. Christian Wolff’s earlier volumes had covered the same ground, as had most, if not all earlier systems belonging to what we currently consider the history of philosophy.5 Metaphysics as a discipline had been under attack in Britain, France, and the German-speaking provinces for nearly a century in Kant’s time, and, thanks to vigorous activity in translation and reviewing, Kant’s readers were well aware of these attacks.6 Radical challenges to religious orthodoxy as well had been present in Germany since the late 17th century (Wunderlich 2016, 2017). Pierre Bayle had found merit, absurdity, and undecidability in every system of philosophy and theology whether received or heterodox (Bayle 1741). Although Hume was initially better known and admired on the Continent as a historian than feared as a destroyer of metaphysics, his iconoclasm increasingly made itself felt (Gawlick and Kreimendahl 1987). In reducing causality to the habit of expecting consequences, he appeared to throw the concept of a law of nature and the basis for science and divine rationality into disarray. Hume’s other offenses to metaphysics were the reduction of the self to a series of experiences, a notion incompatible with the notion of the self as a locus of agency, and his rejection of the physico-theological argument as well as the familiar a priori arguments for the existence of God. Etienne de Condillac’s Locke-inspired Traité des systèmes (1749, German translation 1777), cited by Kant (28: 156), attacked the systems of Descartes, Leibniz, and Spinoza, declaring all reasoning about the terms being, substance, essence, nature, attribute, property, mode, cause, effect., liberty, and eternity, unreliable (Condillac 1777: 24). Jean-Baptiste Boyer d’Argens, mentioned by Kant, along with Bayle and Toland, as an unanswerable sceptic (16: 450), was the libertine author of Philosophie du bon sens which appeared in 1737 (d’Argens 1737). Bon sens was translated into German as Philosophie der gesunden Vernuft in 1756 and presented a sceptical treatment of historical knowledge, metaphysics, and astrology. For Julien Offray de La Mettrie, whose Oeuvres philosophiques were published in Berlin in 1774, we have no real knowledge regarding the nature of the soul, or infinity, or the existence of God, and it is ‘folly to torture ourselves so much about what we cannot know and what would not make us any happier if we did manage to know it’ (La Mettrie 1996: 23). In L’homme machine of 1748, he urged the reader to descend from ‘the top of that magnificent theological machine’ to the ‘physical ground

Kant’s Promise of a Scientific Metaphysics 135 from where, seeing all around you only eternal matter and forms perpetually taking each other’s place and perishing, you will admit in embarrassment that all animated bodies are destined for complete destruction’ (Ibid. 147). ‘Who knows after all’, he asked, whether the reason for man’s existence is not his existence itself? Perhaps he was thrown by chance on a point of the earth’ s surface without anyone being able to say how or why, but simply that he has to live and die, like mushrooms which appear from one day to the next, or flowers which grow beside ditches and cover walls. (Ibid. 23) Voltaire, in a work translated anonymously into German in the year of its appearance in French, said that we know nothing of matter or mind, nothing of ourselves, only that we have movement, life, and feeling without knowing how this comes about (Voltaire 1784: 36). In the ‘Preliminary Discourse’ to the Encyclopédie of 1751, Jean Le Rond d’Alembert claimed that the ‘essence of matter, and the manner in which we form an idea of it, will remain forever covered in clouds. We may conclude from our sensations that something exists outside of us, but this being that we call “matter” – is it similar to the idea that we form of it?’ (d’Alembert 1751: 58–59). We will never know whether the soul always thinks, how the union of soul and body takes place, or whether animals have awareness, as much as the “bold metaphysicians” try to inform us on these matters. ‘Over all these questions the supreme being has drawn a curtain that our weak view tries in vain to pull back’ (Ibid. 63). Kant owned this work in the German translation of 1761. Astute, scientifically gifted underminers, as well as mockers, and scoffers were everywhere, even in the German universities, and in Frederick the Great’s court. Beginning in the 1770s, the Göttingen school attacked metaphysics and commended in its place the study of psychology, the history of institutions, and anthropology. Frederick, who ruled from 1740 to 1786, wrote Epicurean poetry and moral philosophy and collaborated with d’Argens on an extract of Bayle’s Dictionary (1767). The scandalous ideas of empiricists and materialists reverberated in Prussia with the francophile monarch’s invitations to La Mettrie, Maupertuis, Diderot, and Voltaire to join the Berlin Academy and to spend time in his company. His court was seen as a hotbed of moral decadence. According to Wilhelm Tennemann’s History of Philosophy, the thought and behaviour of the French were characterised by ‘immortality, levity, the taste for pleasure and a high degree of intellectuality’. This mentality arose out of ‘vanity and frivolity, out of the avid search for the new and brilliant, scorn for the classic and established customs’ (Tennemann 1819, 11: 312).

136  Catherine Wilson 5.3  Philosophical Disorientation and Kant’s Reaction Kant too viewed this situation with alarm. As more and more writers resort to print, he complained, the ever-widening reading public becomes ever more engaged with popular philosophy. The real source of lack of faith in God, freedom, and immortality is, however, dogmatic metaphysics (Bxxx). The purported proofs of the metaphysicians, ‘arrived at through subtle but ineffectual distinctions’ are persuasive neither to the untutored public nor to the learned (Bxxxii–xxxv). Religious scepticism threatened to erode the foundations of morality and unleash criminality. Evil-doers could no longer be restrained by fear of divine punishment. In sensitive persons, mortalism produced disorientation and even desperation. A number of texts of the period show us how the hedonic moral philosophy of seizing the day and its transitory pleasures since death was really the end of everything provoked fear and anguish. In 1808, Kant’s admirer Heinrich Jung-Stilling blamed the mechanical philosophy introduced by Copernicus, Descartes, and Newton for existential despondency and the corruption of morals. Now, he lamented, there existed only matter and material forces working through eternal and unchangeable laws (1808: 15). Philosophers recognised only God with no influence on the world – the world could have existed from eternity and be her own God (Ibid. 21). They claimed that the human soul depended on the brain or that it was an incorporeal being that had no influence except on its own body (Ibid. 22). The world of Aufklaerung is the product of an ‘ice-cold forlorn rationality. … Is it any wonder that one who adopts this system puts a bullet through his head?’ (Ibid.). J.J. Spalding, in a book reprinted many times between 1748 and 1794, described his voyage from a disillusioned Epicureanism to religion and a Stoic morality.7 Friedrich Carl Casimir von Creuz poured out his feelings of forlornness in poetry (Wilson 2016). Hermann Samuel Reimarus expressed his moral confusion in the face of the writings of Lucretius, Rousseau, Buffon, La Mettrie, and Maupertuis. We are like ‘persons carried on board a ship in a profound sleep’. Awaking, one naturally asks ‘with some emotion, Where am I? How came I here? What am I to do in this place? Whither am I going? Who has the management of the vessel’? The ‘small company… and few trifling amusements’ on board do not provide an answer (Reimarus 1766: 461). Given this background, Kant’s desire to see metaphysics reconstructed in systematic form and dismantling the image of nature presented in the unsettling works of unorthodox French philosophers is understandable. Only a metaphysics that placed the human being above the rest of nature and oriented it to eternity and divinity could supply a keenly felt need. In the Metaphysics Lectures taken down by Volckmann in 1784–1785,

Kant’s Promise of a Scientific Metaphysics 137 Kant maintains that transcendental questions are inescapable, not only for educated philosophers but also for those growing up in an untutored environment. One may ask, why occupy oneself with metaphysical questions. But to renounce them is impossible as they are woven into our rationality, and we cannot teach our understanding to reject them for we cannot modify her nature. Did the world have a beginning? Is there anything in us that is indestructible? Will there be a future life? No human being can escape these questions. (28: 378) His frequently repeated claim that every human being longs for answers to these questions was not to be understood as expressing a need for proof that God exists and created the world, that the soul is immortal and will be judged for its objective moral failings in the afterlife, etc. All that was required – and here Kant appeared satisfied with what he had accomplished8 – were answers that could bring reason – vacillating between the alternatives described in the Antinomies and torn by doubts – out of conflict with herself. Kant’s intention to rescue the supersensible from sceptical and libertine critique can be read off from the 1787 second Preface to the CPR, in which, as in the Prolegomena, he attempted to clarify his aims and correct the misunderstandings aroused by the first edition of 1781. Evidently, its readers had not grasped his purpose; certainly not the first reviewers, Johann Georg Heinrich Feder and Christian Garve, who treated the book as a complicated rehash of what everyone already knew: that the world as it appears to our senses is not the world as it is in itself and who suspected Kant of naively promoting a version of Berkeleyan idealism (Garve 1783: 59–80). Moses Mendelssohn, by contrast, in the Preface to his Morgenstunden described Kant as the ‘all destroyer’ (Mendelssohn 2017: 3). To set the record straight, Kant removed certain rather mystical sounding passages (cf. A357) and distinguished his transcendental idealism sharply from Cartesian and Berkeleyan idealisms (B274). In the Preface, he explained that his actual purpose in the work was to cut off, ‘at the very root, materialism, fatalism, atheism, freethinking lack of faith, fanaticism, and superstition’ (Bxxxiv; cf. 4: 363). The contested terms fall into two categories. Materialism and atheism, along with what Kant calls ‘indifferentism’, were associated with empiricist philosophy, while fatalism (determinism) was associated with the metaphysics of Spinoza, Leibniz, and Wolff as well as with La Mettrie’s doctrine of “machine man”. By contrast, fanaticism and superstition, Kant’s Schwaermerei, pertained to Pietist excess, Swedenborgian dreams

138  Catherine Wilson and hallucinations, and the popular enthusiasm for ghost stories which were presented as empirical evidence for the incorporeal soul and its survival after the death of the material body. The failure of metaphysics to prove and communicate to the public the existence of God, and the incorporeality and immortality of the soul, had, in Kant’s view, pushed popular culture towards these absurd forms of religiosity. Kant’s strategy for setting the human being outside of nature and slaying these dragons was to approve investigations in natural science and to agree that mechanism should be assumed as far as possible in the explanation of all phenomena, including human behaviour, but to insist that Nature – as studied by the natural scientist – does not exhaust reality. Newtonian physics systematises the behaviour of the physical bodies that appear to us in the only way we can apprehend the world: as a set of causally interacting objects in space and time. But as bodies are only appearances, materialism and ‘thinking matter’ are metaphysically untenable (A357; cf. B519). Determinism, which implies that the world outside the mind is a vast array of rigid causal sequences in space and time, applies only to the appearances, insofar as space and time are only ‘forms of intuition’ (A536/ B564; 5: 95–97). In terms of the critical philosophy, materialism, fatalism, and atheism qualify as ‘speculative’ doctrines that may legitimately be opposed with other speculative doctrines such as the immortality of the soul (A779–80/B807–8). Kant agreed with the sceptics that one could not prove the existence of God, the freedom of the will, and the immortality of an incorporeal soul, the ‘principal truths’ that had long served to provide existential orientation. Nevertheless, he concluded, ‘The transcendental ideas … serve, if not to instruct us positively, at least to negate the impudent assertions of materialism, naturalism, and fatalism which constrict the field of reason, and in this way they serve to provide moral ideas with space outside the field of speculation’ (4: 363). The Canon of Pure Reason was the crowning achievement of the CPR as far as Kant was concerned. God and the soul were thought entities, whose existence was rationally and obligatorily, though subjectively, affirmed. There are a number of reasons for doubting that Kant could have contributed more to a constructive philosophy addressed to the worries of so many of his contemporaries than the relegation of empirical natural science to the systematic, explanatory regimentation of appearances; the fideistic avowals of the Canon;9 and the reconstruction of metaethics that constituted his contributions up to the late 1780s. First, the Architectonic sections of the CPR do not suggest that Kant himself really meant to address these topics fully and systematically. He concluded the Critique with a hopeful statement that before the end of the 18th century he – or someone else – would be able to help to turn

Kant’s Promise of a Scientific Metaphysics 139 the critical ‘path’ he had cleared into a ‘highway’ and to ‘bring human reason to complete satisfaction in what has always – although thus far in vain – engaged its desire to know’ (A856/B884). In a disconcerting, ironicsounding fashion, he went on to remark in the Appendix to the Prolegomena, that, where the production of a scientific metaphysics was concerned, the present moment is ‘not unfavourable to my expectation’, insofar as the Germans don’t seem to know what else to do with their time. However, he says, ‘[t]o discover how the endeavors of the learned may be united in such a purpose, I must leave to others’ (4: 382). Now that he has shown the way, ‘it cannot fail that a system, albeit not my own, shall be erected’ for grateful future generations. In both versions of the CPR and in the Prolegomena, then, Kant appeared to concede that he was not in a position to construct a full scientific metaphysics, including a metaphysics of nature, during his own lifetime. And strangely, in both editions of the CPR, a ringing denunciation of the very idea of a Wolffian Lehrgebaeude sits side by side with the description of the architectonic of a future metaphysics ‘[S]omeone who has … learned a system of philosophy (e.g., the Wolffian system)’, Kant declared, even if he has in his mind all the principles, explications, and proofs, … and could count it all on his fingers – would still have none but a complete historical cognition of the Wolffian philosophy. He knows and judges only as much as was given to him. Dispute one of his definitions, and he will not know whence to take another. He moulded himself according to the reason of another person. But the reproductive power is not the productive. (A836/B864) His knowledge is ‘merely historical’, Kant repeats. Though he has understood and remembered what he read, he is only ‘the plaster cast of a living human being’. If there is to be a new metaphysics to replace those current in the universities, it is hard to see how it can function without principles, explications, and proofs or, if they are to supply the format, how the new metaphysics can serve as the stimulus to the productive use of one’s own reason. Second, Kant’s requirements for a critical metaphysics of nature leave it unclear exactly what sorts of propositions concerning exactly what sorts of objects would be found in it. His own division of metaphysics in the Architectonic is quadripartite in its main headings and follows Baumgarten’s closely.10 He lists ontology, rational physiology, rational cosmology, and rational theology (A846/B874), and he gives us an elaborate account of its subparts and sub-subparts. According to his scheme, the metaphysics of nature divides into transcendental philosophy devoted to a priori

140  Catherine Wilson principles of the understanding, which Kant has already treated definitively, according to his own statement in the second Preface, and a ‘physiology of pure reason’, which he has not. This physiology, which deals with actual objects, not the understanding, is in turn divided into a physical or ‘immanent’ branch and a hyperphysical or ‘transcendental’ branch. The physical branch is in turn divided into rational physics, dealing with corporeal nature and rational psychology, dealing with thinking. The hyperphysical branch is divided into an ‘inner’ subbranch dealing with the world as a whole and an ‘outer’ subbranch dealing with the relationship of God to the world. Unlike in Baumgarten’s scheme, empirical psychology would not be discussed. Nor would the metaphysics contain any mathematical physics, but only the a priori principles of psychology and physics. There are, then, a number of formal ‘spaces’ for these classificatory headings. The problem is content: what can possibly occupy these spaces?11 Kant did not seem to have ideas about this. Rational cosmology and rational psychology were shown in the Antinomies and Paralogisms to have no cognitively accessible content. Once again, Kant points to the difficulty of filling these spaces. ‘People initially required more from metaphysics than can legitimately be demanded … they ultimately found their hopes betrayed; and thus metaphysics fell into general contempt’. Nevertheless, he insists, metaphysics must always remain the ‘bulwark’ if not the foundation of religion (A849/B877). Human reason, he continues, returning to his repeated refrain, cannot be satisfied with a naive sense-based empiricism (‘naturalism’), and the ‘empirical cognition of the human being’, i.e., anthropology, psychology, and natural history, without ‘the mediation of a rational cognition from mere concepts’, leads to ‘devastation’ in morality and religion (A850/B878). As he told his students in his lectures of 1782–1783, ‘The human understanding is … impelled by natural needs to know where all of its ends lead. … It is not satisfied with what the sensible world delivers to it; rather it must know what the future has in store for it – whoever believes that everything ends with death must have a low concept of his life. … The interest in the afterlife is indeed stronger than the interest in nature’ (29: 757). Earlier in the CPR, he had insisted that the ‘cold-blooded’ Hume and the materialist Joseph Priestley had taken too narrow a view of the ‘interest of reason’ by supposing that reason should stay close to the ‘laws of material nature’ (A745–6/B773–4). The interest of reason goes beyond scientific coherence and closure in this sense. A ‘lawless speculative reason’ must be prevented from leading away from –perhaps even spoiling by implication – the ‘general happiness’ (A856/B884). The sorrows and anxieties of Kant’s despondent contemporaries clarify this reference to the threat to the general happiness. But the question remains, why didn’t Kant think that the fideism of the Canon of Pure Reason

Kant’s Promise of a Scientific Metaphysics 141 was the perfect stopping point for philosophy, less than had formerly been hoped for by philosophers and their audiences, but fully adequate to secure social order and the general happiness? My suggestion is that the ‘daring and fertile’ (A851/B879) but nevertheless dangerous works of Kant’s contemporaries included not only impossibleto-substantiate metaphysical claims but also the ‘daring adventures of reason’ (cf. 5: 419, footnote 5) of contemporary natural science. To be clear here, a ‘scientific’ metaphysics, in Kant’s sense, had only to be rigorous; it did not have to deal much, or at all with the empirical sciences: physics, chemistry, theory of generation, astronomy, etc., and Kant specifically screens these off as not belonging to the metaphysics of nature. However, a ‘metaphysics of nature’ would have to present a systematic view of external nature, not just of human cognition of external nature. And the metaphysics of external nature propounded by the most active and prolific of French natural philosophers was at once too enticing and too depressing to be allowed to go unchallenged, while agnostic fideism was too weak to serve as the required ‘bulwark’ for morality and religion in this regard. Only a rival metaphysics of nature that restored the universal lawfulness of nature and the passivity of matter while allowing for free will, an incorporeal soul, and developmental and providential directedness would do. But how was this to be written? 5.4  The – Daring and Fertile – Metaphysics of Nature Jung-Stilling’s references to Copernicus, Descartes, and Newton as underminers of faith and moral confidence were out of date. The 18th-century literature is replete with general theorising about nature and with ‘systems of nature’ that gave far more cause for concern. Philosophers posited an all-embracing nature that did not include God, the immortal soul, or any other spiritual substances within it. Their systems were materialistic in ontology, sensualistic in epistemology, hedonistic in moral theory, and incompatible with Kant’s retributivist views on punishment. Though often characterised as Spinozistic (Israel 2001), they were more closely related in content to the Epicurean philosophy of Lucretius’s De Rerum Natura than to the writings of the historical Spinoza. J.C. Zabuesnig published an account of chiefly British and French “new philosophers” in 1777 and 1779.12 Many of them were available in German translation by mid-century. John Toland’s Letters to Serena (1704) defended active matter, urging thereby the gentlest possible criticism of Spinoza. Before d’Holbach’s terrifyingly deterministic physiochemical Système de la Nature of 1770, Pierre Louis Maupertuis had published his hylozoistic Système de la Nature in 1751 and Jean-Baptiste Robinet, his gloomy, Manichaean De la Nature in

142  Catherine Wilson 1761. A grand perspective on the place of the human being in nature was supplied in even greater detail by the Comte de Buffon. In his Histoire Naturelle, he presented nature as chaotic and classificatory systems as absurd, life as emerging from eternally existing ‘organic molecules’, and the earth as fated to cool down to a block of ice extinguishing all of living nature.13 Human and ape, he argued in 1766, are emotionally and cognitively disparate, but it is not certain that these differences really reflect the possession of an incorporeal soul in one but not the other (Buffon 1749–1804, 14: 32). Kant’s former pupil, Johann Gottfried Herder, who enjoyed good intellectual relations with the French, presented a vitalistic system that Kant found repellent (8: 45–66); and Georg Forster’s evolutionary ideas were equally distasteful (8: 179). Human and animal, said La Mettrie, are made of the same dough; the human brain is larger and more convoluted than that of the animal, but ‘nature has merely changed the yeast; she expends as much effort on the meanest insect as on man’. Men are ‘haughty vain beings who wish to exalt themselves, but they are only … vertically crawling machines’. Life resides in the smallest fibres of the animal body. Man is ‘only an animal or a construction made of springs which all wind each other up’ (de La Mettrie 1996: 31). ‘Nature’, he maintained, ‘created us all solely to be happy – yes, all, from the worm crawling on the ground to the eagle soaring on high’ (Ibid. 22). Virtues and especially vices were hereditary, according to La Mettrie; evil-doing was caused by madness or simply reflected the poor construction or the poor functioning of the machine. ‘We are no more criminal than the Nile is because of its floods, or the sea because of the ravages it causes’ (Ibid. 103). Punishment was required in the interests of society, but ‘it is much to be wished that excellent physicians might be the only judges’ (La Mettrie 1912: 119). The enlightened person will ‘pity the wicked without hating them; … and consider them as no more than mis-made men’ (Ibid. 147–148). The Baron d’Holbach too attacked the doctrine of the incorporeal soul as morally pernicious. ‘Man, in making a spiritual substance of his soul, has contented himself with administering to it spiritual remedies, which either have no influence over his temperament, or do it an injury. The doctrine of the spirituality of the soul has rendered morals a conjectural science, that does not furnish a knowledge of the true motives which ought to be put in activity in order to influence man to his welfare’ (d’Holbach 1783/1820–1821, I: 206). A metaphysics of nature that actually addressed Kant’s worried contemporaries would have had to take into account the developments in natural philosophy of the mid-18th century by physicians, physicists, chemists, embryologists, and anatomists who supported such radical claims. It would have had to provide the following: a general ontology of elementary

Kant’s Promise of a Scientific Metaphysics 143 particles or other units, forces, powers, and qualities; an account of space and time; a cosmology and theory of the earth; an account of living nature, its origins, its constitution, and its ultimate fate; an account of the human being as a member of the animal kingdom but with special characteristics, including or not including an incorporeal and immortal soul; a position on the mechanisms of generation, and the possibility of the transformation of species or their extinction. A systematic metaphysics of nature would also have to take a position on the topic of the relationship of God to Nature, in case God was held to exist, or on the self-sufficiency of Nature in the absence of God. Although Kant discussed all these natural philosophy topics in his essays, articles, and reviews, and in the Critique of Judgment, and even propounded his own physiological theory of African skin colour as caused by over-phlogisticated blood (8: 103) as well as the nebular hypothesis of the Universal Natural History, the only truly “constitutive” beliefs about what nature was and how nature operated, expressed without critical reservation and epistemic caution, pertained to the analysis of space, time, matter, and causation. For Kant was hostile to Newtonian science as the historian of science understands it and largely out of step with his scientific contemporaries (Zammito 1992: 189–190). The 18th-century Newtonian universe was indeed “daring and fertile”, It was a universe of active powers, embracing electrical and magnetic forces, a vital aether, formative forces in embryology, and even animal magnetism. Kant remained persuaded that mechanical explanations – as the 17th century had favoured – were uniquely suited to the human mind; all else was hypothetical and provisional. Where the universal applicability of ‘the mere mechanism of nature’ is concerned, ‘[U]nless we presuppose it in our investigation [of nature] we can have no cognition of nature at all in the proper sense of the term’ (5: 387). Kant allowed Blumenbach’s formative Bildungstrieb a role in embryology, but only because it was incapable of imparting organisation to dead matter and because, in Kant’s view, it left ‘an indeterminable and yet unmistakable share to natural mechanism’ (5: 424). Not only then was Kant’s commitment to the personal fulfilment of the project of a scientific metaphysics unsteady, the resources offered by the critical philosophy in conjunction with his scientific knowledge and prejudices were inadequate to this task. In the Metaphysical Foundations of Natural Science, Kant presented a metaphysics of corporeal nature in the form of principles he considered to be a priori synthetic. These dealt with the concepts of matter, space, and time and causality that were meant to be the necessary presuppositions of any system of physics. But the Metaphysical Foundations are clearly not the metaphysics of nature that was promised in the CPR. The metaphysics of nature, in order to fill out all of the Architectonic’s divisions and to address the cultural crisis that had

144  Catherine Wilson sparked Kant’s efforts, would have needed to provide a compelling alternative to the dangerous ‘isms’, presenting a non-deterministic, deistic, semiimmaterialistic system of nature that was free of any taint of enthusiastic mysticism. It would have had to anticipate later forms of Naturphilosophie. Providing a fuller account of what Kant called the “transition” from metaphysics to science remained an elusive goal (Butts 1984). It became, however, a project of his advanced old age, and the Opus postumum, with its return to the transition problem, its reflections on God, the soul, and the world, and its references to a universe-pervading caloric, a morphogenetic world spirit, and the fate of humanity, has a better claim to the title of an attempted metaphysics of nature. But bringing its material into the once promised Wolffian form and according to the Baumgarten-derived architectonic would have been a task, not just for Kant’s successors, but for the angels. And those who thought they understood the point of the Critique of Pure Reason regarding the true interests of reason, and who were satisfied by Kant’s account of morality and moral motivation in the Groundwork of the Metaphysics of Morals and the Critique of Practical Reason, might have been baffled. Just at this point, the elderly Kant’s own mental faculties were in their final downward spiral.14 His biographer Johann Gottfried Hasse described the scene in Kant’s household: For many years, a handwritten manuscript lay on his desk of more than a hundred folio pages, densely written, under the title ‘System of pure philosophy in its total presentation (ganzen Inbegriff) ….’ He allowed me several times to have a look at it and to leaf through it. There I saw that he concerned himself with the most important topics: philosophy, God, freedom, and, as he told me, chiefly with the transition from physics to metaphysics [sic]. … Kant confided this was his most important work, his ‘chef d’oeuvre’ and asserted that it was his complete system of everything completely finished and requiring only to be edited, which he hoped himself to do. (Hasse 1804: 21–22) The publisher would have to take care in this regard, Hasse adds, ‘since Kant in his last years often replaced better passages with something worse and added irrelevancies such as the meal plan for the day’ (Ibid. 22, footnote). Kant’s friend and partial heir, Johann Friedrich Gensichen, was taxed with the responsibility of getting it published, but another close friend, Johann Friedrich Schultz, told Hasse that ‘he had found nothing in the work corresponding to the title or that it would be advisable to publish’ (Ibid.). Kant’s promise of a complete future metaphysics was nothing more than the fond hope that something other than sceptical indifference,

Kant’s Promise of a Scientific Metaphysics 145 the raving of ghost-obsessed “enthusiasts”, or the empiricist image of the human being as wholly enclosed within nature must be possible for someone to deliver. Notes 1 Karl Leonhard Reinhold, who grasped Kant’s intentions and philosophical strategy well, emphasised the constructive aspects of the critical philosophy in his 1789 Letters on the Kantian Philosophy aiming to make it more intelligible to the average reader and succeeding well. Even Kant’s most philosophical friends and correspondents, including Herder and Mendelssohn, confessed to finding him difficult to understand, and some readers were shocked by what they did understand. See the letters of the furious Johann August Schlettwein, accusing Kant of arrogance and despotism (12: 372–370). 2 Kant describes popular philosophy in the Groundwork of the Metaphysics of Morals as ‘a disgusting hodgepodge of patchwork observations and halfrationalized principles … now perfection, now happiness, here moral feeling, there fear of God, a bit of this, and also a bit of that in a marvellous mixture’ (4: 409–410). See Crample-Casnabet 2002: 113–114. 3 The criticism applies to the modern followers of Epicurus, who in his Letter to Pythocles explained that ‘The size of the sun and the other heavenly bodies relative to us is just as big as it appears. But relative to itself it is either bigger or a bit smaller than it is seen as being, or just the same size’ (Epicurus 1994: 20). 4 For a more favourable estimate of Kant’s success – based on a different conception of what this task required – see De Boer 2020 and Franks 2022. 5 On the sources and meaning of a metaphysics of nature’ (Metaphysica Naturalis), see Freuler 1991: 384–385. 6 For further discussion, see Wilson 2022: Ch. 1 et. passim. 7 On the importance of this work for Kant, see Brandt 2003: 43; 2007. 8 Though in the Critique of Practical Reason, and even to some extent in Foundations of the Metaphysics of Morals that appeared between the two editions of the CPR, he sought to detach moral obligation more fully from fear of punitive divine judgement and to attach it more firmly to ‘reason’ and hopefulness (4: 409–410; 5: 126–140). 9 ‘Since the moral precept is thus at the same time my maxim, … I will inexorably believe in the existence of God and a future life, and I am sure that nothing can make these beliefs unstable, since my moral principles themselves, which I cannot renounce without. becoming contemptible in my own eyes, would thereby be subverted’ (A828/B856). 10 On its relationship to Baumgarten’s schema, see Switzer 2014. 11 An unfilled conceptual space (for a ‘thing’ treated as a ‘person’ as opposed to a ‘person’ treated as a ‘thing’, i.e., a servant) occurs in at least one other Kantian text (6: 358). 12 Volume Two contained presentations of many of the best-known and most scandalous authors: Argens, Bayle, La Beaumelle, Blount, Bollingbroke, Boulanger, Bruno, Chubb, Collins, Diderot, Dolet, the Encyclopedia, Freret, Helvetius, Herbert of Cherbury, Hobbes, Hubert, Maillet, Marmontel, Marsais, Meslier, La Mettrie, Montesquieu, Morgan, Prades, Premontval,

146  Catherine Wilson Rousseau, Servetus, Spinosa, Tindal, Toland, Toussaint, and Vanini. See also Anonymous 1780. 13 German translations of the Histoire Naturelle began to appear in 1750. Kant generally disapproved of Buffon (7: 221), but he nevertheless used parts of the Histoire as the basis for some of his lectures on physical geography and in composing his Universal Natural History. 14 Kant appeared to be suffering from a neurological disease at least from his retirement from teaching in 1796, though possibly from an earlier date. On his likely malady, see Fellin and Blè 1997, Marchand 1997, and Miranda et al. 2010.

Bibliography Intertextual citations are to Immanuel Kant, Gesammelte Schriften, Ed. Akademie der Wissenschaften (Berlin: 1902–). References to the Critique of Pure Reason (volume III, 2nd B ed., 1787, and volume IV, 1st A ed., 1781) are given in the usual A/B format. The translations cited of the Critique of Pure Reason are those of Werner S. Pluhar (Indianapolis: Hackett, 1996), (passim), and Paul Guyer and Alan Wood (Cambridge: Cambridge University Press, 1998), (p. 2). Other English translations cited are shown below. Volume and page numbers refer to the Academy edition cited above. Prolegomena to Any Future Metaphysics (1783). Translated and edited by Gary Hatfield (Cambridge: Cambridge University Press, 2004) Groundwork of the Metaphysics of Morals (1785). In Practical Philosophy, translated and edited by Mary J. Gregor (Cambridge: Cambridge University Press, 1996), pp. 37–108 Critique of Practical Reason (1788). In Practical Philosophy, translated and edited by Mary J. Gregor (Cambridge: Cambridge University Press, 1996), pp. 133–272 Anonymous (1780) Vernunftmäßige Prüfung des neuesten Systems der Natur (Frankfurt am Mayn: Andreä) Baumgarten, Alexander Gottlieb (2013) Metaphysics: A Critical Translation. Translated and edited by Courtney D. Fugate and John Hymers (London: Bloomsbury) Bayle, Pierre (1741) Historisches und Critisches Woerterbuch…uebersetzt…versehen von Johann Christoph Gottscheden… (Leipzig: Christoph Breitkopf) Brandt, Reinhard (2003) “The Guiding Idea of Kant’s Anthropology and the Vocation of the Human Being.” In Brian Wallace Jacobs and Patrick Kain (Eds.) Essays on Kant’s Anthropology (Cambridge: Cambridge University Press), pp. 85–104 Brandt, Reinhard (2007) Die Bestimmung des Menschen bei Kant (Hamburg: Felix Meiner) Buffon, George-Louis Leclerc, Comte de (1749–1804) Histoire Naturelle, Générale et particulière, 44 vols. (Paris: Imprimerie Royale) Buffon, George-Louis Leclerc, Comte de (1750–1785) Allgemeine Historie der Natur nach allen ihren besonderen Theilen abgehandelt (Hamburg and Leipzig: Grun and Holle)

Kant’s Promise of a Scientific Metaphysics 147 Butts, Robert E. (1984) “Kant’s Philosophy of Science: The Transition from Metaphysics to Science,” Proceedings of the Biennial Meeting of the Philosophy of Science Association, 685–705 Condillac, Etienne Bonnot de (1777) Herrn Abt Condillacs Unterricht aller Wissenschaften (Bern: Typographische Gesellschaft) Crampe-Casnabet, Michèle (1995) “Rousseau et Kant: une philosophie peut-elle être populaire.” In Popularité de la philosophie, Eds. Philip Beck and Denis Thouard (Lyon: ENS), pp. 108–141 d’Alembert, Jean le Rond (1751) Discours Préliminaire de l’Encyclopédie, Ed. Louis Ducros (Paris, 1893), repr. Leopold Classic Library, n.p. n.d. d’Alembert, Jean le Rond (1767) Essai sur les Éléments de Philosophie ou sur les Principes des connoissances humaines. In Mélanges de Littérature, d’Histoire, et de Philosophie, 4th ed. (Amsterdam: Zacharie Chatelaine et Fils), IV: 7–187 d’Argens, Jean-Baptiste Boyer (1737) Philosophie du bon sens: Réflexions philosophiques sur l’incertitude des connoissances humaines, l’usage des cavaliers et du beau sexe (London: Au dépens de la compagnie) d’Argens, Jean-Baptiste Boyer (1756) Philosophie der gesunden Vernuft, oder Philosophische Betrachtungen ueber die Ungewissheit der menschlichen Erkaentniss: Zum Gebrauch vornehmer Standes-Personen beiderlei Geschlechts (Breslau: Peitsch) De Boer, Karin (2011) “Transformations of Transcendental Philosophy: Wolff, Kant, and Hegel,” Hegel Bulletin, 32(1–2), 50–79 De Boer, Karin (2020) Kant’s Reform of Metaphysics: The Critique of Pure Reason Reconsidered (Cambridge: Cambridge University Press) Epicurus (1994) The Epicurus Reader. Translated and edited by Brad Inwood and L.P. Gerson (Indianapolis: Hackett) Fellin, Renato and Alessandro Blè (1997) “The Disease of Immanuel Kant,” Lancet 350, 1771–1773 Ferrari, Jean (1980) Les Sources Francaises de la Philosophie de Kant (Paris: Klincksiek) Ferrari, Jean (1992) “Kant, Lectuer de Buffon”. In Buffon 88: actes du Colloque International pour le Bicentenaire de la Mort de Buffon, Ed. Jean Gayon (Paris: Vrin), pp. 155–162 Franks, Paul (2022) “Reform and/or Revolution? Comments on Karin de Boer, Kant’s Reform of Metaphysics,” Kantian Review, 27(1), 127–132 Freuler, Leo (1991) “L’origine et la fonction de la metaphysica naturalis chez Kant,” Revue de Métaphysique et de Morale, 96, 371–394 Friedrich II (“The Great”) (1750–1762) Oeuvres du Philosophe de Sans Souci, 3 vols. (Berlin: Schneider) Garve, Christian (1783) “Review of Kant’s Critique of Pure Reason.” In Brigitte Sassen, Tr. and Ed., 2000. Kant’s Early Critics: The Empiricist Critique of the Theoretical Philosophy (Cambridge: Cambridge University Press), pp. 259–280 Gawlick, Gunter and Lothar Kreimendahl (1987) Hume in der Deutschen Aufklärung: Umrisse einer Rezeptionsgeschichte (Stuttgart: Frommann-Holzboog) Gurbert, Guillaume Pigeard de (1995) “Kant, lecteur de Boyer d’Argens.” In Popularité de la philosophie, Eds. Philip Beck and Denis Thouard (Lyon: ENS), pp. 176–189

148  Catherine Wilson Hasse, Johann Gottfried (1804) Letzte Äußerungen Kant’s von einem seiner Tischgenossen (Koenigsberg: Nikolovius) Herder, Johann Gottfried (1784) Ideen zur Philosophie der Geschichte der Menschheit, 4 vols. (Riga and Leipzig: J.F. Hartknoch) d’Holbach, Paul Henri Thiery (1783 System der Natur, oder von den Gesetzen der Physischen und Moralischen Welt (Frankfurt: s.n). d’Holbach, Paul Henri Thierry (1820–1821) The System of Nature (London: Helder) (facsimile reprint, n.p. n.d.) Hume, David (2003) A Treatise of Human Nature, Eds. David Fate Norton and Mary Norton (Oxford: Oxford University Press) Israel, Jonathan (2001) Radical Enlightenment: Philosophy and the Making of Modernity, 1650–1750 (New York: Oxford University Press) Jung-Stilling, Heinrich (1808) Theorie der Geister-Kunde in einer Natur-Vernunftund Bibelmäßigen Beantwortung der Frage: Was von Ahnungen, Geschichten, und Geistererscheinigungen geglaubt und nicht geglaubt werden müßte (Nürnbeg: Raw) La Mettrie, Julien Offray de (1912) Man a Machine. Translated by Gertrude Carman Bussey (La Salle IL: Open Court) La Mettrie, Julien Offray de (1996) Machine Man and Other Writings. Translated by Ann Thomson (Cambridge: Cambridge University Press) Marchand, J. C. (1997) “Was Emmanuel Kant’s Dementia Symptomatic of a Frontal Tumor?” Revue Neurologique, 153, 35–39 Mendelssohn, Moses (2017) Morgenstunden oder Vorlesungen über das Dasein Gottes (Stuttgart: Frommann-Holzboog) Miranda, Marcelo, Andrea Slachevsky, and Diego Garcia-Borreguero (2010) “Did Immanuel Kant Have Dementia with Lewy Bodies and REM Behavior Disorder?” Sleep Medicine, 11, 586–588 Reimarus, Hermann Samuel (1755) Die vornehmsten Wahrheiten der natürlichen Religion (Hamburg: J. C. Bohn) Reimarus, Hermann Samuel (1766) Principal Truths of Natural Religion. Translated by R. Wynne (London: B. Law) Reinhold, Karl Leonhard (1789) Briefe ueber die Kantische Philosophie (Mannheim: Bender) Robinet, Jean-Baptiste (1764) Von der Natur, …Aus dem Franzoezischen Uebersetzt (Frankfurt and Leipzig: Kraus) Spalding, Johann Joachim (1748) Betrachtung über die Bestimmung des Menschen (Greifswald: H.J. Struck) Switzer, Adrian (2014) ‘The Traditional Form of a Complete Science: Baumgarten’s Metaphysica in Kant’s “Architectonic of Pure Reason,”’ Philosophica: International Journal for the History of Philosophy, 22, 149–164 Tennemann, Wilhelm Gottlieb (1819) Geschichte der Philosophie, 11 vols. (Leipzig: Barth) Voltaire, Francois Marie Arouet (1784) La vie privée du roi de Prusse, ou Mémoires pour servir à la vie de M. de Voltaire, écrits par lui-même (Amsterdam: Rey) Voltaire, Francois Marie Arouet (1784) Das Privatleben des Königs von Preußen, oder Nachrichten zum Leben des Hrn. von Voltaire, von ihm selbst geschrieben, n.p.

Kant’s Promise of a Scientific Metaphysics 149 Wilson, Catherine (2016) “The Presence of Lucretius in Eighteenth-Century French and German Philosophy.” In Lucretius and Modernity, Eds. Liza Blake and Jacques Lezra (New York: Palgrave Macmillan), pp. 107–132 Wilson, Catherine, (2022) Kant and the Naturalistic Turn in 18th Century Philosophy (Oxford: Oxford University Press) Wunderlich, Falk (2016) “Materialism in Late Enlightenment Germany: A Neglected Tradition Reconsidered,” British Journal for the History of Philosophy, 24(5), 940–962 Wunderlich, Falk (2017) “Mortalismus und Materialismus in der deutschen Aufklärung,” Aufklärung, 29, 193–212 Zabuesnig, J. C. (1779)  Historische und kritische Nachrichten von dem Leben und den Schriften des Herrn von Voltaire und anderer Neuphilosophen unserer Zeiten (Augsburg: Veith) Zammito, John H. (1992) The Genesis of Kant’s Critique of Judgment (Chicago: University of Chicago Press)

6

Can Metaphysics Become a Science for Kant? Gabriele Gava

6.1 Introduction Famously, Kant presents the main question of the Critique of Pure Reason by saying that it intends to determine whether metaphysics can become a science (Bxv–xvi; B21–22).1 He answers positively to this question. He writes that the ‘experiment’ carried out in the text ‘promises to metaphysics the secure course of a science in its first part, where it concerns itself with concepts a priori to which the corresponding objects appropriate to them can be given in experience’ (Bxviii–xix). Therefore, the Critique establishes that metaphysics can become a science as far as it collects a priori cognitions regarding objects of possible experience. It is quite otherwise for its presumed capacity to obtain cognitions regarding unconditioned objects. In this respect, ‘there emerges a very strange result, and one that appears very disadvantageous to the whole purpose with which the second part of metaphysics concerns itself, namely that with this faculty we can never get beyond the boundaries of possible experience, which is nevertheless precisely the most essential occupation of this science’ (Bxix). The “strange” result is that we cannot obtain “theoretical” cognition of those very objects that seem the most important for metaphysics, like God and immortality.2 Nevertheless, Kant clearly states that metaphysics can become a science, although in a more modest fashion than we had initially in view.3 Since Kant is explicit on this issue, my question in this chapter does not concern whether Kant claimed that metaphysics could become a science. Rather, the question arises because even though Kant asserts this possibility, he provides an account of what philosophers can legitimately claim regarding their attempts to realise metaphysical systems, where this account seems to imply that metaphysics cannot attain one of the conditions of science that Kant himself sets. The condition that I have in mind is what Kant calls ‘architectonic unity’.4 Kant claims that a body of cognitions needs to enjoy this kind of unity in order to become a science. Having DOI: 10.4324/9781003187561-7

Can Metaphysics Become a Science for Kant? 151 architectonic unity, however, does not simply mean to have organised cognitions systematically. In addition to this, the cognitions in questions must be organised according to one “idea” given a priori by reason. Therefore, in order to be able to claim that a body of cognitions can become a science, it seems that we should at least be in the position to claim that we can grasp the idea of a science and that we can organise its body of cognitions according to that idea. The problem is that Kant seems to maintain that no philosopher can ever legitimately claim to have grasped the idea of metaphysics. This is suggested by his contention that nobody can claim to have matched the idea of the philosopher. But this generates a problem for Kant’s claim that metaphysics can become a science. In Section 6.2, I will start by having a closer look at the problem I have just sketched. In Section 6.3, I will propose a first attempt at a solution, one that builds on the claim that it is indeed possible to grasp the idea of metaphysics, even though nobody can ever be sure that they have done that. This claim captures a line of thought that is present in the Architectonic of Pure Reason. However, I do not think it identifies the main reason why Kant states that no philosopher can legitimately claim to have matched the idea of the philosopher. In Section 6.4, I submit that we can make sense of the latter contention by distinguishing between two ways of understanding the tasks of philosophy: according to the first way, philosophy is simply a system of “philosophical” cognitions that collects both “theoretical” and “practical” cognitions. In this picture, metaphysics is the part of philosophy that comprises philosophical cognitions that are a priori. By contrast, according to the second way, philosophy must provide practical guidance regarding how we can attain virtue. To illustrate this distinction, I will focus on three further distinctions: that between the “school concept” and the “worldly concept” of philosophy, that between philosophy as a doctrine of skills and philosophy as a doctrine of wisdom, and that between the “philodox” and the “misologist”. Finally, in Section 6.5, I will suggest that framing Kant’s demand of modesty towards philosophers from the perspective of these distinctions provides a new meaning to the problem from which we started. However, it is not clear whether Kant has a solution for the problem even when it is construed in this way. 6.2  Architectonic Unity and the Demand of Modesty Famously, in the Architectonic of Pure Reason, Kant submits that systematicity is a condition of science. Accordingly, he writes that ‘systematic unity is that which first makes ordinary cognition into science’ (A832/ B860). He further distinguishes between “technical” and “architectonic” unity. The former arises empirically, ‘in accordance with aims occurring contingently’. The latter arises ‘from the chief end of reason’, where this

152  Gabriele Gava end is an idea given a priori by reason (A833/B861). It is only architectonic unity that can provide the status of science to a body of cognitions: ‘What we call science … cannot arise technically, … but arises architectonically’ (A833/B861). Therefore, while technical unity is a form of systematic unity, it is insufficient to attain the status of science because the principle according to which it organises cognitions is contingent and arbitrary. By contrast, architectonic unity is achieved when we not only organise cognitions systematically but also adopt a principle of organisation that has its seats a priori in reason.5 These claims are puzzling in many ways. For starters, it is not clear what the content of the idea that gives unity to a science should be.6 Moreover, Kant does not provide any justification for the contention that reason admits only one ordering of cognitions as non-arbitrary. I will not discuss these difficulties here.7 What is important for my purposes is that Kant appears to be committed to the following theses: 1 There is one correct ordering of the cognitions belonging to a science. 2 In order to be considered a science, the body of cognitions belonging to a science must be organised according to its correct ordering. 3 We can have a priori access to the idea of reason that prescribes this ordering. 4 We can organise the cognitions belonging to a science according to their correct ordering, so that they can become a science.8 Now, these claims have the following consequences for the “scientists” who attempt to establish a science: a A scientist can grasp the idea prescribing the ordering of the cognitions of a science. b A scientist can organise the cognitions of a science according to that idea. Therefore, because “philosophers” are the “scientists” who attempt to establish metaphysics as a science, when we claim that metaphysics can become a science, it appears that Kant must accept the following contentions: a A philosopher can grasp the idea prescribing the ordering of the cognitions of metaphysics. b A philosopher can organise the cognitions of metaphysics according to that idea. However, some passages in the Architectonic conflict with both (a) and (b), so it seems that a condition for metaphysics to become a science

Can Metaphysics Become a Science for Kant? 153 cannot be met. In a first passage, Kant links the idea of philosophy to an “archetype” (Urbild) that constitutes the standard according to which the correctness of our attempts to build philosophical systems should be judged: Now the system of all philosophical cognition is philosophy. One must take this objectively if one understands by it the archetype for the assessment of all attempts to philosophize, which should serve to assess each subjective philosophy. … In this way philosophy is a mere idea of a possible science, which is nowhere given in concreto, but which one seeks to approach in various ways until the only footpath, much overgrown by sensibility, is discovered, and the hitherto unsuccessful ectype, so far as it has been granted to humans, is made equal to the archetype. (A838/B866) The passage focuses on philosophy, whereas we have so far directed our attention towards metaphysics. However, since metaphysics is a part of philosophy (A840–841/B868–869), what applies to philosophy in general should apply to metaphysics in particular, too. Notice two things about the passage. Firstly, in a way similar to any idea of a science, the archetype of philosophy provides a standard of correctness for assessing concretely existing philosophical systems. Secondly, Kant does not rule out that existing philosophical systems can realise the archetype. He writes that no presently existing system has matched the archetype yet, but he leaves open the possibility that this might be possible in the future. If we apply this thought to metaphysics and its relationship to its idea, it is possible that we get to a point where the cognitions belonging to it are ordered according to the idea that prescribes their correct ordering. So, no problem for Kant’s claim that metaphysics can become a science here. However, a few lines later, Kant uses again the concept of the archetype in relation to the idea of philosophy, but this time he links it to a particular understanding of the figure of the philosopher, characterised as ‘the legislator [Gesetzgeber] of human reason’ (A839/B867). In this context, he submits that ‘[i]t would be very boastful to call oneself a philosopher in this sense and to pretend to have equaled the archetype, which lies only in the idea’ (A839/B867). Here, Kant submits not only that it would be pretentious to claim to have realised the archetype of the philosopher, but also that that archetype must remain an idea. We find similar contentions in student notes from Kant’s lectures. For example, in the Lectures on Philosophical Encyclopedia, Kant states: ‘An archetype remains an archetype no more if it can be reached. It should serve merely as a guide [Richtschnuur]. The philosopher is only an idea’ (29: 8).

154  Gabriele Gava Understanding what it means that the idea of the philosopher must remain an archetype will be a central aim of this chapter. For now, it is sufficient to see why this claim can generate a problem for Kant’s contention that metaphysics can become a science. The claim might be taken as a request of modesty directed towards any philosopher: that is, no philosopher can legitimately maintain that she has said the last word regarding philosophy; she cannot legitimately maintain that she has grasped and realised the ultimate philosophical system. In one sense, the claim might only concern what a philosopher can maintain. Leaving aside the question whether she can in fact realise the ultimate system of philosophy, the claim would submit that she cannot maintain that she has done it. But even if we understand the claim in this modest form, it seems that it causes problems for Kant’s contention that metaphysics can become a science. For if nobody can ever say that she has realised philosophy as a science, where this, of course, also applies to metaphysics as a part of philosophy, in which sense can we say that metaphysics can become a science? However, since Kant submits that the idea of the philosopher must remain an idea, the claim can be taken more strongly, as implying not only that no philosopher can legitimately maintain that she has realised the ultimate system of philosophy, but also that she cannot actually realise it. 6.3  Reason’s Lack of Self-Transparency One first way to solve the problem is to embrace the first way of understanding Kant’s request of modesty and try to make plausible the idea that metaphysics can indeed become a science even though nobody can legitimately claim that one has realised metaphysics as a science. Maintaining that metaphysics can become a science on this basis means maintaining, on the one hand, that an individual philosopher is perfectly capable of realising metaphysics as a science and, on the other, that she cannot however legitimately claim that she has done that, even when she did. But how could we justify this contention? I believe that it can be justified when we focus on reason’s lack of selftransparency.9 We know that metaphysics collects a priori cognitions of reason. Clearly, we are capable to this kind of cognitions. However, given Kant’s account of reason, it is also clear that we do not have an immediate and transparent grasp of the rational cognitions that are in our reach. Take Kant’s account of transcendental illusion. According to it, our reason has a natural tendency to make claims it is not justified in making, where this leads us to believe that we are able to attain rational cognitions that in fact are impossible for beings with the cognitive capacities that we happen to have. The claim that reason has not a clear grasp of the cognitions that are in its reach does not only apply to reason in the narrow sense and its

Can Metaphysics Become a Science for Kant? 155 contentions concerning the unconditioned. Rather, something similar can be said regarding valid rational cognitions that concern objects of possible experience. If we take the a priori principle of causality, this is clearly a valid a priori cognition of reason (understood in the broad sense as the faculty of a priori cognitions). However, it is not the case that we have a clear and immediate grasp of the nature and limits of this cognition. Rather, we have a natural tendency to take the principle as applying to things in themselves. As a consequence of this, we misrepresent the conditions of its valid application. Given reason’s lack of self-transparency, it is possible to make sense of the idea that an individual philosopher is perfectly capable of realising metaphysics as a science while she cannot legitimately claim that she has done that. She can realise metaphysics as a science because metaphysics comprises rational cognitions that are in her power. She cannot ever legitimately claim that she has realised metaphysics as a science because, given reason’s lack of self-transparency, she cannot ever be completely sure that the system of cognitions she is proposing matches the correct system of metaphysical cognitions.10 6.4  The Unrealisability of the Idea of the Philosopher Even though I believe that it is correct, on Kantian grounds, to advance a request of modesty towards philosophers along the lines I just sketched, I do not think that this is what Kant has in mind when he says that the idea of the philosopher is unrealisable. In my view, it is no coincidence that Kant’s claim concerns the unrealisability of the idea of the philosopher and not philosophy. Moreover, both in the Critique of Pure Reason and in the passage from the Lectures on Philosophical Encyclopedia that we quoted, Kant speaks of the idea of the philosopher only after having introduced the distinction between the “school” concept and the “worldly” concept of philosophy. These clues point to a different understanding of the claim, according to which the latter means that no philosopher can ever become an example of perfect morality. To show this, I will begin by focusing on three relevant distinctions: that between the school concept (Schulbegriff) and the worldly concept (Weltbegriff) of philosophy, that between philosophy as a doctrine of skills and philosophy as a doctrine of wisdom, and that between the “philodox” and the “misologist”. 6.4.1  The School Concept and the Worldly Concept of Philosophy

In both the Critique of Pure Reason and various lectures notes, Kant links the unrealisable archetype of the philosopher to the idea of the philosopher as a legislator (Gesetzgeber) of human reason (see A839/B867; 9: 24;

156  Gabriele Gava 29: 7–8). This is contrasted with the idea of the philosopher as an “artist of reason” (Vernunftskünstler) (9: 24; 29: 7–8). What is distinctive of the former is that it takes into consideration the supreme ends of human reason (29: 8). This also characterises what Kant calls the worldly concept of philosophy. While according to its school concept philosophy is a ‘system of cognition that is sought only as a science without having as its end anything more than the systematic unity of this knowledge’ (A838/B866), according to its worldly concept, ‘philosophy is the science of the relation of all cognition to the essential ends of human reason’ (A839/B867). It is no wonder that it is exactly this second concept of philosophy that is often ‘personified and represented as an archetype in the ideal of the philosopher’ (A838–9/ B866–7). Kant explicitly stresses that the philosopher as a legislator of human reason pursues the worldly concept of philosophy (A839/B867). Kant does not clarify what the essential or supreme ends of reason are, but he submits that there is only one final end at the top of them (A840/ B868). According to a student note, the final end in question is the highest good (29: 948). This identification is consistent with the general outlook of Kant’s system. Kant explicitly links the final end of philosophy according to its worldly concept to morality (A840/B868) and it is plausible to consider the highest good the chief end of reason from the perspective of Kant’s moral philosophy. As is well known, the concept of the highest good postulates a proportionality between moral worth and happiness. Therefore, it seems that both philosophy according to its worldly concept and the idea of the philosopher as a legislator of human reason give centre stage to the thought of this proportionality. What is still not clear is why Kant stresses that the concept of philosophy according to its worldly concept is often personified in the idea of the philosopher. 6.4.2 Philosophy as a Doctrine of Skills and Philosophy as a Doctrine of Wisdom

Kant often links the worldly concept of philosophy to the idea of philosophy as a “doctrine of wisdom” and the school concept of philosophy to the idea of philosophy as a “doctrine of skills”. The idea behind this distinction is that insofar as philosophy according to its school concept does not take into consideration which ends we should pursue, the cognitions it collects are made available for a multiplicity of ends. In this sense, the philosopher ‘gives rules for the use of reason for any sort of end one wishes’ (9: 24). By contrast, philosophy as a doctrine of wisdom provides guidance regarding how we can attain our most fundamental ends and, among these, the highest good in particular. But how can a doctrine of wisdom show us how we can pursue the highest good? To answer this question, we must first keep in mind that

Can Metaphysics Become a Science for Kant? 157 of the two elements composing the highest good, virtue and happiness, it is only the former that is in our control. Happiness is only something we can hope for once we have made ourself worthy of it. Because a doctrine of wisdom can only guide us towards the highest good for the part of it that is in our power, it seems that it should first of all show us how we can become virtuous. This clarifies why philosophy, when understood according to its worldly concept and as a doctrine of wisdom, is often personified in the figure of the philosopher. For Kant believes that it is only by being an example of morality that we can really teach anybody how to become virtuous. In the Jäsche Logic, he writes that a ‘teacher of wisdom’ delivers his teachings ‘through doctrine and example’ [my emphasis] (9: 24). In the Vienna Logic, we find a characterisation of the figure of the philosopher which also emphasises its capacity to be an example of morality: ‘[t]he cause of the fact that we esteem highly someone who arranges his actions according to the strictest laws of morality, and who never departs from the straight path, is perhaps that morals is in fact always the end toward which all speculations tend. Morals constitutes a unity of all cognition of reason, and only he who follows its rules can be called a philosopher’ (24: 798–799). In the Critique of Practical Reason, Kant explicitly links the contention that no philosopher can legitimately claim to have attained the archetype to the need to be an example of perfect morality: ‘philosophy, as well as wisdom, would itself always remain an ideal, whereas subjectively, for a person, it is only the goal of his unceasing endeavours; and no one would be justified in professing to be in possession of it, so as to assume the name of philosopher, unless he could also show its infallible effect in his own person as an example (in mastery of himself and the unquestioned interest that he preeminently takes in the general good), which the ancients also required for deserving that honorable title’ (5: 108–109). At this point, one might ask why a philosopher as a teacher of wisdom must provide both a “doctrine” and an “example”. Why is it not sufficient that she provides a principle to determine what is the right thing to do in a particular situation? After all, Kant’s practical philosophy does exactly that. It delivers a principle that enables us to check whether the maxims that guide our actions are morally acceptable. Clearly, Kant’s categorical imperative is able to determine what is morally right and morally wrong. However, we are not perfectly rational beings and often, even though we recognise what is morally prescribed in a particular case, we fail to act accordingly. Given this aspect of our nature, guidance regarding how to be perfectly virtuous should not only determine what is morally right in each particular situation. It must also show us how we can bring ourselves and our partly sensible nature to infallibly comply with the principle of

158  Gabriele Gava morality.11 But this cannot be done by providing an additional set of rules. Rather, it can only be done by providing an example of how attaining perfect morality has been possible for a particular person. This account of the philosopher as a teacher of wisdom provides a completely new meaning to the claim that idea of the philosopher is unrealisable. The latter claim does not mean that no particular philosopher can ever realise – or claim to have realised – the ultimate system of philosophy. Rather, it means that no particular philosopher can ever be – or claim to be – an example of perfect morality for other human beings. 6.4.3  The Philodox and the Misologist

One issue that we have not considered yet is how the two understanding of philosophy that we have analysed so far are related to one another. How does philosophy according to the school concept and philosophy according to the worldly concept relate to one another? Some passages suggest that Kant regarded these two senses of philosophy as reciprocally dependent on each other.12 Accordingly, he calls philodox the philosopher who is only concerned with erecting a system of philosophical cognitions and does not consider essential ends of reason. ‘The artist of reason, or the philodox, as Socrates calls him, strives only for speculative knowledge, without looking to see how much the knowledge contributes to the final end of human reason’ (9: 24).13 By contrast, the misologist is the philosopher who only cares about wisdom and is not interested in building a system of philosophical cognitions as a science. ‘He who has hatred toward all the sciences, and who pretends that wisdom alone is to be esteemed, is called a misologist’ (24: 800; see also 9: 26). Therefore, the misologist is characterised by giving up any attempt to build philosophy as a science, as a system of valid cognitions of reason, while she only focuses on providing practical guidance on how to attain virtue and the highest good. Both philodoxes and misologists are mistaken in their endeavours. The former are mistaken in thinking that philosophy can do without considering essential ends of reason. These ends are simply fundamental from the perspective of our human nature and philosophy is the only discipline that can take them into account. The latter are mistaken in thinking that wisdom can be attained without science. On the one hand, this means that trying to provide practical guidance on how to attain virtue or the highest good without first determining what these are is hopeless. On the other hand, if we have to provide guidance towards attaining those goals, it is important that we first figure out whether and how they are possible.

Can Metaphysics Become a Science for Kant? 159 6.5 Metaphysics as a Science and Philosophy as a Doctrine of Wisdom In the last section, I have suggested that Kant’s claim that the idea of the philosopher is unrealisable must not be read as we first imagined. The claim means that no particular philosopher can ever be – or claim to be – an example of perfect morality for other human beings. What consequences does this have for the problem from which we started? Does it mean that, after all, the claim does not generate particular difficulties for Kant’s contention that metaphysics can become a science? Indeed, Kant’s distinction between philosophy according to its school concept and philosophy according to its worldly concept might be read as confirming that the problem disappears. For one might argue that whether metaphysics can become a science is a problem that only concerns philosophy according to its school concept. Metaphysics as a science, as a part of philosophy as a science, must provide a system of a priori cognitions of reason, where, one might continue, this system does not need to include a doctrine of wisdom, understood as a discipline providing practical guidance on how to become virtuous. This approach finds confirmation in Kant’s account of the misologist of reason. As we have just seen, the misologist only seeks wisdom and ignore the pursuit of science, which suggests that “science” and “wisdom” are sharply separated. True, Kant contends that both the philodox and the misologist are wrong in thinking that they can pursue either science or wisdom in separation from one another. But this does not necessarily imply that in order to attain metaphysics as a science, we must also perfectly embody the idea of the philosopher as a teacher of wisdom. Therefore, from the fact that nobody can ever become – or claim to have become – a teacher of wisdom, it does not follow that we cannot attain metaphysics as a science. The solution to our problem is not so simple, however. In fact, what the new understanding of the claim regarding the unrealisability of the idea of the philosopher provides is a new way of picturing the problem from which we started. This new version of the problem arises because Kant appears to think that philosophy can only attain the status of science when it is construed according to its worldly concept.14 Accordingly, in a footnote in the Architectonic chapter, Kant writes: ‘A worldly concept here means one that concerns that which necessarily interests everyone. I determine the aim of a science in accordance with school concepts if it is regarded only as one of the skills for certain arbitrary ends’ (A839/B867n). As we have seen in Section 6.2, it is technical unity that is achieved in relation to arbitrary ends. By contrast, architectonic unity is achieved in relation to reason’s supreme ends and ideas, where these provide a principle of

160  Gabriele Gava organisation that is not arbitrary. Since, as we saw, architectonic unity is necessary for science and philosophy according to its school concept can at best achieve technical unity,15 it follows that philosophy according to its school concept cannot become a science. This has at least two consequences for how we should address the problem of this chapter. Firstly, we cannot stress that whether metaphysics can become a science is a problem that only concerns philosophy according to its school concept. So it seems that the solution that we sketched at the beginning of this section is not an option. Secondly, since philosophy according to its worldly concept is often personified in the figure of the philosopher as a “legislator of human reason” and a “teacher of wisdom”, and since it is according to this concept that philosophy can attain architectonic unity, it seems that the question of whether this idea of the philosopher can be realised is indeed relevant for determining whether philosophy can become a science. At this point, one might point out that even though Kant believes that it is only according to its worldly concept that philosophy can become a science, this does not necessarily apply to metaphysics, as a part of philosophy. Kant might contend that while it is possible for metaphysics to become a science even without considering essential ends of reason and how to attain them, this issue is only central for philosophy as a whole. I do not think that this suggestion works, for at least two reasons. Firstly, Kant’s discussion in the Architectonic is clearly focused on a priori cognitions of reason, which belongs to metaphysics. Secondly, the idea of the highest good and the question regarding whether we can attain it play an important role within metaphysics for Kant. If this is right, it seems that Kant’s idea of metaphysics as a science is paradoxical.16 On the one hand, he tells us that metaphysics can only achieve the status of science when it is construed according to a particular understanding of philosophy, one according to which philosophy should provide guidance regarding how we can achieve virtue and the highest good. On the other hand, given some characteristics of that same understanding, namely that it requires that somebody become an example of perfect morality, where this is impossible, it seems that metaphysics cannot ever be completely realised. Now that we have presented the problem in its new form, let me sketch a tentative solution to it. A possible way to solve the paradox might lie in distinguishing two different ways in which we can consider the highest good as a fundamental end of reason. In a first sense, saying that metaphysics must consider the highest good as the final end of reason might simply mean that it must take into account that it is in the nature of human beings to have this final end. A consequence of having this end is that we have a particular interest in the truth of some propositions that have

Can Metaphysics Become a Science for Kant? 161 always had a central place within metaphysics, like those asserting that God exists and the soul is immortal. Being conscious that we have this interest and that it could bring us to making unjustified contentions in metaphysics can be a means to avoid those contentions, which is in turn instrumental to achieving the status of science. More generally, taking into account that it is in the nature of human beings to have the highest good as a final end could mean to explicitly consider whether it is achievable and whether it conflicts with theoretical cognitions we have. In a second sense, saying that philosophy must take into consideration the highest good as our supreme end might instead mean that it should provide guidance regarding how this end can be concretely achieved. It is according to this second sense that philosophy must show, through the example of the philosopher, how one can achieve perfect morality in one’s actions. When we distinguish between these two ways of taking the highest good into consideration, we can accept the claim that metaphysics can only become a science according to its worldly concept without implying that this requires that a philosopher must be able to become an example of perfect morality. Rather, saying that metaphysics should be construed according to its worldly concept would only mean that it must take into account that it is in the nature of human beings to have the highest good as a final end. By contrast, trying to approach the archetype of the philosopher in our actions would be what we should do when we pursue philosophy as a doctrine of wisdom.17 6.6 Conclusion In this chapter, I started from a problem that appeared to threaten Kant’s contention that metaphysics can become a science. The problem arises because Kant’s account of science requires that the cognitions belonging to a science are ordered according to an idea of reason that prescribes their correct ordering. However, Kant submits that the idea of the philosopher is unrealisable, where this can be taken as maintaining that nobody can ever grasp and realise the correct idea of philosophy and metaphysics (as a part of philosophy). I have suggested that this is not the correct way of understanding Kant’s contention that the idea of the philosopher is unrealisable. The claim should be understood from the perspective of the worldly concept of philosophy and philosophy as a doctrine of wisdom. From this point of view, the claim means that nobody can ever become – or claim to have become – an example of perfect morality. However, this new understanding of Kant’s claim does not provide a solution to our problem. It only put the problem in a different way. This is the case because Kant claims that it is only according to the worldly concept that philosophy (and metaphysics) can become a science. Therefore, it seems that Kant’s conception of

162  Gabriele Gava metaphysics as a science is paradoxical. On the one hand, he tells us that metaphysics can only achieve the status of science when it is construed according to a particular understanding of philosophy. On the other, given some characteristics of that same understanding, it seems that metaphysics cannot ever be completely realised. I have sketched a tentative solution to the problem, when it is understood in this latter form, one that rests on a distinction between two ways in which metaphysics and philosophy can consider the highest good as a fundamental end of reason. Whether this solution is successful or not, one advantage of understanding Kant’s claim that the idea of the philosopher is unrealisable along the lines I have suggested is that it allows us to draw a strong link between Kant’s Transcendental Doctrine of Method in the first Critique and Kant’s “practical” doctrines of method in the second Critique and the Doctrine of Virtue.18 In the second Critique, Kant sharply separates their approaches. He writes: The doctrine of the method of pure practical reason cannot be understood as the way to proceed (in reflection as well as in exposition) with pure practical principles with a view to scientific cognition of them, which alone is properly called method elsewhere, in the theoretical (for popular cognition needs a manner but science a method, i.e., a procedure in accordance with principles of reason by which alone the manifold of a cognition can become a system). Here the doctrine of method is understood, instead, as the way in which one can provide the laws of pure practical reason with access to the human mind and influence on its maxims, that is, the way in which one can make objectively practical reason subjectively practical as well. (5: 151) The question regarding how metaphysics can become a science in accordance with principles of reason is clearly a central issue in the passages of the Architectonic that we have analysed. By contrast, the question regarding how we can make objective practical reason subjectively practical is very close to our characterisation of philosophy as a doctrine of wisdom. If it is true that, as we have suggested, the idea of the highest good is central to both these questions, even though from different but complementary perspectives, “theoretical” and “practical” doctrines of method might not be as far apart as Kant suggests.19 Notes 1 I refer to Kant 1781 and Kant 1787 with A and B, respectively, and to Kant 1900– by indicating volume and page numbers. Translations are from Kant 1992–.

Can Metaphysics Become a Science for Kant? 163 2 However, Kant hints at the possibility to find a place within metaphysics for a commitment towards those objects from a “practical” perspective. ‘[W]hat still remains for us is to try whether there are not data in reason’s practical data for determining that transcendent rational concept of the unconditioned, in such a way as to reach beyond the boundaries of all possible experience, in accordance with the wishes of metaphysics, cognitions a priori that are possible, but only from a practical standpoint’ (Bxxi). 3 In his contribution to this volume, “Two Models of Critique of Metaphysics: Kant and Hegel”, Dietmar Heidemann argues that Kant believes that metaphysics cannot become a science. Of course, the disagreement between our views could be mainly terminological. For example, if, with the term “metaphysics”, one specifically understands a discipline that describes supersensible or “unconditioned” objects, it is clear that Kant believes that we cannot have cognitions regarding those objects, which appears to imply that they cannot constitute a science. However, Heidemann’s characterisation of metaphysics is broader. He takes his claim that, for Kant, metaphysics cannot become a science to apply to both general and special metaphysics. In his view, they both attempt to develop cognition from “concepts” alone, but this is impossible, since we also need intuition to obtain cognition. I agree that one criticism that Kant moves against traditional metaphysics is that it proceeds by concepts alone and so cannot establish synthetic claims – a version of this reading has been defended by Lanier Anderson (2015). However, this does not rule out that metaphysics can become a science when it adopts a new self-understanding, for example, one in which, in order to provide a priori cognition of nature (just to focus on the theoretical side), we need to identify both a priori concepts and corresponding rules that a priori determine the synthesis of intuition. The result would be a metaphysics that is more modest (it could not comprise cognition of objects that cannot be given in sensible intuition), but not necessarily one for which it would be impossible to become a science. 4 On architectonic unity in Kant, see La Rocca 2003: Ch. 6; Manchester 2003; Manchester 2008; Sturm 2009: Ch. 3; Sturm 2020; Ypi 2011; Ypi 2021; Gava 2014; Gava 2023: Ch. 1; Ferrarin 2015: Ch. 1. 5 I elaborate further on the claim that architectonic unity cannot simply be systematicity in Gava 2023: Ch. 1. An implication of this claim is that we must separate Kant’s discussion of reason’s contribution to the systematicity of empirical cognition in the Appendix to the Transcendental Dialectic from his analysis of architectonic unity in the Doctrine of Method. Interpreters that see a stronger connection between these two discussions are, for example, Sturm 2009: Ch. 3, and Ypi 2021. It might seem odd to claim that the principle of organisation of the cognitions belonging to a science has its seat a priori in reason. However, the sciences that Kant discusses in the Architectonic chapter comprise a priori cognitions. When we take this fact into consideration, it seems plausible to maintain that the principle according to which these cognitions are to be organised should also be accessible a priori, where this does not mean that it is immediately transparent to us (see Gava 2016; Gava 2023: Ch. 1). 6 In Gava 2023: Ch. 1, I claim that the idea of a science is a correct description of the body of cognitions that form a science and its parts–whole relationships. 7 See Gava 2023: Ch. 1.

164  Gabriele Gava 8 Since architectonic unity is required for science and since it is achieved when a body of cognitions is organised according to its proper idea, it seems that the ideas that are at stake in architectonic unity are realisable, otherwise Kant could not consider some existing disciplines as proper sciences. For a different reading that stresses that ideas must remain a focus imaginarius, see Strum 2009: 168. 9 On this point, see also Gava 2016. 10 One problem for this line of reasoning is that Kant sometimes claims that certainty is a condition for rational cognition (see A822-3/B850-1), where this seems to imply that a philosopher, in order to realise metaphysics as a science, should know that she knows. Since, in this chapter, I will argue that Kant’s claims regarding the unrealisability of the idea of the philosopher do not mainly rest on reason’s lack of self-transparency, I will not deal with this issue. For an attempt to depict Kant’s requirement of certainty in fallibilist terms, see Chignell 2021. 11 Notice that this is totally different from identifying extra-moral incentives to be moral, as Kant’s position in the Canon of the first Critique sometimes seems to do (see A811/B39; A812/B840). Rather, the relevant incentive is moral through and through. The doctrine of wisdom simply provides an example of how we can infallibly follow that incentive. 12 I believe it is more correct to say that the worldly concept of philosophy incorporates what is correct within the school concept, that is, the search for systematic cognition. I here pursue this different way of understanding their relationship because it appears to provide a path to solve our problem, one according to which attaining the status of science is only a question for the school concept of philosophy. 13 One might here wonder whether the school concept of philosophy is inherently philodoxical. After all, if the philosopher following this concept is interested in the systematicity of philosophical cognitions, and if construing philosophy in relation to the essential ends of reason is the only way to attain this systematicity, the school concept would naturally develop into the worldly concept. This seems correct, but Kant seems to characterise the school concept both for what it does (furthering systematicity) and for what it does not (considering essential ends). In this sense, the concept of philosophy we attain at the end cannot be described as a school concept, even though it incorporates some of its features. 14 I provide a full defence of this claim in Gava 2014 and Gava 2023: Ch. 1. 15 In Gava 2023: Ch. 1, I argue that philosophy according to its school concept cannot in fact achieve any kind of unity because it cannot achieve coherence. For the point that I am making here, it is sufficient to show that philosophy according to its school concept can at best achieve technical unity. Still, my position is different from that of Tonelli (1994: 272) and Ypi (2011: 144), who maintain that philosophy according to the school concept does have technical unity. La Rocca (2003: 221) and Ferrarin (2015: 81) have still different views. La Rocca claims that philosophy according to its school concept has architectonic unity, while Ferrarin contends that it has a unity that is neither technical nor architectonic. 16 A hint towards this paradoxical nature is also given by the fact that while Kant describes philosophy according to its worldly concept as an idea of philosophy as a science, in his description of the misologist and the philodox, he regards philosophy as a science and philosophy as a doctrine of wisdom as different but complementary.

Can Metaphysics Become a Science for Kant? 165 17 Notice that this attempted solution squares well with the two ways in which Kant describes philosophy as an archetype in the Architectonic. As we saw in Section 6.2, Kant equates the archetype of philosophy, firstly, to the ‘idea of a possible science’ (A838/B866) and, secondly, to the figure of the philosopher as ‘the legislator of human reason’ (A839/B867). While the first understanding of the archetype would require considering the highest good from the perspective of metaphysics as a science, the second sense points towards philosophy as a doctrine of wisdom. 18 On practical Doctrines of Method, see Bacin 2002 and Bacin 2010. 19 I thank the participants to the conference “Metaphysics as a Science in Classical German Philosophy” (16–17 July 2021, over Zoom) and the editors of this volume for useful feedback on earlier drafts of this chapter.

Bibliography Anderson, R. Lanier (2015) The Poverty of Conceptual Truth: Kant’s Analytic/ Synthetic Distinction and the Limits of Metaphysics (Oxford: Oxford University Press) Bacin, Stefano (2002) “Sul rapporto tra riflessione e vita morale in Kant: Le Dottrine del metodo nella filosofia pratica” Studi Kantiani 15: 65–91 Bacin, Stefano (2010) “The Meaning of the ‘Critique of Practical Reason’ for Moral Beings: The Doctrine of Method of Pure Practical Reason” In Kant’s Critique of Practical Reason: A Critical Guide, edited by Andrews Reath and Jens Timmermann (Cambridge: Cambridge University Press), pp. 197–215 Chignell, Andrew (2021) “Kantian Fallibilism: Knowledge, Certainty, Doubt” Midwest Studies in Philosophy 45, pp. 99–128 Ferrarin, Alfredo (2015) The Powers of Pure Reason (Chicago: University of Chicago Press) Gava, Gabriele (2014) “Kant’s Definition of Science in the Architectonic of Pure Reason and the Essential Ends of Reason.” Kant-Studien 105(3), pp. 372–393 Gava, Gabriele (2016) ‘The Fallibilism of Kant’s Architectonic’ In Pragmatism, Kant and Transcendental Philosophy, edited by Robert Stern and Gabriele Gava (New York: Routledge), pp. 46–66 Gava, Gabriele (2023) Kant’s Critique of Pure Reason and the Method of Metaphysics (Cambridge: Cambridge University Press) Kant, Immanuel (1781) Critik der reinen Vernunft, 1st edition (Riga) Kant, Immanuel (1787) Critik der reinen Vernunft, 2nd edition (Riga) Kant, Immanuel (1900–) Kants gesammelte Schriften (Berlin: De Gruyter, Reimer) Kant, Immanuel (1992–) The Cambridge Edition of the Works of Immanuel Kant, edited by P. Guyer and A. Wood (Cambridge: Cambridge University Press) La Rocca, Claudio (2003) Soggetto e mondo. Studi su Kant (Venezia: Marsilio) Manchester, Paula (2003) “Kant’s Conception of Architectonic in Its Historical Context” Journal of the History of Philosophy 41(2), pp. 187–207 Manchester, Paula (2008) “Kant’s Conception of Architectonic in Its Philosophical Context” Kant-Studien 99(2), pp. 133–151 Sturm, Thomas (2009) Kant und die Wissenschaften vom Menschen (Paderborn: mentis)

166  Gabriele Gava Sturm, Thomas (2020) “Kant on the Ends of the Sciences” Kant-Studien, 111, pp. 1–28 Tonelli, Giorgio (1994) Kant’s Critique of Pure Reason within the Tradition of Modern Logic (Hildesheim: G. Olms) Ypi, Lea (2011) “Practical Agency, Teleology and System in Kant’s Architectonic of Pure Reason” In Politics and Metaphysics in Kant, edited by Sorin Baiasu, Sami Pihlström, and Howard Williams (Cardiff: University of Wales Press), pp. 134–152 Ypi, Lea (2021) The Architectonic of Reason (Oxford: Oxford University Press)

7

Scientific Metaphysics and Metaphysical Science The Demand for Systematicity in Kant’s Transition Project Michael J. Olson

7.1 Introduction Charting the boundaries of the disciplines of metaphysics and natural philosophy such that a mutually beneficial division of labour might be established between them was central to Kant’s intellectual efforts from his earliest writings to his last unpublished notes. 1 Philosophy, mathematics, natural science, and religion can each flourish, on his view, only when they operate within their appropriate domains. Religious claims about the nature of the soul, mathematical claims about the infinite divisibility of matter, and denials of human freedom on the basis of mechanistic natural science all miss the mark. Thus, Kant argues, ‘It is of the utmost importance to isolate cognitions that differ from one another in their species and origin, and carefully to avoid mixing them together with others with which they are usually connected in their use. What chemists do in analysing materials, what mathematicians do in their pure theory of magnitude, the philosopher is even more obliged to do’ (CPR A842/B870; see also MFNS 4: 524). Natural science, mathematics, and philosophy differ according to the origins of their cognitions and must be distinguished on that basis, but these domains nonetheless share – or ought to share – an important formal characteristic. Under the inescapable influence of reason, all knowledge, Kant claims, aspires to attain a systematic organisation that clarifies the compatibility, interconnection, and completeness of its claims. ‘Since systematic unity is that which first makes ordinary cognition into science’ (CPR A832/B861), reason strives to unite all cognition – a priori as well as empirical – into a system.2 Although reason’s demand for systematic unity applies to all domains of knowledge, it does so in different ways. With metaphysics, which Kant understands to be a discipline of a priori cognitions through concepts, the principle that guides its systematic organisation and underwrites its completeness is also a priori (CPR A81–83/B107–109). The

DOI: 10.4324/9781003187561-8

168  Michael J. Olson metaphysics authorised by transcendental critique is thus a paradigmatic science for Kant. The accomplishment of systematic unity and elevation to the true dignity of a science occurs differently with empirical disciplines. Since knowledge in these disciplines is won only on the basis of experience, discovering a principle to organise and unify cognitions into an interconnected whole is a more complicated affair involving both a prospective decision to treat nature as if it were a systematic whole and retrospective reflective efforts to discern the principle that unites existing knowledge in some field and to suggest paths for further development.3 Reason’s injunction that we search for unity in our cognitions, which is constitutively achieved in the a priori science of metaphysics, can be approached only asymptotically in empirical domains and the idea of the systematic unity they pursue has a merely regulative or heuristic status. That is, the unity of speculative metaphysics is rooted in the unity of the table of categories, the universal objectivity validity of which Kant claims to establish apodictically in the Transcendental Deduction. In chemistry, however, the most general principles used to organise empirical observations and suggest further experimental investigation are themselves drawn from empirical observation, and so they can never attain the apodictic certainty characteristic of metaphysics. That, anyway, is Kant’s view throughout the 1780s. Empirical cognition strives to become scientific by assuming that systematic unity is possible, if always only on the horizon. The complete system of metaphysics, including the metaphysics of nature and the metaphysics of morals, is secured in its general outline by transcendental critique, though Kant seems to think its final articulation is something of a chore. Metaphysics achieves its scientific status through the rigorous separation of the a priori and a posteriori components of cognition. Natural science aspires to something similar under the banner of a heuristic principle, but is prevented from ever fully achieving it due to the inescapably empirical nature of its cognitions. Though Kant’s view on these issues changes in interesting ways in the 1780s, this core remains consistent. That makes a later shift in his writing about the systematic unity of empirical cognitions remarkable. In the notes that make up the Opus postumum, Kant appears to reject this way of thinking entirely.4 Rather than holding that the search for systematicity operates only heuristically, he begins to work through the idea that reason provides a constitutive principle for physics. That is, Kant flirts with the view that the unity of the empirical cognitions that make up our scientific knowledge of the physical world requires a real, given a priori principle rather than the merely regulative one he had previously appealed to. Since this principle must be a priori if it is to provide the fully systematic and scientific characteristic of metaphysics, he comes to think, the unifying principle of our empirical study of nature must be given by

Scientific Metaphysics and Metaphysical Science 169 reason. In other words, in stark contrast to his general insistence on isolating them from each other, in the Opus postumum Kant comes to think that the natural sciences can only finally become scientific by becoming metaphysical. The goal of this chapter will be to understand what led Kant to change his mind on this issue in the mid-1790s. More specifically, the claim of the chapter is that Kant came to think that empirical cognitions could not achieve the kind of scientific status he took them to deserve through the heuristic art of system-building alone, at least partly because of his sense of the disarray of the physics of his own day. The optimism of the first Critique, I argue, is deflated by the realities of actual scientific practice. Before turning to a consideration of why Kant changed his mind, we will first sketch Kant’s view of the systematicity of empirical cognitions in the 1780s and how it changes in the 1790s. 7.2  Systematicity in the Critical Philosophy Kant’s most significant discussion of the organisation of empirical cognition under the idea of a unified system occurs in the Appendix to the Transcendental Dialectic of the first Critique. After locating the source of so much fruitless metaphysical speculation in the transcendent use of the ideas of reason, Kant underscores a different, productive role reason plays in cognition. In its regulative use, reason’s tendency to create concepts of objects is curtailed and its ability to order and unify existing cognitions is brought to the fore. Reason unifies cognitions, Kant explains, by presupposing the idea of ‘the form of a whole of cognition, which precedes the determinate cognition of the parts and contains the conditions for determining a priori the place of each part and its relation to the others’ (CPR A645/B473). The idea of a complete system of cognitions guides reason in the ordering of cognitions through the application of three principles: homogeneity, specification, and the continuity of forms. The assumption of the first of these principles leads us to believe that despite the extraordinary diversity of appearances, there is sufficient similarity in nature to validate the conceptual cognition of distinct individuals under common species. At a higher level, this same principle leads us to identify the shared features of distinct species that define their common genus. Where homogeneity presupposes commonality, the principle of specification assures us of natural diversity. Despite the real similarities in nature, there are also always differences to be found such that in familiar species we can still discriminate meaningful subspecies. The final principle, the continuity of forms, asserts that in the complete systematisation of cognition, any two concepts are connected if one only ascends to a sufficiently general genus. It is this last

170  Michael J. Olson principle that underscores the unity produced through the idea of a systematic ordering of cognitions. With these principles in hand, reason goes to work on empirical cognitions. When reflecting on the forces governing natural phenomena, for example, the principle of homogeneity will prompt us to consider whether the cause of the warmth of fermenting beer has the same cause as the warmth of a living body, glowing coals, or bales of fresh hay. Meanwhile, the principle of specification will lead us to look for differences in the causes of the warmth in warm- and cold-blooded animals, for example. All the while, the principle of continuity assures us that no matter how far down we ascend in our search for further subspecies in one domain or how much commonality we can find in the genera of another domain, all the concepts involved cohere into a single hierarchy of systematic cognition. The warmth of a cask of fermenting beer and of the dog sleeping at my feet are connected somehow, even if not directly. The real issue Kant considers in the Appendix is the epistemic status of these principles and the idea of a systematic unity of cognition that underwrites them. He distinguishes between a logical and a transcendental application of such principles. The first takes this unity to be a device ‘for the benefit of reason’ (CPR A649/B677) rather than a feature of nature itself. On this view, Kant explains, ‘systematic unity … is only a projected unity, which one must regard not as given in itself, but only as a problem; this unity helps us to find a principle for the manifold and particular uses of the understanding, thereby guiding it even in those cases that are not given and making it coherently connected’ (CPR A647/B675). The transcendental application of the demand for unity, by contrast, ‘pretends to objective reality’ (CPR A650/B678). Though the illegitimate transcendental application of ideas is at the heart of Kant’s rejection of transcendent metaphysical claims in the Dialectic, one cannot, he argues simply, adopt the logical principle while rejecting the transcendental. Reason’s demand for unity is not satisfied by the merely logical search for unity because the motivation to search for that unity in our cognitions is undermined ‘when reason is free to admit that it is just as possible that all powers are different in kind, and that its derivation of them from a systematic unity is not in conformity with nature’ (CPR A651/B679). That is, I cannot earnestly seek to order my cognitions and to leverage that order to guide future investigations if I do not assume that the world really is ordered systematically. The logical principle of systematic unity thus presupposes the transcendental principle of systematic unity. The systematic unity of nature we must presuppose, since it is an idea, cannot itself provide any determinate cognition of nature. Moreover, ‘experience never gives an example of perfect systematic unity’ (CPR A681/ B709). Nevertheless, Kant claims the principle of systematic unity has an

Scientific Metaphysics and Metaphysical Science 171 indeterminate or indirect objective validity.5 This is so, he argues, because the idea of the systematic unity of cognition ‘serves as a rule of possible experience, and can even be used with good success, as heuristic principles, in elaborating it’ (CPR A663/B691). So while the idea of a complete systematic ordering of cognitions is of a merely heuristic use and cannot itself determine anything in natural phenomena, reason must nonetheless presuppose the systematicity of nature in order to set about organising our cognitions in good faith and this presupposition is not irrational insofar as empirical cognitions regularly conform to and the pursuit of knowledge benefits from it. Though it is rational to presuppose the unity of empirical cognitions, Kant is clear that the objective validity of such a presupposition cannot be proved. After claiming an indirect objective validity for reason’s principles in the Appendix to the Dialectic, in the Canon of Pure Reason Kant revisits the idea in the context of the practical use of reason. There he writes, ‘whereas the systematic unity of nature in accordance with speculative principles of reason could not be proved’, ‘the principles of pure reason have objective reality in their practical use, that is, in the moral use’ (CPR A807/B835). Given his argument that the logical application of the principle of unity depends on the more fundamental presupposition that there really is systematic unity to be found in empirical cognitions, Kant’s insistence that there is no proof that the world conforms to reason’s demand in this regard foreshadows Kant’s repeated reconsideration of the grounds supporting claims regarding the unity of empirical cognitions. Indeed, Kant revisits the basis of our expectation of systematicity in empirical cognitions five years later in the Metaphysical Foundations of Natural Science. What we see there is that he seeks to leverage the constitutive objective validity of the table of categories for the purposes of shoring up the rationality of our expectation of systematic unity in the natural sciences more generally. That is, he borrows some of the apodictic systematicity of metaphysics to secure the systematic foundations of the sciences. The relevant analysis comes in the Preface to the Metaphysical Foundations. There he distinguishes between natural science (Naturwissenschaft) and a doctrine of nature (Naturlehre) such that the former is ‘a whole of cognition ordered according to principles’ (MFNS 4: 468) and the latter is just ‘systematically ordered facts about natural things’ (Ibid.). On the basis that a science is a systematic whole of cognitions, Kant distinguishes a narrow or proper sense of natural science, which restricts itself entirely to a priori principles and their attendant apodicticity, from a more capacious or improper sense of the term, which also treats empirical laws. He then insists that ‘one easily sees why natural science must derive the legitimacy of this title’ (Ibid.) from the narrow, a priori sense of the term alone.

172  Michael J. Olson Distinguishing these senses of natural science has a couple of advantages, Kant notes, including shedding light on how so-called proper natural science relates to the broader Critical project. ‘In everything that is called metaphysics,’ Kant explains, ‘one can hope for the absolute completeness of the sciences, of such a kind one may expect in no other type of cognition’ (MFNS 4: 473). This is true, he continues, ‘whether it be of nature in general, or of corporeal nature in particular’ (Ibid.). That is, the systematic completeness of the table of categories secures the systematic completeness of the pure principles of the metaphysics of nature in general elaborated in the first Critique just as much as it does the pure principles of the metaphysics of corporeal nature in particular, which are the topic of the Metaphysical Foundations. By showing how a metaphysics of corporeal nature grows out of the table of categories, Kant also addresses a crucial interest of transcendental philosophy, namely securing the objective validity of a priori representations of objects of experience. Kant explains, ‘in order to make possible the application of mathematics to the doctrine of body, which only through this can become natural science, principles for the construction of the concepts that belong to the possibility of matter in general must be introduced first’ (MFNS 4: 472; see also CPR A847–848/B875–876). That is, the objective validity of the mathematical representations inherent to the empirical natural sciences must be demonstrated through the pure construction of the most general object of the natural sciences, namely matter as the movable in space. So, in 1786, Kant understands the systematic unity of the a priori core of the natural sciences – that is, what he also calls proper natural science or the metaphysics of corporeal nature – to be necessary for the systematicity of the natural sciences more generally. In other words, the constitutive or determinative systematicity of the unity of the metaphysics of corporeal nature is a condition of the regulative or heuristic systematicity of empirical scientific cognitions. In short, the empirical sciences lean on metaphysics for their systematic unity and the scientificity that depends on it. The Metaphysical Foundations establishes a point of contact between the a priori objective validity of the categories and the empirical concept of matter and Kant marshals this connection to justify our expectation that nature conform to reason’s demand for systematic unity across the empirical natural sciences. Nevertheless, Kant is careful to separate the proper natural science of the metaphysical foundations of empirical natural science from empirical natural science itself. The features of all matter in general insofar as they can be mathematically constructed according to the interaction of the original attractive and repulsive forces exhaust the metaphysics of corporeal nature. Those powers do not, however, exhaust the moving forces of matter.

Scientific Metaphysics and Metaphysical Science 173 Those forces that do not act universally on all matter but vary in their effect or intensity from one kind of matter to another cannot be grasped by an a priori construction of matter in general. Accordingly, these forces, which include gravitation, cohesion, elasticity, chemical affinities, and many others, can only be understood empirically. Kant’s a priori analyses of the forces of original attraction and repulsion are thus not valuable for natural science as a constitutive contribution to its cognitions, but ‘only for the purpose of guiding natural philosophy’ (MFNS 74; 4: 534), that is as a heuristic. Specifically, the Metaphysical Foundations motivates scientists to pursue dynamic rather than corpuscularian theories by providing a rational framework for constructing a concept of matter that does not depend on atoms and the void to explain variations in density. Thus, in the end, Kant’s a priori development of an empirical concept of matter retains the familiar Critical separation of domains. The formal assurance that empirical cognitions are grounded in an apodictic systematic unity is provided by metaphysics; how the empirical laws governing material phenomena fit into and grow out of such a unity can only be discovered in experience. This will not always be Kant’s position, however. 7.3 The Transition Project, the Opus postumum, and the Systematicity of Physics By 1792, Kant had come to see a serious problem with the metaphysics of corporeal nature elaborated in the Metaphysical Foundations.6 That view maintains that the impenetrability of a body is the effect of the equilibrium of the attractive and repulsive forces within it. When the attractive force is strong in relation to the repulsive force, the body is relatively dense; when it is weak, the comparative density of the body is as well. The problem becomes clear enough when we recall that what determines the quantity of the attractive force is the quantity of matter: the more matter, the greater the force. As Kant forthrightly explains in response to a letter from J.S. Beck, the view defended in the Metaphysical Foundations is circular ‘because the attractive power depends on the density, but the density depends on the attractive power’ (C 11: 361–362). Since the systematic results of the Metaphysical Foundations – the rudiments of rational physiology, the demonstration of the objective validity of the categories vis-à-vis objects of outer sense, the metaphysical guarantee of the laws of Newtonian mechanics – all hinge on the relation between the attractive and repulsive powers, circularity in his understanding of how this relation determines density reveals the inadequacy of the whole view. Though this problem arises independently of the programmatic remarks in the Preface we considered earlier, it has important ramifications for how Kant explains the systematicity of empirical cognitions. After all, as we have seen, Kant argues the

174  Michael J. Olson regulative unity of natural science depends upon the constitutive systematic unity of the metaphysics of corporeal nature. Circularity in the latter thus threatens the former. Given that, providing a new explanation of how the various forces that determine material bodies can find apodictic certainty and systematic completeness in the a priori conditions of cognition, on the one hand, while accommodating wide empirical variation between material bodies, on the other, becomes a desideratum for Kant in the mid-1790s. As we will see, Kant’s pursuit of this explanation under the title of a Transition from the Metaphysical Foundations of Natural Science to Physics in the Opus postumum will lead him to reconsider his earlier insistence that the rational demands for systematicity in the empirical sciences can only ever be regulative or reflective rather than constitutive or determinative.7 Focusing on the collection of notes written between July 1797 and July 1798, we see a notable shift in Kant’s thinking about the unity of empirical cognitions. The goal of the Transition project remains the same as that described in the Appendix in the first Critique, namely the asymptotic pursuit of systematic organisation. Kant writes in the Opus postumum that the task of the Transition is ‘to classify the real objects of nature according to a principle, and to bring the empirical study of nature ever closer to a system – although it never attains such completeness, which cannot be expected from experience’ (OP 21: 477). Whereas the principle guiding this task according to the first Critique would be reason’s regulative principle of systematic unity, now Kant understands the regulative assumption of unity in empirical cognition to be insufficient. Thus, ‘There is not a merely regulative, but also a constitutive principle, existing a priori, of the science of nature, for the purpose of a system’ (OP 22: 240).8 Even under the guidance of reason’s regulative use of the principle of unity, Kant now claims, ‘Merely empirical nature can never amount to a system, but, at best, a fragmentary, ever-increasing aggregate’ (OP 39; 21: 474). Since Kant no longer believes the regulative application of reason’s demand is up to the task of securing the unity reason demands, an alternative approach is required. The Transition from the Metaphysical Foundations of Natural Science to Physics is intended to be just this. The idea is that the Transition will possess features characteristic of each of the domains it connects. Like the metaphysical foundations of science, it is a priori; like physics, it depends on experience. The a priori dimension of the Transition is derived from the table of categories such that the constitutive a priori principle at the heart of the project, when analysed according to the concepts of quantity, quality, relation, and modality, generates a formally complete sketch of the way the universal moving forces of matter – attraction and repulsion – generate through their various relations the full range of empirical moving forces. Thus, an analysis of the original moving

Scientific Metaphysics and Metaphysical Science 175 forces through the lens of quantity issues in an explanation of weight and density; quality issues in explanations of fluidity, solidity, and heat; and relation in cohesion, dissolution, and crystallisation.9 The Transition draws on experience insofar as the forces at work in cohesion, crystallisation, magnetism, etc. are not derived a priori but first discovered and investigated in nature.10 We briefly discussed above how in the Metaphysical Foundations of Natural Science Kant drew on the table of categories to provide an a priori foundation for the natural sciences. In that text, he restricted the scope of the contribution metaphysics was thought to make to the empirical sciences. The metaphysical principles elaborated there are not themselves meant to amount to empirical knowledge of how real material bodies interact. It was instead a sketch of the fundamental principles characteristic of all matter in general, abstracted from all the properties that make material bodies the specific bodies they are. Such a sketch was intended to orient empirical studies of material phenomena by ruling out corpuscular matter theories, but it remained detached from questions about specific empirical laws and phenomena. A little more than ten years later, however, Kant is exploring the possibility of organising physical and chemical forces under the a priori structures of the table of categories. If that line of thinking were successful, there would be a historically specific account of the empirical forces governing material bodies whose objectivity and completeness is secured by the table of categories and the resulting metaphysics of nature. Metaphysics would, in other words, intervene directly in natural science in order to furnish the systematicity it depends on but cannot provide for itself. Though Kant repeatedly explains that the a priori constitutive principle at the heart of the Transition project is necessary to secure the systematicity of empirical cognitions of the laws of the moving forces of matter, he is not clear about what led him to conclude that his long-standing view that we can appeal only to a merely regulative principle in our efforts to order scientific knowledge. What led Kant to think the view he held throughout the 1780s was mistaken? After so regularly insisting on maintaining a bright line between the subjective necessity of reason’s demands and the objective necessity of knowledge, why do the notes from the late 1790s pursue a blurring of this line? One explanation of this change points to Kant’s recognition of the failure of the Metaphysical Foundations. Since the explanation of the grounding of material bodies in the original attractive and repulsive powers proved to be circular, a new construction of the universal features of matter in general is required. This view, which sees the Transition project as a product of conceptual problems internal to the Critical philosophy is presented most prominently by Förster.11 The first page of Kant’s

176  Michael J. Olson Opus postumum notes reproduces a critical review of the Metaphysical Foundations of Natural Science, which clearly indicates the genetic role of the shortcomings of that text in his early thinking about the Transition. Moreover, Förster convincingly demonstrates that an important goal of the Opus postumum is to establish the objective validity of the categories in relation to matter as the object of outer sense. What he does not explain as incisively, however, is what we have focused on here, namely Kant’s new belief that a merely regulative idea of the systematic unity of physics is insufficient. Why is a Transition necessary not just as a replacement for the results of the Metaphysical Foundations but also, as Kant repeatedly says in the Opus postumum, for the sake of physics as an empirical pursuit? An alternative explanation of why Kant came to think a Transition was necessary comes from Hein van den Berg. He argues that ‘the necessity of the transition project … arises in part from the fact that the metaphysical foundations and physics have the same object of investigation. Because both sciences are concerned with a common object, it is necessary to show not only that these disciplines are consistent with each other, but also how they are related to each other’ (van den Berg 2014: 175). Clearly, delineating the boundaries of distinct epistemic domains is of course central to the Critical philosophy in general. Moreover, Kant specifically highlights the illicit mixing of empirical and metaphysical claims as an important impediment to systematising physics (OP 21: 256). The question this explanation raises, however, is why Kant did not think such a transition was necessary to clarify the relation between the empirical results of physics and either the a priori conditions of cognition in general (in the first Critique) or the a priori metaphysical foundations of natural science (in the Metaphysical Foundations). These also involve two distinct cognitive approaches to the same objects of inquiry. In fact, in the first Critique, Kant uses the idea of organising the forces of matter under the idea of a system as an example of how to think about the proper, regulative use of the principle of unity (CPR A650/B478). Although Kant was generally concerned to coordinate the domains of a priori and empirical cognition, that does not explain why in the mid-1790s he takes the Transition project to be necessary to secure this coordination. The great strength of van den Berg’s analysis of the Opus postumum is his careful attention to Kant’s understanding of the empirical physics of the time. In this regard, van den Berg follows Michael Friedman, who argues that Kant drew an optimistic enthusiasm from his interest in the newest results in experimental physics (Friedman 1992: 265). If we follow their lead, we can gain some insight into why Kant changed his thinking. A quick survey of the sources we know Kant had access to in his study of the state of physics and chemistry in the 1790s and the implications that study had on his assumption that the results of physics would slowly

Scientific Metaphysics and Metaphysical Science 177 converge on the kind of systematic order reason demands illustrates why Kant had reason to despair in this regard. Because of its importance in the broader sweep of the Opus postumum, I focus here on the concept of ether. The 18th-century debates about the existence, nature, and explanatory power of ether have their origin in the Queries of Newton’s Opticks.12 There Newton suggests investigating whether ether, a ‘medium exceedingly more rare and subtile than the Air, and exceedingly more elastic and active’ (Newton 1721: 324), might play a role in the phenomena of heat, gravity, electricity, magnetism, and the function of the nervous system (Newton 1721: 324–328). Leonhard Euler later denied that ether played a role in magnetism, but proposed that ether was essential to explaining light as well as the explosive power of gunpowder (Euler 1795: 2: 260–269, 277; 1: 86). Johann Samuel Traugott Gehler’s eight-volume Physicalisches Wörterbuch (1787–1795)was especially influential for Kant; it contains entries that present ether as entirely hypothetical and without direct evidence (in the entries on “Aether,” published in 1787, and “Masse,” in 1790) and later as clearly established as caloric (according to the entry on “Wärme” published in 1791). He acknowledges that other concepts, including light matter, phlogiston, magnetic matter, and electrical matter are often associated with ether but are not known through clear experience (“Masse,” 1790), while also noting that those who deny the existence of light matter should ‘become better acquainted with chemistry’ (“Licht,” 1789). Two authors with whom Kant had a personal connection further muddied the waters. Christoph Girtanner’s Anfangsgründe der antiphlogistischen Chemie (1792), which was decisive in communicating Lavoisier’s ideas to German readers, agrees with Gehler’s eventual conclusion that caloric is real, though Girtanner notes that ‘We don’t know what caloric really is’ (Girtanner 1792: 35). Despite this, he rejects Newton’s account of light and claims, ‘It is most likely that light is not its own matter, but just a modification of caloric by which the latter becomes capable of making a certain impression on the eyes’ (Girtanner 1792: 17). Karl Gottfried Hagen, Kant’s colleague in Königsberg, agreed that caloric was real but rejected theories of the relation between caloric and light of any kind as ‘merely hypothetical’ (Hagen 1796: 23). Tiberius Cavallo, whose Complete Treatise on Electricity (1777) Kant owned in translation, sums up the state of affairs succinctly: As to the identity of the Electric, and the ethereal fluid, it seems to me quite an improbable, or rather a futile and insignificant hypothesis; for this ether is not a real, existing, but merely an hypothetical fluid, supposed by different Philosophers to be endued with different properties, and to be an element of several principles. Some suppose it to

178  Michael J. Olson be the element of fire itself, others make it the cause of attraction, others again derive animal spirits from it, &c.; but the truth is, that not only the essence, or properties, of this fluid, but even the reality of its existence is absolutely unknown. (Cavallo 1796, 1: 121; emphasis added) In short, there was little certainty regarding how or whether some of the most important material phenomena investigated by 18th-century physicists were or could be related. There was only slightly more consensus about how to proceed. With this brief historical sketch in view, we can see the importance of a key passage in the Opus postumum. After comparing the Transition project to architectural blueprints, which provide a plan but cannot anticipate all the details involved in the actual construction of the building, Kant returns to the topic of why a Transition is necessary. He writes, ‘It is indeed a common illusion that one may hope, using nothing but mathematics, to produce a philosophical system of physics without prior metaphysical foundations; results show, however, that in this fashion, everything is treated fragmentarily and that a satisfactory whole, or even the plan of one cannot emerge’ (OP 21: 527; emphasis added).13 Kant’s earlier optimism that physics would in time approximate a systematic whole on its own is dashed, he suggests, by the disjointed reality of physics research. Perhaps this is why Kant’s colleague and former student Karl Ludwig Pöschke observed that Kant’s study of physics around this time ‘disturbed him internally’ (Malter 1990: 595). Consensus about the relations between heat, light, magnetism, electricity, chemical affinities, and all the rest is nowhere on the horizon. This is at least in part a result of a confusion of the proper relation of metaphysical and empirical inquiry. The proximity of metaphysical and empirical matters in fundamental questions about the nature of matter and the forces that govern it, he claims, ‘creates doubts and contains difficulties which should be embarrassing for physics …. For the admixture or insertion of one into the other, as commonly occurs, is dangerous; not just to its elegance, but even to its thoroughness’ (OP 21: 526). Debates about the existence of empty space and the necessity of an exceedingly expansive and subtle matter to fill out the plenum are certainly fertile ground for a blending of physics and metaphysics.14 In addition to the problems internal to the Critical philosophy, Kant was keen to address in his final work, it seems his understanding of the importance of Transition project cannot be limited to those interests. We have good reason to think, then, that Kant’s reconsideration of his view of the systematic unity of empirical cognitions and so his conception of the Transition project as a whole is motivated by a loss of faith in the capacity of the empirical investigations of physicists to develop this unity

Scientific Metaphysics and Metaphysical Science 179 on their own, even with the regulative idea of the systematic unity of a science guiding their efforts. It is not surprising that Kant might despair of the spontaneous convergence of empirical physics on a unified, systematic science of the forces governing material phenomena when faced with the tangle of claims and hypotheses sketched above. That is not, however, to contest Friedman’s claim that Kant drew inspiration and confidence from these sources as he sought to formulate an a priori proof of the existence of ether. Kant could well have been confident of the existence of ether on the basis of his reading of these sources while at the same time not expecting the disjointed efforts of those and other investigators of nature to coalesce around a shared conception of the interconnections of the forces determining material phenomena. In the absence of hope that physics could order its results in a way that would reflect the demands of reason, meet the criteria of a science, and support the furtherance of the empirical investigation of physical forces, Kant concluded that physics required a metaphysical crutch. Whereas he consistently maintained in the 1780s that the empirical sciences would pursue the systematicity characteristic of the title by empirical hypotheses alone, in the late 1790s he argues reason provides both the demand for unity and the organising principle required for systematic ordering. That is, in the Opus postumum, Kant adopts the view that the idea of systematic unity in natural science is constitutive, not merely regulative. The irony of this view is that the apodictic systematic unity that makes metaphysics paradigmatic of science in general for Kant is compromised once it is hitched so tightly to empirical questions. Kant had long sought formal metaphysical foundations for natural science, but he had also been principled in his insistence that the material content of scientific cognition come from experience rather than speculation. When his hope that scientific cognitions would converge on an integrated, unified view of the moving forces of matter in general, he leaned more heavily on metaphysics. If the table of categories could map the relationships between the forces of cohesion, elasticity, crystallisation, and the rest, then physics might attain the scientific status it sorely lacked. That is, the natural sciences appeared to need a bit more metaphysics in order to become the sciences reason required them to be. But if metaphysics could transmit some of its lustre to natural science, certainly it would lose some of its own shine in the process. Yoking the a priori table of categories to what clearly appeared to Kant to be an unstable field of empirical debate in defence of one natural scientific view against others compromises the notion that the universal objective validity of the categories places it beyond the vagaries of empirical research and theory formation. Blurring the boundaries between metaphysics and natural science in order that the latter might take on some more of the systematicity required to make it properly scientific, in other

180  Michael J. Olson words, also weakens the claim that metaphysics is systematic precisely because of its separation from the merely hypothetical results of empirical cognition. We should recall that Kant of course never published the Opus postumum and we should accordingly be cautious about drawing systematic conclusions about his Critical philosophy more generally from these notes alone. What I hope to have shown in this chapter is that in the mid-1790s Kant reconsiders his long-standing view of the status of the idea of systematic unity in empirical cognitions. Throughout the published works of the 1780s, he consistently maintained that reason’s demand for unity in cognition was fully realised in the organisation of the metaphysics of nature under the organisation of the table of categories. In the empirical cognition of nature, however, the idea of a fully integrated system of cognitions could only ever serve as a hypothetical or regulative idea that helped orient empirical investigation without itself supplying any cognitive content. As Kant’s earlier optimism about the likelihood of the spontaneous convergence of natural scientific research on a single system of empirical cognition was deflated by his study of late 18th-century physics (as well as by some systematic problems in his metaphysics of corporeal nature, though that has not been our focus here), he explored the possibility that the metaphysical idea of unity might have a more direct role to play in empirical research and debate. The notes that make up the Opus postumum do not establish that Kant abandoned or rejected his earlier views. These late writings do establish, however, that Kant continued to think until the end of his working life about a central problem of the Critical philosophy, namely mapping the proper domains of our mental faculties and exploring means by which these faculties can come into productive relationships without overstepping the boundaries that distinguish them. Thus, he distinguishes understanding from intuition in order to see more clearly how they cooperate in cognition. Similarly, he distinguishes the a priori cognitions of metaphysics from the empirical claims of the natural science while also insisting that even empirical cognitions depend on a priori principles, including the demand for systematic unity. Explaining how this a priori demand could be met by the teeming variety of empirical experience such that it would be more than just a subjective psychological requirement was a problem Kant returned to often. Although his understanding of the metaphysics of corporeal nature shifted in the 1780s, he remained committed to maintaining a bright line between the metaphysics of nature and natural science such that the demand for unity in cognition, which he took to be constitutively achieved in the former, could only ever be a heuristic ideal in the latter. Distinguishing the a priori domain of metaphysics from the empirical domain of natural science and guarding against the threat of subreption is a constant refrain in Kant’s thinking from the 1770s. That

Scientific Metaphysics and Metaphysical Science 181 he was willing to blur that line in the Opus postumum in order to secure the scientific character that flows from the complete systematic ordering of cognitions for the natural sciences underlines how central the idea of a science was to his understanding of both metaphysics and the empirical investigation of the natural world. Notes 1 References to Kant’s writings in this chapter use the following abbreviations: Critique of Pure Reason: CPR; Metaphysical Foundations of Natural Science: MFNS; Opus postumum: OP; Correspondence: C. Page references follow the Academy edition (Kant 1910– ) by volume and page number, except for the CPR, which is cited according to the customary A (1781) and B (1787) edition pagination. English translations follow the Cambridge editions included in the Bibliography. 2 Kant distinguishes between internal and external systematicity, where the former describes the systematic interconnection of claims within some particular field like physics or mathematics and the latter concerns the interconnection of these distinct sciences into a single whole of knowledge in general. See Sturm 2009: 135–136. This chapter focuses on internal systematicity. 3 See Gava 2014: 383–384. 4 Terrence Thomson 2019 also argues that the Transition project of the Opus postumum marks a shift in Kant’s understanding of systematicity. Whereas Thomson considers how this change affects the systematic division of Kant’s metaphysics, I focus here on the distinct question of the systematicity of empirical cognitions. See Thompson 2019. Alexander Rueger identifies a change in Kant’s thinking about the systematicity of empirical cognitions in the Opus postumum, though he claims the demand for systematic unity remains regulative. We will see below that the latter claim is mistaken. See Rueger 2009. 5 See Zuckert 2017. 6 Eckart Förster argues that Kant had already begun to think about the Transition in 1789 or 1790 (Förster 2002: 3). 7 There is considerable discussion of the systematicity of nature and our warrant for assuming it as part of Kant’s analysis of natural purposiveness in the third Critique. I take Kant’s return to the topic in 1790 to be an indication that he found neither the account from the Appendix to the Dialectic nor the narrower account of the Metaphysical Foundations adequately to ground the systematic unity of empirical cognitions. I leave the topic of purposiveness and systematicity in the third Critique to the side here because it ultimately supplements the view defended in the first Critique. Thomas Teufel’s summary of the third Critique view captures this nicely: ‘When fully analysed, then, the transcendental principle of nature’s purposiveness emerges as an, in effect, doubly regulative principle. It is a reflecting-judgment-determining (hence, regulative; specifically, heautonomous) demand for a determining-judgment-determining (hence, regulative) assumption of nature’s purposiveness’ (Teufel 2017: 119). 8 Michael Friedman argues that what Kant comes to see while working on the Transition is that constitutive principles of the systematic unity of cognition are necessary ‘for each and every particular empirical judgement’ (Friedman 1992: 263). Regardless of whether he is right about the broader ramifications of the Transition project for the Critical philosophy, Kant’s focus in the Transition

182  Michael J. Olson is on the systematic unity and resulting scientificity of the physical sciences in particular and the present analysis is confined to the relation between the systematic unity of metaphysics, on the one hand, and the natural sciences, on the other. 9 The sketch of how these forces would be rendered in a complete outline of the system of moving forces in matter remains unclear in the notes and the details change from sheet to sheet. 10 In the first Critique, Kant reflected on how to draw the line between the metaphysics of nature and natural science: ‘Does the concept of that which is extended belong to metaphysics? You answer, Yes! But what about that of body? Yes! And that of fluid body? You are stumped, for it goes on this way, everything will belong to metaphysics’ (CPR A843–844/B871–872). The key to distinguishing metaphysics from natural science, Kant concludes there, lies in identifying the origin of cognitions: metaphysics is a priori and natural science a posteriori. That view remains in place in the Opus postumum, though his loss of faith in the sufficiency of the merely regulative demand for unity in empirical now leads Kant to try to connect the domains he has separated according to the origins of their cognitions. 11 See Förster 2002. For a different analysis that also locates the genesis of the Transition project in problems internal to the Critical philosophy, see Emundts 2004. 12 The first edition contained 16 queries, to which 7 more were added in the 1706 Latin edition, and 8 more in the second English edition in 1717. 13 Compare this with a passage from the first Critique that acknowledges the initial disorganisation of sciences whose systematicity “seems to have been formed, like maggots, by a generatio aequivoca” but are eventually capable of a more thorough unification that ‘would not merely be possible but would not even be very difficult’ (CPR A835/B863). 14 Kant addressed precisely this issue in the Metaphysical Foundations, where he insisted on maintaining a bright line between the a priori analyses of transcendental philosophy and the empirical studies of natural science. See MFNS 4: 534–535.

Bibliography Cavallo, Tiberius (1783) Vollständige Abhandlung der theoretischen und praktischen Lehre von der Elektricität nebst eignen Versuchen (Leipzig). Cavallo, Tiberius (1796) A Complete Treatise on Electricity, in Theory and Practice; with Original Experiments, 3rd ed. (London) Emundts, Dina (2004) Kants Übergangkonzeption im Opus postumum (Berlin: De Gruyter) Euler, Leonhard (1795) Letters of Euler to a German Princess, on Different Subjects in Physics and Philosophy. Translated by Henry Hunter (London: H Murray) Förster, Eckart (2002) Kant’s Final Synthesis: An Essay on the Opus postumum (Cambridge, MA: Harvard University Press) Friedman, Michael (1992) Kant and the Exact Sciences (Cambridge, MA: Harvard University Press) Gava, Gabriele (2014) “Kant’s Definition of Science in the Architectonic of Pure Reason and the Essential Ends of Reason” Kant-Studien 105(3): pp. 372–393

Scientific Metaphysics and Metaphysical Science 183 Gehler, Johann Samuel Traugott (1787–1795) Physikalisches Wörterbuch oder Versuch einer Erklärung der vornehmsten Begriffe und Kunstwörter der Naturlehre (Leipzig) Girtanner, Christoph (1792) Anfangsgründe der antiphlogistischen Chemie (Berlin: J.F. Unger) Hagen, Karl Gottfried (1796) Grundsätze der Chemie, durch Versuche erläutert (Königsberg: Nicolovius) Kant, Immanuel (1910–) Gesammelte Schriften. Edited by Akademie der Wissenschaften (Berlin: Reimer, later DeGruyter) Kant, Immanuel (1993) Opus postumum. Edited by Eckart Förster, translated by Eckart Förster and Michael Rosen (Cambridge: Cambridge University Press) Kant, Immanuel (1998) Critique of Pure Reason. Translated by Paul Guyer and Allan W. Wood (Cambridge: Cambridge University Press) Kant, Immanuel (1999) Correspondence. Translated and edited by Arnulf Zweig (Cambridge: Cambridge University Press) Kant, Immanuel (2002) Metaphysical Foundations of Natural Science. Translated by Michael Friedman, in Theoretical Philosophy after 1781, edited by Henry Allison and Peter Heath (Cambridge: Cambridge University Press), pp. 171–270 Malter, Rudolf (ed.) (1990) Immanuel Kant im Rede und Gespräch (Hamburg: Felix Meiner Verlag) Newton, Isaac (1721) Opticks: or a Treatise of the Reflections, Refractions, Inflections and Colours of Light, 3rd ed. (London: Walter and John Innys) Rueger, Alexander (2009) “Brain Water, Ether, and the Art of Constructing Systems.” Kant-Studien 86(1): pp. 26–40 Sturm, Thomas (2009) Kant und die Wissenschaften von Menschen (Paderborn: Mentis) Teufel, Thomas (2017) “Kant’s Transcendental Principle of Purposiveness and the ‘Maxim of the Lawfulness of Empirical Laws.’” In Kant and the Laws of Nature, edited by Michela Massimi and Angela Breitenbach (Cambridge: Cambridge University Press), pp. 108–127 Thomson, Terrence (2019) “A Suspicion of Architectonic in Kant’s Transition Project” Angelaki 24(5): 11–28 van den Berg, Hein (2014) Kant on Proper Science: Biology in the Critical Philosophy and the Opus postumum (Dordrecht: Springer) Zuckert, Rachel (2017) “Empirical Scientific Investigation and the Ideas of Reason.” In Kant and the Laws of Nature, edited by Michela Massimi and Angela Breitenbach (Cambridge: Cambridge University Press), pp. 89–107

8

Kant, Reinhold, and the Problem of Philosophical Scientificity Karin de Boer and Gesa Wellmann

8.1 Introduction As is well known, Kant presents the Critique of Pure Reason as a propaedeutic discipline intended to determine under which conditions metaphysics can be ‘brought onto the secure path of a science’ (Bxxiii). Many of his early critics, however, were not convinced that the path taken by the Critique could solve the problems they were most concerned with. From 1789 onwards, Reinhold undertook to defend the Critique against misunderstandings and criticisms by presenting what he saw as its main results in a truly scientific form.1 Only in this way, he believed, could he convince Kant’s empiricist and rationalist critics alike of the truths contained in this work.2 Referring to himself, Reinhold writes in the preface to his 1789 Essay on a New Theory of the Human Capacity for Representation: The more he compared the writings of the two parties [the supporters and opponents of Kantian philosophy], the more firmly he was persuaded that their dispute … could no more be ended than the dispute between earlier dogmatic systems themselves, … because it was conducted with completely opposed basic concepts and principles about questions, which, if treated without the most perfect agreement on principles and the utmost sobriety of speculation, could only result in useless subtleties and more than scholastic niceties. (Essay 60–61, translation modified)3 Yet the strategies Reinhold employed to rescue Kant’s work from the hands of his critics without estranging Kant’s followers came at a price. As we see it, Reinhold’s reading of the Critique produces subtle yet significant shifts of meaning that are not always easy to identify.4 And since Reinhold’s way of framing Kant’s achievements and failures was highly influential, even contemporary historians of modern philosophy often assume that Reinhold’s understanding of the Critique of Pure Reason was by and large correct.5 DOI: 10.4324/9781003187561-9

Kant, Reinhold, and the Problem of Philosophical Scientificity 185 In this chapter, we want to challenge this assumption by confronting Rein­hold’s appropriative interpretation of the Critique of Pure Reason with the distinctively critical task carried out in this work, namely, establishing the limits within which pure reason can generate valid a priori cognitions of objects. Reinhold repeatedly states that his Elementarphilosophie is in line with the critical standards set by Kant.6 Yet we aim to show that the meaning he gave to concepts such as “science”, “propaedeutic”, “principle”, and “system” obliterates the critical context within which they were used by Kant and, hence, distorts the intentions of the Critique. The texts we will focus on were published shortly after the 1789 Essay, namely, the first part of the Contributions to the​Correction of Previous Misunderstandings of Philosophers (1790) and The Foundation of Philosophical Knowledge (1791).7 In these texts, Reinhold’s attempts at clarifying and defending Kant’s work become increasingly entangled with metaphilosophical reflections on the requirements of a properly scientific philosophical system and the role of the Elementarphilosophie in this regard.8 While this chapter focuses on Reinhold’s distortion of key aspects of Kant’s Critique, we believe that the ‘philosophy of philosophy’ (Con I 55) he developed in response to Kant gave a new and exciting direction to the debates among his contemporaries.9 Accordingly, we hope to contribute to a more precise understanding of the point at which Reinhold’s attempted defence of Kant turned into a crucial stage of early post-Kantian philosophy. We will proceed as follows. In Section 8.2, we argue that Reinhold obfuscated the properly critical strand of the Critique by assessing its content – in particular, the Transcendental Analytic – in terms of what he took to be the requirements of a truly scientific philosophy. Section 8.3 argues that Reinhold was not justified in purporting to derive his own metaphilosophical conception of these requirements from Kant’s Critique and, thus, to present his own foundationalism as a means to achieve an aim set by Kant in a way he would have endorsed. Turning to the positive strand of Reinhold’s engagement with Kant, Section 8.4 shows how his conception of Elementarphilosophie emerged from his contentious account of the Critique. Finally, we argue in Section 8.5 that Reinhold did not abandon Kant’s distinction between propaedeutic and system, but intended to elaborate a post-critical version of both parts. 8.2 Reinhold’s Metaphilosophical Account of Kant’s Critique of Pure Reason In this section, we discuss elements of Reinhold’s engagement with Kant’s largely implicit metaphilosophy, in particular as regards Kant’s distinction between critique and system and the standards of scientificity Reinhold

186  Karin de Boer and Gesa Wellmann attributes to Kant. We aim to show that Reinhold misconstrued Kant’s own views on this issue and, on this basis, overstated the continuity between his own project and Kant’s Critique. As Kant puts it in the Introduction to this work, transcendental critique provides a ‘propaedeutic to the system of pure reason’ (A11/B25) that prepares the ground for an exhaustive account of all a priori elements of cognition. It does so by determining under which conditions a priori cognition of objects is warranted. Accordingly, transcendental critique investigates the pure understanding in view of these conditions: This investigation, which we can properly call not doctrine but only transcendental critique, … and which is to supply the touchstone of the worth or worthlessness of all cognitions a priori, … is accordingly a preparation, if possible, for an organon, and, if this cannot be accomplished, then at least for a canon of these cognitions, in accordance with which the complete system of the philosophy of pure reason … could in any case at least some day be exhibited. … For that this should be possible … can be assessed in advance from the fact that our object is not the nature of things, which is inexhaustible, but the understanding, which judges about the nature of things, and this in turn only in regard to its a priori cognition, the supply of which … in all likelihood is small enough to be completely recorded, to be assessed with regard to its worth or worthlessness, and to be subjected to a correct appraisal. (A12–13/B26, cf. A841–842/B869–870) In 1790, Kant had not published the metaphysical system he anticipates here and elsewhere. In fact, he never actually carried out his plans in this regard. For this reason, the propaedeutic function of the Critique of Pure Reason received little attention from early commentators.10 Reinhold, for his part, repeatedly acknowledged the distinction between the critique carried out in the Critique of Pure Reason and the metaphysical system Kant intended to elaborate on its basis, noting, for example, that the Critique ‘is a propaedeutic to metaphysics and is defined as such by its very author’ (Con I 278, cf. FPK 115–116). The following passage shows that he was well aware of the critical end pursued in the Transcendental Analytic and the Critique of Pure Reason as a whole: [The Critique of Pure Reason] did not intent at all to provide the science of the faculty of representation or even the faculty of cognition. The aim of this work … consisted in investigating that which is possible through pure reason and in particular the possibility of metaphysics qua alleged science of the super-sensible. … In order

Kant, Reinhold, and the Problem of Philosophical Scientificity 187 to determine the capacity of pure reason, the latter had to be distinguished from sensibility and … the understanding. … The course of these investigations had to be determined and, hence, limited, by its end, which was none other than a critique of pure reason. Whereas, in this way, Kant discovered the forms of the representations of sensibility, of the understanding, and of reason, he developed them only in view of his end, namely, to show that objective cognition is possible only of appearances and not of things in themselves. (Con I 274–275, emphasis in original, cf. FPK 72–73) Actually, however, Reinhold does not dwell on the properly critical strand of the Critique. More specifically, he does not focus on Kant’s effort in the Transcendental Analytic to demonstrate that a priori cognition of objects is limited to possible objects of experience, and largely ignores that Kant’s investigation into the various faculties of the human mind is part and parcel of his transcendental critique. Reinhold rather frames the Transcendental Analytic in terms of Kant’s attempt to identify the a priori elements of any cognition of objects, i.e., space and time, the categories, and the ideas of reason.11 This is to say that he highlights Kant’s preliminary elaboration of his projected metaphysical system in the Critique itself at the cost of the critical reflection on the very possibility of metaphysics that the Critique elaborates in tandem with the latter.12 Reinhold further claims that the work does not meet Kant’s own elevated standards of scientificity. By doing so, he can present his own philosophy as the truly scientific alternative to Kant’s merely provisional propaedeutic to metaphysics. This strategy is brought out very clearly in a passage from On the Foundation of Philosophical Knowledge: But that the Kantian propaedeutic is a preparatory science, if only with regard to metaphysics, can only be admitted if one takes the term ‘science’ in the broad and wavering sense in which it, at least in philosophy, … ought to be used the least. Whenever Kant refers to philosophy as a science, he himself requires systematic form, and the thoroughgoing unity of a manifold of cognitions under a single principle. When he puts forward the plan of a science, i.e., of the metaphysics of sensible nature, this is done by a precise indication of the systematic foundation (Grundlage). Surely he would not have lost sight of the latter condition of science, which he required and heeded (befolgte) himself, if he in his critique had intended to deliver the science that precedes metaphysics, or even all theoretical and practical philosophy, rather than the mere preparation of such a future science. … If the critique of pure reason should already have

188  Karin de Boer and Gesa Wellmann been a strict and proper science, what are the principles that, related to a single highest principle, would constitute its foundation? (FPK 116–17, italics added)13 In this dense passage, Reinhold suggests that Kant deliberately refrained from presenting his preliminary investigation into the human mind in a scientific form. He also suggests that Kant’s own criteria for a properly scientific philosophy converge with his own. To support this view, Reinhold must account for the discrepancy between what he takes to be Kant’s ‘unscientific’ practice, Kant’s own remarks on his practice, and the meta-philosophical criteria of scientificity he attributes to Kant. Reinhold does so by suggesting that Kant used the term “science” merely in a loose sense in relation to the propaedeutic task carried out in the Critique of Pure Reason. The following passage indeed suggests that Kant did not consider his propaedeutic as a science in the strict sense of the term: [W]e can regard a science of the mere estimation of pure reason, of its sources and boundaries, as the propedeutic to the system of pure reason. Such a propaedeutic would not have to be called a doctrine, but merely a critique of pure reason, and its utility would really be only negative. (CPR A11/B25, translation modified) However, Kant elsewhere suggests that the results of the Critique itself are apodictically certain and thus already meet the highest standards of scientificity: Every cognition that is supposed to be certain a priori proclaims that it wants to be regarded as absolutely necessary, and this holds true even more of a determination (Bestimmung) of all pure cognitions a priori that is to be the standard and thus even the example of all apodictic (philosophical) certainty. Whether I have performed what I have just pledged in that respect remains wholly to the judgment of the reader. (CPR Axv) Reinhold can be said to respond to the final part of this passage by claiming that Kant failed to elaborate his propaedeutic in such a way that its results must be regarded as apodictically certain, i.e., failed to meet his own standards of philosophical scientificity. As we see it, Reinhold is certainly right to point out that Kant does not spell out why he considers the Critique itself to be a science and, more

Kant, Reinhold, and the Problem of Philosophical Scientificity 189 specifically, which criteria of scientificity he takes this work to meet at least potentially.14 Kant’s metaphilosophical reflections concern his projected metaphysical system rather than the status of the preliminary investigations carried out in the Critique of Pure Reason itself, and Reinhold rightly addresses this gap.15 This metaphilosophical gap allows Reinhold to interpret the Critique according to his own concerns. The core of his strategy consists in arguing, firstly, that the content of this work must be cast in a properly scientific mould and, secondly, that this can and must be done by taking recourse to Kant’s own standards of scientificity. However, we hold that Reinhold’s account of the Critique fails to do justice to the critical intentions of this work. The next section illustrates this by clarifying the divergence between Kant and Reinhold’s conception of first principles. 8.3  Reinhold and Kant on the Function of First Principles As was discussed in the previous section, Reinhold claims in Foundation that a proper science must be able to derive its contents from a single indubitable principle and suggests, moreover, that he derived this requirement from Kant’s Critique of Pure Reason itself. Although, on his account, the Critique does not establish ‘a first principle (Grundsatz) of all philosophy’ (Con I 273; cf. FPK 62, 67, 114), he considers various passages in the Critique that deal with “local” first principles to imply that, for Kant, any scientific elaboration of a philosophical discipline requires a solid foundation.16 Thus, Reinhold claims that when Kant presented the plan of his ‘metaphysics of sensible nature’, he did so by specifying its ‘systematic foundation (Grundlage)’ (FPK 116, cf. 55–56). However, it is not very clear what Reinhold means by “foundation” in relation to Kant. To clarify this, we briefly turn to Reinhold’s discussion of what he calls Kant’s metaphysics of sensible nature in Foundation and Contributions I and compare it with relevant passages from the Transcendental Analytic. Reinhold appears to use the title “metaphysics of sensible nature” to refer to Kant’s Metaphysical Foundations of Natural Science (1786) and he clearly considers the chapter of the Critique titled ‘System of all Principles of the Pure Understanding’ (A148/B187) to ground this work (cf. FPK 67, Con I 89). Drawing on the table of categories, Kant in this chapter provides a systematic account of the synthetic a priori principles that any empirical cognition of objects presupposes. In Foundation, Reinhold comments on this strand of the Critique as follows: The Kantian foundation of the metaphysics of sensible nature, or the science of knowable real objects, consists of what Kant calls the principles of the pure understanding, [i.e.,] the metaphysical laws of nature

190  Karin de Boer and Gesa Wellmann that are subordinated to a first principle: ‘Any knowable object is subject to (steht unter) the formal and material conditions of experience’. (FPK 114)17 The “first principle” Reinhold mentions in this passage refers to what Kant calls the supreme principle of all synthetic judgements: The supreme principle of all synthetic judgments is, therefore: Every object is subject to (steht unter) the necessary conditions of the synthetic unity of the manifold of intuition in a possible experience. (A158/B197, translation modified)18 This principle summarises the results of the Transcendental Analytic so far. Kant takes himself to have established, firstly, that any cognition of objects consists in grasping the unity of a sensible manifold, secondly, that any a priori cognition of objects requires the synthetic activity carried out by the pure imagination, and, thirdly, that a priori cognition of objects can only be achieved with regard to possible objects of experience. Accordingly, Kant asserts that ‘[t]he possibility of experience is … that which gives all of our cognitions a priori objective reality’ (A156/B195). This assertion arguably represents a simplified version of the supreme principle of all synthetic judgements. Relying on these passages, Reinhold repeatedly claims that Kant conceived of the possibility of experience as the “foundation” of his metaphysics of sensible nature: [Kant] has not limited himself … to establishing the possibility of experience … as the ultimate foundation in general terms. … He has shown what the possibility of experience consists in by means of an exhaustive analysis of the faculty of cognition. (FPK 56–57, translation modified)19 Moreover, Reinhold considers Kant’s supreme principle of all synthetic judgements to ground the manifold of principles of the pure understanding. As he puts it: This principle (Satz) stands at the head of all principles, theorems, and corollaries of the metaphysics of sensible nature. It grounds their demonstrability and it defines the scope of the science. (FPK 69, cf. 68–69) While Reinhold’s comments are somewhat vague, he seems to conceive of Kant’s supreme principle of all synthetic judgements as a ‘systematic foundation’ (FPK 116) in the sense that it ought to be possible to deduce all subordinate principles of the pure understanding from it.20

Kant, Reinhold, and the Problem of Philosophical Scientificity 191 Kant can be said to invite this line of thought by presenting the supreme principle of all analytic judgements (A150–53/B189–93) and the supreme principle of all synthetic judgements side by side (A154–58/B193–97). This presentation clearly intimates the Wolffian account of the principles of non-contradiction and sufficient reason.21 However, Kant’s actual discussion of the supreme principle of all synthetic judgements makes it clear that he assigned a very different function to this principle than to that of the classical principle of sufficient reason.22 For Kant, this principle does not represent a foundation but rather establishes the limit within which the human mind can obtain a priori cognitions of objects. It follows from the supreme principle that such cognitions are valid insofar as they pertain to possible objects of experience and invalid insofar as they pertain to quasi-objects such as the soul, the world as such, and God. Seen in this way, Kant intended to establish the objective validity of the categories not in order to establish the a priori foundation of empirical cognition, but in order to undergird his critique of former general and special metaphysics. In other words, Kant conceives of the possibility of experience as a criterion and not as a foundation: After what has been shown in the deduction of the categories, hopefully no one will be in doubt about how to decide the question, whether these pure concepts …, as conditions of a possible experience, relate a priori solely to appearances, or whether, as conditions of the possibility of things in general, they can be extended to objects in themselves. (A139/B178, cf. A401) Kant nowhere claims that the warranted synthetic a priori cognitions he called principles of the pure understanding can be derived from the supreme principle of all synthetic judgements: this principle is key to the properly critical strand of the Transcendental Analytic and merely restricts the use of the categories to possible objects of experience.23 Ignoring this point, Reinhold surreptitiously replaces Kant’s critical conception of the conditions that any warranted a priori cognition of objects must meet by a foundationalist account of the conditions of possibility of experience and, moreover, suggests that this account is already contained in the Transcendental Analytic itself. On his account, Kant was merely unable to elaborate the content of the Critique as a whole according to the foundationalist model instantiated – in a limited manner – by his metaphysics of sensible nature.24 As a result, Reinhold largely obscured both the properly critical strand of the Critique of Pure Reason and the non-foundationalist method Kant employed to develop it. This twofold obfuscation constitutes a core premise of Reinhold’s own Elementarphilosophie, to which we now turn.

192  Karin de Boer and Gesa Wellmann 8.4  Reinhold’s First Principle So far, we have argued that Reinhold held that the elements identified in the Critique ought to be cast in a properly scientific mould and, more­ over, that this can and must be done by taking recourse to Kant’s own standards of scientificity. Drawing on this result, this section returns to the problem of the first principle to clarify how Reinhold’s Elementarphilosophie emerges from his attempt to eliminate the alleged shortcomings of the Critique. As seen, Reinhold takes the Critique to conceive of the possibility of experience as the foundation of his metaphysics of sensible nature. By assessing Kant’s account in these terms, Reinhold can subsequently maintain that Kant’s candidate fails to meet the requirements of a foundational principle. Such a principle, Reinhold claims in the Foundation, should be both universally valid (allgemeingültig) and universally accepted, i.e., command universal assent (allgemeingeltend).25 Similarly, Contributions I states that ‘either a universally accepted principle must be possible or philosophy is impossible as a science’ (Con I 367). Referring to an ongoing debate among Kant’s critics, Reinhold states that the Critique fails to provide a convincing account of experience.26 Kant’s conception of experience cannot be universally accepted, he claims, since it presupposes concepts Kant does not determine, for instance, the concepts “representation”, “sensible perception”, “object”, “interconnection”, and “necessity”.27 Accordingly, he denies that the possibility of experience can function as the ultimate ground of philosophy as such: The fate of the new metaphysics that Kant invented depends, therefore, on the development (Entwicklung) of the possibility of experience, i.e., the original structure (Einrichtung) of the faculty of cognition. Although this metaphysics can be universally valid (allgemeingültig) as regards its principles, theorems, and corollaries, it cannot become universally accepted (allgemeingeltend) until that upon which its foundation is grounded itself is constituted by principles (Grundsätze), theorems, and corollaries that are universally accepted. (FPK 70, translation modified)28 Thus, Reinhold took it upon himself “to develop and determine the ultimate principles that Kant, although erecting the Critique of Pure Reason upon a thorough foundation … had left undeveloped and undetermined”.29 As was mentioned in Section 8.2, Reinhold attributes this under-determination to the limited task carried out in the Critique. More specifically, he attests to the Critique’s limitation in two regards. Firstly, Reinhold takes

Kant, Reinhold, and the Problem of Philosophical Scientificity 193 the Critique to be limited in that it examines the faculties merely with regard to the question as to how they bring about cognitions of objects: The critique of reason has investigated sensibility, the understanding, and reason merely with regard to the cognitions they make possible and, more specifically, only with regard to those cognitions that concern objects in the strictest sense. (Con I 275, cf. 277) Secondly, Reinhold extends this criticism to Kant’s account of the a priori forms of cognition, i.e., the ‘forms of the representation of sensibility, understanding and reason’.30 By these forms he means space and time, the pure concepts, and the ideas of reason.31 As Reinhold points out, Kant treats these forms only in view of his limited purpose, namely to show that cognition proper is only possible of objects of experience.32 In order to undo this twofold limitation, Reinhold undertakes to examine the human capacity to represent prior to its attempts at cognising objects. According to Reinhold, all faculties involved in particular cognitive acts rest on the act of representing.33 Philosophy ought to account for this elementary activity so as to examine ‘sensibility, understanding and reason not only in view of that which can be (objectively) cognized but rather with regard to that which can be represented as such’.34 As regards the forms of cognition, Reinhold similarly considers the term ‘representation’ to pertain to ‘that which sensation, thought, intuition, concept and idea have in common’.35 In this case as well, he takes Kant’s Critique to tacitly presuppose this common genus but to fall short of determining it.36 Seeking to overcome this limitation, Reinhold intends to demonstrate that ‘space, time, the twelve categories, and the three forms of the ideas are originally nothing but properties of mere representations’,37 in other words, that representation is the true ‘ground of the species’.38 Although Reinhold considers the capacity to represent to be the genus of all cognitive capacities, he does not conceive of it as the ultimate ground of all philosophy.39 He maintains that the concept of representation cannot be determined through itself but ‘can only be drawn from consciousness qua fact (Tatsache)’. On his account, ‘this fact alone must ground the foundation of the Elementarphilosophie’.40 By reflecting on this fact, Reinhold writes, one obtains the so-called principle of consciousness, which is articulated in the following proposition: In consciousness, the representation is distinguished from both that which is being represented and that which represents, and is referred to both. (Con I 144)41

194  Karin de Boer and Gesa Wellmann As we see it, Reinhold establishes this principle because it allows philosophy, at least in principle, to study the various types of cognitive acts by means of which the human mind refers representations to objects in an exhaustive manner.42 How exactly Reinhold arrived at the fact of consciousness and from there at the principle of consciousness, and how he adjusted his view on the matter throughout the period of his Elementarphilosophie, has been the subject of many studies and need not be discussed here.43 For our purposes, it suffices to conclude that in order to remedy the perceived shortcomings of the Critique – including its allegedly defective foundation – Reinhold seeks to abolish the restricted aim of a critique and to elaborate what he conceives of as a more elementary philosophy. However, as we have seen, he claimed that the latter, in accordance with Kant’s own meta­ philosophical principles, merely deviates from the Critique by developing its account of the principles of cognition in a properly scientific manner, i.e., by deriving them from a single first principle. 8.5 Reinhold’s Elementarphilosophie and the Philosophy without Surnames Although Reinhold considered his Elementarphilosophie to supersede Kant’s conception of critique, we argue in this section that he nevertheless held on to a version of Kant’s distinction between a propaedeutical discipline and a metaphysical system. In order to do so, it is useful to return to a passage from Foundation discussed in Section 8.4, now cited in full: The fate of the new metaphysics that Kant invented depends, therefore, on the development (Entwicklung) of the possibility of experience, i.e., the original structure (Einrichtung) of the faculty of cognition. Although this metaphysics can be universally valid (allgemeingültig) as regards its principles, theorems, and corollaries, it cannot become universally accepted (allgemeingeltend) until that upon which its foundation is grounded itself is constituted by principles, theorems, and corollaries that are universally accepted, in other words, until the propaedeutic to metaphysics itself has been elevated to the science of the capacity to cognize (Erkenntnisvermögen). (FPK 70, translation modified, cf. Essay 212–219, Con II 416) As the final sentence makes clear, Reinhold held that Kant’s Critique of Pure Reason could be turned into a proper science by developing the various elements of Kant’s theory of cognition from the single capacity of the human mind to represent. That is to say, if the results of the Critique were organised according to the true ground of all philosophy, the theory

Kant, Reinhold, and the Problem of Philosophical Scientificity 195 of cognition contained in this work would meet the scientific standards that Reinhold attributes to Kant. In this sense, the science of the capacity to cognise would amount to an improved version of the Critique of Pure Reason. As its name already indicates, however, the science of the capacity to cognise merely deals with a specific mode of the faculty of representation, namely cognition. Reinhold states that this science must be grounded on a more fundamental science, namely the science of the capacity to represent as such as well as the various types of representation produced by the human mind.44 Operating at this fundamental level, the science of the faculty of representation – or, as Reinhold also calls it, the Elementarphilosophie – demonstrates the correctness of the premises on which the Critique of Pure Reason is based.45 This demonstration, Reinhold holds, can no longer be understood as a critical enterprise: The endeavor (Geschäft) of the critical philosophy could only … be concluded with the absolute root explanation (Grunderklärung) of the representation. However, once philosophy has arrived at this root explanation, it stops being critical. Arrived at this point, the science of the foundation of the philosophy without surnames (Philosophie ohne Beinamen), [i.e.,] the Elementarphilosophie, begins. (FPK 104–105, translation modified) As we have seen in Section 8.4, Reinhold holds that Kant’s critical philosophy does not conceive of the very capacity to represent as the ground or principle of the various cognitive activities carried out by the human mind and thus cannot derive these various activities and their products from this capacity. By providing a “root explanation” of the latter, the Elementarphilosophie moves beyond the critical, i.e., non-scientific analysis of the human mind carried out in the Critique of Pure Reason. Thus, although Reinhold introduces his “root explanation” as a means to make critical philosophy coherent, this very explanation turns his philosophy into an early instance of post-critical philosophy.46 The aim of keeping the Kantian results while engaging in a strategy that is clearly non-Kantian becomes particularly clear in a passage of Reinhold’s Essay on a New Theory of the Human Capacity for Representation (1789). Referring to this earlier text in Contributions I, Reinhold states: Unless this essay has completely failed, it establishes the whole critical elementary philosophy anew, independently of the grounds on which this philosophy rests in the Critique of Pure Reason. Since it leads to the same results through a completely different path, it

196  Karin de Boer and Gesa Wellmann serves – in the same way as alternative calculation methods (Rechnungsproben) do – as a confirmation of Kant’s discoveries. (Con I 262–263) Despite Reinhold’s claim that the Elementarphilosophie is no longer critical, the passage quoted above (FPK 104–105) makes clear that Reinhold continued to conceive of the Elementarphilosophie as a propaedeutic rather than a self-contained system. Contrary to most commentators, we take him more specifically to distinguish between the Elementarphilosophie and a philosophy that has overcome the one-sidedness of all preceding systems and therefore should be called ‘neither critical, nor empirical, rational or skeptical’ but rather a ‘philosophy without surnames’.47 This means that Reinhold, following Kant, maintained that philosophy should ultimately elaborate a metaphysical system – the philosophy without surnames – that presupposes a propaedeutic investigation into the various types of representations produced by the human mind, i.e., the Elementarphilosophie. On Reinhold’s view, the solid foundation of his projected philosophical system would preclude its association with a particular school or movement and, therefore, its entanglement in the controversies that divided his contemporaries. Its scientific nature would thus prevent it from meeting the same fate as Kant’s Critique. Just like Kant, however, Reinhold postponed the realisation of the philosophical system he envisioned: The only possible future system of the scientific philosophy without surnames, of which the Kantian critique established the proper and determined topic, and that, in this sense, could only become possible through a single independent thinker (Selbstdenker), can only be gradually realized through the cooperation of many. (Con II iii, cf. FPK 129) Due to the preparatory work carried out in the Elementarphilosophie, however, Reinhold is confident that such a philosophical system can ultimately be established according to the true nature of philosophy. The latter consists, for Reinhold, in a systematic investigation into the faculty of representation and the a priori elements generated by the various types of representing: The concept of philosophy … of which I believe that it is the only one about which future universal agreement is not impossible is contained in the following declaration: Philosophy is the science of that which is determined by the faculty of representation. (Con I 59, cf. FPK 133)

Kant, Reinhold, and the Problem of Philosophical Scientificity 197 The future philosophy without surnames would be universally agreed upon because of the scientific foundation provided by the Elementarphilosophie. Since, as seen, the latter derives the various types of representation and the principles by which they are guided from a single axiom, i.e., the principle of consciousness, the philosophy without surnames can be established upon this axiom as well and, thus, attain scientific certainty.48 As was shown in Section 8.2, Kant ascribes distinct tasks to the Critique and the metaphysical system: whereas the former reflects on the very possibility of metaphysics, the latter offers an exhaustive account of all a priori elements of cognition. As we have argued, Reinhold pays little attention to the properly critical aspect of Kant’s propaedeutic. Instead, he assesses the critical strand of Kant’s project primarily in view of its allegedly defective method. His Elementarphilosophie rejects the label ‘critical’ in order to emphasise that it proceeds according to a scientific, i.e., deductive, method. Accordingly, on Reinhold’s account, propaedeutic and system can hardly be distinguished. Their only difference consists in the fact that the Elementarphilosophie treats the principle of consciousness and the cognitive capacities, concepts, and principles that can immediately be derived from it, whereas the philosophy without surnames carries out the derivation of the principles of all specific theoretical and practical sciences. 8.6 Conclusion We have argued that Reinhold tried to bring philosophy onto “the secure path of a science” by claiming that Kant’s Critique of Pure Reason re­ presented only a first step on this path. In order to achieve a truly scientific philosophical system, Reinhold undertook to develop the content of the Critique according to an axiomatic-deductive method. In the texts he published around 1790, Reinhold tried to convince his readers that the path he promoted and had taken himself was the very path that Kant had discovered and pursued up to a point. Reinhold does so, we have shown, by borrowing extensively from Kant’s account of the faculties and the a priori elements of cognition. In this chapter, however, we have focused on Reinhold’s appropriation of key aspects of Kant’s metaphilosophy, namely his distinction between propaedeutic and system and his critical conception of the role of principles. While it is clear that Reinhold was deeply indebted to Kant, we have argued that his presentation of the Critique is misguided in three respects. Firstly, Reinhold’s presentation marginalises the critical strand of the Transcendental Analytic and the Critique of Pure Reason as a whole. Secondly, he attributes to Kant a foundationalism that is at odds with Kant’s conception of critique and the principles it establishes. Thirdly, while Reinhold, in line with Kant, regarded the Elementarphilosophie as propaedeutic to a scientific philosophical system, the idea that

198  Karin de Boer and Gesa Wellmann this propaedeutic ought to deduce all cognitive capacities, concepts, and principles from a single foundational principle is squarely non-Kantian. Thus, we hope to have shown how Reinhold’s metaphilosophical innovations caused the idea of a critique of pure reason to disappear from the scene of early post-Kantian philosophy.49 Abbreviations CPR Critique of Pure Reason (1781/1787) Con I Contributions I (1790) Con II Contributions II (1794) FPK The Foundation of Philosophical Knowledge (1791) Essay  Essay on a New Theory of the Human Capacity for Representation (1789) Notes 1 The texts Reinhold published between 1789 and 1794 are usually considered as partial elaborations of what he refers to as his Elementarphilosophie. Reinhold himself uses the term in most of these texts, but nowhere clearly indicates which of his works exactly are to be understood as belonging to Elementarphilosophie. Fabbianelli, however, has provided good reasons to count at least the first volume of the Contributions to the​Correction of Previous Misunderstandings of Philosophers (1790) and The Foundation of Philosophical Knowledge (1791) as concerned with Elementarphilosophie. For a discussion of the works included in this phase, see Fabbianelli 2003. 2 Reinhold held that the Critique of Pure Reason itself had already united what is true in the philosophies of Locke, Leibniz, and Hume, but that both Kantians and anti-Kantians had failed to understand this achievement and held on to the very assumptions that had fuelled earlier controversies. See Essay 12–23, FPK 60–61, 131. 3 Following the most recent translation of Reinhold’s 1789 Essay (2011), we will refer to the book title as Essay on a New Theory of the Human Capacity for Representation. However, we will translate Vorstellungsvermögen as ‘capacity to represent’ in all other cases. 4 We agree with Beiser (1987: 240) that Reinhold’s strategy represents an early and implicit form of immanent critique. However, we do not share Beiser’s more specific claim that this critique is “strictly immanent” in that it is derived from Kant’s own ideals. For a consideration of Hegel’s early conception of immanent critique, see De Boer 2012. 5 Noting that Reinhold’s engagement with Kant wavers between an ‘apologetic completion and a firm correction of Kantian thoughts’ (37), Bondeli (1995) almost completely disregards Reinhold’s distortions of ideas he derived from the Critique. Similarly, Bernecker (2010: 226–227) writes that ‘Reinhold’s program to derive the specific content of the Kantian system from a first principle takes its inspiration from Kant himself’, including his conception of the ‘highest principle of all synthetic judgments’ (A158/B197), without pointing to what we take to be Reinhold’s misconstrual of the role Kant attributed to this

Kant, Reinhold, and the Problem of Philosophical Scientificity 199 principle (see Section 8.3). As far as we are aware, authors who do address the differences between Kant and Reinhold do not focus on the issue of critique (see Pinkard 2002: 100–01; Franks 2005: 217–18), or do so only to a very limited extent (see Onnasch 2009: XIX). Unlike Baum (1974: 93–95), moreover, we do not hold that Reinhold’s criticism of Kant is warranted. We also dis­ agree with Sauer (1982: 96), according to whom Reinhold might not have been aware of the differences between Kant’s theory and his Elementarphilosophie. In a more critical vein, Di Giovanni (1985) considers Reinhold’s introduction of a new method to undermine the very idea of Kant’s Critique as regards the distinction between thought and sensibility (cf. 19). Ameriks (2000: 2) claims that the Critique ‘was seriously misunderstood by those who seemed to stand closest to it’, but does not discuss or assess Reinhold’s reading of particular elements of the first Critique in this light. Our chapter converges with Goh (2014: 248), who argues that Reinhold distorts certain ideas of Kant’s Critique of Pure Reason with regard to the problem of scientificity. Yet while Goh approaches this problem from the perspective of Reinhold’s philosophy, our chapter investigates the idea of scientific philosophy from both a Kantian and a Reinholdian perspective. 6 See, e.g., FPK 63, 73–74; Con I 262–263, 185–186, 309. However, these passages do not carry much weight in his actual arguments, which are mostly concerned with the lack of scientificity of the Critique of Pure Reason. 7 The fifth essay in Contributions I, titled “On the Relationship of the Theory of the Faculty of Representation to the Critique of Pure Reason” (Con I 257– 338), is particularly relevant to our purposes. The translations of passages from Contributions I and II, as well as the parts of Foundation not translated in Di Giovanni 1985 (FPK 37–138), are our own. 8 In the texts mentioned above, Reinhold both presents his metaphilosophical ideas about the method, form, and content of his Elementarphilosophie and carries out parts of his plan. This is to say that he never published a complete version of the philosophy designated by this title. In Foundation, he explicitly states that the ‘future Elementarlehre must first establish what is assumed as established in the Critique of Pure Reason’ (FPK 132). On this, see Bondeli 1995: 39. 9 See Ameriks 2000: 84. Unlike most commentators, we treat Reinhold’s texts in view of his engagement with the Critique of Pure Reason rather than his contribution to the development of German idealism (cf. Breazeale 1982, Ameriks 2000, 2004; Silva 2017: 80–82). Whereas commentators generally affirm the shifts effected by Reinhold and other early post-Kantians, we believe that further investigations into early post-Kantian philosophy would profit from paying more attention to their distorting reading of Kant’s work. 10 According to Reinhold, at least Kant’s followers considered the Critique of Pure Reason already to contain Kant’s system (see FPK 62–63, 115). 11 Reinhold writes that ‘the main moments of the critique of reason are the forms of intuitions, of concepts, and ideas’ and that this work completely enumerates them ‘insofar as they are determined entirely a priori in the nature of sensibility, the understanding, and of reason, and cannot be attributed to things in themselves’ (Con I 263; cf. FPK 72–73). Note that Reinhold disregards Kant’s intention to complement his account of the categories with a discussion of the derivative pure concepts he calls ‘predicables of the pure understanding’, including “force”, “action”, and “change” (CPR A82/B108). 12 See CPR A13–14/B27–28. Admittedly, Kant does not explain to what extent this critical reflection itself can already proceed systematically. Moreover, his

200  Karin de Boer and Gesa Wellmann account of the relationship between transcendental critique (A12/B26) and transcendental philosophy proper, i.e., considered as first part of metaphysics, is not crystal clear. On the relationship between these two strands in the Transcendental Analytic, see De Boer 2020: 73–100. 13 Cf. FPK 133; Con I 278. 14 See De Boer 2021 for a recent discussion of this problem. 15 Clearly, Reinhold could draw on Kant’s own reflections on this issue: in the Architectonic of the Critique of Pure Reason, Kant considers systematicity to be a fundamental condition of scientificity (CPR A832/B860). In contrast to Reinhold, however, Kant does not claim that a system must deduce its contents from a single first principle; he rather suggests that any system should be teleologically organised (cf. CPR A835/B863; A832–833/B860ff; A839/B867). On this issue, see Wellmann 2018: 88–91, 133–138. Thus, even though it remains unclear whether Kant took the criteria developed in the Architectonic chapter to apply to the Critique of Pure Reason itself, Reinhold’s conception of systematicity importantly deviates from Kant’s own. 16 We translate both “Grundsatz” and “Satz” as “principle” in order to indicate their proximity to classical metaphysical principles. Accordingly, we translate “Satz des Bewusstseins” as ‘principle of consciousness’ rather than “proposition of consciousness”. 17 Reinhold’s account clearly draws on passages from the Critique such as the following: ‘Although we learn many laws through experience, these are only particular determinations of yet higher laws, the highest of which … come from the understanding itself a priori, … provide the appearances with their lawfulness and by that very means make experience possible. … Thus, as exaggerated and contradictory as it may sound to say that the understanding is itself the source of the laws of nature, and thus of the formal unity of nature, such an assertion is nevertheless correct and appropriate to the object, namely experience’ (CPR A126–127). 18 See FPK 68, where Reinhold cites Kant’s description of the supreme principle of all synthetic judgments at A158/B197. See also FPK 62, 69. 19 Cf. FPK 114. In Beitrage I, Reinhold summarises this point by stating that ‘experience is … the properly supreme ground, the foundation (Fundament), upon which the great edifice of the Critique of Pure Reason is erected’ (Con I 278). Baum (2004: 106) suggests that Reinhold might have taken over this idea from Maimon’s Versuch über die Transzendentalphilosophie (1790: 186), which indeed puts forward a very similar interpretation of Kant’s notion of experience and its role in the Critique of Pure Reason. 20 It is hard to tell whether Reinhold considered Kant himself to have held this view. In Beitrage I, he at one point provides an explanation that is in line with Kant’s own. Reinhold here states with regard to space and time and the categories: ‘On the basis of the original determination of these forms as ‘making possible experience’, it is shown that their objective reality is limited to objects of experience’ (Con I 278). However, remarks such as these are marginal to Reinhold’s interpretation of the Critique. 21 For a helpful discussion of this account, see Jakušić’s contribution to this volume. 22 While Kant’s language and presentation clearly intimate the Wolffian tradition, both the context in which he presents the supreme principle of all synthetic judgements and the meaning of the principle itself make it clear that Kant conceived of the function of this principle in very different terms. Reinhold, for his part, also criticises philosophies that give too much weight to either the principle

Kant, Reinhold, and the Problem of Philosophical Scientificity 201 of non-contradiction (FPK 29–31, 48–49, 110–111) or the principle of sufficient reason (FPK 33–42; Con I 28–42). This notwithstanding, he obviously embraced the idea, usually attributed to Leibniz and Wolff, that philosophy should derive all of its content from a single principle (cf. FPK 27). On Reinhold’s debt to Wolffianism, see Mensen 1974; Sauer 1982: 57–97; and Frketich 2016a. 23 In our view, Kant deliberately separated the so-called metaphysical deduction from the transcendental deduction. Whereas the former is said to generate the table of categories from ‘a common principle, namely, the capacity to judge’ (A80–81), the latter seeks to determine, in a critical fashion, the conditions under which the use of these categories is warranted (A85/B117). On this, see Frketich’s contribution to the present volume. Clearly, Reinhold extrapolates Kant’s idea that the categories can be derived from a single principle to the other parts of the Critique, including Kant’s account of the various faculties involved in the cognition of objects. 24 ‘The principle that it presents in the system of principles of the pure understanding as the highest one … is merely the basic principle of the use of the understanding in relation to experience … and presupposes … the correct theories of sensibility, the understanding, and the faculty of representation’ (Con I 273). 25 FPK 69–70. We adopt the clause ‘command universal assent’ from Di Giovanni 1985: 11. On Reinhold’s conception of a universally recognised axiom, see Ameriks (2004: 84), who discusses this concept in light of Reinhold’s endeavour to popularise philosophy. 26 See Maimon 1790: 73 f; cf. FPK 69, 130. In Contributions I, 278–281, Reinhold repeats a common reproach to Kant, namely that experience can only serve as a foundation in the Critique of Pure Reason by using circular reasoning: ‘The concept of experience, insofar as it is the basis (Basis) of the Kantian system, can in no way be based on this system and cannot even be clarified by dint of the latter without [creating] a circle’ (Con I 281, cf. 428). A discussion of the role Kant assigns to experience in the Critique falls outside the scope of this chapter. 27 Con I 281. 28 FPK 70, cf. 62; Con I 281. 29 Con I 333, cf. 281. Evidently, Reinhold ultimately aimed to derive the specific principles assumed in the Critique and all other philosophical disciplines from a single principle (see FPK xiii-xiv). On this, see Mensen 1974: 118–119. 30 Con I 275. 31 Con I 263. 32 FPK 70. 33 FPK 71–73. Reinhold stresses that the capacity to represent cannot simply be abstracted from the above-mentioned faculties, since, as he holds, ‘sensibility, understanding and reason, taken together, constitute the extension, not the content of the concept of the capacity to represent’ (Con I 177; cf. Essay 203–204). He elsewhere points out that the capacity to represent can only be clarified by analysing its effects, namely representations (Con I 179). 34 Con I 276–277; cf. Con I 280. 35 Essay 214. 36 Con I 267–269; 357–358. To be sure, Kant’s so-called Stufenleiter also considers the concept of representation to be a highest genus (CPR, A320/B376– 377). However, the concept does not play as central a role as Reinhold believes it ought to play (see Con I 275–277). See Bondeli 2008: 66 and Goh 2014: 252 for helpful discussions of Kant and Reinhold’s notions of representation. 37 FPK 72–73.

202  Karin de Boer and Gesa Wellmann 8 Con I 271. 3 39 Con I 267, 270, 272. 40 FPK 77–78. 41 Cf. Versuch 200; FPK 78. As Bondeli (2003: 16) points out, Reinhold’s principle might have been inspired by the following passage from the Transcendental Dialectic: ‘Now what is universal in every relation (Beziehung) that our representations can have is 1) the reference (Beziehung) to the subject, 2) the reference to objects, and the latter either as appearances, or as objects of thinking as such’ (CPR, A333–334/B390–391, translation modified). On this, see Goh 2014: 252. In Contributions I, Reinhold claims accordingly that ‘[c]onsciousness is the source of all principles of the Elementarphilosophie, and those principles are propositions that express nothing else but consciousness’ (Con I 162). 42 Reinhold held that volitional acts, i.e., the various types of acts by means of which the human mind refers representations to the subject, also rely on the principle of consciousness and that the latter can therefore provide practical philosophy with a solid foundation as well (Con I 86–87, 252; FPK v-xviii, 75–76). On this issue, see Lazzari 2004. 43 See Mensen 1974: 122–123, Bondeli 1995: 41–107, Breazeale 2004: 50, Frketich 2016b, and Frketich’s contribution to this volume. 44 As mentioned in note 42, Reinhold considered the principle of consciousness to ground the theoretical as well as the practical part of his projected system: ‘The science of the capacity to cognize must be preceded … by another one, namely, the science of sensibility, the understanding, and reason, insofar as they together … ground the capacity to cognize (yet not this capacity alone, but also the capacity to desire (Begehrungsvermögen))’ (FPK 71, cf. Con I 277). On this, see Bondeli 1995: 78–79. 45 FPK 72–75. 46 This is not to say that, after Reinhold, early post-Kantian philosophy repudiated the critical impetus of Kant’s Critique of Pure Reason in all regards. In Fichte and Hegel’s works, critique is rather transformed into an inherent element of the method by means of which each particular cognitive activity, concept, or principle that does not establish the unity of its contrary determinations in an adequate way is shown to be one-sided and thus to call for its overcoming. See Wellmann 2021: 59–73 and De Boer 2004: 801–803 for considerations of this issue in Fichte’s Wissenschaftslehre and Hegel’s Science of Logic. 47 FPK 132, cf. Con I 84–86. In both cited passages from Foundation (FPK 104– 105 and 132), Reinhold defines the Elementarphilosophie as the science of the foundation of the philosophy without surnames. Thus, he clearly does not identify this philosophy with the Elementarphilosophie, as commentators such as Breazeale (1982: 795), Fabbianelli (2004a: xlvi–xlvii), and Bernecker (2010: 221) suggest. On this, see Wellmann 2018. 48 As we see it, the principle of consciousness determines the projected system at two levels. Firstly, it directly determines each of the main subjects treated in the system. Thus, Reinhold divides pure philosophy – that is, the whole of the disciplines derived from the Elementarphilosophie – into theoretical and practical philosophy. He defines theoretical philosophy as the science that deals with the relation between representations and the object that is being represented. By contrast, practical philosophy is defined as the science that deals with the relation between representations and the representing subject (Con I 87). Secondly, the principle of consciousness indirectly determines the internal organisation of each of the sciences that together make up the philosophy without

Kant, Reinhold, and the Problem of Philosophical Scientificity 203 surnames: ‘Since it [the first principle] is the indirect ground (Grund) of necessity of all other principles …, the whole edifice of science, which becomes systematic only in virtue of this principle, achieves its stability through the latter. This stability can only be achieved by the thoroughgoing connection of all propositions and by tracing them back to a single proposition’ (Con I 119). Logic, for instance, is said to be based on the law of non-contradiction. The law of non-contradiction presupposes, Reinhold argues, the concept of thinking and the latter, in turn, presupposes the concept of representation. Since this concept is determined by the principle of consciousness, the first principle of logic is indirectly determined by the former (FPK 84–86). 49 We would like to thank Elise Frketich and Kienhow Goh for their helpful comments on a draft of this chapter.

Bibliography Primary Sources Kant, Immanuel (1998) Kritik der reinen Vernunft (Hamburg: Meiner) Kant, Immanuel (1998) Critique of Pure Reason. Translated by Paul Guyer and Allen W. Wood (Cambridge, MA: Cambridge University Press) Reinhold, Karl Leonhard (1978) “Über das Fundament des philosophischen Wissens.” In Gesammelte Schriften. Edited by Wolfgang Schrader (Hamburg: Meiner) Reinhold, Karl Leonhard (1985) “The Foundation of Philosophical Knowledge” in Between Kant and Hegel: Texts in the Development of Post-Kantian Idealism. Edited and translated by George Di Giovanni and H.S. Harris (Albany: State University of New York Press) Reinhold, Karl Leonhard (2003) Beiträge zur Berichtigung bisheriger Mißverständnisse der Philosophen (I) Erster Band, das Fundament der Elementarphilosophie betreffend. Edited by Faustino Fabbianelli (Hamburg: Meiner) (Contributions to the​Correction of Previous Misunderstandings of Philosophers) Reinhold, Karl Leonhard (2004) Beiträge zur Berichtigung bisheriger Mißverständnisse der Philosophen (II) Zweiter Band, die Fundamente des philosophischen Wissens, der Metaphysik, Moral, moralischen Religion und Geschmackslehre betreffend. Edited by Faustino Fabbianelli (Hamburg: Meiner) (Contributions to the​Correction of Previous Misunderstandings of Philosophers) Reinhold, Karl Leonhard (2010) Karl Leonhard Reinhold, Versuch einer neuen Theorie des menschlichen Vorstellungsvermögens, Teilband 1, mit einer Einleitung und Anmerkungen. Edited by Ernst Onnasch (Hamburg: Meiner) Reinhold, Karl Leonhard (2011) Essay on a New Theory of the Human Capacity for Representation. Translation by Tim Mehigan and Barry Empson (Berlin: De Gruyter) Other Primary Source Maimon, Salomon (1790/1963) Versuch über die Transzendentalphilosophie. Mit einem Anhang über die symbolische Erkenntnis und Anmerkungen (Darmstadt: WBG)

204  Karin de Boer and Gesa Wellmann Secondary Sources Ameriks, Karl (2000) Kant and the Fate of Autonomy. Problems in the Appropriation of the Critical Philosophy (Cambridge, MA: Cambridge University Press) Ameriks, Karl (2004) “Reinhold über Systematizität, Popularität und die historische Wende.” In M. Bondeli (Ed.) Philosophie ohne Beinamen. System, Freiheit und Geschichte im Denken Karl Leonard Reinholds (Basel: Schwabe Philosophica), pp. 303–333. Baum, Manfred (2004) “Die Möglichkeit der Erfahrung und die analytische Me­ thode bei Reinhold.” In Philosophie ohne Beinamen. System, Freiheit und Geschichte im Denken Karl Leonard Reinholds, edited by Martin Bondeli (Basel: Schwabe Philosophica), pp. 104–118 Beiser, Frederick (1987) The Fate of Reason: German Philosophy from Kant to Fichte (Cambridge, MA: Harvard University Press) Bernecker, Sven (2010) “Reinhold’s Road to Fichte: The Elementary-Philosophy of 1795/96.” In Karl Leonhard Reinhold and the Enlightenment, edited by George di Giovanni (Dordrecht: Springer), pp. 221–240 Bondeli, Martin (1995) Das Anfangsproblem bei Karl Leonhard Reinhold. Eine Systematische und Entwicklungsgeschichtliche Untersuchung zur Philosophie Reinholds in der Zeit von 1789 bis 1803 (Frankfurt am Main: Vittorio Klostermann) Bondeli, Martin (2003) “Reinholds Kant-Kritik in der Phase der Elementarphilosophie.” In Die Philosophie Karl Leonhard Reinholds, edited by Martin Bondeli and Wolfgang H. Schrader (Amsterdam: Brill), pp. 1–24 Breazeale, Daniel (1982) “Between Kant and Fichte: Karl Leonhard Reinhold’s ‘Elementary Philosophy’” The Review of Metaphysics, 35, No. 4, pp. 785–821 De Boer, Karin (2004) “The Dissolving Force of the Concept: Hegel’s Ontological Logic” The Review of Metaphysics, 57, No. 4, pp. 787–822. De Boer, Karin (2012) “Hegel’s Conception of Immanent Critique: Its Sources, Extent, and Limit” In Conceptions of Critique in Modern and Contemporary Philosophy, edited by Karin de Boer and Ruth Sonderegger (London: Palgrave), pp. 83–100 De Boer, Karin (2020) Kant’s Reform of Metaphysics: The Critique of Pure Reason Reconsidered (Cambridge, MA: Cambridge University Press) De Boer, Karin (2021) “Does the Investigation Kant Carries Out in the Critique of Pure Reason Amount to a Science?” In Philosophie als Wissenschaft, edited by Nora Schleich (Hildesheim: Olms), pp. 31–46 Di Giovanni, George; Harris, H.S (1985) Between Kant and Hegel. Texts in the Development of Post-Kantian Idealism (Indianapolis: Hackett Publishing Company) Fabbianelli, Faustino (2016) Karl Leonhard Reinhold’s Transcendental Psychology (Berlin: De Gruyter) Franks, Paul (2005) All or Nothing (Cambridge, MA: Harvard University Press) Frketich, Elise (2016a) “Reinhold’s Elementarphilosophie: A Scholastic or Critical Philosophical System?” Kant Yearbook 8, No 1, pp. 17–38 Frketich, Elise (2016b) “The Starting Point of Reinhold’s Elementarphilosophie” Workshop of the Research Network Transcendental Philosophy/German Idealism February 18–20, 2016. University of Southern Denmark, Odense Gerten, Michael (2010) “Bürgerliche Aufklärung und streng systematische Philosophie. Zum Verhältnis von Leben und Philosophie bei K. L. Reinhold.” In

Kant, Reinhold, and the Problem of Philosophical Scientificity 205 Karl Leonhard Reinhold and the Enlightenment, edited by George di Giovanni (Dordrecht: Springer), pp. 65–85 Goh, Kienhow (2014) “Reinhold and the Transformation of Philosophy into Science.” In The Palgrave Handbook of German Idealism, edited by Matthew Altman (New York: Macmillan), pp. 392–416 Heinz, Marion and Stolz, Violetta (2008) “Vernunft und Geschichte: Von Kant zu Reinhold.” In Am Rande des Idealismus, edited by Wolfgang Kersting and Dirk Westerkamp (Paderborn: Brill), pp. 159–185 Lauth, Reinhard, ed (1974) Philosophie aus einem Prinzip. Karl Leonard Reinhold (Bonn: Bouvier) Lazzari, Alessandro (2004) Das Eine, was der Menschheit Not ist. Einheit und Freiheit in der Philosophie Karl Leonhard Reinholds (1789–1792) (StuttgartBad Cannstatt: Frommann-Holzboog) Mensen, Bernhard (1974) “Reinhold zur Frage des ersten Grundsatzes der Philosophie.” In Philosophie aus einem Prinzip, edited by Reinhard Lauth (Bonn: Bouvier), pp. 109–128 Onnasch, Ernst-Otto (2009) “Einleitung” In Versuch einer neuen Theorie des menschlichen Vorstellungsvermögens (Hamburg: Meiner), pp. xi–cxxxiii Pinkard, Terry (2002) German Philosophy. 1760–1860. The Legacy of Idealism (Cambridge, MA: Cambridge University Press) Rumore, Paola (2009) “Wolff und die neue Theorie des menschlichen Vorstellungsvermögens. Das Wolffianische Erbe in der Elementarphilosophie Reinholds.” In Christian Wolff und die europäische Aufklärung. Akten des 1. Internationalen Christian-Wolff-Kongresses, Halle (Saale), Bd.5, edited by Jürgen Stolzenberg (Hildesheim: Olms), pp.163–176 Sauer, Werner (1982) Österreichische Philosophie zwischen Aufklärung und Restauration: Beiträge Zur Geschichte des Frühkantianismus in der Donaumonarchie (Amsterdam: Rodopi/Brill) Schrader, Wolfgang (1979) “Philosophie als System – Reinhold und Fichte.” In Erneuerung der Transzendentalphilosophie im Anschluss an Kant und Fichte, edited by Klaus Hammacher (Stuttgart Bad Cannstatt: Frommann-Holzboog), pp. 331–344 Silva, Fernando (2017) “‘Das Eine, was der Philosophie Not ist’: Reinhold’s argument concerning the absolute principle of philosophy.” In Filosofia Unisinos. Unisinos Journal of Philosophy 18, No. 2, pp. 79–86 Wellmann, Gesa (2018) The Idea of a Metaphysical System: Lambert, Kant, Reinhold, and Fichte. PhD thesis, University of Leuven. Wellmann, Gesa (2021) “Kritik und Philosophie als Wissenschaft – das Verhältnis der Wissenschaftslehre zu Kants kritischer Metaphysik.” In Philosophie als Wissenschaft, edited by Nora Schleich (Hildesheim: Olms), pp. 59–73

9

Reinhold on the Deduction of the Categories Elise Frketich

9.1 Introduction Kant famously undertakes a critique of pure reason to place metaphysics ‘upon the secure course of a science’ (Bvii). Reinhold shares Kant’s goal, and while he does not think that Kant has achieved it, he praises Kant for putting metaphysics on the path towards becoming a science.1 A science, for Kant and Reinhold, is a system of philosophy. A critical system of philosophy, according to Reinhold, is one in which the a priori forms of the human mind determine the first principles of logic, metaphysics, and morality. Reinhold thinks that Kant has identified the correct forms of the human mind, namely space and time, the categories, and the ideas (Beiträge 263−264). However, he does not think that Kant has proved them in the Critique of Pure Reason. Thus, they cannot yet ground any necessary principles of metaphysics. For this reason, Reinhold aims at proving that these are indeed the forms of the human mind, and he takes himself to do so in the “Elementarphilosophie”, his own propaedeutic to a critical philosophical system.2 His strategy is to ground these forms in what he takes to be a scientific account of representation in general. This is also his strategy with respect to his deductions of the categories, which are the focus of this chapter. To understand the aims of these deductions, we must first look to Kant. The first deduction of the categories that Kant carries out is the so-called metaphysical deduction (hereafter: MD), in which he aims to identify the concepts that make up the table of categories (A79/B105–A80/B106). This table lists the fundamental ways of cognising an object. The second is the famous transcendental deduction (hereafter: TD), in which he aims to establish these categories as a priori conditions for the possibility of experience (A94/B126). The MD and TD are crucial for Kant’s metaphysics because they provide the fundamental categorial structure of the phenomenal world.3 Reinhold agrees with Kant that the MD and TD are crucial for metaphysics. However, he takes issue with the deductions themselves. While he does not criticise the categories that Kant identifies, nor does he take issue with DOI: 10.4324/9781003187561-10

Reinhold on the Deduction of the Categories 207 Kant’s claim that the categories are a priori conditions for the possibility of experience, he is concerned that Kant does not rule out the possibility of additional categories in the MD and that the TD does not prove that the categories that Kant identifies are necessary for experience, leaving a fundamental aspect of Kant’s metaphysics vulnerable to rejection or scepticism.4 For these reasons, and in service of Kant’s and Reinhold’s shared goal, Reinhold attempts to provide a genuinely necessary MD and TD. In this chapter, I will first take up Reinhold’s criticisms of Kant’s deductions of the categories, connecting them to criticisms of these deductions today (Section 9.2). Here we will see that, along with several contemporary Kant scholars, Reinhold does not think that Kant provides a completeness proof for the table of categories, and he holds that Kant’s MD is grounded in a vicious circle. Second, to remedy these perceived issues, I hold that Reinhold reverses the order in which Kant proceeds. While Kant begins with the MD and proceeds to the TD, I will show, against current interpretations, that Reinhold begins with his own version of a TD, one whose method differs from Kant’s (Section 9.3), and only then proceeds to his own version of an MD (Section 9.4).5 That is, he first attempts to show that a Kantian category is an a priori condition for the possibility of experience, i.e., one that originates in a fundamental act of the understanding, and only then can he provide a completeness proof for the table of the categories by exhausting the fundamental acts of the understanding. While reversing the order of Kant’s deductions alone does not make any difference to their success, Reinhold uses this order reversal as well as additional principles to avoid the issues that he thinks plague them. 9.2  Kant and His Interpreters Criticisms of Kant’s deductions of the categories today arise depending on how one responds to the following two questions, questions that Reinhold also addresses: 1 Does Kant provide a completeness proof for the table of judgements, and, thereby, for the table of categories in the MD? 2 Is Kant’s MD grounded in a vicious circle? Kant gives rise to the first question when he asserts that a completeness proof for the table of categories is mandatory: Transcendental philosophy has the advantage but also the obligation to seek its concepts in accordance with a principle (Princip), since they spring pure and unmixed from the understanding, as absolute unity, and must therefore be connected among themselves in accordance

208  Elise Frketich with a concept or idea. Such a connection, however, provides a rule by means of which the place of each pure concept of the understanding and the completeness of all of them together can be determined a priori, which would otherwise depend upon whim or chance. (A67/B92) A completeness proof shows that the 12 categories that Kant identifies make up the table of categories. Such a proof, he explains, requires a principle according to which the table of categories is determined. This principle not only ensures that the correct number of categories has been found, but it also ensures the accuracy of the classification of these categories into four sections. While Kant clearly thinks that a principle is required to prove that the table of categories is complete, he is not forthcoming about what that principle might be. However, he comes closest to articulating it in the following passage, which follows directly upon his table of categories: Now this is the listing of all original pure concepts of synthesis that the understanding contains in itself a priori, and on account of which it is only a pure understanding; for by these concepts alone can it understand something in the manifold of intuition, i.e., think an object for it. This division is systematically generated from a common principle, namely the faculty for judging (which is the same as the faculty for thinking), and has not arisen rhapsodically from a haphazard search for pure concepts, the completeness of which one could never be certain, since one would only infer it through induction, without reflecting that in this way one would never see why just these and not other concepts should inhabit the pure understanding. (A80/B106−A81/B107) Kant not only announces that he has provided a completeness proof for his table of categories in the last sentence of this passage, but he also equates the principle that guarantees this proof with “the faculty for judging” or “the faculty for thinking”. However, if “principle” here refers to a proposition, for Kant, as is commonly thought today, then we must ask, how can a faculty be a principle? Contemporary Kant scholars are torn about how to answer this question and, as a result, how to understand Kant’s claim that his table is complete. While some hold that Kant did not provide a completeness proof for his table of categories at all,6 others argue that he indeed provided one and that he did so by seeking the categories in accordance with a principle (or at least that he provided all the information necessary to reconstruct such a proof)7. However, representatives of the latter group are at odds

Reinhold on the Deduction of the Categories 209 about exactly what that principle is. While the most promising candidate seems to be the principle of apperception, which Kant puts forth in the TD, there is no agreement on the general version of this principle that ought to be employed, i.e., the synthetic or the analytic principle of apperception.8 Where does Reinhold fit in? He belongs to the first group. He does not think that Kant provides a completeness proof for the table of categories. He puts forth this view in various ways. He does so indirectly when he asserts that ‘the Critique of Pure Reason has not established a first principle of all philosophy, … nor has it established one for the theory of sensibility, nor one for the theory of the understanding’ (Beiträge 273). For Reinhold, without a first principle or a principle for the theory of the understanding, it is impossible to provide a completeness proof for the table of categories. In a further quotation, he makes this point directly: The completeness of these forms [i.e., the forms of judgement] themselves must be proved. It must be shown that only the four indicated moments (of quantity, quality, relation, modality) and, in each of these moments, that only three forms of judgement no more and no less are possible. To my knowledge, this has not happened in the Critique of Reason. (Beiträge 315) After clearly explaining what a completeness proof is, Reinhold simply states that he does not believe that Kant has provided one for his table of categories. This brings us to the second question that has been a cause of contention: is Kant’s MD grounded in a vicious circle? The charge of circularity today is roughly that although Kant grounds the table of categories in the table of the judgements, these two tables actually presuppose one another. Thus, the latter cannot ground the former without vicious circularity.9 While Reinhold likewise raises a charge of circularity here, his charge is different from the one that is raised today. Instead of holding that the vicious circle is internal to the MD, he argues that it is internal to the TD. This poses a problem for the MD, according to Reinhold, because the MD is grounded in the TD, and, thus, the MD is grounded in a proof that is inherently and viciously circular. The MD presupposes the TD, in his view, because Kant must first establish that categories are concepts that are necessary for the possibility of cognising objects and that they originate in these very acts of cognising. At stake here is the kind of completeness proof that is appropriate given the nature of the categories. If the categories simply denote substances and qualities (qua things in themselves), as we see in the Aristotelian tradition,

210  Elise Frketich then the suitable completeness proof would be sought, e.g. by asking questions about these substances like “what is it?” (answer: substance) and “where is it?” (answer: place), as J.L. Ackrill explains with respect to how Aristotle establishes his list of categories (1963, 78–79). However, Kant rejects this tradition. He does not hold that the categories denote substances and qualities qua things in themselves. Instead, he holds that they directly signify substances and qualities qua appearances and indirectly signify the fundamental acts of cognising of the understanding that determined them. For this reason a completeness proof for his table of categories must be sought by exhausting these acts of the understanding or, as he refers to them in the MD, the ‘functions’ of the understanding.10 Reinhold puts this point in terms of where critical philosophers versus ‘knowers of things in themselves’11 locate what he calls the ‘objective ground of the connection of external intuitions’: [The knowers of things in themselves] deny neither the necessity nor the priority of this connection [i.e., the subjective ground of the connection of external intuitions] insofar as it is determined in mere representations. But they deny that the objective ground of the a priori determined connection is to be found anywhere else than in things in themselves. They deny that this connection would occur between representations and be necessary if it did not occur in things in themselves and was not necessarily in them. They claim that the laws of connection of representations in the representing subject are indeed determined a priori, but that they are determined in this subject in this way only because these laws, and no other ones, are grounded in laws in things in themselves. (Beiträge 310) As Reinhold points out, the knowers of things in themselves locate the objective ground of the connection of empirical intuitions, unsurprisingly, in things in themselves. That is, they hold that our fundamental philosophical concepts, e.g. of a substance connected with its accidents, denote things in themselves along with the actual attributes that they have or the actual causal relations in which they stand. For this reason, a table of categories must be completed by looking to these things and the connections between them, as we saw with the example of Aristotle, or by looking to innate concepts in the human mind that denote things in themselves, as we see with, e.g. some Leibnizians. To counter this line of argumentation, Kant needs to show that the rules of connection contained in the categories must have their objective ground in the mind, that is, in acts of synthesis of the understanding. This, in turn, implies that the categories must be sought by exhausting these acts.

Reinhold on the Deduction of the Categories 211 While it might be said that Kant already does this in the MD when he claims that both the forms of judgements and the categories originate in functions of the understanding,12 this is a minority view in the Kant scholarship, and Reinhold thinks this is a mere stipulation.13 Since this rules out the MD as a source for such an argument, on his account we are left with the TD, in which Kant attempts to establish that the categories of the understanding are necessary for the possibility of experience. Thus, For Reinhold the success of the MD hinges on that of the TD. However, Reinhold argues that the TD ultimately fails to persuade those who are not already convinced that we need to engage in Kant’s epistemological revolution or that Kant puts forth the correct model of human cognition in the Critique of Pure Reason because, in his view, the TD is circular. According to him, the point from which Kant begins the TD is human experience. This is a common interpretation in the Kant scholarship today. However, the prominent view among scholars today is that Kant begins with a thin account of human experience, one that we must all accept, namely the fact that we as human beings have experience.14 By contrast, in Reinhold’s view, Kant begins with a thick account of human experience, namely ‘the representation of the necessary connection of objects of sensible perceptions’ (Beiträge 286). This account already relies on the categories to explain how empirical objects in perception can be represented as substances that have accidents and that are related to other substances within a necessary connection. Reinhold does not here want to make a claim about the conception of experience to which Kant intends to refer. His argument is rather that Kant cannot prove everything he hopes to prove in the TD without assuming a thick account of experience. Thus, if Reinhold is correct, then the starting point of the TD presupposes what Kant sets out to prove, namely that experience requires that objects in intuition be thought through categories that have their objective ground first and foremost in acts of the understanding. Assessing whether Reinhold is right about the starting point of Kant’s TD is beyond the scope of this chapter.15 It suffices to say that this is not a settled issue in the Kant scholarship.16 Furthermore, Reinhold holds that the first Critique does not have the resources to provide the requisite arguments for this starting point: ‘the concept of experience, insofar as it is the basis of the Kantian system, … cannot be explained through this system without [vicious] circularity’ (Beiträge 281). For this reason, and because Reinhold is so thoroughly convinced of Kant’s critical philosophy on a whole, he aims to win over its opponents with his Elementarphilosophie.

212  Elise Frketich 9.3  Reinhold’s Transcendental Deduction To successfully show that the categories are necessary for the possibility of experience without a vicious circularity, Reinhold’s strategy in his TD is first and foremost to show that they originate in the mind. Once he has shown this, he can pinpoint their origin in functions of the understanding. Reinhold calls the unity produced by these functions ‘objective unity’ (Versuch 443). For him, ‘objective unity’ co-signifies an object in intuition and the concept through which we cognise this object (Versuch 430). Since objective unity is produced by a function of the understanding, which, as we will see, is a kind of act of synthesis of the understanding for Reinhold, its objective ground is such an act of synthesis. Kant explains objective unity in somewhat different terms. According to him, objective unity occurs when intuitions are related to the transcendental apperception.17 However, to my mind, Reinhold’s explanation is in line with Kant’s because, as we know from Kant’s TD, intuitions can only be related to the transcendental apperception when they are synthesised according to the pure concepts of the understanding. While the knowers of things in themselves agree that we require the concept of an object to cognise objects, they do not agree that the objective ground of this concept is an act of the understanding instead of things in themselves. Thus, Reinhold’s task is to prove them wrong. Only by showing that the objective ground of objective unity is an act of synthesis of the understanding and that categories are modifications of objective unity can Reinhold prove to the knowers of things in themselves that the categories must be sought in functions of the understanding and not in things in themselves, thus, paving the way for his MD. Objective unity, in Reinhold’s TD, is founded on a single and what he takes to be self-evident first principle. This principle neither expresses nor presupposes a thick conception of experience. Thus, it enables him, in his view, to avoid the charge of vicious circularity that he raises against Kant’s TD. Reinhold calls his first principle the ‘principle of consciousness’ (Satz des Bewußtseins), and he expresses it as follows: ‘in consciousness, the representation is distinguished from the subject and the object and is related to both by the subject’.18 He repeats this principle in various discussions of his system of philosophy, sometimes with ‘by the subject’ and sometimes without it.19 However, on my interpretation, all iterations of it amount to the same thing, which is to say that every representation is identical neither to the object nor to the subject, is about something, is represented by something, and occurs in consciousness.20 Although this principle is intuitively correct, and although Reinhold repeatedly asserts that it is selfevident, his contemporaries, e.g. Solomon Maimon, G.E. Schulze, and J.G.

Reinhold on the Deduction of the Categories 213 Fichte, reject the latter claim.21 Nevertheless, some German idealists, e.g. Fichte, agree that it accurately expresses our consciousness of a representation and integrate it into their own “critical” systems of philosophy.22 After putting forth the principle of consciousness, Reinhold indirectly derives the table of categories from it.23 This derivation begins with him unpacking what he calls the “concept of representation” directly from the principle of consciousness. As we will see shortly, he then derives objective unity and provides a completeness proof for the table of categories from this concept. Since each step builds upon the previous one without presupposing what is only proved at a later stage, Reinhold takes himself to avoid a vicious circularity. However, since this derivation ultimately rises and falls with the principle of consciousness, in the end, although it inspired friends of critical philosophy to further its ends in their own ways, it was no more successful at convincing those who oppose critical philosophy than were Kant’s own MD and TD. Nevertheless, it is instructive to look at Reinhold’s derivation more closely. To unpack the concept of representation from the principle of consciousness, Reinhold focuses on the various relations indicated in this principle and asks himself how these relations are possible, i.e., how they can come about. The first relation that he takes up is ‘the representation is related to the object’, which, as we saw above, is just to say that an actual representation is about something, e.g. my current representation of a dog on the street.24 He attempts to explain this relation by asking himself how a representation can be about something. His answer is that there must be something in the representation that refers it to the object (Beiträge 182). This something is the representative of the object in the representation, acting as evidence that the representation has been related to the object and making it possible that the representation be about the object, i.e., making it possible that my current representation is of a dog. Reinhold calls this representative of the object in the representation ‘matter’ (Stoff) (Beiträge 182).25 Now his response here may seem ad hoc. However, it allows him to explain how part of what makes up a sensible representation is given from without, and, for this reason, how a representation can be about something that exists independently of it. What this ‘matter’ is, of course, changes with each type of representation, as we will see shortly. However, using the same word to denote the information that we are given allows Reinhold to track it through the stages of mental processing that it undergoes. And by integrating the term “matter” into his derivation of the concept of a representation, he gives it a philosophical definition that is unique to the Elementarphilosophie. Once Reinhold has explained how a representation can be related to the object, he does the same with the subject. That is, he asks himself how the representation can be related to the subject. His answer is that there must

214  Elise Frketich be something in the representation that refers it to the subject. In other words, just as something in the representation acts as a representative of the object, so too does something in the representation act as a representative of the subject. This representative, according to Reinhold, is what he calls ‘form’ (Beiträge 183). Like the matter in a representation, what the form actually is changes whether the representation is in sensibility, the understanding, or reason. However, in general, we can understand it as being responsible for bringing the matter of a representation into a format such that it can be represented by the subject, that is, it ensures that matter adopts the form of a unity. Reinhold continues to unpack the concept of representation by building upon what we have seen thus far. He does so by next asking how matter can be contributed to the representation. His answer is that it must be given to the representing subject, that is, that the subject must be affected from without (Beiträge 189). It is easy for us to accept that matter in sensibility, i.e., sensations, is given to us from without. However, this claim becomes more difficult to accept when we recall that the representation stands for an intuition in sensibility, a concept in the understanding, or an idea in reason. For, according to this claim, just as matter in sensibility is given from without, so too is matter in the understanding and in reason “given from without”. Although odd sounding, it just means that matter given to the understanding and to reason, respectively, originates in a different faculty than the one in question, i.e., matter given to the understanding originates in sensibility, and matter given to reason originates in the understanding. Next, Reinhold does the same for form, i.e., he asks how form can be contributed to the representation. His answer is that it must originate in the representing subject, and, since he claims that it must be opposed to given matter in the representing subject, it must be spontaneously produced by the same (Beiträge 189). The “representing subject” stands for any of the three faculties of the human mind. It follows from this that, for Reinhold, sensibility also spontaneously produces the form of intuition. Yet, this might be difficult for a Kantian to accept as, for Kant, sensibility is a passive faculty. However, for Reinhold, any formatting of given matter in a representation is an activity, which includes formatting given sensations according to space and time in sensibility. Therefore, even sensibility, for Reinhold, has an active component. The representation can thus be viewed as what I call a kind of “hylomorphic compound”. It is a unity of given matter and produced form. I use this label to make clear that, for Reinhold, every representation must have the same structure, i.e., every intuition, concept, and idea must be a unity of given matter and produced form (Beiträge 177). Note, however, that this is the only connotation of “hylomorphic compound” that should be carried over into Reinhold’s

Reinhold on the Deduction of the Categories 215 concept of representation. Any further implications, e.g., that matter and form cannot exist independently of the compound, should not be applied to Reinhold’s representation. Although Reinhold ultimately fails to convince knowers of things in themselves to accept a critical account of human cognition, there are advantages to his line of argumentation. For example, it makes clear that if one accepts a kind of representationalist theory of human cognition, which knowers of things in themselves did,26 then there is good reason to view a mental entity as a kind of hylomorphic compound, and this still benefits the critical philosophical project because it advocates the “two stems of cognition” tenet of Kant’s critical philosophy in a way that is harder for its opponents to deny. Furthermore, Reinhold takes it to avoid a vicious circularity, and, for this reason, in his view, is an improvement upon the line of argumentation in Kant’s TD. Now that we have a clear idea of what Reinhold views as the concept of representation, we can move onto its species, most importantly for our purposes here: intuition and concept. He does so by developing what he calls the ‘types of consciousness’ in his theory of the mind.27 Each type of consciousness signifies having a representation that is one step further removed from the object that affects our senses. In the first type of consciousness, a representation is directly related to a mind-independent object, i.e., the thing in itself (Beiträge 185). Reinhold, unsurprisingly, calls this representation ‘intuition’ (Beiträge 233−235). Since any intuition is a representation, for Reinhold, it too must be a hylomorphic compound made up of given matter and produced form. The given matter of an intuition originates in affection from without. That is, we are affected by the thing in itself, and sensibility receives these affections. The result is given matter in sensibility (Beiträge 237). While this explains one half of the hylomorphic compound in question, we must now turn to the second half, namely produced form. For Reinhold, any produced form of any representation is responsible for bringing a disparate manifold into the form of a representation, which he thinks always means bringing some kind of unity into some kind of manifold. In the case of intuition, the given matter is the manifold, and the produced form is space and time (Beiträge 295−296). Here, of course, Reinhold follows Kant in viewing space and time as a priori forms of intuition. While it seems correct to say that, for Kant, these forms are produced by the thinking subject, without altering the passive status of sensibility on a whole, Kant does not emphasise this point to the extent that Reinhold does.28 So much for the first type of consciousness, in which the representation is directly related to the object that affects our senses. Now we can turn to the second type, in which the representation is one step removed from this object. This is a representation that is brought forth by the understanding,

216  Elise Frketich otherwise known as a concept (which is produced by a judgement, as we will see shortly). Since any concept is a representation, for Reinhold, it too must be a hylomorphic compound. The object to which the representation is referred in the understanding gives us the matter of this representation, and it is no longer the object that affects our senses, but rather an intuition. That is, sensibility contributes a spatiotemporally ordered manifold to a representation in the understanding, and the understanding spontaneously produces its form. This spontaneously produced form is the unity of the manifold in intuition, which Reinhold claims produces a representation that he calls ‘objective unity’.29 This representation is made up of a matter component—a manifold of intuition that has been unified into an object— and a form component: the concept “objective unity”. With the representation of objective unity, we arrive at the critical moment in Reinhold’s TD. Here, we must look closely at how, for Reinhold, a representation of objective unity is produced. According to him, since it must be spontaneously produced by the representing subject, as we just saw, it is produced by a function of the understanding. However, we saw that there are two main ways to understand a function of the understanding in the Kant scholarship, namely as an action or as a concept, i.e., the product of an action. How are we to understand what Reinhold means by this term? Reinhold unambiguously calls a function of the understanding an action. However, he also clearly explains the relation between this action and a concept, which, to my mind, helps shed light on why it is difficult to clearly distinguish the one from the other in Kant’s critical philosophy. For Reinhold, the manifold of intuitions is unified by a spontaneous act of the understanding (Versuch 431). This spontaneous act is a synthetic judgement, and he calls this act a ‘function’ (Versuch 443). This initial synthetic act or judgement, for Reinhold, explains how there can even be a unified manifold in intuition, i.e., a perceived object.30 However, this does not yet explain how this object is cognised as object. Cognising this object as object requires both the predicate “objective unity” and a further act of the understanding, in which this object in intuition is presented as the subject of a judgement and “objective unity” is predicated of it. For Reinhold, the initial synthetic judgement unifies the manifold of intuition into an objective unity and, at the same time, produces the concept “objective unity”, which can then be used as a predicate (Versuch 432). The unity contained in this concept directly signifies the objective unity in intuition and indirectly signifies the act/function of the understanding that produced it (Versuch 443). And the second act that predicates this concept of the subject, such that the object in intuition can be thought as object, is, for Reinhold, an analytic judgement (Versuch 442). These two steps, according to him, show first that a function (or synthesis) that originates in the mind produces both an objective unity in intuition and the predicate “objective unity” and second how this unity in intuition is cognised as an object.

Reinhold on the Deduction of the Categories 217 Reinhold calls objective unity the ‘concept in general’ as well as the ‘concept a priori’ (Versuch 434). All categories that make up the table of categories are modifications of this general and a priori concept, i.e., they are its specific determinations (Versuch 440–441). Thus, they likewise indirectly signify the functions, i.e., synthetic judgements, of the understanding that produce them, and they are likewise predicated of objects in intuition in analytic judgements. As we have seen, “objective unity”, for Reinhold, directly co-signifies an objective unity in intuition and the concept through which we cognise this object, and it indirectly signifies the synthetic a priori judgement of the understanding that produced them. By showing that objective unity originates in an act of the understanding, and by explaining that the categories are modifications of it, Reinhold takes himself to have systematically proved that the categories must be sought in the functions of the understanding and that these categories are necessary for the possibility of any type of cognition, including experience. That is, he has provided an argument for this view against knowers of things in themselves where he thinks that Kant has not, and he takes himself to have done so without a vicious circularity. This is one respect in which Reinhold views his TD as making an important contribution to the critical philosophical project. 9.4  Reinhold’s Metaphysical Deduction As we saw with Kant, for Reinhold, a completeness proof for the table of categories must be guided by a principle. The principle guiding Reinhold’s proof is the principle of consciousness. As we saw in the previous section, Reinhold takes himself to have successfully established, on the basis of this principle, that objective unity originates in an act of synthesis of the understanding and is necessary for the possibility of experience, and all of this without a vicious circularity. Since he takes the categories to be modifications of objective unity, they likewise originate in functions of the understanding and are necessary for the possibility of experience. For this reason, the completeness proof for the table of categories can now be provided by exhausting these functions and by subsuming them under objective unity in his MD, and this without concern that additional categories could be sought elsewhere (i.e., in things in themselves). Reinhold derives the table of categories from objective unity via the table of judgements, as he explains in the following passage: The modifications of the most general form of judgement determined by the nature of the understanding or the combination in objective unity are particular logical forms of judgement, and just as the most

218  Elise Frketich general form of the conception of objects results from the most general form of judgement, so too are there just as many particular forms in which to conceive objects determined in the nature of the understanding or just as many attributes of objects in general, which are dependent on the understanding resulting from the particular logical forms of judgement. We must then seek out the logical forms of judgement. (Versuch 441−442, translation slightly modified) Here, Reinhold explains that the most basic function of the understanding produces the most general concept of the same, that is, as we have seen, the concept of objective unity. Just as the most basic function of the understanding produces the most general concept, so too do its possible variations produce the possible determinations of pure concepts, i.e., the categories. Thus, in line with Kant, Reinhold claims that we must discover the table of judgements by exhausting the functions of the understanding. We can then generate the table of categories based on these functions, namely by matching a category to the function that it signifies. For these reasons, we must begin by listing the possible modifications of objective unity. To do so, Reinhold states that we must look to the syntax of any judgement “S is P” (in line with Kant).31 This move should not be controversial because, during Reinhold’s time, thinkers agreed that the understanding is the faculty of judgements and that judgements have the form “S is P”. Regarding this form, we can first ask ourselves: what are the possible variations of the subject term? The only possible variations are that the subject can be one, some, or all individuals (Versuch 444). This exhausts the possible variations of the subject term in a judgement and results in the judgements of quantity, which are singular (where singular can only mean one), particular (where particular means some), and general judgements (which means all subjects) (Versuch 444). The categories whose unities denote the forms of these judgements are the categories of quantity, namely unity, multiplicity, and universality (Versuch 449). Next Reinhold turns to the predicate term and likewise exhausts its possible variations, i.e., the ways that it can be related to the subject. The predicate can be said of the subject, not said of the subject, or, finally, both said of the subject in one way and not said of it in another (Versuch 444−445). This exhausts the possible variations of the predicate term and results in the judgements of quality, namely affirmatory, negatory, and indeterminate (Versuch 449). When we affirm a predicate of a subject in a judgement, we say something real about that subject. When we negate a predicate of a subject, then we say something about what the subject is not. Finally, when we affirm a predicate of the subject in one respect and negate it of the same in

Reinhold on the Deduction of the Categories 219 another, then we judge that the predicate holds of the subject in a qualified way. In these ways, the judgements of quality give us the categories of the same, namely reality, negation, and limitation (Versuch 449). Continuing our analysis of the syntax of a judgement, we can now look at the copula that connects the subject and the predicate. In Reinhold’s terms, it can connect them ‘internally’, ‘externally’, or both (Versuch 445). In the first case, the subject and the predicate form one object and the latter is bound to the former in a categorical judgement (Versuch 445). In the second case, the subject and the predicate are separate objects and the former grounds the latter in a hypothetical judgement (Versuch 445). Finally, the predicate is bound to various subjects and not to others and the judgement is disjunctive (Versuch 446). These judgements of relation produce the categories of the same, namely substantiality, causality, and concurrence (Versuch 449). In the first case, the predicate is an accident of the subject, which, in turn, denotes a substance (Versuch 453). In the second case, the predicate denotes an effect of what the subject denotes (Versuch 454). Finally, in the third case, the subject denotes a group of objects which is thought through the predicate as a community (Versuch 454). Finally, Reinhold must look at the relation of consciousness to a judgement. According to Reinhold, and following Kant, this relation is likewise internal, external, or both (Versuch 446). In the first case, consciousness is directly related to the representation, the act of combining occurs in the object itself, and the judgement is assertory (Versuch 446).32 In the second case, consciousness is not directly related to the representation, the act of combining occurs in the form of thinking alone, and the judgement is problematic (Versuch 447). Finally, consciousness is both internally and externally related to the representation, the act of combination in the form of thinking determines the act of combination in the representation in a judgement, and the judgement is apodictic (Versuch 447). This produces the final section of the table of categories, namely those of modality. When consciousness relates internally to the representation in a judgement, then the combination thought therein goes through the channels proper to cognition, and the category of actuality is produced (Versuch 455). When consciousness relates externally to the representation in a judgement, then the combination thought therein does not go through the channels proper to cognition, but rather goes through those proper to the form of thinking alone, and the judgement is merely thinkable (Versuch 455). This produces the category of possibility. Finally, when consciousness is related both externally and internally to the judgement, that is, when thinkability determines its reality, then the category produced is necessity (Versuch 455). This was Reinhold’s deduction of the categories from start to finish, i.e., from the principle of consciousness in his TD to his MD.

220  Elise Frketich 9.5 Conclusion Reinhold raises important issues regarding Kant’s deductions of the categories, issues that are still debated today. These issues arise when one searches the Critique of Pure Reason for answers to such questions as: does Kant provide a completeness proof for his tables of judgements and categories, and is the MD grounded in a vicious circle? In this chapter, I have presented Reinhold’s answers to these questions, as well as his solutions to the issues that he sees in Kant’s own deductions. First, Reinhold begins by making explicit his own principle of consciousness, which, for him, is the self-evident and first principle of philosophy. Second, on my account, he reverses the order of his MD and TD. In contrast to Kant’s order of presentation, Reinhold first aims to show that the categories both originate in the mind and not in experience and are necessary for experience. He does so in an attempt to convince those who oppose critical philosophy that they must seek the categories in the actions of the understanding and not in things in themselves. Only in this way can the MD work. That is, only once this has been established does it make sense to complete the table of categories based on an exhaustive list of the most fundamental acts of synthesis of the understanding.33 Notes 1 In his words, ‘the Critique of Reason has delivered all the data necessary to solve the greatest problems’, by which he means that the first Critique, as the propaedeutic to a philosophical system, has identified the a priori forms of the human mind, namely space and time, the categories, and the ideas of reason, and these forms must then be used to determine the principles of logic, metaphysics, and morality in a scientific philosophical system (Beiträge 263−264). 2 Although Reinhold developed three major systems of philosophy over the course of his career, I will only treat his first major system here, which he calls the “Elementarphilosophie” (1789−1794). In it, he restructures Kant’s critical philosophy by deriving its concepts and principles from a first principle according to the axiomatic-deductive method (i.e., the method described in Aristotle’s Posterior Analytics and employed in Euclid’s Elements). The name “Elementarphilosophie” refers to a system of philosophy, not to a specific text, and Reinhold put forth a few iterations of it in the works that he published between 1789 and 1794. In this chapter, I will focus on the two of these works that are most relevant to the topic at hand, namely Essay on a New Theory of the Human Capacity for Representation (1789) (hereafter: Versuch) and Contributions toward Correcting the Previous Misunderstandings of Philosophers, Volume 1 (1790) (hereafter: Beiträge). All translations of untranslated texts are mine. 3 For example, in the Metaphysical Foundations of Natural Science, Kant states that ‘the schema for the completeness of a metaphysical system, whether of nature in general, or of corporeal nature in particular, is the table of categories’ (MAN AA IV 473). He then proceeds to determine the concept of matter according to the four types of categories, namely quantity, quality, relation, and modality.

Reinhold on the Deduction of the Categories 221 4 However, Reinhold does not believe that Kant aimed at convincing those who oppose critical philosophy in the Critique of Pure Reason. Rather, he thinks that Kant’s aim was to investigate the possibility of metaphysics: Kant ‘did not have the intention to deliver the science of the capacity for representation or the science of the capacity for cognition in his work. The intention of this work, which is already indicated determinately enough by the title, is to investigate what is possible through pure reason, and particularly to investigate the possibility of metaphysics as an alleged science of the supersensible’ (Beiträge 274). 5 There are two approaches to Reinhold’s deductions of the categories in the current Reinhold scholarship: (i) to claim that Reinhold collapses the MD and TD into one deduction (e.g. Fabbianelli 2016: 89), and (ii) to focus solely on the MD that Reinhold provides in the Versuch (e.g. Onnasch 2006: 94−101). Against (i), I hold that the MD and TD are clearly distinguishable in the Elementarphilosophie, as I show in Sections 9.3 and 9.4, and I build upon (ii) by showing that Reinhold’s MD, which is presented only in his Versuch, must be grounded in his TD, which is only presented in the Beiträge. The former must be grounded in the latter because Reinhold does not yet present the principle that he puts forth as the first principle of all philosophy in the Versuch. In this work, Reinhold merely puts forth a preliminary version of the principle of consciousness (see Bondeli 1995: 56−58; Fabbianelli 2003: xix). He only asserts his first principle in his next published work, namely the Beiträge. Since this principle bears the weight of grounding any and all systematicity in the Elementarphilosophie, I rely on both texts. 6 See, e.g. Longuenesse 2005: 100; Brandt 1991: 85−86; Allison 1983: 128; Krüger 1968: 333−337. 7 See, e.g. Schulting 2018: 1; Wolff 1995; and Reich 1992: 5−6. 8 For example, Klaus Reich holds that the MD follows from the synthetic principle of apperception (1992: 45−46), and Dennis Schulting holds that it follows from the analytic principle of the same (2018: 6). 9 Henry Allison, for example, charges Kant with circularity in this sense (1983: 128). By contrast, Dennis Schulting resists it (2018: 7). 10 See: ‘If we abstract from all content of a judgment in general, and attend only to the mere form of the understanding in it, we find that the function of thinking in it can be brought under four titles, each of which contains under itself three moments’ (A70−B95, translation slightly modified). Kant then again appeals to this function in the transition from the table of judgements to the table of categories: ‘The same function that gives unity to the different representations in a judgment also gives unity to the mere synthesis of different representations in an intuition, which, expressed generally, is called the pure concept of understanding’ (A79/B104−105). What Kant means by “function” is not a settled issue in the Kant scholarship. Some hold that it is another name for concept, that is, the result of an activity (e.g. Guyer 2010: 125; 127; Henrich 1976: 21−22). Others hold that it is the action that brings forth a concept or a judgement (e.g. Schulting 2018: 121; Hoeppner 2011: 196; Allison 1983: 123). I agree with the latter group. 11 For example, Leibniz, Wolff, and Eberhard; see Beiträge 286−287; 310; 312. 12 Or it might be said that Kant already takes steps towards establishing this stance in the Transcendental Aesthetic when he argues that things in themselves cannot appear in space and time; he here only attempts to rule out the possibility of things in themselves appearing in space and time, he does not attempt to rule out that we can cognise or think them through the categories

222  Elise Frketich (A26/B42). Furthermore, Reinhold thinks that more arguments are required than those provided in the Transcendental Aesthetic to prove that things in themselves cannot appear in space and time (Beiträge 296−297). 13 Longuenesse makes this point when she asks ‘But why should there be syntheses of parts and aspects of objects presented to our sensibility? Why should it not be the case that empirically given objects just do present themselves as spatiotemporal, qualitatively determined wholes that have their own presented boundaries? Kant does not really justify this point in section three of the Leitfaden chapter’ (Longuenesse 2005: 101). See further Guyer 2010: 128 and Horstmann 2018: 15. 14 See, e.g. Pereboom 1995: 2. 15 Martin Bondeli argues that Reinhold has a point, that the actuality of synthetic a priori judgements is grounded on a vicious circle between Kant’s principles of the pure understanding and his conception of experience (1995: 71). Manfred Baum, by contrast, argues that Reinhold erroneously reads what he views as the correct method of philosophy, namely the axiomatic-deductive or ‘analytic’ method, into Kant’s TD (Baum 2004: 107). See Karin de Boer’s and Gesa Wellmann’s contribution to this volume for a new take on this latter reading. 16 Reinhold is not alone in thinking that Kant begins the TD with a thick conception of human experience. Contemporary Kant scholars who hold this view include, e.g. Ameriks (1978: 281), Allais (2015: 262−263), and Sommerlatte (2016: 450−457). 17 Kant explains objective unity in this way in the following passage: ‘this pure, original, unchanging consciousness I will now name transcendental apperception. That it deserves this name is already obvious from this, that even the purest objective unity, namely that of the a priori concepts (space and time) is possible only through the relation of the intuitions to it’ (A107). 18 Beiträge 167. For a presentation and discussion of the variations of Reinhold’s principle of consciousness from his Versuch through to the second volume of the Beiträge, published in 1794, see Bondeli 1995: 56−58. 19 Compare, e.g. Beiträge 144 with Beiträge 167. 20 My interpretation here is quite close to Frederick Beiser’s (1987: 253−254). When I say that the representation is “about” something, I mean that it refers to a mind-independent object, not that the represented object is identical to the mind-independent object, as is often thought today. 21 See further Frketich 2021: 67−69. 22 See further Frketich 2021: 69−70. 23 Although I do not detail the method that Reinhold uses in his “derivation”, Stefan Lang might oppose my use of “derive”, as he argues that not all concepts and principles that make up the Elementarphilosophie can be derived from what Reinhold views as the self-evident first principle of this system of philosophy (2018: 13−14). However, I think that this view misunderstands the method that Reinhold employs. Since all concepts and principles that make up Reinhold’s system of philosophy are either directly or indirectly determined by the first principle, we must view them as derived from it in the sense of Aristotle’s axiomatic-deductive method. 24 Recall that when I say that a representation is “about” something, I mean that it refers to a mind-independent object, not that the represented object is identical to the mind-independent object. 25 Reinhold follows Kant in employing the matter-form distinction. Kant applies this distinction to experience, which ‘contains two very heterogeneous elements, namely a matter for cognition from the senses and a certain form

Reinhold on the Deduction of the Categories 223 for ordering it from the inner source of pure intuiting and thinking, which, on the occasion of the former, are first brought into use and bring forth concepts’ (A86−B118). Reinhold, however, goes beyond Kant by establishing form and matter as essential elements of any representation. 26 For example, Baumgarten does not mince words about being a representationalist: ‘Thoughts are representations. Therefore my soul is a power for representing (§505)’ (Metaphysics §506). 27 Reinhold outlines the types of consciousness in his theory of the mind in the section he calls ‘theory of consciousness’ (Beiträge 218−223). He calls the first type clear consciousness and the second type distinct consciousness (Beiträge 221−222). Each of these types can then also have three possible objects in consciousness, i.e., the representation, the object, and the subject (as per the principle of consciousness) (Beiträge 221−222). I provide a highly simplified version of these types in the main text, one that ignores the distinctions Reinhold draws between clear and distinct types of consciousness, and I do so for the sake of readability. Reinhold also discusses cognition as the third type of consciousness. I do not frame cognition in these terms here. For a more detailed discussion of Reinhold’s presentation of the three types of consciousness, see Alfred Klemmt 1958: 83−91. Bondeli reads these types of consciousness in light of a ‘genetic deduction of consciousness’, i.e., the ‘becoming of consciousness’, through its different types (Bondeli 1995: 136). For Bondeli, this introduces elements of circularity and self-reference into Reinhold’s deduction (1995: 136−137). This reading is especially interesting in view of Reinhold’s influence on Fichte. However, I instead read it in light of Reinhold’s response to the German rationalists. While, for Reinhold, in the wake of Kant, the different objects of consciousness are different in kind, for the German rationalists, they are merely different in degree (gradually different from one another). 28 See Julian Wuerth 2014: 132, for more on Kant on space and time as actively produced in intuition. 29 On Reinhold’s interpretation, Kant first tries to prove that the categories are necessary for the possibility of experience in the TD and then, on this basis, that they are a priori. By contrast, Reinhold first attempts to prove that the categories originate in functions of the mind, i.e., they are a priori and then, on this basis, they are necessary for human cognition. See Beiträge 278−279. 30 Reinhold further explains that, for him, ‘The unity produced by the binding of the manifold represented by intuition is called objective unity, and is the most general form according to which the object (of an intuition) is thought’ (Versuch 429). 31 Here I provide an account of the main moves of what I call Reinhold’s MD. For a more detailed account, see Onnasch 2006: 108−113. 32 Note that Reinhold here first identifies the assertory judgement prior to the problematic judgement. This marks a departure from Kant’s own MD, as Kant first identifies the latter and then the former. According to Fabbianelli, this departure is to be traced back to Reinhold’s main goal with the Elementarphilosophie as compared to Kant’s with the Critique of Pure Reason. Whereas Reinhold, according to Fabbianelli, aims to provide a theory of representations in general, i.e., a theory of the marks that any representation must have to be considered a ‘real’ or well-formed representation, Kant aims to establish ‘the normativity which regulates the cognition of the world’ (Fabbianelli 2016: 78). While Fabbianelli explains this shift in focus in terms of Reinhold responding to sceptics, whose main position, as Fabbianelli clearly presents it, ‘is to doubt the agreement between the representation and the object and therefore [to think] that the things

224  Elise Frketich outside of the mind cannot be demonstrated in their existence’ (Fabbianelli 2016: 80), I explain it in terms of responding to the knowers of things in themselves. 33 I thank Robb Dunphy and Toby Lovat for inviting me to be a part of this project and for the insightful comments on this chapter. I also thank Karin de Boer and David Bartha for further helpful comments. This chapter was written within the context of a research fellowship at the ‘Human Abilities’ Centre for Advanced Studies in the Humanities (Berlin).

Bibliography Ackrill, J. L. (1963) Aristotle: Categories and De Interpretatione (Oxford: Clarendon Press) Allais, Lucy (2015) Manifest Reality: Kant’s Idealism and His Realism (Oxford: Oxford University Press) Allison, Henry (1983) Kant’s Transcendental Idealism (New Haven: Yale University Press) Ameriks, Karl (1978) “Kant’s Transcendental Deduction as a Regressive Argument” Kant-Studien 69(1), pp. 273–287 Baum, Manfred (2004) “Die Möglichkeit der Erfahrung und die analytische Methode bei Reinhold”. In Philosophie ohne Beynamen, M. Bondeli and A. Lazzari (Eds.) (Basel: Schwabe), pp.104–118 Baumgarten, Alexander (2013) Metaphysics, C. Fugate and J. Hymers (Eds. and trans.) (London: Bloomsbury) Beiser, Frederick (1987) The Fate of Reason: German Philosophy from Kant to Fichte (Cambridge, MA: Harvard University Press) Bondeli, Martin (1995) Das Anfangsproblem bei Karl Leonhard Reinhold; Eine systematische entwicklungsgeschichtliche Untersuchung zur Philosophie Reinholds in der Zeit von 1789 bis 1803 (Frankfurt: Klostermann) Brandt, Reinhard (1991) Die Urteilstafel: Kritik der reinen Vernunft A67–76; B92–101 (Hamburg: Felix Meiner Verlag) Fabbianelli, Faustino (2003) “Einleitung”. In K. L. Reinhold: Beiträge zur Berichtigung bisheriger Mißverständnisse der Philosophen (Hamburg: Meiner), pp. xi–xxxviii Fabbianelli, Faustino (2016) Karl Leonhard Reinhold’s Transcendental Psychology (Berlin: De Gruyter) Frketich, Elise (2021) “The First Principle of Philosophy in Fichte’s 1794 Aenesidemus Review”. The Enigma of Fichte’s First Principles, Fichte-Studien 49, D. Wood (Ed.), pp. 59–76 Guyer, Paul (2001) “Space, Time, and the Categories: The Project of the Transcendental Deduction”. In Idealismus als Theorie der Repräsentation? R. Schumacher (Ed.), 313–338 (Paderborn: Mentis) Guyer, Paul (2010) “The Deduction of the Categories: The Metaphysical and Transcendental Deductions”. In The Cambridge Companion to Kant’s Critique of Pure Reason, P. Guyer (Ed.), 118–150 (Cambridge: Cambridge University Press) Henrich, Dieter (1976) Identität und Objektivität: eine Untersuchung über Kants transzendentale Deduktion (Heidelberg: C. Winter) Hoeppner, Till (2011) “Kants Begriff der Funktion und die Vollständigkeit der Urteils- und Kategorientafel” Zeitschrift für philosophische Forschung 65(2), pp. 193–217

Reinhold on the Deduction of the Categories 225 Horstmann, Rolf-Peter (1997) “Die Funktion der metaphysischen Deduktion in Kants Kritik der reinen Vernunft”. In Bausteine kritischer Philosophie, R-P Horstmann (Ed.) (Bodenheim: Philo), pp. 55–78 Horstmann, Rolf-Peter (2018) “Kant on Imagination and Object Constitution”. In Kant on Freedom and Spontaneity, K. Moran (Ed.) (Cambridge: Cambridge University Press), pp. 13–29 Kant, Immanuel (1998) Critique of Pure Reason, P. Guyer and A. Wood (Eds. and trans.) (Cambridge: Cambridge University Press) Kant, Immanuel (1900–) Gesammelte Schriften. Bd. 1-22 Preussische Akademie der Wissenschaften, Bd. 23 Deutsche Akademie der Wissenschaften zu Berlin, ab Bd. 24 Akademie der Wissenschaften zu Göttingen (Eds.) (Berlin: De Gruyter) Klemmt, Alfred (1958) Karl Leonhard Reinholds Elementarphilosophie: Eine Studie über den Ursprung des spekulativen deutschen Idealismus (Hamburg: Meiner) Krüger, Lorenz (1968) “Wollte Kant die Vollständigkeit seiner Urteilstafel Beweisen?” Kant-Studien 59, pp. 333–356 Lang, Stefan (2021) “Karl Leonhard Reinholds Begriff der Deduktion” Archiv für Geschichte der Philosophie 103(3), pp. 531–561 Longuenesse, Beatrice (2005) Kant on the Human Standpoint (Cambridge: Cambridge University Press) Onnasch, Ernst-Otto (2006) “Vorüberlegungen zur Herleitung der Urteilsformen und Kategorien in Reinholds Theorie des Vorstellungsvermögens”. In Am Vorhof des Idealismus, P. Valenza (Ed.) (Pisa: Archivio di Filosofia), pp. 93–113 Pereboom, Derk (1995) “Self-understanding in Kant’s Transcendental Deduction” Synthese 103(1), pp. 1–42 Reich, Klaus (1992) The Completeness of Kant’s Table of Judgments (Stanford: Stanford University Press) Reinhold, Karl Leonhard (2003) Beiträge zur Berichtigung bisheriger Mißverständnisse der Philosophen, erster Band, F. Fabbianelli (Ed.) (Hamburg: Felix Meiner Verlag) Reinhold, Karl Leonhard (2011) Essay on a New Theory of the Human Capacity for Representation, T. Mehigan and B. Empson (Eds. and trans.) (Berlin: De Gruyter) Schulting, Dennis (2018) Kant’s Deduction from Apperception: An Essay on the Transcendental Deduction of the Categories (Berlin: De Gruyter) Sommerlatte, Curtis (2016) “Empirical Cognition in the Transcendental Deduction: Kant’s Starting Point and His Humean Problem” Kantian Review 21(3), pp. 437–463 Wolff, Michael (1995) Die Vollständigkeit der kantischen Urteilstafel: mit einem Essay über Freges Begriffsschrift (Frankfurt: Klostermann) Wuerth, Julian (2014) Kant on Mind, Action, and Ethics (Oxford: Oxford University Press)

10 Schulze’s Scepticism and the Rise and Rise of German Idealism Robb Dunphy

10.1 Introduction Widespread among both pre-Kantian German rationalists and postKantian German Idealists was the goal of developing a systematic, scientific metaphysics on the grounds of secure first principles. As Fichte points out, progress in the defence of projects of this kind was often prompted by sceptical objections levelled at the ‘precariousness’ of the grounds on which attempts at such systems were built (Fichte 2000: 137). Indeed, on occasion these sceptical objections were developed, or at least endorsed, by notable German Idealist themselves, at least when they were directed at the systems of their forbears and competitors.1 One of the most influential works of sceptical criticism in the period of post-Kantian German Idealism, however, was developed not by a competing systematic metaphysician, but by a self-professing sceptic: G.E. Schulze. The work was his Aenesidemus, or On the Foundations of the Elementary Philosophy Issued by Professor Reinhold in Jena, Along with a Defence of Scepticism Against the Pretensions of the Critique of Reason.2 The story of the impact of Schulze’s Aenesidemus on the development of post-Kantian idealism has been told several times, and although certain elements of the common interpretation of both Schulze’s criticisms and of their influence upon German Idealist philosophers, Fichte most notably,3 have been called into question in some recent scholarly work, my primary intention in this chapter is not so much to directly contribute to that discussion as it is to respond to a rather different case that has recently been made for the significance of Aenesidemus: that it is a mistake to consider Schulze’s work from a perspective which focuses on those of its arguments to which Fichte and others were moved to respond, because the real significance of the work lies in its rehabilitation of Pyrrhonian Scepticism in a manner that was entirely missed by post-Kantian idealists, and to which their subsequent systematic, metaphysical endeavours remain fundamentally vulnerable. These claims DOI: 10.4324/9781003187561-11

Schulze’s Scepticism and the Rise and Rise of German Idealism 227 have recently been defended by Jessica Berry, in a piece entitled “Sextan Scepticism and the Rise and Fall of German Idealism”. The title of this chapter, accordingly, illustrates the gist of the response that I will provide. While sympathetic to elements of Berry’s account, I will suggest that the attribution of a “Sextan” scepticism to Schulze remains to some extent questionable, and furthermore that it is not clear that such a scepticism is quite so problematic for the German Idealists as Berry suggests that it is. I shall proceed as follows: In Section 10.2, I provide a brief introduction to Schulze’s Aenesidemus and to recent scholarship on the work. In Section 10.3, I present the key elements of the interpretation developed by Berry in her article on Schulze’s scepticism. In Section 10.4, I argue that, for all that Aenesidemus presents itself at times as a reanimation of Pyrrhonian Scepticism, the case for reading Schulze as a Pyrrhonist is not as straightforward as Berry suggests. Finally, in Section 10.5, I argue that Schulze’s scepticism should not be understood to be as troubling for the German Idealists as Berry’s account suggests. In presenting this argument, I hope to go some way towards improving our understanding of the nature of Schulze’s criticisms of Kant and Reinhold, the nature of Schulze’s own scepticism and its relation to Pyrrhonism, and his significance in the development of German Idealism. 10.2 Introducing Aenesidemus Aenesidemus, as its title suggests, is intended to make a case for a certain scepticism against what the author takes to be the governing dogmatic positions of the time, i.e., those of Kant and Reinhold, just as the historical Aenesidemus is taken to have rebelled in the name of Pyrrhonian Scepticism against what he took to be the dogmatism ascendent in the Academy during his time.4 As the remainder of its title goes on to suggest, Aenesidemus is principally concerned with what Schulze takes to be the fundamental claims defended by Kant in his Critique of Pure Reason, and with Reinhold’s attempt to guarantee the scientific status of Kant’s conclusions by deriving them from a single, foundational principle in the “Elementary Philosophy” that he began developing in 1789 (and would continue to defend until his public adoption of Fichte’s idealism in 1798). The work takes the form of a series of letters, introduced by an anonymous editor, between Hermias, a recent devotee of Reinhold’s Elementary Philosophy, and the titular Aenesidemus, who attempts to convince the former that both Reinhold’s and Kant’s efforts remain problematically vulnerable to sceptical criticism. He makes this case both in his letters, and in a large appendix to the third letter dedicated to a more systematic examination and critique of Reinhold and Kant.

228  Robb Dunphy My intention in this section is to briefly acknowledge some of what are usually taken to be the central elements of both Schulze’s Reinhold critique and his Kant critique, before saying a few words about the dominant attitude towards Schulze in the secondary literature on Classical German Philosophy. The goal is not to provide a detailed interrogation of this material, but to prepare the ground for the examination of Berry’s more radical interpretation of Aenesidemus in the remainder of the chapter. The success of Reinhold’s supposedly scientific restatement of the fundamental claims of Kant’s critical philosophy depends, ultimately, on the case he makes for his “principle of consciousness” as the ‘first and universally accepted principle of all philosophy’ (Reinhold 2003: 3), from which a scientific treatment of representation in general can be derived which can then ground a science of metaphysics and other philosophical disciplines.5 Little wonder, then, that Schulze begins his account of Reinhold’s Elementary Philosophy with a sequence of criticisms all directed precisely at this principle (A 51−63). The principle, as Reinhold expresses it, states that ‘representation is distinguished in consciousness from the subject and from the object and is related to both’ (Reinhold 2003: 144). Schulze directs a number of distinct criticisms towards the idea that Reinhold’s principle of consciousness can serve as an irrefutable, selfevident, foundational principle for a scientific philosophical system. Some of these are rather less devastating that Schulze himself seems to think,6 but others may amount to genuine challenges to Reinhold’s principle. Schulze notably argues, for example, that Reinhold’s case for the principle of consciousness being universally accepted and self-explanatory looks suspect, since it is not clear what is involved in the activities of “distinguishing” the representation from and “relating” it to the subject and the object, to which the principle appeals (A 54−58). Perhaps even more problematically for Reinhold, Schulze goes on to argue that the attempt to ground the principle on the “fact” of conscious experience itself is to render the principle itself an abstraction, known a posteriori. Yet if it is known a posteriori, then it is difficult to see how Reinhold can hold it up as universal, certain, and necessary, as he seeks to (A 62−63). Aenesidemus contains criticisms not only of Reinhold’s putative fundamental principle, but also of various other aspects of his Elementary Philosophy, including his characterisations of the nature of representation and of self-consciousness. Schulze also appears to develop an early version of the criticism often directed towards modern transcendental arguments against scepticism. He suggests, for example, that Reinhold ‘only wants to show how we must think about representations, the faculty of representation, the origin of our cognitions, the constitution of their components, etc.’, but argues that such a project remains compatible with the possibility that ‘the entirety of our cognition is only an empty illusion, only a mere

Schulze’s Scepticism and the Rise and Rise of German Idealism 229 play of thoughts, from which many other perfections can come, but not reality and truth’ (A 159).7 Another distinctive criticism from Schulze is his suggestion that, since Reinhold’s own Elementary Philosophy holds that ‘the faculty of representation is the cause and ground of the actuality of representations’ (A 76), this faculty itself is therefore being treated as an object of a knowledge claim, the justification for which is not apparent. Since it is this faculty that Reinhold will treat first and then use as a foundation for treatments of other disciplines such as metaphysics, Schulze’s move is effectively to extend sceptical worries from ordinary, first-order knowledge claims to higher-order claims about the conditions of the possibility of ordinary knowledge. It is in this sense that Beiser will characterise Schulze’s scepticism as ‘meta-critical’ (Beiser 1987: 268).8 Much more could be said about the details of all of these criticisms of Reinhold, but they are not the primary focus of this chapter.9 Kant’s own critical philosophy, although it receives less explicit attention than Reinhold’s Elementary Philosophy, is also the target of Aenesidemus’ scepticism. Schulze takes aim at a fairly wide variety of Kant’s positions and arguments, and, just as with Reinhold earlier, I cannot address them in any detail here. One of the most prominent targets is Kant’s transcendental deduction, which seeks to demonstrate the legitimacy of the application of a priori concepts to objects of experience. Schulze takes the central purpose of Kant’s argument to be the refutation of Humean scepticism: the view that, since concepts such as causality are (arguably) not derived from empirical experience, their application to objects of experience is unjustified.10 On this basis, Schulze claims (rather questionably) that Kant begs the question against Hume by seeking to explain the legitimacy of the application of a priori concepts to experience simply by supposing that the mind, according to the organising role it plays in experience, itself is the cause of the fact that the ‘actual’ is ‘combined according to the law of causality’ (A 103). But, argues Schulze, this procedure’s reliance on the notion of causality itself is to ‘presuppose as already certain and agreed those principles against the reliability of which Hume directed all of his sceptical doubt’ (A 99).11 Schulze is also an early proponent of the criticism that Kant’s appeal to the thing in itself in order to explain what is ‘supposed to have delivered the materials of intuitions by way of our sensibility’ (A 184) is effectively to attribute the category of “cause” to the thing in itself, treating the content of sensible intuitions as its effects. If this is a fair reading, then it is clearly problematic since one of the primary conclusions of Kant’s transcendental deduction is that the fundamental a priori categories of the understanding, causality included, can be applied legitimately only to objects as they appear to us, and not to things in themselves (KRV B166). In fact, Schulze argues that Kant’s appeal to things in themselves as the

230  Robb Dunphy source of the content of our sensible intuitions is itself straightforwardly problematic even without any discussion of the application of the category of causality, simply because he takes this appeal to conflict with Kant’s position that ‘cognition of things in themselves … is everywhere impossible for us’ (A 181).12 As is the case with the sceptical criticisms that Schulze directs at Reinhold, a great deal more could be said about these criticisms and Kant’s ability to successfully answer them, but such an investigation is again not the primary purpose of this chapter. Although disagreements exist about several of the specifics of the arguments that Schulze levels at Kant and Reinhold in Aenesidemus,13 the general tendency among readers of Schulze has been to see him as offering interesting challenges which are either of use in bettering one’s understanding of Kant by coming to see how Schulze’s criticisms miss their mark14 or, where those criticisms are taken to be rather more serious, of use when it comes to understanding how responding to them effectively played a role in the development of German Idealism in the 1790s and into the 19th century.15 Jessica Berry has recently suggested that this tendency can lead to Schulze being read as part of a ‘vindicating history’ of German Idealism which, because the primary interest of its proponents is in charitably reconstructing the positions and arguments of the German Idealists, tends to result in a rather dismissive attitude towards Schulze and to the downplaying or even completely missing of the force of his sceptical challenges and the strength of his sceptical position (Berry 2020: 155−156). A fine example of such a dismissive attitude is provided in Kroner’s classic Von Kant bis Hegel, which states that it is only for the sake of [an understanding of] the historical circumstances of the development of Fichte’s views of Reinhold and the establishment of a fundamental principle for transcendental philosophy that Schulze’s broadly and shallowly written book is to be mentioned. (Kroner 1961: 325) Berry is certainly right to notice that there has been a general trend towards reading Schulze as a stepping stone on the way to the systems of the later German Idealists, and that varying degrees of dismissiveness tend to feature in discussions of the arguments of Aenesidemus in the secondary literature on Classical German Philosophy. Whether this has prevented readers from appreciating some more powerful sceptical criticisms, or a more attractive sceptical position in that work, remains to be seen, but Berry’s concern should not be dismissed out of hand. Another element that seems to be fairly consistent across most engagements with Aenesidemus is the opinion that Schulze, despite his choice of pseudonym, is perhaps less interested in reanimating positions that might

Schulze’s Scepticism and the Rise and Rise of German Idealism 231 be genuinely attributed to ancient Pyrrhonists, as represented primarily by Sextus Empiricus, than he is in countering what he sees as modern dogmatism with primarily modern, Humean sceptical resources.16 And there is no doubt that a defence of specifically Humean scepticism from the popular claim that Kant has refuted it takes up a significant proportion of Aenesidemus (A 84−129). There has, accordingly, been far more attention paid to Schulze’s defence of Hume than to his claim to endorse ‘the genuine character of the sceptical philosophy … that once bloomed in Greece’, or that he takes ‘the explanations of the epochē that Sextus provides as the essential characteristic of scepticism’ and sets out to ‘neither embellish nor worsen’ its ‘genuine spirit’ in his own presentation of scepticism (A 27).17 It is precisely by emphasising and seeking to remedy this lack of attention paid to Schulze’s endorsement of Pyrrhonian Scepticism that Berry has recently argued that the “vindicating history” of German Idealism is indeed guilty of downplaying or even completely missing both the genuine strength of the challenges Schulze provides and the attraction of the alternative, sceptical position that he develops. These are significant claims. They involve a far more sweeping revision of our attitude towards Aenesidemus than a disagreement about the specifics of any particular argument he directs against this or that Kantian or Reinholdian doctrine could motivate.18 Instead, Berry’s argument is that the most important aspects of the sceptical position developed in Aenesidemus are yet to be properly appreciated, and that these aspects in fact call into question the tendency to read Schulze as little more than a step on the way to the improved systematic metaphysical projects of later German Idealists. Indeed, she suggests that the scepticism of Aenesidemus even poses deep problems for such subsequent projects, to which it is not clear that Fichte or Hegel have answers. Because of the significance of these claims, the remainder of this chapter will be dedicated to assessing Berry’s case for Schulze’s unrecognised Pyrrhonism and its implications for German Idealism. 10.3  Berry’s Sextan Schulze In her recent article, Berry defends three distinctive claims on the basis of a reading of Aenesidemus, none of which are found in the standard accounts of the significance of that work addressed earlier. These claims are as follows: (1) Schulze’s work incorporates the endorsement of a distinctively Pyrrhonian Scepticism which went unappreciated by the German Idealists who responded to Aenesidemus and has continued to be neglected by more recent scholarship on the reception of Kant’s critical philosophy and the development of German Idealism; (2) Kant and the German Idealists are vulnerable to criticisms from this Pyrrhonian position, especially in terms of the uncritical attitude they take towards their metaphilosophical

232  Robb Dunphy commitment to the generation of systematically organised metaphysical knowledge, grounded in basic principles; and (3) Pyrrhonian Scepticism of the kind endorsed by Schulze amounts to a viable philosophical project in its own right, rather than merely a challenge for post-Kantian philosophers to overcome (Berry 2020: 154−155). It seems to me that each of these claims deserves serious consideration, since their truth or falsity will have serious implications for how we understand and evaluate Schulze’s work, and for how we evaluate the defence of the systematic metaphysical projects provided by the German Idealists who reacted to him. In this section, I summarise the key elements of Berry’s argument for each of the three claims just introduced. Then, in Section 10.4, I will provide some evaluative remarks on the first claim, with reflections on the second claim, and far briefer reflections on the third, following in Section 10.5. Berry’s first claim is that Schulze’s Aenesidemus, rather than being only a reinvigoration and restatement of Humean scepticism against Kant and Reinhold, also contains central elements which belong rather to the Pyrrhonian Sceptical tradition. I think that there are five major elements to Berry’s case for the presence of Pyrrhonism in Schulze’s text:19 (S1) Schulze’s criticism of Kant’s transcendental deduction is an application of arguments derived from the five modes of Agrippa.20 Specifically, the suggestion is that Kant begs the question against Hume by supposing that the mind plays a causal role in structuring experience, while the principle of causality was precisely what Hume called into question. This is not the same as merely restating the Humean case for scepticism (Berry 2020: 158−159).21 (S2) Schulze’s scepticism is reminiscent of Pyrrhonism in that he indicates that he has no ‘philosophical commitments’ of his own (Berry 2020: 159). Instead, his case against Kant and Reinhold makes use of their own rational commitments in order to undermine their position (Berry 2020: 165). Aenesidemus might therefore be said to be arguing dialectically, as Sextus presents the Pyrrhonists proceeding against the Dogmatists. (S3) Schulze’s scepticism is characterised in terms of a commitment to ongoing inquiry and the rejection of a “negative dogmatism” which declares knowledge to be impossible (Berry 2020: 159−160). These characteristics are explicitly those of Sextus’ account of Pyrrhonism, and are often appealed to distinguish Pyrrhonism from modern scepticisms like that of Hume, which declare (certain kinds of) knowledge impossible. (S4) Aenesidemus’ scepticism, like that of Sextus, is not based on the upholding of any epistemic or rational norms and finding knowledge claims to fall short. Instead, following Williams’ (2010: 299) characterisation

Schulze’s Scepticism and the Rise and Rise of German Idealism 233 of Sextus, Berry suggests that Aenesidemus is ‘not so much an epistemological sceptic as a sceptic about epistemology’; he has an entirely different, primarily psychological rather than rational approach to the matter of philosophical arguments (Berry 2020: 161). (S5) Aenesidemus’ scepticism is Pyrrhonian in that it can plausibly be understood not merely as a set of challenges to be overcome, but as a philosophical orientation that one can adopt and happily live with (Berry 2020: 167−169). Berry suggests that these elements of Schulze’s scepticism have largely been overlooked, both by his contemporaries such as Fichte and Hegel, and by recent scholarship on Classical German Philosophy. There are two principal explanations that she offers for this oversight on the part of readers of Schulze. Firstly, Berry points out that their attention has almost exclusively been focused on the lengthy appendix to Aenesidemus’ third letter to Hermias (Berry 2020: 157−158), while the more strongly Pyrrhonian elements of the text, she suggests, are to be found in the letters that frame this appendix. Secondly, she appeals to the existence of what she takes to be the widespread tendency of readers to approach Aenesidemus in the context of the “vindicating narrative” of the development of post-Kantian idealism which I discussed earlier. Berry’s suggestion is that the Pyrrhonian elements of Schulze’s text do not fit this vindicating narrative as well as the Humean elements, and thus tend to be passed over. I shall now introduce the other two of Berry’s central claims. The second claim is that, perhaps in part due to the fact that they fail to properly attend to the Pyrrhonian elements of Aenesidemus, later German Idealists in fact remain vulnerable to some of the criticisms put forward in that work, especially those concerning their central metaphilosophical commitments. What Berry finds in Aenesidemus, then, is not a hitherto unnoticed argument which would call into question the success of Fichte’s case for his ‘absolutely first foundational principle’ (Fichte 2021: 200) or the arguments by which Schelling or Hegel will later claim to develop forms of idealism without Kant’s supposedly problematic reliance upon the “thing in itself”, for example. Instead, the ‘Greek scepticism uniquely outfitted to cut to the very heart of German Idealism’ that Berry (2020: 155) finds in Aenesidemus is intended, she thinks, to challenge basic presuppositions that continued to go unquestioned by later post-Kantian idealists, especially the presupposition that the fundamental philosophical goal should be the development of an idealist ‘system, fully articulated and grounded absolutely in a single first principle’ (Berry 2020: 156), because achieving this goal, completing the dogmatic investigation, ‘the satisfaction of reason’s demands’, is the recipe for peace and well-being (Berry 2020: 164). What Schulze offers, in this context, Berry suggests, is ‘an insightful and

234  Robb Dunphy accurate diagnosis of an epistemic pathology’ (Berry 2020: 165). The suggestion is that both Kant and Reinhold, before Aenesidemus, but just as much Fichte or Hegel after, have unquestioningly supposed that there is an absolute value to satisfying the demands of reason, and that the enthusiasm with which they pursue their metaphysical system-building belies an equally unquestioned fear of the misery and immorality that they suppose would accompany a life without the accomplishment of scientific certainty.22 These unquestioned presuppositions led them to waste their time on a hopeless quest for absolute knowledge that Berry compares to Carroll’s “Hunting of the Snark” (Berry 2020: 155).23 Schulze does not share these presuppositions, on her reading, but this element of his dispute with Kant and Reinhold has been disregarded both by his immediate audience and by contemporary scholars in favour of concentrating on those arguments he offers which can be engaged with merely as interesting challenges entirely within the project of developing a truly scientific metaphysical system, without ever calling the value of that project itself into question (Berry 2020: 162). They thus miss and remain vulnerable precisely to Schulze’s deeper criticism of the assumptions structuring their projects. Berry’s third claim is that the Pyrrhonism she detects in Schulze’s work amounts to a viable and attractive philosophical orientation in its own right, quite distinct from the grandiose metaphysical system-building of the German Idealists. Such a scepticism is neither the “sickness of the age” nor merely a “healthy provocation” towards an improved dogmatism.24 Instead, Berry suggests that Schulze should be read as defending the viability of a sceptical way of life which is dedicated to ‘figuring out how to live well with uncertainty, and how to get along – how to flourish, even – in a “post-certainty” world’ (Berry 2020: 169), a task with which, presumably, the German Idealists are of no help.25 Having presented the three central claims Berry defends in her case for a Pyrrhonian reading of Schulze, the next thing to do is to provide some critical evaluation of them. The next section discusses strengths and limitations of Berry’s first claim: that distinctively Pyrrhonian elements are present within Aenesidemus, which have been neglected by its readers. 10.4  Schulze and Pyrrhonism I think that Berry’s case for recovering little-noticed but important Pyrrhonian elements in Schulze’s Aenesidemus is a valuable addition to scholarly work on that text. Questions remain, however, about the nature of Schulze’s understanding of Pyrrhonism and the extent of his endorsement of this kind of scepticism in Aenesidemus. In this section, I shall briefly run through the five elements addressed in the previous section in order to indicate the degree to which I think we should agree with Berry on this score.

Schulze’s Scepticism and the Rise and Rise of German Idealism 235 (S1), I think, should probably be dismissed when it comes to making the case for Schulze’s Pyrrhonism. The argument that Kant begs the question against Hume in his transcendental deduction is not enough to establish specifically Pyrrhonian argumentative credentials. The Pyrrhonists, after all, do not have a monopoly on the charge of begging the question or any of the other arguments which make up the Agrippan modes. The options making up the “Agrippan trilemma” of arbitrary hypothesis, reciprocity, and infinite regress appeared in Aristotle’s Posterior Analytics long before we suppose Agrippa to have been born, and a version of the charge that Kant’s deduction begs the question against Hume had been made before Schulze by Maimon, who does not present his own scepticism as a Pyrrhonian one.26 (S2) is an interesting case. The first example on which Berry focuses occurs when Schulze is suggesting that Reinhold and Kant dogmatically assume the reality of various cognitive faculties, the existence and nature of which is really doubtful. Schulze continues: But does the Sceptic deny, one will ask, the existence of sensibility, the understanding, and reason in humans? He is always appealing to grounds of the understanding or of reason in his disputes with dogmatism. How can he therefore deny or doubt their existence without refuting himself? (A 78) Schulze’s answer is that the Sceptic acknowledges that we have representations of such faculties, but that it is not clear that they have any genuine existence beyond our representations. Thus, he adds, ‘when the sceptic makes use of the words reason or understanding this is done merely with the intention of expressing themselves in a way that is commonly understood’ (A 79), i.e., according to the sense in which the sceptics’ interlocutors make use of the terms. This does indeed appear to recall elements of Sextus’ presentation of the Pyrrhonists’ argumentative procedure. Time and again Sextus argues that the Pyrrhonists carry out their characteristic philosophical investigation without themselves holding opinions or beliefs (e.g. PH I: 12, 15, 24, 206, 223, 226, 231) and he is commonly understood to frequently present the Sceptics as arguing by dialectically adopting positions and argumentative strategies which their dogmatic opponents endorse, rather than necessarily endorsing them themselves.27 Schulze appears to abstain from any theoretical commitments concerning the existence of various cognitive faculties, and to use the relevant terminology precisely to cause trouble for philosophers like Kant who rely heavily upon conceptions of such faculties. This sounds very much like a Pyrrhonian manoeuvre, although it occurs in one fairly specific place in the

236  Robb Dunphy text, where Schulze is developing what Beiser calls his ‘meta-critical’ scepticism, directed towards the account of cognitive faculties which forms the basis of Kantian critique.28 The question is whether this Pyrrhonian approach generalises across the work. I think, across the work in general, there are clear limits to the claim that Schulze has no theoretical commitments of his own and argues only dialectically. Even in the context of the passage just cited, for example, he holds that the ‘existence [of] and differences’ between distinct representations – intuitions, concepts, and ideas – is ‘a fact’ (A 78). What is more, in the third letter, which precedes the lengthy appendix on Reinhold and Kant, Schulze has Aenesidemus put forward two fundamental principles upon which all of his sceptical arguments rest, and which, he claims, no sceptic would ever doubt.29 These appear very much to be Schulze’s own theoretical commitments, not the positions of others, adopted for the purposes of dialectical argument. The second of these principles, in particular, amounts to little more than affirming the laws of traditional, general logic as ‘the touchstone of all truth’ (A 40). Schulze adds that he thinks that no Sceptic has really ever doubted ‘the certainty of the syllogistic’ (A 41). Yet this appears to be precisely to break with Sextus, who concludes his account of traditional syllogistic proofs in the Outlines of Pyrrhonism by saying that ‘we must suspend judgement about proofs’ in general (PH II: 192). The other of Berry’s examples of Aenesidemus arguing dialectically is his appeal to the ‘eternal and unchanging law of our reason, to hold nothing to be true without sufficient grounds’ (A 29). Berry suggests that this law is appealed to in order to show that ‘by the Idealists own lights’ they ought not to make any claim to have established universally valid principles with certainty (Berry 2020: 165). Yet there are two things to notice here. The first is that Schulze himself does not in fact seem to be suggesting that Aenesidemus is arguing dialectically here. Just as with his two fundamental principles mentioned earlier, Aenesidemus appears genuinely to be committed to this “eternal and unchanging law” himself. This suggests again that there are clear limits to the extent that Schulze can be taken to have no philosophical commitments of his own. The second thing, however, is that this is perhaps not to break with Pyrrhonian Scepticism at all. In Against the Logicians, for example, when Sextus is opposing the Dogmatists’ practice of arguing on the basis of hypotheses, he attributes to ‘our side’, that is, to the Sceptics, the view that ‘nothing ought to be assumed all by itself’ (AL II: 373). So here Schulze does indeed appear to be echoing Sextus, but in a way that appears to challenge the view that Berry has of both Schulze and Sextus, according to which they have no philosophical commitments of their own and argue only dialectically. (S3), likewise, certainly seems to capture something important and distinctive about Schulze’s scepticism, but I think that there are some

Schulze’s Scepticism and the Rise and Rise of German Idealism 237 limitations to the extent to which we can take this aspect of his argument as evidence for a genuine return to Pyrrhonism in Schulze’s thought. Berry is right to point out that Schulze distinguishes his scepticism from Kant’s outright rejection of the possibility of knowledge of things in themselves by claiming that ‘[f]rom the fact that human reason has not yet achieved something, despite all of its attempts and efforts, it cannot indeed be concluded with certainty that it will never be able to achieve it full stop’, and thus that ‘no mind that thinks for itself can legitimately hold off from striving after the cognition of things in themselves’ (A 112). This replacement of the denial of the possibility of knowledge with an attitude of suspension of judgement concerning its possibility, coupled with a project of openended inquiry, very much resembles the distinction Sextus draws between the Sceptics ‘who are still investigating’ and the negative dogmatism of those who ‘have asserted that things cannot be apprehended’ (PH I: 3). It is tempting to say that Schulze is to Kant (and Reinhold) as Sextus is to the Cyrenaics: ‘we suspend judgement … about externally existing things, while the Cyrenaics assert that they have an inapprehensible nature’ (PH I: 215). Schulze’s scepticism might therefore be thought to more closely resemble the scepticism of the ancients than modern scepticism, not least because modern scepticism is often understood precisely to involve something approximating negative dogmatism about some topic; the claim that the existence of the external world is unknowable, for example.30 And yet, in other respects, Schulze’s scepticism does look very modern. He goes so far as to define scepticism, not in terms of the Pyrrhonian practice of openended inquiry with an attitude of suspension of judgement, but rather in terms of ignorance of things in themselves, a decisively modern focus: Scepticism is nothing other than the claim that in philosophy nothing has been established according to indisputably certain and universal principles either concerning the existence and non-existence of things in themselves and their properties or concerning the boundaries of human cognitive powers. (A 26) For all that his approach at times recalls Pyrrhonism then, he seems focused solely on modern sceptical problems. And in fact, when he takes himself to endorse ‘the genuine character of the sceptical philosophy … that once bloomed in Greece’, he immediately adds that it is this character ‘that in modern times found such a sharp-witted defender in David Hume’ (A 27). He appears not to clearly acknowledge any important distinction between Pyrrhonian Scepticism and the scepticisms of modern philosophy.31 This suggests that there are clear limits to the extent that we can

238  Robb Dunphy suppose Schulze to be breaking with the philosophical concerns of his day and returning to a classically Pyrrhonian approach. (S4) – the claim that Schulze’s scepticism is Pyrrhonian precisely because it does not turn on upholding rational standards for the evaluation of knowledge claims and finding them wanting – is, I think, the most interesting aspect of Berry’s case for Schulze’s Pyrrhonism. It connects her reading of Schulze to one of the central debates concerning how to understand Sextus’ own Pyrrhonism. That debate, which I can only gesture towards here, concerns the nature of the Pyrrhonist’s appreciation of the equal strength of arguments for opposing claims, and their subsequent suspension of judgement on the basis of finding those arguments to be equally strong.32 While various positions have been taken up on this topic, I think that it is fair to say that they mostly lie between two poles. At one pole, the Sceptic is understood to be interested in rationally evaluating the strength of the arguments and evidence that they consider, and their subsequent suspension of judgement after finding the arguments in question to be equally strong is understood to be motivated by the Sceptic’s concern for believing, disbelieving, and suspending judgement only as they rationally ought to. The most noteworthy defence of this view is probably the one provided by Perin (2010: 33−58).33 At the other pole, the Sceptic is taken precisely not to have any of the normative commitments that the rational evaluation of arguments or the idea that one is rationally required to suspend judgement in the face of equally strong arguments involve. Instead, the Sceptic’s appreciation of the force of arguments is merely a matter of finding them psychologically persuasive, rather than evaluating their rational merits, and their subsequent suspension of judgement in the face of equally strong arguments is psychologically caused by the perception that the arguments in question are equally strong, rather than being something that the Sceptic does because suspending is what one ought to do in the face of equally strong arguments, if one is to be rational. Giving up traditional, rational norms, as Berry acknowledges, makes this position one of scepticism about epistemology, rather than epistemological scepticism. Williams (1988, 2010), whom Berry cites here, is a prominent advocate of this reading. This is not the place to attempt to decide what the stronger interpretation of Sextus is, but it is clearly important to see what Schulze’s position on this matter is. Exactly where Berry’s Pyrrhonian interpretation of Schulze stands on this topic is not completely clear to me. She explicitly endorses Williams’ rejection of readings of Sextus, which take sceptical inquiry to involve the upholding of rational norms in the evaluation of arguments and the attempt to satisfy the demands of reason and suggests that Schulze’s scepticism should be read in the same way (Berry 2020: 161−164). However, she also suggests that the Sceptic is an ‘honest player of the game of rational

Schulze’s Scepticism and the Rise and Rise of German Idealism 239 inquiry’ and cites Perin’s account in support of this claim. This seems to me to risk attaching incompatible accounts of Pyrrhonism to this interpretation of Schulze. Berry is clear, however, in rejecting Beiser’s (1987: 269−272) suggestion that Schulze’s scepticism is motivated by a demand that we maximally improve our own cognition, or strive to perfect our reasoning, taking this interpretation to uncritically attribute to him the very metaphilosophical assumptions of Kant and Reinhold that she takes him to criticise. Berry defends this rejection of the idea that Schulze’s scepticism turns on attempting to satisfy the demands of reason by noting that Beiser takes the claim that the Sceptic is committed to ‘the never-ceasing perfectibility of philosophising reason’ and ‘always strives for that even more perfect knowledge in philosophy’ (A 8−10) not from the letters of Aenesidemus, Schulze’s mouthpiece, but from the preface to the work, attributed to an anonymous “editor”. Berry thinks that we should not be so quick to attribute what the editor says to Schulze himself. I myself think that it is reasonable to attribute these views to Aenesidemus and thus to Schulze. The character of the editor writes as if Aenesidemus has made his position and his motivation for publishing the letters that make up the body of the work very clear to him, so I think they can probably be taken as an accurate picture of Aenesidemus’ views. I would say much the same thing about a passage in Hermias’ first letter to Aenesidemus where, referring to previous discussions the two are supposed to have had, he says: As far as I know, you have never wanted, by means of scepticism, to create that peace and undisturbedness of mind which the sceptics declare to be the noblest fruit of their unsystematic system, and with the possession of which one can only deceive oneself for as long as one has completely misunderstood the essential needs of human reason. Rather, you always say that you really only apply scepticism to ensure that your mind is sensitive to the knowledge of the truth, and to prevent the sight of your reason from being weakened by ungrounded dogmatism. (A 19) Putting aside the fact that this passage also seems to suggest that Aenesidemus has at some point explicitly distinguished his scepticism from that of the Pyrrhonists(!), it clearly states that the scepticism Aenesidemus, and thus Schulze, defends is on the basis of an interest in satisfying the demands of reason. And in his response to Hermias’ letter, Aenesidemus does not dispute his friend’s characterisation of his scepticism. Instead, he writes, ‘the essence of sceptical philosophy consists really in nothing other than that practice most unique to human reason’ (A 25).34 Regardless of

240  Robb Dunphy whether or not Williams is right to suggest that Sextus rejects a model of sceptical argumentation which is undertaken on the basis of the demands of reason, then, I think that Berry is wrong to attribute such a position to Schulze. Of course, if one thinks, against readers like Williams, that Sextus’s Sceptic in fact also holds off from claims to knowledge precisely in the name of observing the demands of reason more thoroughly than the “rash” dogmatists do, then this would support the claim that Schulze’s attitude is continuous with that of ancient Pyrrhonism, but not in the way that Berry understands it. (S5), meanwhile, attributes a Pyrrhonism to Schulze on the grounds that, unlike Kant, for whom, as Berry acknowledges, scepticism is ‘no dwelling place for permanent residence; for the latter can only be found in complete certainty’ (KRV A761/B789),35 he seems to understand scepticism as position that one can adopt and remain with. I agree that this might seem to be an important similarity between Schulze and Sextus, although I am not sure how far this similarity can be pushed, or how much evidence it provides for attributing a Pyrrhonian Scepticism to Schulze. Sextus, for example, famously defends the possibility of a sceptical way of life, at least in part, by suggesting that the sceptic can undogmatically observe the ‘laws and customs’ of their society ‘without holding any opinions’ (PH I: 23−24), while Schulze seems to think that scepticism about the supposedly “universally valid principles” to which Kant and Reinhold appeal is a genuinely liveable option precisely because he takes his scepticism not to apply to several of the most important aspects of life. He says, for example, that ‘the moral legislation of reason can just as little be doubted as its logical legislation, and we can just as little misunderstand the demands of the former than the commands of the latter’ (A 281). Whether there is room to see in Schulze’s exempting of moral reasoning from sceptical doubt an echo of Sextus’ suggestion that Pyrrhonian Scepticism is compatible with adherence to various practical laws, or whether Schulze’s position rather suggests a far more limited and modern conception of scepticism is something I cannot satisfactorily explore here, so I will simply concede that Schulze does at the very least superficially appear to resemble Sextus in supposing that a good life is possible without reaching absolute certainty when it comes to the traditional objects of metaphysical inquiry. To conclude this section then, it seems to me that Berry is to some extent right: attending to degrees of continuity between Schulze’s scepticism and that of Sextus Empiricus is helpful when it comes to a proper appreciation of the nature of the former. But there are, I have suggested, clear limitations to the extent of Schulze’s Pyrrhonism. In various ways he is very much a modern sceptic. And the suggestions that Schulze’s endorsement of Pyrrhonism means breaking with the attempt to satisfy the demands of reason, or that he has no theoretical commitments of his own,

Schulze’s Scepticism and the Rise and Rise of German Idealism 241 I have suggested, should be abandoned. Whether the interpretation of Sextus upon which such suggestions rely is correct cannot be decided here, but I think that it is clear that they are not supported by Schulze’s own text. 10.5  Schulze, Pyrrhonism, and German Idealism What does this reconsideration of the nature and extent of Schulze’s scepticism mean for Berry’s other two central claims? The first of these was that the German Idealists remain problematically vulnerable to Schulze’s Pyrrhonian criticism of their metaphilosophical commitments to the satisfaction of the demands of reason by the construction of a scientific, systematic metaphysics on the basis of fundamental principles. The immediate thing to say here, I think, is that if the account of Schulze’s scepticism developed in the previous section is correct, then the claim that Schulze is targeting the rationalist metaphilosophical commitments of Kant and the postKantian idealists by abandoning the idea that philosophy must proceed on the basis of rational norms that guide one towards the attempt to provide an ultimate proof of the validity of one’s philosophical claims is mistaken. Regardless of what this means for his proximity or distance from Sextus, Schulze does not seem to be interested in diagnosing the Kantian and German Idealist attempts to satisfy the demands of reason as symptoms of an epistemic pathology. His scepticism rather shares with his opponents a commitment to only believing what it is rational to believe, and proceeds just as they do on the basis of what he takes to be fundamental principles. This is not to say, of course, that the German Idealists might not be vulnerable to such a line of criticism, even if it turns out not to have been developed by Schulze, although I cannot address this concern in depth here. Berry’s suggestion is that their interest in thoroughly satisfying the demands of reason by developing a scientifically certain metaphysical system is motivated by horror at the prospect of a life without certainty; the epistemological ambition of Kant and the German Idealists, she suggests, ‘begins in fear’ (Berry 2020: 169) – a fear which is dogmatic precisely because it ignores the possibility, which the Pyrrhonists exemplify, of a good life without certainty. I am not so sure of the accuracy of this diagnosis. A different angle from which to consider the goal of elaborating a scientifically certain metaphysical system is the one that considers the standards of justification that are endorsed in such an endeavour. In striving for scientific certainty rather than, say, accepting a metaphysical thesis on the basis of empirical evidence that appears to make it probable, the idealists are raising those standards as high as they can. But, as a more recent epistemologist puts it, ‘[i]f you are working hard to avoid dogmatism, your standards for justification tick up. If you’re working to avoid scepticism, your standards tick down’ (Aikin 2011: 48).36 In their search for certainty,

242  Robb Dunphy the idealists exhibit their anti-dogmatic credentials more clearly than their anti-sceptical ones. In doing so, they open the door to sceptical rejoinders far more willingly (and, perhaps, boldly) than they would if they endorsed metaphysical claims as instances of genuine knowledge on far weaker grounds. Indeed, here they are at their closest to Schulze, who differs from them only in that he does not think that they have succeeded in proving the universal validity of their fundamental principles and accompanying systems (a complaint that, I noted earlier, the idealists frequently directed towards one another). I would expect a philosophical project that really begins in fear of the possibility of a life without metaphysical knowledge to emphasise anti-scepticism over anti-dogmatism, but this is not what one finds in the work of Kant, Reinhold, Fichte, or Hegel. According to this suggestion, then, the metaphilosophical commitments of Kant and the German Idealists reflect ambition, or aspiration, certainly, but not necessarily dogmatism. In line with this suggestion, it is worth at least gesturing towards a few instances where the German Idealists are not so quick to assume that they are in possession of indubitable fundamental principles or a system of scientifically certain metaphysical claims, where haste in this regard might indicate a dogmatic fear of uncertainty. Fichte, for example, rather than always dogmatically assuming the possibility of a scientific philosophy grounded in a single, foundational principle, at least acknowledges the possibility that our various philosophical claims might receive no ultimate justification and thus remain fundamentally uncertain, as well as the possibility that we might instead ground those claims on a number of distinct foundations which resist organisation into the form of a unified system (Fichte 2021: 167−168). And for all that he takes the foundational principle of the self-positing “I” to be guaranteed by a peculiar, practical kind of intellectual intuition, he does not take this to be sufficient, firstly, to prove that it is in fact the fundamental principle of the entire system of philosophy, nor, secondly, to guarantee the possibility of developing a unified, adequate system on its basis. The second task, he argues, can only be accomplished by exhaustively exploring the extent to which a system of philosophical propositions can be derived from the principle under consideration, and finding in turn that from these the first principle can again be derived, “completing the circuit” and thus demonstrating the unity of the system (Fichte 2021: 173). It is in this sense, rather than dogmatically assuming that the demands of reason will be satisfied with the provision of a scientific metaphysical system, he says that ‘[e]verything depends on the experiment’ (Fichte 2021: 169). In the case of the first task, that of proving that one has in fact identified ‘the fundamental principle of human knowledge as such’ (Fichte 2021: 173), the aspirational element of the case is still more explicit, since he seems

Schulze’s Scepticism and the Rise and Rise of German Idealism 243 to allow that one can never rule out the possibility that future experience might lead one to a philosophical claim the validity of which cannot be demonstrated on the basis of one’s putative fundamental principle, thus calling its fundamentality into question. There is no option, he thinks, but to proceed on the assumption that one is charting the system of human knowledge as such, even though one cannot rule out the possibility of encountering some new claim that will call for the revision of its foundations (Fichte 2021: 173−175). In Fichte’s acknowledgement that ‘[n]ot only is this foundational principle … incapable of proof, it also cannot be proven that it is the foundational principle of all knowledge’ (Fichte 2021: 169), the metaphilosophical commitments that Berry suggests might be vulnerable to scepticism seem far less dogmatic.37 And in the case of Hegel, even if the aspirational elements of the presentation of his system are less marked than Fichte’s, there is a sense in which he might be taken to go still further in this direction, precisely because he rejects the possibility of establishing a fundamental principle which grounds and provides unity to the system at its beginning, by means of intellectual intuition or otherwise.38 Instead, he insists repeatedly that such a principle could only be the result of an exhaustive, philosophical inquiry (W 3: 24, 5: 30, 6: 245). And nothing entitles one setting out on such a project to suppose in advance that such a result will be delivered. Although the relation between the post-Kantian idealists and Pyrrhonism is not the topic of this chapter, I think it is worth noting that here their respective metaphilosophical commitments do not appear to be so very different. If, as both Sextus and Schulze agree, scepticism involves a commitment to open-ended inquiry rather than a negative dogmatism that denies the possibility of metaphysical knowledge altogether, this is compatible with the attempt to develop a systematic, scientific metaphysics in an aspirational vein. To deny outright the possibility that the idealists will locate their ‘fantastic, transcendental quarry’ (Berry 2020: 170) would be to risk slipping into making the kind of negatively dogmatic claim that neither Schulze nor Sextus would make. Berry’s third major claim was that Pyrrhonism, which she detects in Aenesidemus, represents a genuine philosophical alternative to the attempt to secure an account of the good life on the basis of the development of a metaphysical system, precisely because it is concerned with ‘figuring out how to live well with uncertainty, and how to get along – how to flourish even – in a “post-certainty” world’ (Berry 2020: 169). I do not have the space here to give this claim the consideration that it requires, so I shall conclude with three remarks. The first is that I think that we should concede the viability and possible value of such a project, regardless of the extent to which we consider it to have been effectively carried out by the Pyrrhonists. The second is that Schulze, as I have presented him in

244  Robb Dunphy the previous section, is not obviously concerned with this “Pyrrhonian” project so much as he is interested in using sceptical arguments to call into question the perceived dogmatism of his idealist interlocutors, and the extent to which he considers scepticism to be compatible with the good life seems to turn largely on the fact that he (rather dogmatically) considers moral reasoning, just as much as logical reasoning, to be immune to sceptical argumentation. Finally, it is not obvious, I think, that Kant or the German Idealists would reject the possibility of living well without demonstrating the validity of one’s beliefs on the basis of fundamental principles. Kant, for example, acknowledges that the task of elaborating a metaphysics of morals is ‘capable of a great degree of popularity and suitability for the common understanding’ even without deriving it from its foundational principle (GMS 4: 391), while Hegel’s picture of ethical life is famously not just a matter of deriving moral imperatives from pure practical reason a priori, but allocates a crucial role to ‘custom’ and the ‘habitual practice of ethical living’, despite the fact that these elements of one’s ‘second nature’ fall short of explicit, rational judgement concerning the good (W 7: §151). Here again, perhaps, the idealists are not so far from the virtues Berry finds in Pyrrhonism as she seems to think.39 Abbreviations A: Schulze, Aenesidumus AL: Sextus Empiricus, Against the Logicians GMS: Kant, Groundwork to the Metaphysics of Morals KRV: Kant, Critique of Pure Reason PH: Sextus Empiricus, Outlines of Scepticism W: Hegel, Werke in Zwanzig Bänden Notes 1 Examples would include Reinhold’s own concern that Kant has not provided an adequate foundation for his critical philosophy (e.g. Reinhold 1978: 116−117), Fichte’s endorsement of some of Schulze’s criticisms of Reinhold’s principle of consciousness, or Hegel’s claim, later, that ‘The deficiency in the Schellingian philosophy’ is that the fundamental principle of the indifference of subjectivity and objectivity upon which it rests ‘is posited as absolute, without it being proven that it is the truth’ (W 20: 435), which recalls Sextus’ use of the mode of hypothesis (PH I: 168). 2 Schulze would continue to produce work in the same vein as Aenesidemus, notably his two-volume 1802 Critique of Theoretical Philosophy, which has as its primary focus a much more painstakingly detailed sceptical examination of Kant’s first Critique. My principal focus will be on Aenesidemus rather than Schulze’s later work, however, since it is that text that is the focus of the recent interpretation of Schulze’s significance to which I wish to respond here.

Schulze’s Scepticism and the Rise and Rise of German Idealism 245 3 It might be thought that, in addressing Schulze’s Critique of Theoretical Philosophy at length in his “On the Relationship of Scepticism to Philosophy”, Hegel too is illustrating the influence of Schulze’s thought on his own. The text quickly reveals that this is not the case, however. Hegel is almost entirely dismissive of Schulze’s work in that essay, entirely unlike Fichte’s response to Aenesidemus, which expresses a careful and at least sometimes appreciative response to those of Schulze’s criticisms that Fichte considers perspicacious and significant. A stronger case can be made for the influence of Schulze’s satirical Aphorisms on the Absolute upon Hegel’s later decision to compose the Phenomenology of Spirit as an introduction to his system in which sceptical objections to his position are overcome. I will not address that work here, but see Westphal 2020 for discussion. 4 See Hankinson 2010 for a discussion of Aenesidemus’ scepticism, his rejection of Academic dogmatism, and the relation of his position to the later Pyrrhonism of Sextus. 5 Of course, the success of Reinhold’s Elementary Philosophy does not depend only on the principle of consciousness. The method according to which he derives his results must stand up to scrutiny, and those results themselves must amount to unified and complete system of cognitions, to provide two further conditions. Nevertheless, since he takes the principle of consciousness to be the foundation of that system, if it should turn out to be fundamentally dubious, then it does not matter, e.g. how carefully and rigorously he derives his results from it. If the principle fails, then it takes the system with it. Reinhold acknowledges as much when he claims that it is only by way of criticisms of his fundamental principle that his Elementary Philosophy can effectively be disputed (Reinhold 2003: 247−248). I hasten to add that I do not suppose that Kant would agree with Reinhold that his critical philosophy requires grounding in such a fundamental principle. See Chapter 8 by Karin de Boer and Wellmann in this volume. 6 The significance of Schulze’s first charge, that the principle of consciousness cannot amount to the highest principle of a philosophical system because it itself must stand under the principle of non-contradiction (A 52−53), is certainly contestable. Breazeale (2016: 159), following Fichte’s review of Aenesidemus, suggests that this criticism is misguided in that it fails to observe the distinction between a ‘real or material’ first principle and a ‘merely formal or logical one’, and points out the principle of consciousness is not derived from the principle of non-contradiction, even if it must not violate that formal principle. Since it is the derivation of a philosophical system that is at stake, for Reinhold, Schulze misses the mark. Fichte himself, in his Foundation of the Entire Wissenschaftslehre, seems to think that formal principles too can be derived, albeit not directly, from the fundamental principle which he himself endorses (Fichte 2021: 187, 207−210). 7 It should be acknowledged, however, that he appears to make this argument on the basis of a fairly dogmatic-sounding transcendental realism: ‘We can only ascribe reality and truth to the representations in which our cognition consists insofar as they stand in relation and connection with a certain something that is distinct from themselves, and all searching for the truth of our cognitions aims at detecting the connection of our representation and the characteristics occurring within them with something which exists independently of them’ (A 158). 8 For Schulze’s Pyrrhonian-sounding response to the objection that it is not really possible to extend sceptical doubts to the subject matter of one’s own cognitive faculties, see A 78−79.

246  Robb Dunphy 9 See Breazeale 2016: 155−158 for a brief but informative summary of the major criticisms that Schulze directs towards Reinhold’s Elementary Philosophy in Aenesidemus. 10 Kant clearly does intend his first critique to achieve certain anti-sceptical goals. Nevertheless, it has been suggested that reading that work entirely through the lens of responding to Hume is liable to warp the interpretation of Kant’s arguments somewhat. See, for example, Ameriks 2000: 55−63. 11 See Beiser 1987: 282−283 for the suggestion that Schulze’s overly psychologistic understanding of Kant’s project of identifying the conditions of the possibility of knowledge robs this criticism of much of its force. 12 See Fincham 2000: 99−109 for further discussion of Schulze’s criticisms of Kant’s account of causality and of his appeals to the notion of the thing in itself. 13 The most notable example concerns the attribution of a certain regress argument to Schulze against Reinhold’s principle of consciousness. This argument suggests that Reinhold’s principle itself gives rise to a certain regress because it states that any representation is distinguished by the subject from both the subject and the object, yet this implies that the consciousness of the subject and the object on which the principle relies can only be saved from contradicting the principle itself if they too are distinguished as representations by the subject from a subject and an object, which must then in turn also be so distinguished, and so on. Versions of this argument are attributed to Schulze in the discussions of Aenesidemus in Neuhouser 1990, Martin 1997, and Baur 1999, for example. But as Franks (2005: 219−224), Messina (2011), and Breazeale (2016) have argued, it is not clear that Schulze in fact levels this argument at Reinhold, and Reinhold anyway seems to think that our knowledge of the subject and the object in his principle of consciousness is not representational. 14 Examples include Beiser 1987: 280−284, or di Giovanni 2000: 20−27. 15 Examples include Baur 1999: 75−85, Bristow 2007: 122−133, or Förster 2012: 158−166. 16 The complicated nature of Hume’s own debt to Pyrrhonian Scepticism is something I cannot address here, unfortunately. For a good recent discussion, see Ribeiro 2021: 97−111. 17 Hegel’s influential ridiculing of Schulze’s reading of Sextus in his review of Schulze’s Critique of Theoretical Philosophy has no doubt played a not insignificant part in the disinterest among scholars of German Idealism in Schulze’s alleged Pyrrhonism (W 2: 213−272). 18 See Berry 2020: 158. 19 There are several smaller pieces of evidence to which Berry appeals, for example the suggestion that Schulze’s portrayal of Hermias might be intended to resemble the Pyrrhonists’ dogmatic opponents, but I cannot consider these for reasons of space. 20 I will attribute to Schulze the arguments of the character of Aenesidemus within the work. This is not controversial. 21 Curiously, Berry suggests that begging the question amounts to opening oneself to either the mode of infinite regress or of hypothesis. It seems to me that the natural Agrippan mode to reach for is that of reciprocity, but this is of no real significance here, since the claim is simply that Aenesidemus is making use of the Agrippan modes in his criticisms of Kant. One could disagree about exactly which mode is being made use of while still agreeing with the broad claim.

Schulze’s Scepticism and the Rise and Rise of German Idealism 247 22 This clearly resonates with the account of Kant’s motivations when developing his critical philosophy presented in Chapter 5 by Catherine Wilson in this volume. 23 This is not the first time such a comparison has been made. Ward (1919−1920: 9) suggests that Fichte’s attempt to develop a scientific transcendental philosophy on the basis of the first, fundamental principle of the “I” might be seen to be ‘a case of hunting the snark and finding it a boojum’, although in the specific context of a discussion of Kant’s reaction to the Wissenschaftslehre. While some might wish to extend this analogy in the Popperian spirit of emphasising the perceived dangers of the inheritance of German Idealism by pointing out that one hunter dies when the snark turns out to be a boojum and another goes mad during the hunt; it is perhaps also worth noting that two of the hunters become great friends, and at least one learns a great deal. This last element, in particular, accords with the aspirationalist aspects of German Idealism that I emphasise in Section 10.5. 24 See Bondeli, Chotaŝ, and Vieweg 2016. 25 Berry seems to suggest that it is Nietzsche who represents a clear breakthrough when it comes to this kind of philosophical task (Berry 2020: 169). See Berry 2011 for her in-depth account of Nietzsche’s relation to Pyrrhonian Scepticism. 26 Aristotle 1984: 72b5−32; see Maimon 2010: 42−43 for the suggestion that Kant’s deduction is circular, and Maimon 2000: 171−172 for the suggestion that his own scepticism is not an “ancient” one. 27 See Machuca 2019 for a good discussion of the Sceptic’s dialectical argumentation. 28 See Beiser 1987: 268−272. 29 Schulze acknowledges that the offering of alternative fundamental principles is demanded by Reinhold from those who wish to challenge his Elementary Philosophy, but gives no indication that he is doing so merely in order to argue dialectically. Indeed, Schulze seems to recognise the legitimacy of arguing according to such a procedure, and says that Reinhold’s ‘demand is most just’ (A 39). 30 The emphasis on this topic is frequently taken to be one of the clear indicators of the difference between Pyrrhonian and modern scepticism. See Williams 1988: 550, for example. See Forster 1989: 9−35 for another case to this effect, and for Hegel’s appreciation of just this matter. 31 It is for this reason that Hegel is so critical of Schulze’s portrayal of scepticism in his later Critique of Theoretical Philosophy (W 2: 225−226, 230−233, 249−269). I by no means mean to suggest that we should endorse Hegel’s interpretation of Sextus, which is in various ways questionable, but in the way that he distinguishes Pyrrhonian from modern scepticism, at least, it is Hegel rather than Schulze who is closer to the contemporary orthodoxy on this topic. 32 See Eichorn 2020: 188−197 for an excellent overview of this debate. 33 But see also Vogt 2011 or Dunphy 2022. 34 I do not think that Schulze’s commitment to the practice of reasoning is incompatible with his claim, discussed above, only to use the term “reason” in a manner comprehensible to his interlocuters, since that claim concerns only the objective existence of “reason” as a distinct cognitive faculty, not the idea that philosophical inquiry is bound to observe certain rational norms. 35 See Berry 2020: 167. 36 It might be that this distinction between the focus on anti-scepticism and on anti-dogmatism roughly tracks the distinction Schulze makes in the preface

248  Robb Dunphy to Aenesidemus, between the philosophical party that always assumes that it knows the truth, and the party that always strives to improve its philosophical reasoning (A 8), in which case Kant and the German Idealists would be closer to Schulze than he himself thinks. 37 See Scribner 2014 for an example of a reading of Fichte’s idealism that emphasises its more aspirational aspects. 38 See W 5: 65−66 for an instance of Hegel’s rejection of beginning with a principle arrived at by means of intellectual intuition. 39 The research that went into this chapter was largely carried out while the recipient of a postdoctoral research fellowship from the Irish Research Council (IRC). I am grateful to the IRC, and to Brian O’Connor and the School of Philosophy at University College Dublin, for hosting me during that fellowship. Thanks are also due to Toby Lovat and to Charlotte Baumann for helpful feedback on drafts of this chapter.

Bibliography Aikin, S. (2011) Epistemology and the Regress Problem (London: Routledge) Ameriks, K. (2000) Kant and the Fate of Autonomy: Problems in the Appropriation of the Critical Philosophy (Cambridge: Cambridge University Press) Aristotle (1984) The Complete Works of Aristotle (2 vols.), J. Barnes (Ed.) (Princeton, NJ: Princeton University Press) Baur, M. (1999) “The Role of Skepticism in the Emergence of German Idealism” in M. Baur and D. Dahlstrom (Eds.) The Emergence of German Idealism (Washington, DC: The Catholic University of America Press), pp. 63–91 Beiser, F. (1987) The Fate of Reason: German Philosophy from Kant to Fichte (Cambridge, MA: Harvard University Press) Berry, J. (2011) Nietzsche and the Ancient Skeptical Tradition (Oxford: Oxford University Press) Berry, J. (2020) “Sextan Skepticism and the Rise and Fall of German Idealism”. In K. Vogt and J. Vlasits (Eds.) Epistemology after Sextus Empiricus (Oxford: Oxford University Press), pp. 154–172 Bondeli, M., Chotaŝ, J., and Vieweg, K. (Eds.) (2016) Krankheit des Zeitalters oder heilsame Provokation? Skeptizismus in der nachkantischen Philosophie (Paderborn: Wilhelm Fink Verlag) Breazeale, D. (2016) “Reinhold/Schulze/Fichte: A Re-Examination”. In M. Bondeli, J. Chotaŝ, and K. Vieweg (Eds.) Krankheit des Zeitalters oder heilsame Provokation? Skeptizismus in der nachkantischen Philosophie (Paderborn: Wilhelm Fink Verlag), pp. 151–179 Bristow, W. (2007) Hegel and the Transformation of Philosophical Critique (Oxford: Oxford University Press) di Giovanni, G. (2000) “The Facts of Consciousness”. In G. di Giovanni and H. Harris (Eds.) Between Kant and Hegel: Texts in the Development of PostKantian Idealism (Indianapolis, IN: Hackett), pp. 2–50 Dunphy, R. (2022) “From Proto-sceptic to Sceptic in Sextus’ Outlines of Pyrrhonism” Apeiron Volume 55 Issue 3, pp. 455–484 Eichorn, R. (2020) “Sextus Empiricus on Isostheneia and Epochē: A Developmental Model” Sképsis Volume 11 Number 21, pp. 188–209

Schulze’s Scepticism and the Rise and Rise of German Idealism 249 Fichte, J. (2000) “Review of Aenesidemus”, trans. G. di Giovanni. In G. di Giovanni and H. Harris (Eds.) Between Kant and Hegel: Texts in the Development of Post-Kantian Idealism (Indianapolis, IN: Hackett), pp. 136–157 Fichte, J. (2021) Foundations of the Entire Wissenschaftslehre and Related Writings (1794−95). Translated by D. Breazeale (Oxford: Oxford University Press) Fincham, R. (2000) “The Impact of Aenesidemus upon Fichte and Schopenhauer” Pli Volume 10, pp. 96–126 Forster, M. (1989) Hegel and Skepticism (Cambridge, MA: Harvard University Press) Förster, E. (2012) The Twenty-five Years of Philosophy: A Systematic Reconstruction.Translated by B. Bowman (Cambridge, MA: Harvard University Press) Franks, P. (2005) All or Nothing: Systematicity, Transcendental Arguments, and Skepticism in German Idealism (Cambridge, MA: Harvard University Press) Hankinson, R. (2010) “Aenesidemus and the Rebirth of Pyrrhonism” in R. Bett (Ed.) The Cambridge Companion to Ancient Scepticism (Cambridge: Cambridge University Press), pp. 105–119 Hegel, G. (1986–) Werke in Zwanzig Bänden. Edited by E. Moldenhauer and K. Michel (Frankfurt am Main: Suhrkamp) Kant, I. (1998a) Critique of Pure Reason. Translated by P. Guyer and A. Wood (Cambridge: Cambridge University Press) Kant, I. (1998b) Groundwork to the Metaphysics of Morals. Translated by M. Gregor (Cambridge: Cambridge University Press) [GMS] Kroner, R. (1961) Von Kant bis Hegel (Tübingen: Mohr Siebeck) Machuca, D. (2019) “Pyrrhonian Argumentation: Therapy, Dialectic, and Inquiry” Apeiron Volume 55 Issue 3, pp. 199–221 Maimon, S. (2000) “Letters of Philaletes to Aenesidemus”, trans. G. di Giovanni. In G. di Giovanni and H. Harris (Eds.) Between Kant and Hegel: Texts in the Development of Post-Kantian Idealism (Indianapolis, IN: Hackett), pp. 158–203 Maimon, S. (2010) Essay on Transcendental Philosophy. Translated by N. Midgley, H. Somers-Hall, A. Welchmann, and M. Reglitz (London: Bloomsbury) Martin, W. (1997) Idealism and Objectivity: Understanding Fichte’s Jena Project (Stanford, CA: Stanford University Press) Messina, J. (2011) “Answering Aenesidemus: Schulze’s Attack on Reinholdian Representationalism and Its Importance for Fichte” Journal of the History of Philosophy Volume 49 Issue 3, pp. 339–369 Neuhouser, F. (1990) Fichte’s Theory of Subjectivity (Cambridge: Cambridge University Press) Perin, C. (2010) The Demands of Reason: An Essay on Pyrrhonian Scepticism (Oxford: Oxford University Press) Reinhold, K. (1978) Über das Fundament des philosophischen Wissens (Hamburg: Felix Meiner Verlag) Reinhold, K. (2003) Beiträge zur Berichtigung bisheriger Mißverständnisse der Philosophen (I) Erster Band, das Fundament der Elementarphilosophie betreffend (Hamburg: Felix Meiner Verlag) Ribeiro, B. (2021) Sextus, Montaigne, Hume: Pyrrhonizers (Leiden: Brill) Schulze, G. (1996) Aenesidemus oder über Fundamente der von dem Herrn Professor Reinhold in Jena gelieferten Elementar-Philosophie (Hamburg: Felix Meiner Verlag)

250  Robb Dunphy Scribner, F (2014) “A Plea for (Fichtean) Hypothetical Idealism: Exosomatic Evolution and the Empiricism of the Transcendental”. In D. Breazeale and T. Rockmore (Eds.) Fichte and Transcendental Philosophy (Basingstoke: Palgrave Macmillan), pp. 314–330 Sextus Empiricus (2000) Outlines of Scepticism.Translated by J. Annas and J. Barnes (Cambridge: Cambridge University Press) Sextus Empiricus (2005) Against the Logicians. Translated by R. Bett (Cambridge: Cambridge University Press) Vogt, K. (2011) “The Aims of Skeptical Investigation”. In D. Machuca (Ed.) Pyrrhonism in Ancient, Modern, and Contemporary Philosophy (Dordrecht: Springer), pp. 33–49 Ward, J. (1919−1920) “In the Beginning…” Proceedings of the Aristotelian Society Volume 20, pp. 1–24 Westphal, K. (2020) “Aphorisms on the Absolute: Editorial Introduction” The Owl of Minerva Volume 51, Issue 1/2, pp. 1–10 Williams, M. (1988) “Scepticism Without Theory” The Review of Metaphysics Volume 41 Number 3, pp. 547–588 Williams, M. (2010) “Descartes’ Transformation of the Sceptical Tradition”. In R. Bett (Ed.) The Cambridge Companion to Ancient Scepticism (Cambridge: Cambridge University Press), pp. 288–313

11 The I and I The Pure and the Empirical Subject in Fichte’s Science of Science Kienhow Goh 11.1 Introduction With the historic triumph of absolute idealism, what enthusiasm could be summoned for the philosophic thought of Johann G. Fichte is blunted by its alleged affiliation to an extreme, unmitigated form of subjectivism.1 Though not without basis, the caricature misleads so long as subjective is broached in terms of what is abstract and formal, or what is arbitrary and contingent. Far from thinking of the subjective in either way, Fichte pled for a conception of the subject which is lawfully necessary and concretely material. In my view, the perennial tendency to think otherwise is a result of the failure to draw the boundary between the pure, absolute I – here I speak not only of the infinite, unlimited I, but also of the finite, limited I – and each of us qua an empirical, individual I. Though the distinction of the finite, limited I from the empirical, individual I has not gone unrecognised, its differences have not yet been expressly addressed and their implications adequately understood. In what follows, I set out to clear up potential misunderstandings by offering a discussion of several prominent aspects of the better-known versions of the philosophical science which were developed by Fichte during his Jena period (1794–1799). My discussion is organised into four sections. Section 11.2 highlights the primarily epistemological nature of Fichte’s overall philosophical project. It identifies one of the project’s chief impetuses as the defence of the scientific status of Kant’s Critical philosophy against the challenge of Maimon’s scepticism. Section 11.3 examines the concept of scientificity that informs Fichte’s concept of philosophical science. Fichte apparently analyses the scientificity of a body of knowledge in terms of its certainty and systematicity. But it is in connection with his conception of a system of philosophical science as a documentation of an autonomous “system of a human mind” that we are able to make sense of the systematicity and certainty that are at play in the concept. Section 11.4 considers the features of philosophical science that pertains DOI: 10.4324/9781003187561-12

252  Kienhow Goh to its scope. The philosophical−scientific system falls short of meeting the Maimonian sceptical challenge so long as it remains incomplete. Section 11.5 investigates the relationship of the human mental system presented by philosophical science to actual experience. It makes clear that the human mental system is no empirical subject at all, but a pure subject that maps the possible perceptions and actions which an empirical subject has the potential to realise for itself. 11.2  The Project of a Science of Science In an effort to distinguish his philosophical thought from previous bodies of philosophical teachings which he deemed to fall short of being a science (Wissenschaft), Fichte christened it Wissenschaftslehre. The term, which literally translates as “Doctrine of Science”, indicates its primary objective to be the epistemological one of furnishing a ground for scientific knowledge. Convinced that the Doctrine of Science is the first body of philosophical teachings to attain scientific status, Fichte refers to it as the ‘science of a science in general [Wissenschaft von der Wissenschaft überhaupt]’ (SW I: 43).2 In striking contrast to Descartes, Fichte does not take the defeasibility of extant human knowledge for the starting point of such a science. He begins with the belief that extant ordinary knowledge and the scientific knowledge which derives from it are by and large sound and reliable (and able to autocorrect when they are not) and undertakes to vindicate this belief by demonstrating that (and how) they are based upon a set of lawful (and hence necessary and universally valid) acts of the human mind (menschlichen Geistes). Each of these acts is lawful insofar as it is subject to a higher act (or what is the same, accomplished in accordance with a higher law). Immediately or mediately, each of them is subject to ‘that act that subsumes all others and furnishes the highest law’ (SW I: 71). Through this “highest act”, the acts subject to it make up a whole which Fichte dubs the “system of a human mind”, and sometimes the ‘system of human knowing [menschlichen Wissens]’.3 It is on account of its unity (i.e. its being itself) that the acts are lawful. The task of the Doctrine of Science is then to demonstrate the acts’ lawfulness by developing a presentation (Darstellung) of the whole system. As such, it amounts to a ‘pragmatic history of the human mind’ (SW I: 77, 222). The “highest act” is none other than “the I” or “I am”, expounded in the opening paragraphs of the student handbook Foundation of the Whole Doctrine of Science (1794/95) in terms of a “fact-act” (Tathandlung) of self-positing. According to Fichte, this I, which is “determined purely and unconditionally by itself”, is also opposed by a not-I and to this extent, not absolutely self-determining. It is the pure or the absolute I in the former

The I and I 253 respect and the theoretical I or “intelligence”4 in the latter respect. This ‘sharpest contradiction of all’ (SW I: 366) between two respects of one and the same I is resolved upon the transformation of the absolute I into a practical I, which can be viewed either as the unattainable but indispensable goal of intelligence to determine itself absolutely or as an endless but futile effort by it to realise the goal. In either case, the theoretical and the practical I are shown to be different sides of the same I. One does not exist apart from, or operate independently of, the other. To this extent, it is apt to describe the I as a theoretico-practical I. The absolute I seems to be no longer operative as such in being transformed into the practical I. In truth, it is aufgehoben in the Hegelian sense of being preserved at the same time as it is being cancelled. Most obviously, it is preserved in the theoretico-practical I’s unceasing effort to determine itself absolutely as it is cancelled in its constant failure to do so.5 More importantly, it continues to be operative in the incomprehensible unity of the I’s theoretical and practical activities (that makes possible the procedure of Fichte’s synthetic method) and the a priori-unknowable coherence and consistency of the material content of experience (that make possible the empirical sciences). These systematising functions of the absolute I are by no means of mere architectonic interests. As Neuhouser has argued, Fichte’s occupation with systematicity ‘springs from his conviction that systematization holds the key to refuting skeptical challenges raised against Kant’s accounts of both knowledge and morality’ (Neuhouser 2014: 306).6 One of the main sceptical challenges Fichte is occupied with is that posed by the elusive and wayward Kant enthusiast and critic Salomon Maimon. According to Maimon, Kant’s transcendental deduction of the categories fails to provide a satisfactory answer to his own question of the quid juris, of how pure concepts of the understanding and sensible intuitions can come together to yield a cognition.7 A self-professed admirer, Fichte was goaded by Maimon’s argument to ‘advance an idealist metaphysics, one in which the content of cognition is taken to be constituted by an act of the absolute I’ (Thielke 2001: 102). As I see it, the absolute I is Fichte’s equivalent for Kant’s intuitive understanding (intellectus archetypus), that is, a cognition that proceeds in cognising an object from the intuition of a thoroughly determinate whole to each part of the whole (Kant 1902: 5: 407). In appropriating the Kantian notion, Fichte is careful to build into it features that forestall a return to pre-Kantian dogmatic rationalism. Fichte does not deny the discursivity of our understanding, but accounts for it in a way that wards off Maimon’s charge of the lack of adequate proof for the applicability of its pure concepts to empirical objects. Though the theoretico-practical I is individuated to a measure that allows each of us to speak of his or her own I, it does not of itself amount to an individual I. It can be thought of, I will

254  Kienhow Goh show, as standing midway in a range of generality/individuality between the absolute I at one extreme and the individual I at the other. 11.3  The Concept of a Science of Science In the opening section of the programmatic tract Concerning the Concept of the Doctrine of Science (1794), Fichte reaffirms the common view that philosophy should be a science (Wissenschaft) and considers what it means for philosophy to be one. Like any of the sciences in circulation, philosophical science should have systematic form and sound content. A science has a systematic form when the propositions comprising it ‘are joined together in a single first principle’ such that they form a whole (SW I: 38).8 A science has sound content when the propositions comprising it are known with certainty (Gewissheit). As a preliminary observation, Fichte notes that soundness of content is more basic and essential to science than systematicity of form: while unsystematic knowledge is still knowledge (albeit of a methodologically defective sort), unsound knowledge is no knowledge at all. Sound content is valued for its own sake, whereas systematic form is valued only as a means to ensuring soundness of content (SW I: 39). On Fichte’s view, propositions are either certain of themselves or certain on account of their ‘position within and relation to the whole’ (SW I: 40). Propositions that are certain of themselves are ‘purely and simply certain’ (SW I: 41). Propositions that are certain on account of their systematic position are certain by virtue of their connection with other propositions; their certainty is derived ultimately from that of some purely and simply certain proposition. Fichte terms a proposition of the latter ‘sort, one which is certain prior to and independent of the association with others, … a first principle’ (SW I: 41). It is unclear at this point of the inquiry as to what the exact nature of the “connection” of the other propositions with first principles is. What is clear is that they take the form of a ‘specific kind of inference by which we infer the certainty of other propositions from the certainty of the first principle’ (SW I: 43). If a science S1 is to have sound content, its first principles must be certain prior to and independent of it. As such, they cannot be established from within its own domain. They ‘should and must be certain in advance. Their certainty cannot be demonstrated within the systems themselves; on the contrary, every proof which is possible within these systems presupposes the certainty of these first principles’ (SW I: 43). But even so, they could be established from within the domain of a higher science S2. The distinction between propositions that are certain of themselves and propositions that are certain on account of their systematic position is thus relative to a scientific domain: a proposition that is certain of itself in S1 could be certain on account of its

The I and I 255 systematic position in S2. Also, if a proposition is certain, it must not only have a ground simpliciter, but also a cognisable ground. A non-cognisable ground is nothing to us. The possibility of philosophy turns on ‘whether the whole of our knowledge has a cognizable fixed ground’ (SW I: 44). Philosophical science proceeds only on the assumption that our knowledge has such a ground. It aims to provide all the other sciences with their first principles by having the propositions established from within its own domain. In this sense, it is a science of science. So far, Fichte’s view of scientific knowledge seems very much a standard epistemological foundationalist one. However, striking differences quickly emerge from his discussion of the first principle. For a start, it is noteworthy that Fichte takes the connection between propositions comprising a science to be more tight-knit than what we would expect of an epistemological foundationalist. By latter’s account, those propositions which are certain of themselves cannot also be certain by virtue of their connection with the other propositions. Otherwise, it would be impossible to differentiate between foundational and non-foundational propositions. Fichte’s view is startlingly contrary: ‘as all propositions of that system are inseparably connected with one another, if any one of them is true, all are true necessarily, if any one of them is false, all should be false necessarily’ (SW I: 61). The statement conjures up a scenario where the certainty of all the propositions is derivable from that of any of them, i.e. where no proposition is certain prior to and independent of another. It leaves us to wonder what the whole point of defining a first principle is. Does it not follow from the fact that all the propositions are first principles that none of them is? The conclusion is inevitable only if we assume that Fichte thinks of the connection between the propositions in purely logical terms. While he certainly does admit that they stand in a biconditional relation to each other, he regards them as doing so only on the basis of an extra-logical relation, namely the mental relation of certainty. In Fichte’s eyes, the certainty imparted by one proposition to another is not so much a matter of objective evidence as of subjective feeling.9 This is evident from Fichte’s curious insistence on the singularity or uniqueness of the first principle of a science: [T]here can in each science only be one proposition that is certain and established prior to the connection between the propositions. Were there several such propositions, then either they would have no connection at all with the other proposition that is certain and established prior to the connection, or else they would be connected to it. In the first case they would not then be part of the same whole, but would constitute one or more separate wholes. But the only way in which propositions is supposed to be connected to each other is

256  Kienhow Goh by sharing a common certainty, so that if one is certain then another one must also be certain, and if one is uncertain then the other one must also be uncertain; and only this connection of their certainty to one another can determine their relation. But a proposition which possesses its own certainty independently of the other propositions could not be connected with them in this manner. If its certainty is independent, then it remains certain even if the others are not. Consequently, such a proposition would not be connected via certainty with the other propositions at all. (SW I: 41, emphases added) For Fichte, a science must have not only at least one first principle, but also no more than one first principle. The systematic form of a science is a form that turns on the unity or oneness (Einheit) of its first principle. Admittedly, Fichte’s characterisation of it as ‘the systematic form of the whole that arises from the individual parts’ (SW I: 42) fails to capture this characteristic of it. Nonetheless, it is evident from his claim of all the propositions “sharing a common certainty” that he thinks of it as the form of an original whole that is prior to and independent of all its parts, and that subordinates all its parts to itself via the certainty it imparts to them. In this way, the singularity of the first principle is maintained vis-à-vis its biconditional relations to the propositions. As we have noted from the outset, Fichte believes that extant human knowledge is by and large sound and reliable. More precisely, he views it as sound and reliable insofar as it is based upon a ‘system of a human mind’ that is implicitly operative prior to and independently of philosophy (SW I: 76–77). It is not philosophy’s task to replace the human mental system by a better one, but to make it explicit. The system that is characteristic of philosophy as a science is supposed to re-enact the human mental system. It is not the human mental system, but a presentation of the human mental system. The human mental system is present in advance of any philosophising on the philosopher’s part (SW I: 70, 71). By contrast, the philosophical−scientific system is a product of his philosophising (SW I: 46). The philosopher’s task is not to prescribe laws for acts of the human mind but to describe the laws to which the acts are subject. As Fichte puts it, philosophers ‘are not the legislators of the human mind but rather its historians’ (SW I: 77).10 A philosopher who sets out to present the human mental system has immediate access only to the acts that make up the human mental system – not the system. Thus, Fichte specifies that philosophical science has the human mental system for its ‘object [Object]’, but the human mind and its acts for its ‘content [Gehalt]’ (SW I: 70). The content has a form of its own prior to and independent of philosophical−scientific system – namely the human mental system. And owing to the strictly

The I and I 257 descriptive relation of philosophical science to its object, its systematic form is determined not so much by the logical relations of its propositions as by the mental relations of its propositions’ content. The singularity of philosophical science’s first principle stems from the supremacy of the human mind’s highest act; the unity of the former’s system stems from the latter’s highest act’s subsumption of all its other acts. If philosophical science is to serve its role as a science of science (viz. of furnishing a ground for all the other sciences), its first principle cannot be established, neither from within its own domain, nor from within the domain of any higher science. It must be ‘certain in itself, through itself and for its own sake’ (SW I: 48). It is distinguished from the first principles of the other sciences as the ‘highest and absolute first principle’ (SW I: 54). By the same token, philosophical science is – inasmuch as it ‘contains all possible human knowing’ (SW I: 62) – distinguished from the other sciences as the “universal science” (allgemeine Wissenschaft). On the other hand, the other sciences are distinguished from philosophical science (and from each other) as a “particular science” (besondere Wissenschaft). Like the first principle of philosophical science, the first principles of the other sciences cannot be established from within their own domain. Nevertheless, they can be established from within the domain of a higher science, namely philosophical science. Put differently, they are deducible from the first principle of philosophical science. But if so, the question arises as to how the other sciences can be considered distinct and separate from philosophical science, rather than an integral part of it, in the first place. The key to Fichte’s answer lies in the above-mentioned distinction between the pure, rational subject and the empirical, natural subject. The objects of philosophical science are a product of the pure, rational subject, and are to this extent ‘necessary acts’; the object of a non-philosophical science is a product of the empirical, natural subject, and is ‘itself a free act’ (SW I: 72). Given the principles legitimised by philosophical science, the power of judgment [Urteilskraft] still retains its complete freedom to apply these laws at all or not; or from among the multiplicity of laws and of objects, which law it will to apply to an arbitrary object, for example, to regard the human body as raw material, as organized material, or as animate material. But as soon as the power of judgment has been given the task of observing a particular object according to a particular law, in order to see whether and to what extent the object conforms or fails to conform to it, then the power of judgment is no longer free, but is subject to a rule. And accordingly, we are no longer in the Doctrine of Science, but instead are in the area of another science called “doctrine of nature. (SW I: 64–65)

258  Kienhow Goh Unlike philosophical science, a non-philosophical science stands in a prescriptive relation to its object. It freely takes upon itself “the task of observing a particular object according to a particular law”, and of enacting the law as a law of the object to the extent that it observes the latter conforms to the former. In this way, geology enacts the principle of reciprocal interaction as a law of inorganic matter, botany enacts the principle of organisation as a law of inanimate organisms, and zoology enacts the principle of articulation as a law of non-human animate organisms. As the first principles of a science, these propositions cannot be established from within the domain of their science. Nevertheless, they can be established from within the domain of the higher universal science. But regardless of whether they are so established, the power of judgement is free to apply them as a rule by which it reflects on some freely delimited object. This is what Fichte means when he says that the propositions must be combined with ‘something supervening’ (etwas hinzukommen) in order to be raised to the status of the first principle of a particular science (SW I: 62). Due to the freedom of the power of judgement to choose which law to apply as a rule of its reflection and which object to reflect on by the rule, a particular science is able make infinite, unending progress in a way the universal science is not able to. 11.4  The Scope of the Science of Science As the science of science, philosophy is supposed to provide an account of ‘human knowing in general’ (menschliche Wissen überhaupt). By this, I take Fichte to mean the system of acts that furnishes a basis for every genuine instance of human knowing, qualifying it as such. This is none other than the “system of a human mind”, which in this context is more aptly designated by the “system of human knowing”. Ordinary human knowledge is scientific insofar as it is grounded in the acts that make up the system; it is established as scientific insofar as it is shown to be grounded in them. Philosophy endeavours to discover the set of propositions that portray the human mental system. Its immediate task is not so much to establish ordinary human knowledge as scientific as it is to establish itself as the philosophia perennis. It does so by showing that the propositions it sets forth are those that portray such a system. As we have seen, the philosopher has no immediate access to the human mental system and hence no advance knowledge of whether there is indeed such a system. What he has immediate access to are acts of his own he believes to make up such a system. This belief cannot be vindicated except upon the actual presentation of the system in its totality. From this angle, we can appreciate Fichte’s emphasis on the need for a system of philosophical science (viz. a presentation of the human mental system) to

The I and I 259 be “complete” (vollständig) and the length to which he goes to spell out a criterion for it (SW I: 58). A philosophical−scientific system that shows that its principles portray a human mental system can do no less than to ‘encompass the whole system’ (SW I: 333). Short of being complete (viz. presenting the human mental system in its totality), it fails to demonstrate the possibility of such a system (and hence of scientific knowledge). Therefore, it is incumbent on the philosopher, over and above deriving one act from another, to show that the derived acts taken together make up a system. Despite what the term “human knowing in general” suggests, it is a mistake to interpret the human mental system in one-sidedly abstract and formal terms. In doing so, we dispossess the Doctrine of Science of its capacity to address the doubt it is conceived to address. As we have noted, Fichte’s philosophical project was envisaged with an eye to addressing a doubt Maimon had raised against Kant’s deduction of the categories. Maimon’s scepticism is aimed primarily at Kant’s demonstration (or lack thereof) of the applicability of the pure concepts of the understanding to sensible intuitions (Thielke 2001: 108–114, Beiser 2002: 248–252, Franks 2005: 176–189). Maimon is ready to concede to Kant that the categories are pure concepts of the understanding by means of which we think an object. But objects we receive by means of sensible intuitions might or might not answer to them. ‘For Kant’s account to be convincing, Maimon claims, the categories must not only apply to the formal features of intuitions, but must also discriminate between the different contents of intuitions’ (Thielke 2001: 113). So long as it stops short of doing so, Kant’s account fails to rule out the possibility of sensible intuitions falling outside the domain of thought. Viewed in this light, the rationale behind Fichte’s move to extend the scope of the concept of a human mental system from the forms of experience to encompass its material content is apparently one of revising the long-standing account of experience in terms of the relation of something falling outside of the human mental system to the system – a view that gives rise to questions concerning the lawfulness of the relation and our ability to know it. The strategy can be gleaned from his statement: ‘Nothing enters human beings from the outside’ (SW II: 478). It is not as if I first exist as a mind (Gemut) in isolation from objects and then come to know objects by entering into some relation with them. As an I, my very ‘being is a knowing’ and my knowing entails my being related to ‘a determinate system of other things for [me]’ (SW II: 478). This knowing can be considered either from the standpoint of philosophy or from the standpoint of experience. From the standpoint of philosophy, I am a pure subject, which has ‘the system of the whole of experience’ (SW II: 478) for its object. As such, philosophy ‘anticipates the whole of experience and thinks of it only

260  Kienhow Goh as necessary, and it is to this extent, by comparison with actual experience, a priori’ (SW I: 447). From the standpoint of experience, I am an empirical subject which ‘exists only in experience (a posteriori)’ (SW II: 478) and relates to its object in just the way one object of experience relates to another – contingently. The strategy then is to extend the scope of the a priori to encompass the empirical – not by getting rid of the a priori/a posteriori divide, but by reinterpreting it in terms of the standpoint from which one and the same things are considered. While the strategy for Fichte’s response to Maimon’s scepticism in his move to include the material content of experience is already adumbrated in the concept of a human mental system, its details are worked out only in section VII of Outline of the Distinctive Characteristics of the Doctrine of Science (1795), where a genetic account of actual objects by which the categories are shown to ‘arise together with the objects’ can be found (SW I: 387).11 Fichte is well-known for ascribing the productive process to an exercise of the power of imagination (Einbildungskraft). As Maimon also considers objects to be products of the power, this elicits the impression that he succumbs to Maimonian scepticism – an impression reinforced by his own admission that the Critical philosopher ‘concedes everything the sceptic demands and usually even more’ (SW I: 389). In fact, Maimon and Fichte are working with very different conceptions of the power of imagination: for Maimon, the power is a Humean reproductive one whose function is to retain content it passively receives; for Fichte, it is a Kantian productive power whose function is to generate content according to rules it actively applies. Fichte’s ascription of objects to the power of imagination also fosters the myth, as Beiser has observed, that he sees the imagination as ‘creat[ing] ex nihilo the entire world’ (Beiser 2002: 254). In truth, he fleshes out the process (which he describes as ‘the most important part of our present inquiry’ (SW I: 354)) in terms of a series of lawful, overlaying syntheses accomplished through the I’s tendency to reflect upon itself as limited. At the heart of the account is the I’s ‘limitedness’ (Beschränkheit) whose thoroughgoing ‘determinacy appears as what is absolutely contingent, and delivers the sheer empirical in our cognition’ (SW I: 489). Upon being reflected upon, this limitedness is first posited as a feeling, which is Fichte’s equivalent for Kant’s sensation (SW I: 490). The I is then necessitated by laws of its own nature to posit an opposing intuition whose limiting action explains the feeling and to carry over (übertragen) the feeling to the intuition, which is constituted thereby as an object (SW I: 490). The role of the power of imagination is confined to ‘fabricat[ing] extended matter through the intuition’ (SW I: 490) and forming an image (Bild bilden) from out of the feeling. What of the categories? Fichte illustrates how they are deduced by means of the genetic account of actual object with the examples of the concepts of substance and cause. As the I has to unify the necessary not-I

The I and I 261 and a contingent image of it without undoing the former’s necessity and the latter’s contingency, it can do so only by relating in terms of substantiality, i.e. the image to the not-I as an accident of it and the not-I to the image as a substrate of it. Furthermore, as the unification of the accident with the substrate is itself contingent and the I has to posit what is contingent as originating from the action of another, it can do so only by relating the unification to something else in terms of efficacy, i.e. the unification to another as its effect and another to the unification as its cause (SW I: 385–386). In a word, Fichte’s explicit response to Maimon’s scepticism is this: categorial principles are applicable to empirical objects because empirical objects are first produced in accordance with them. To be clear, Fichte does not claim to be able to know a priori what the material content of experience is. As a philosopher’s presentation of the human mental system – as opposed to the system itself – the philosophical−scientific system is conditioned by the philosopher’s own finite, discursive intelligence. Consequently, the claim in the hypothetical treatment of philosophical science in §1 of Concerning the Concept that ‘in each science there can be only one proposition that is certain and established prior to the connection between propositions’ (SW I: 41) turns out be unsustainable. Fichte candidly admits that the highest act does not of itself give the philosopher any headway in a genetic deduction of (self-)consciousness. And simply because no consciousness comes about through this mere act, we may indeed infer further to another act, whereby a non-I arises for us; only so can we make progress in our philosophical argument, and derive as required the system of experience. (SW I: 459) In the first place, it is due to the finite discursivity of human reason that human knowledge (philosophical knowledge included) takes the form of propositions. Every proposition is about something and says something about what it is about. Fichte terms what a proposition is about its content and what the proposition says about what it is about its form. For example, the proposition “The sky is blue” is about the sky and blueness and says something about them, namely that they are related as a subject is related to its predicate. Granted that every proposition has a content and a form, there are three ways in which a proposition could qualify as a “first principle”: (i) the proposition can be unconditioned with respect to both its form and content (i.e., absolutely unconditioned); (ii) it can be unconditioned with respect to its form but conditioned with respect to its content; (iii) it can be conditioned with respect to its form but unconditioned with respect to its content. This gives rise to the possibility of more than one first principle (up to three first principles) for a scientific system.

262  Kienhow Goh Therefore, a scientific system could have an absolutely unconditioned first principle, a second first principle whose form is unconditioned but whose content is conditioned by the absolutely unconditioned first principle, and a third first principle whose content is unconditioned but whose form is conditioned by the absolutely unconditioned first principle. Indeed, in Division One of the Foundation, “Foundational Principles of the Doctrine of Science”, Fichte advances a total of three foundational principles (Grundsätze): (i) “The I posits itself without condition [schlechthin]”; (ii) “The I posits a not-I in opposition to itself”; and (iii) “The I posits the not-I as determining itself and itself as determining the not-I”. To the extent that the second and the third principles are conditioned by the absolutely unconditioned first principle, no proposition of the system is not deducible from the latter. In this way, the systematicity of the system is, despite a plurality of first principles, held together by the unity of a singular absolutely unconditioned first principle. Therefore, Fichte does not reject Kant’s discursivity thesis concerning our cognition, but assimilates it into a unified, comprehensible account of reason. By this account, theoretical reason (the I’s positing the not-I as determining itself) and practical reason (the I’s positing itself as determining the not-I) are interdependent sides of one system of reason (the I’s positing itself and the not-I). Reason falls short of the whole of itself, of its own nature, as it were. This falling-short of itself is expressed on the practical side of a theoretico-practical I as the limitation of a striving and on the theoretical side of the same as the inability of an intelligence to know a priori the thoroughgoing determinacy of the limitation: ‘Here’, Fichte writes, ‘all deduction comes to an end’ (SW I: 490). For him, as it is for Kant, there is a component of experience that is simply given. The Streitpunkt between them is not whether there is a component of experience that is simply given, but whether what is given has a source that is external to the human mental system, or whether the system stands in a merely determinable or a thoroughly determinate relation to what is given. So long as a source external to the system for what is given, or an indeterminate relation of the system to what is given, is admitted, the Maimonian query over the lawfulness of the relation of what is pure (whether it be a concept or an object) to what is derived from the source will be raised. It is with a mind to satisfying such a query that Fichte insists that what is given has an a priori basis in the I’s limitedness. 11.5  The Metaphysics of the Science of Science For Fichte, philosophy is both metaphysics and metaepistemology: metaphysics inasmuch as it has for its material content acts that are preconditions of experience and hence beyond experience; metaepistemology

The I and I 263 inasmuch as the acts it has for its material content are constitutive of knowing. Thus, he notes in his second introduction to the second Jena exposition of the Doctrine of Science that philosophy is metaphysics, since in ‘[establishing] the ground of experience, it elevates itself over the [sphere of experience]’ (Fichte 1994: 13). He understands by metaphysics then not ‘a theory of the so-called things in themselves’, but ‘a genetic deduction of what we find in our consciousness’ (SW I: 32). The philosopher sets out to present the system of a human mind by observing the acts that make up the system (rather than the system itself). The acts he is able to observe are his own; the system he is able to present is one that is made up of his own acts. Granted that the acts are constitutive of experience, the philosopher’s ability to reflect upon them presupposes his having ‘already run through the whole course of experience’ (SW I: 459). The acts are fait accompli for the philosopher in the sense that he has already performed them. It is in this light that Fichte refers to the human mental system as a ‘system of facts’ (SW I: 331). By facts, he has in mind subject-objective “fact acts” (Tathandlungen) rather than plain objective facts. Whatever these are, they are not Reinholdian “facts of consciousness”, i.e. facts straightforwardly accessible through self-conscious introspection. As preconditions of (inner) experience, the facts are by definition beyond the reach of (inner) experience (SW I: 334). In this sense, they are aptly described as preconscious. Still, it does not follow from his having performed the acts that he has attended to them. To accomplish the later, he needs to be attending to them as and when he is performing them. He needs as it were to ‘catch [himself] in the act’ (SW I: 362, SW III: 23). According to Fichte, very few are in fact able to attend to their own acts with the clarity and distinctness that is required of them by the Doctrine of Science. On the ladder of mental development, the nearest is in fact the furthest away: we proceed from our being directed outwards towards objects to our being directed inwards towards ourselves. Following Kant, Fichte considers determining acts of judging to be subsuming and abstracting acts of judging to be reflecting. The former proceeds by mechanical necessity, whereas the latter proceeds by deliberative freedom. Taking his cue from the distinction Maimon draws between concept formation and reflection, Fichte further regards necessary, subsuming acts of judging as synthetic, and free, reflecting acts of judging as analytic (Maimon 1790, 93–94). Although both classes of acts are subordinate to laws, the former is subordinate to theoretical laws (viz. laws according to which they are necessitated), whereas the latter is subordinate to practical laws (i.e. laws according to which they are merely urged). As best as I can make out, subsuming, synthetic acts of judging are those constitutive of theoretical activities (e.g. representing, cognising, knowing) and

264  Kienhow Goh pertain to the determination of objects, including ourselves qua an object (namely our corporeal body [Körper] and its potential to be organised [organiziert], sensible [empfindlich], and articulated [leiblich]). On the other hand, reflecting analytic acts of judging are those constitutive of practical activities (e.g. willing, realising). Through them, our practically conditioned relationships to other objects (namely our actual bodily organisation, sensibility, and articulation) are determined. In the first instance, necessary synthetic acts of judging are necessary insofar as they are necessitated by laws enforced from within the system of a human mind. The necessary acts are by definition acts that go to make up the system. Their necessity is merely conditional, that is to say, they must occur on the condition that certain other acts do. Though the acts – given the fact that they do occur – occur in the way they do necessarily, there is nothing necessary about the fact that they occur. According to Fichte, their occurrence is based on the I’s limitedness. As we have seen, this limitedness is the original basis of “the sheer empirical in our cognition”. However, it does not follow from the arbitrary contingency of this content in a metaphysical account of it that the content itself does not conform to any law. While the free acts of an empirical I are merely urged by practical laws of morality, those of the absolute I are necessitated by a law that is higher than both the theoretical laws of nature and the practical laws of morality (SW VIII: 415). On Fichte’s view, the higher law serves to bring nature into harmony with morality by ensuring the realizability of the final end of reason in the sensible world. Inasmuch as the limitations are determined by the law, they determine the content of the world whose idea, as we have seen, is posited in the I as an idea, one whose mechanical and organic laws are perfectly harnessed to realise the moral end. If this is correct, the limitations do not determine the content of (our actual experience of) the sensible world immediately. Rather, they do so by determining the content of (our possible experience of) the intelligible, moral world. If this is right, what, from the standpoint of philosophy, is posited as the higher law is what, from the standpoint of experience, we believe to be “the moral order” of the sensible world – what Fichte infamously identifies with God. In either case, the law serves, first and foremost, to ground the realizability of the final end of reason in the sensible world, i.e. the availability to each of us of a course of action through which the present state of the world is morally improved. In doing so, it also serves to ground the causality of our will in our world. Unlike the Kantian possible experience, the Fichtean possible experience is not merely formally possible but realisable, that is to say, it can, given how everything else is, be occurrent in actual continual experience. As a system of necessary synthetic acts of judging, the human mental system is ‘a system of actions by means of which objects come into being for us’ (Fichte 1994: 22), objects that constitute a whole we call “the

The I and I 265 sensible world”. Inasmuch as its acts occur on the pretext of the I’s limits, the sensible world possesses a moral order. Viewed from this angle, the sensible world is none other than the intelligible moral world. This means, firstly, that everything in it (including we ourselves) is ultimately aimed at the moral perfection of ourselves and everyone else, and secondly, we are free in relation to the world – not determined by anything in it, but able, through determining ourselves, to determine everything in it. In fact, the former furnishes a ground for the latter: it is not because we can realise the end of morality that we aim at it; it is because we (are originally determined to) aim at the end of morality that we can realise it. ‘Actuality’, Fichte asserts, ‘is not inferred from possibility, but the converse’ (SW V: 183). Inasmuch as the limits are determined by a law that brings nature into harmony with morality, they mark our ‘determinate place in the moral order of things’ (SW V: 185). The sensible world, inasmuch as it possesses a moral order, amounts to an originally determined system of dutiful acting, one that contains ‘the sensibilized material of our duty’ (SW V: 185). However, it would be a mistake to equate the objects that arise in and through the system this way with objects of actual experience. For Fichte, there is the system of the whole of experience (the system of necessary representations), which he sees as being necessitated by the essence of a human being qua intelligence, and there is an actual continual experience (aggregate of representations), which he explains as ‘nothing other than the continual analysis of that system grounded through his essence’ (SW II: 478). Correlatively, there is, on the one side, the pure and necessary system of a human mind (system of human knowing), and on the other, the empirical and contingent human mind (sequence of human knowing). Moreover, the acts of the human mental system are accompanied by the pure I’s limitations. In taking the pure I’s limitations into consideration, we overstep the domain of theoretical reason and enter that of practical reason. As I have mentioned, the content of philosophy’s first principle is transformed in the course of its exposition: in positing itself as determining the not-I, the pure I effectively posits itself as an infinite ceaseless striving on the one side and as an idea outside of itself on the other. Owing to the infinity of the pure I’s infinite striving, our experience and knowledge are infinite in extent despite their finitude in kind.12 The pure I’s limitations are nothing but determinate limitedness of its striving, and objects are nothing but products of application of theoretical laws of representations by the pure I to the limiting (upon being limited). Therefore, as long as the pure I strives, it is limited; and as long as it is limited, it produces objects of experience in accordance with theoretical laws of representations. The demand that objects of our representations appear to be independently existing causes of the representations counts among those of the theoretical laws of representations. For this reason, we are entitled to consider them as such.

266  Kienhow Goh The system of the whole of experience is timeless and unchanging; actual continual experience is temporal and changing. From the standpoint of philosophy, the system of (our possible experience of) the intelligible moral world arises immediately from the necessary synthetic acts of judging of the human mental system, whereas (our actual, continual experience of) the sensible world arises mediately through free analytic acts of judging upon fact acts of the system. As I have mentioned, Fichte’s distinction between the two classes of acts is closely modelled after Maimon’s distinction between concept formation and reflection. According to Maimon, concept formation and reflection are opposite ways of considering, and hence two sides of one and the same action: when we form a concept, we begin with the universal and proceed to subsume a particular under it; when we reflect, we begin with a particular and proceed to omit (weglassen) determinations under which it is subsumed. Yet, as Maimon intimates, concept formation is more fundamental than reflection. In Melamed’s words, ‘judgments are in a sense posterior to the formation of concepts, since the synthesis which created the concept of the subject (either by a finite or an infinite understanding) must have preceded the judgment in which we analyze the components of the subject’ (Melamed 2021: 54–55). Correlatively, necessary synthetic acts of judging and free analytic acts of judging are two sides of the same action. Nevertheless, the former is a ground of the latter in a way the latter is not of the former: while acts of analysis are merely those of recollecting the fact acts of synthesis through which the objects have been produced, acts of synthesis are those through which objects are produced in the first place. That system of a human mind which is supposed to be presented by the Doctrine of Science is absolutely certain and infallible. Everything that is based upon this system is absolutely true. It never errs, and anything which has ever been or will ever be necessarily present within the human soul is true. If human beings have erred, the mistake did not lie in something necessary; instead, the mistake was made by free reflective judgment when it substituted one law for another. (SW I: 76–77) As we have seen, Fichte thinks that scientific knowledge is by and large sound and reliable. We can now see that this is so insofar and only insofar as it is based on the human mental system. To be more precise, it is derived from an act of reflection, whose function is merely to recollect acts of synthesis through which objects have been produced. However, since reflection is free, it can fall short of the task. The path of reflection is far from an easy and straightforward one. It amounts in fact to the course of the mental development of humankind.

The I and I 267 From our investigation of the I and its distinctness from, and relationship to, the empirical I, it is clear that its subjectivity is neither abstractly formal nor arbitrarily contingent. Faced with the Maimonian sceptical challenge, Fichte ventured to present a rationalistic system upon which ordinary human knowledge is originally, and scientific knowledge derivatively, based. He termed it the “system of a human mind” and later the “system of the whole of experience”. In its determination as a theoretico-practical I, this human mental system is a subject. This subject is “individuated” inasmuch as it maps a thoroughly determinate system of possible perceptions (or, what amounts to the same thing, actions) that is uniquely our own, as opposed to another’s. For all its thoroughgoing determinacy, however, the theoretico-practical I is not yet an empirical I. The human mental system is ideal rather than real; it constitutes what we could rather than what we do experience. The empirical I, with all its random whims and idiosyncrasies, enters the scene only upon the I’s (intelligence’s) further acts of analysis upon itself (the system). Notes 1 Such a caricature that goes way back to Jacobi’s portrayal of his philosophy as deducing everything from the I (Jacobi 1799: 14). 2 “SW” is employed in this chapter as an abbreviation for the Johann Gottlieb Fichte’s Sämmtliche Werke, edited by Immanuel Hermann Fichte and originally published by Veit und Comp. in Berlin in 1845. 3 In the later part of Jena period, he came to refer to it as a ‘system of the whole of experience [der gesammten Erfahrung]’ (SW I: 446, 458, SW II: 478), a ‘system of necessary representations’ (SW I: 445, 446) and a ‘system of the whole of reason [der gesammten Vernunft]’ (SW I: 466). 4 The I is “intelligence” (Intelligenz) only insofar as it is not absolutely self-determining (SW I: 64, 248). Consequently, Fichte uses the term in the Foundation only towards the end of Part Two “Foundation of the Theoretical Knowledge”, from SW I: 209 onwards. Contrarily, his treatment of the I in the second Jena Doctrine of Science (including First and Second Introduction to the Doctrine of Science (1797/98), as well as the “Doctrine of Science nova methodo”), is couched primarily in terms of intelligence. Assuming that there is no change in its usage, there is a significant distance between the starting point of the first Jena presentation (the fact act) and that of the second (intelligence). 5 This peculiarity of the resolution is underscored in Schwab’s treatment of ‘the problem of transition, namely, the transition from the first principle of philosophy, the self-positing I, to anything opposed to this I, that is, to anything different from the I’ (Schwab 2021: 95). 6 In a recent study, Wellman details how Fichte transforms Kant’s conception of the relationship between the deductions of the categories and the metaphysical system, combining the metaphysical and the transcendental deduction into a single deduction, which he then equates with the whole of the system (Wellmann 2021: 119−121). 7 In a landmark article, Thielke (2001) explores the profound influence of Maimon on Fichte and its significance for the project of the Doctrine of Science. Breazeale further maintains that ‘many of Fichte’s most dramatic departures

268  Kienhow Goh from the “letter” of the Critical philosophy are best understood in the context of his strenuous efforts to respond to the Maimonian challenge’ (Breazeale 2016: 43). 8 Schmidt discerns in the 1801 essay A Crystal-Clear Report an alternative form of systematicity that arises from “a complex of interdependent elements” without privileging any single element over the rest, but does not think that it carries much significance for Fichte’s methodology (Schmid 2019: 76). 9 ‘I know, for example, that I doubt. But on what basis do I know this? Surely I do not know this on the basis of the objective constitution of some judgment I have made. Doubt is something subjective; like its opposite, certainty, it can only be felt’ (SW IV: 167). 10 Fichte’s usage of the term “pragmatic history” was influenced by Maimon, for whom ‘a pragmatic history presupposes a concept of the “object” whose genesis it is describing, and it relates this history strictly in terms of the inner, logical development or systematic unfolding of what is implicit from the start in that whose pragmatic history is being related’ (Breazeale 2016: 77). 11 Fichte also offers an alternative approach, based on the possibility of selfconsciousness, to the deduction in the second Jena exposition of the Doctrine of Science (see Wellmann 2021: 126−130). 12 ‘Human knowing is infinite in its extent, but it is completely determined in its kind through its laws and can be wholly exhausted’ (SW I: 59n).

Bibliography Beiser, Frederick (2002) German Idealism: The Struggle against Subjectivism, 1781–1801 (Cambridge, Massachusetts: Harvard University Press) Breazeale, Daniel (2016) Thinking through the Wissenschaftslehre: Themes from Fichte’s Early Philosophy (Oxford: Oxford University Press) Fichte, Johann G. (1845) Sämmtliche Werke. Edited by Immanuel Hermann Fichte (Berlin: Veit und Comp) Fichte, Johann G. (1994) Wissenschaftslehre nova methodo. Edited by Erich Fuchs (Hamburg: Felix Meiner) Franks, Paul (2005) All or Nothing: Systematicity, Transcendental Arguments and Skepticism in German Idealism (Cambridge, MA: Harvard University Press) Jacobi, Friedrich H. (1799) Jacobi an Fichte (Hamburg: Perthes) Kant, Immanuel (1902–) Kants gesammelte Schriften (Berlin: Königliche Preußischen Akademie der Wissenschaften) Maimon, Salomon (1790) Versuch einer Transzendentalphilosophie (Berlin: Voβ und Sohn) Melamed, Yitzhak Y. (2021) “Salomon Maimon’s Principle of Determinability and the Impossibility of Shared Predicates.” Revue de métaphysique et de morale 109(1), pp. 49–62 Neuhouser, Frederick (2014) “Fichte’s Methodology in the Wissenschaftslehre (1794–95).” In Palgrave Handbook of German Idealism. Edited by Matthew Altman (New York: Macmillan) pp. 300–319 Schmid, Andreas (2019) “Fichte, German Idealism and the Parameters of Systematic Philosophy.” In Palgrave Fichte Handbook. Edited by Steven Hoeltzel (New York: Macmillan), pp. 75–93

The I and I 269 Schwab, Philipp (2021) “Difference within Identity? Fichte’s Reevaluation of the First Principle of Philosophy in 5 of the Foundation of the Entire Wissenschaftslehre.” In The Enigma of Fichte’s First Principles, Fichte-Studien Band 49. Edited by David Wood (Leiden: Brill), pp. 94–118 Thielke, Peter (2001) “Getting Maimon’s Goad: Discursivity, Skepticism, and Fichte’s Idealism.” The Journal of the History of Philosophy 19(1), pp. 101–134 Wellmann, Gesa (2021) “‘The Subsequent Delivery of the Deduction’ – Fichte’s Transformation of Kant’s Deduction of the Categories.” In The Enigma of Fichte’s First Principles, Fichte-Studien Band 49. Edited by David Wood (Leiden: Brill), pp. 119–138

12 The Science of All Science and the Unity of the Faculties Schelling on the Nature of Philosophy Benjamin Berger 12.1 Introduction In his Lectures on the Method of Academic Study, delivered in Jena during the Summer Semester of 1802 and published the following year, Schelling tells his students that philosophy is the ‘absolute science’,1 the ‘science of all sciences’ (SW I/5: 214, 261; 1966, 8, 56), and the ‘knowledge that unifies and is the soul and life of all else’ (SW I/5: 266; 1966, 60). Clearly, Schelling holds something he calls philosophy in high regard. But what exactly does he understand philosophy to be? And in what sense is it an absolute science and science of all sciences? In what follows, I seek to elucidate Schelling’s conception of philosophical science, primarily as it is presented in the Lectures on the Method of Academic Study. Although he has a great deal to say about the nature of philosophy throughout his life, in these lectures he is uncharacteristically straightforward about what philosophy is and how it is related to other scientific endeavours. As will become apparent over the course of this chapter, philosophy, for Schelling, is what is typically understood as metaphysics, inclusive of both general and special metaphysics (although “metaphysics” is a term that Schelling himself seldom uses prior to his later thought on the relationship between positive and negative philosophy). Philosophy, according to Schelling, presents us with the essential being of what there is. Yet before considering this conception of philosophy in more detail, it will be helpful to begin, in Section 12.2, by considering the institutional setting in which philosophical inquiry is pursued: the university. For it is in the context of discussing the structure and purpose of the university that Schelling develops a distinctive understanding of the difference between philosophy and the other sciences. In Section 12.3, I focus on the strictly rational character of philosophy and its holistic vision, and I consider how these features of philosophy are central to its role in unifying the other sciences and making them fully intelligible. I then argue, in Section 12.4, that the unification of the sciences does not primarily depend, as one might expect, on a DOI: 10.4324/9781003187561-13

The Science of All Science and the Unity of the Faculties 271 philosophical grounding of all knowledge, such as that found in Fichte’s Wissenschaftslehre. For Schelling, the unification of the sciences is realised through the selfsame rational insight that is at work, however minimally, throughout all intellectual activity. Finally, in Section 12.5, I offer some brief remarks about how all of this is related to Schelling’s commitment to an anti-critical form of metaphysics. 12.2  Philosophy among the University Faculties Schelling’s Lectures on the Method of Academic Study contributed to debates at the turn of the 19th century concerning the purpose and structure of the German university – debates that also saw influential contributions from Schiller, Fichte, Kant, Schleiermacher, Steffens, and, most famously, Humboldt.2 Schelling’s references to philosophy as the “absolute science” and “science of all science” thus appear within a broader conversation regarding how the university had come to be organised and function as a matter of fact and how the university ought to be organised and function in principle. Since at least Dilthey’s Introduction to the Human Sciences, the academic division with which most scholars are familiar is that between the natural sciences, on the one hand, and the human sciences, on the other3 – with a great deal of work in the social sciences being proof of the appeal of quantitative research methods even in the study of human beings. But the academic division that characterised the university in the period of classical German philosophy was that between the higher and the lower faculties: the higher faculties being medicine, law, and theology, and the lower faculty being philosophy. Although the European research university has its origin in the establishment of professional schools for medicine, law, and theology, there was a need to provide students with a more general education that would make specialisation in these fields possible. Hence the identification of philosophy as the lower faculty, it being the general introduction and, as the saying goes, handmaid to those sciences that are concerned with the actual well-being of the body, the citizen, and the soul. Philosophy so construed does not have the disciplinary specificity we are accustomed to associating with the term today; instead, philosophy as a lower faculty covers all general academic learning as preparation for study within the higher faculties and is equivalent to the arts faculty of the medieval university as well as more recent iterations of liberal arts education. When Schelling and other philosophers of the period began lecturing and writing on the nature and purpose of the university, they did so with the relationship between the higher and lower faculties in mind. A better understanding of this relationship – and a better understanding of the importance of the lower faculty in particular – was seen as a possible solution

272  Benjamin Berger to what had become a moment of crisis for the institution. As Zachary Purvis has pointed out, between the years 1789 and 1815, the number of universities in Europe was reduced from 143 to 83 (Purvis 2016: 92). As others have documented, despite the doubling of German territories between the years 1720 and 1800, annual enrolments dropped from 4,400 to 3,000 (Menand, Reitter, and Wellmon 2017: 84). And those students who did enrol typically entered straightaway into one of the higher faculties, foregoing the philosophical education (i.e., the liberal arts education) that was thought by many to be of little practical benefit (Ibid.). Bureaucrats sought to support the university by doubling down on the idea that research and academic instruction in the lower faculty serves society by providing a general education to those who would become doctors, lawyers, and clergy (Menand, Reitter, and Wellmon 2017: 67−68). To whatever extent philosophical education was regarded as valuable, then, this value was understood in terms of mere utility, doing little to address a crisis of enrolment spurred, in part, by perceptions of philosophy’s negligible instrinsic worth. It is therefore no wonder that so many professional academics pursued another route for defending the relevance of the university:4 these academics argued that, while a broad philosophical education does play an important instrumental role in providing significant background knowledge for one to begin studying the practical fields of medicine, law, and theology, the knowledge gained in a general philosophical education is also an end in itself. Thus, during this period, the intrinsic value of knowledge was championed and the contemplative life defended by philosophers from Kant to Humboldt.5 Schelling’s lectures, in which he, too, argues that knowledge is its own end, were delivered four years after the publication of Kant’s Conflict of the Faculties, and they are responding both implicitly and explicitly to that text. Before turning to Schelling’s own claims about the nature of philosophical study, therefore, it may be helpful to begin with Kant. In providing theoretical support for the standard academic ‘division of labor’ (KGS I/7: 17; 1996: 247), Kant argues that the organisation of the university into various faculties is rationally justified (KGS I/7: 21; 1996, 250). As he writes, ‘the organisation of a university into ranks and classes did not depend entirely on chance’, for each faculty deals with a different and, indeed, essential aspect of the human being (KGS I/7: 21; 1996: 250). Given the different aspects of humanity with which the faculties are concerned, they draw upon different intellectual resources in their research and teaching: in attending to our physical well-being, the medical faculty rightly takes medical regulations to be authoritative; in attending to our civil well-being, the law faculty takes the legal code to be authoritative; and in attending to our spiritual well-being, the theology faculty takes the Bible to be authoritative (KGS I/7: 23; 1996: 251). And in contrast to all of

The Science of All Science and the Unity of the Faculties 273 these, philosophy – which includes philosophy narrowly construed (what Kant calls ‘pure philosophy’) as well as mathematics, history, geography, philology, and natural science (KGS I/7: 28; 1996: 256) – requires that reason be taken as the ultimate authority in our knowledge. Consequently, Kant argues, the philosophy lecturer need not defer to the state or the church in his teaching. Because the philosopher is to be guided by reason in his academic pursuits, he must also be free from government and church interference. If this means that there will be instances of conflict between the faculties, i.e., moments in which the philosopher questions the teaching of the higher faculties, so be it. The lower faculty must be free to criticise received wisdom on matters of national and theological importance. That philosophical knowledge – and, again, “philosophy” here means something quite broad – has its basis in reason suggests that its value is of a different kind than the value of medical, jurisprudential, and theological knowledge. Whereas the higher, professionally oriented faculties are instrumentally valuable and have a direct bearing on society, the lower faculty provides a basic education that is only indirectly utilitarian; yes, the lower faculty prepares students for more specialised education and is in this way useful, but the immediate concern of philosophical learning is knowledge itself – something that is inherently valuable. Schelling follows Kant in all of this. He agrees that the organisation we find in the modern university is rational, not arbitrary (SW I/5: 223; 1966: 17); he remarks that, were the state to have authority over universities, those universities would ultimately be transformed into vocational schools that lack a concern with knowledge as such (SW I/5: 229; 1966: 22); he emphasises the intrinsic connection between rational inquiry and freedom (SW I/5: 222; 1966: 15−16); and he thinks that students must come to see the intrinsic value of learning by overcoming the mistaken yet common assumption that knowledge is a mere means to practical ends (SW I/5: 221, 229; 1966: 14−15, 23). Although they are in agreement about all of this, Schelling also differs from Kant in a subtle yet important way. According to Schelling, the faculty that houses the autonomous search for knowledge as an end in itself should not, in fact, be called the philosophy faculty. Indeed, Schelling thinks that there should be no philosophy faculty at all (SW I/5: 284; 1966: 79). Why does Schelling reject the traditional identification of the lower faculty with philosophy? Answering this question will allow us to begin to reconstruct Schelling’s conception of philosophy as such, i.e., philosophy construed in the narrow sense – although, as will become apparent, there is nothing “narrow” about philosophy conceived as a distinctive science. In fact, the reason philosophy should not be set apart from the “positive” sciences as a faculty unto itself is because ‘that which is all things’ – that is, philosophy understood as the science of all science

274  Benjamin Berger – ‘cannot for that very reason by anything in particular’ (SW I/5: 284; 1966, 79). Thus, there should not be a separate philosophy faculty, on Schelling’s view, because philosophical knowledge is related – in some fundamental sense yet to be explored – to theological, jurisprudential, and medical knowledge, just as it is related, in some fundamental sense, to the other arts and sciences (i.e., those arts and sciences that were traditionally associated with the philosophy faculty). Schelling’s remarks about philosophy being “all things” should not be interpreted to mean that he is simply equating philosophy with all knowledge. This cannot be the case for the simple reason that philosophy, and philosophy alone, is the “science of all science”. It is therefore important to note that Schelling’s understanding of philosophy as universal also determines it as a distinctive way of knowing. In other words, despite the fact that philosophy is fundamentally connected to the other sciences – a point to which I return in more detail below – there is nevertheless an implication in the Lectures on the Method of Academic Study regarding the uniqueness of philosophy: Schelling is bringing out a feature of philosophical thinking that distinguishes it not only from the “positive” sciences, but from all non-philosophical forms of knowledge, including all of the nonphilosophical forms of knowledge traditionally studied in the philosophy faculty. Schelling’s argument that philosophy is a distinctive form of knowing is certainly not without precedent. Not only does he adopt the description of philosophy as science of all science from Fichte – another point to which I return below – but, in addition, there is also reason to think that, by rejecting the identification of the lower university faculty with philosophy, Schelling is in fact furthering Kant’s idea of pure philosophy (KGS I/4: 469; 2004: 5) and its fundamental character with respect to “philosophy” broadly construed (i.e., mathematics, history, geography, etc.).6 Schelling’s discussion of the nature of philosophy in the Lectures on the Method of Academic Study is, from this perspective, one moment within a longer development in the history of idealism and, more broadly still, modern European thought, in which the name “philosophy” comes to mean something far more specific than it had in the past. But Schelling’s role in this history can be difficult to see, especially from our contemporary situation in which academic philosophy is understood by many, both within and outside the discipline, to be most properly philosophical when there is little to no engagement with the other human and natural sciences. In contrast to our own tendency to set philosophy aside from other forms of knowing, for Schelling, it is precisely the impossibility of separating philosophical thinking from any other form of knowledge that distinguishes it as the distinctive science it is: as absolute, as science of all science, and as the highest of the sciences.7

The Science of All Science and the Unity of the Faculties 275 12.3  Philosophy as Absolute Science Focusing on three related features of Schelling’s conception of philosophy will allow us to clarify why he understands it to be an absolute science, the science of all science, and, indeed, a superior form of science. In this section, I consider (i) the strictly rational character of philosophy and (ii) the holistic vision of philosophy. Then, in Section 12.4, I turn to (iii) the presence of philosophical insight in all other sciences. To begin, it is important to be clear about the fact that Schelling does not claim that philosophy is superior to all other human activities. On the contrary, unlike Hegel, for whom philosophical thinking surpasses art and religion in its ability to express the structure of reality, Schelling insists that philosophy is on a par with art and religion. Indeed, at various stages of his intellectual development, he suggests that the greatest form of philosophical science would be one that is permeated with poetry and religion.8 What I am calling the superiority of philosophy, then, only involves the superiority of philosophy with respect to other forms of knowledge, that is, other sciences.9 That Schelling sees philosophy as a superior form of knowledge is not especially surprising given his place in the idealist tradition. As a strictly rational enterprise – and one that surpasses even the rationalism of mathematics (SW I/5: 254−255; 1966, 48) – philosophy provides us with access to the ideal essence of what is. All other sciences are encumbered, in one way or another, with empirical, historical facts about the world. Philosophy, by contrast, attends to what is essentially true, namely the rational structure of the universe and all things in it. Importantly, for Schelling, this is made possible by an act of abstraction through which one’s subjective experience of the objective world is put out of play and one takes on the perspective of reason as such for the first time.10 In 1801, he writes: The thought of reason is foreign to everyone; to conceive it as absolute, and thus to come to the standpoint I require, one must abstract from what does the thinking. For the one who performs this abstraction reason immediately ceases to be something subjective, as most people imagine it. (SW I/4: 114; 2012: 146) And as he puts it in the Lectures on the Method of Academic Study, ‘in true knowledge it is not the individual but reason that knows’ (SW I/5: 224; 1966, 18).11 Because the standpoint of reason is achieved only after abstracting from one’s own subjectivity (and, at the same time, from the object of knowledge as it is ordinarily given to us in experience), this standpoint can also be understood as the indifference of subject and object, or

276  Benjamin Berger their primordial identity.12 And, significantly, because there is no object to be given to a knowing subject here, what is known in philosophy is likewise subject−object identity, or, again, simply “reason”. Exploring subject−object identity and its various modes of expression is meant to elucidate the being of all that is, in the realm of nature and in the realm of spirit (i.e., humanity). For nature, according to Schelling, is subject−object identity that is characterised by a preponderance of objectivity (what he calls the ‘objective subject–object’ [HKA III/2.1: 280; 2012: 45]); spirit is subject−object identity that is characterised by a preponderance of subjectivity (the ‘subjective subject–object’). Consequently, philosophy – which occupies the standpoint of pure reason or subject−object identity without a preponderance of either subject or object – is in a position to shed light upon the being of nature, spirit, and their absolute identity, that is, reason itself, the absolute, or, alternatively, God. And, finally, philosophy allows us to see how these three aspects of subject−object identity are related. Put more simply, philosophy shows us what world, humanity, and God essentially are and how they are connected. Although he doesn’t use the term “metaphysics” during this period of his thought, it should be clear from these brief remarks that Schelling understands philosophy to be what we typically understand by that term (and his project could, from this perspective, be seen as a continuation of at least one aspect of what Kant calls ‘the philosophy of pure reason’, namely the system of pure reason [A841/B869]). The metaphysical view with which Schelling begins is relatively straightforward: everything that is – and therefore everything that can be known – is an expression of one and the same thing, namely reason or subject−object identity. As he puts it, ‘the ultimate law … for all being … is the law of identity … A = A’ (SW I/4: 116; 2012: 147) in which the first “A” signifies the subject and the second “A” signifies the object. Philosophy is superior to other forms of knowing not because it concerns itself with reason or subject−object identity. All forms of knowing do this – or all forms of knowing have in view some aspect of subject−object identity, because that is all there is to know: all that exists are instances of subject−object identity; nothing has being and nothing can be thought beyond this identity. The superiority of philosophy, therefore, does not have to do with the fact that it has in view subject−object identity. Why, then, think that philosophy is a superior form of knowing? One reason, according to Schelling, is that philosophy demonstrates the rational character of what is according to its own strictly rational form. Thus, philosophical thought does not represent truths about a reality that differs from it; it presents the truth that is its own and is that of being at the same time, since thought and being are indistinguishable from the standpoint of reason: what is known in philosophy and the knowing of it are identical. In other words, the rational constructions that take place in

The Science of All Science and the Unity of the Faculties 277 philosophical activity are identical to the constructions of being itself. As Schelling has already argued, when we philosophise, it is not, in fact, us philosophising, but reason or being itself – hence the necessity of abstracting from subjectivity in order to begin to think philosophically. It is perhaps plain to see that this is a quite distinctive form of knowing, but this becomes even clearer when we compare philosophy to the higher faculties or “positive sciences”, medicine, law, and theology, all of which depend in different ways upon empirical history for their comprehension of that which they know (SW I/5: 286; 1966: 82). Since the positive sciences attend to their objects of knowledge empirically, there is a sense in which those objects appear separate from the subject that knows them – a separation that obscures the underlying identity of subject and object. Philosophical knowledge, then, is superior to other forms of knowledge, in part, because it is the form of knowledge in which being knows itself, and this is made possible because philosophy is a strictly rational enterprise – an intellectual procedure that begins and ends with subject−object identity (i.e., reason).13 Relatedly, philosophical knowledge is a superior form of knowing because it is more complete than any other form of knowing. To see this, however, we need to first acknowledge that, although their empirical orientation obscures the underlying identity of subject and object, nonphilosophical sciences do not simply turn away from subject−object identity. In fact, the three positive sciences correspond to the three most general ways that subject−object identity is expressed: however obscurely, medicine, the science of the organism, attends to nature, or objective subject−object identity; law, the science of right, attends to the human spirit, or subjective subject−object identity; and theology, the science of divine being, attends to the absolute identity of subject and object, their point of utter indifference (SW I/5: 283; 1966: 79). Yet, importantly, these positive sciences are partial in their knowledge, for they only focus on a single aspect of subject−object identity. Whereas philosophical science presents the structure of reason in its entirety, with attention paid to nature, spirit, and their absolute indifference, medicine, law, and theology focus only on particular aspects of reason. Philosophy, then, is unlike the positive sciences – and unlike all other sciences – insofar as it has in view reason as a whole. It would be a mistake, however, to think that the superior status of philosophy as a rational, holistic science sets it apart from the other sciences as entirely other than them. For, in fact, the superiority of philosophy is inseparable from its character as encompassing the knowledge of all other science. Indeed, this is central to the holism of philosophical thinking. Put another way, for Schelling, what goes on within the non-philosophical sciences is – in some important sense – an activity that belongs to philosophical knowledge.

278  Benjamin Berger For this reason, Schelling frequently discusses the relationship between philosophy and the other sciences in terms of the relationship between a whole and its parts: ‘since all knowledge is one, and every branch of knowledge is a part of the whole, all sciences and kinds of knowledge are parts of philosophy’ (SW I/5: 218; 1966: 12). This formulation can be misleading if we imagine that philosophy – the whole – is a mere aggregate of all particular forms of knowledge (SW I/5: 230, 1966: 24). Instead, Schelling understands the whole to precede its parts. This does not mean that all sciences have their historical origin in philosophy, as if, empirically speaking, every form of non-philosophical knowledge arises from a more original, philosophical form of knowledge. Rather, according to Schelling, particular forms of knowledge prove to be genuine forms of knowledge only insofar as they are set into intelligible relation to one another by philosophy. In other words, philosophy provides the universal perspective within which any particular form of knowledge is made fully intelligible. For instance, specific facts about plant life become fully intelligible only when they are shown to be rationally connected to other features of organic existence and to the universe more generally. To clarify this point, let us take as an example the relationship between jurisprudence and philosophy. The study of law is the study of the empirical instantiation of reason, or subject−object identity, in the history of political institutions. It is thus concerned with the same essential thing with which philosophy is concerned, namely reason or subject−object identity. But jurisprudence is only concerned with this identity under one aspect – namely under its subjective aspect. The study of law focuses exclusively on humanity; there is no consideration here of nature or God. By contrast – and as we have already seen – philosophy not only involves an uncompromising focus on the rational essence of humanity, apart from its empirical-historical expression; it also considers the human in relation to nature and God. And this means that philosophical knowledge provides the basis for knowledge of law, since the philosophical perspective situates the study of law in its proper place within the whole system of reason. From this perspective, we can see that jurisprudence and other limited sciences ultimately depend upon a form of knowledge that is other than them, namely philosophy: by presenting us with the entire structure of reason – a structure within which each science has its place – philosophy ends up unifying the sciences, organising them into a single, coherent system. 12.4  Philosophical Insight Up to this point, one might think that Schelling is, in large part, following in the footsteps of Fichte, whose understanding of philosophy as the science of all science determines philosophy as the foundation of all

The Science of All Science and the Unity of the Faculties 279 knowledge.14 On a Fichtean account, philosophy should answer the question, ‘how is science possible?’ (FSW I/1: 43; 1988: 105), and it should answer this question in a scientific fashion that will ultimately provide scientific unity to all of the other sciences. Schelling is undoubtedly influenced by this conception of philosophy as science of all science. Not only is he thinking along broadly Fichtean lines when he remarks that philosophical training ought to precede more specialised training in the university (SW I/5: 213, 262; 1966: 7, 56), but he is also endorsing Fichte’s more distinctive conception of philosophy as the science of all sciences to the extent that philosophy, for Schelling, provides rational, systematic unity to those sciences. For both Fichte and Schelling, the intelligibility of all forms of knowledge depends, in some sense, upon philosophical knowledge. Yet, importantly, Schelling also expresses serious reservations about Fichte’s collapse of philosophical science into questions about how knowledge is possible, and this has significant implications for how Schelling ultimately comes to understand philosophy as in fact unifying the sciences. In a letter to Fichte dated November 19, 1800, Schelling conveys his concern about the reduction of philosophy to preparatory, methodological inquiry, remarking that the Wissenschaftslehre is not, in fact, philosophy, and that it is not philosophy because it ‘has nothing to do with reality’ (HKA III/2.1: 280; 2012: 44). Two months later, in On the True Concept of the Philosophy of Nature, he similarly argues that philosophy – the rational investigation into the nature of reality – must be distinguished from the philosophy of philosophy, the latter of which he associates, again, with the Wissenschaftslehre (SW I/4: 85; 2020: 49). In the same text, he complains about those who, following Fichte, think that ‘it is not yet time to speak of a system of philosophy of nature’: ‘how long will we have to wait … for this science to come?’ (SW I/4: 92; 2020: 54).15 This is a running theme in Schelling’s thought. It can even be found in texts that show Schelling at his most similar to Fichte, such as On the Possibility of a Form of All Philosophy, in which we again encounter his frustration with the state of philosophy as merely preparatory or metaphilosophical: ‘May those who seem to have been challenged by philosophy make it their business soon to furnish its completion so that there may be no more need for mere preparations’ (SW I/1: 89; 1980: 40).16 Terry Pinkard helpfully describes Fichte’s science of science as ‘an account of all possible accounts and the ultimate account of making sense of making sense itself’ (Pinkard 2019: 27). From a Schellingian perspective, this an accurate description of Fichte’s conception of philosophy, and it highlights just how far removed from actual knowledge philosophical science becomes in Fichte’s hands. Refusing to discuss what is, philosophy as such is abandoned in the name of more critical, scientific preparations. As George di Giovanni has put it, ‘the Wissenschaftslehre is … essentially a

280  Benjamin Berger science of science rather than of being’ (Giovanni 2005: 38).17 And this, for Schelling, is a grave disappointment. In his own words, this is a form of thinking that ‘never even [leaves] port and whose entire endeavour … consists in not leaving port and in preventing philosophy from ever beginning, by endless philosophising about philosophy’ (SW I/9, 211; 1997: 212).18 As I see it, we need to take into account this general tendency of Schelling’s thought about the nature of philosophy when interpreting his claims about the science of science in the Lectures on the Method of Academic Study. Although he does see the academic study of philosophy as necessary preparation for more specialised forms of study – and although he is certainly interested in the way philosophical thinking grounds other forms of knowledge – focusing on the propaedeutic function of philosophy can obscure the distinctively Schellingian idea of philosophy as science of science. In brief, this idea is that philosophical insight is implicitly at work within all intellectual activity. In this way, philosophy is the truly scientific essence of all science, as opposed to its mere propaedeutic.19 To see this, it may be helpful to think more broadly about the nature of scientific knowledge. All scientific knowledge deals in one way or another with generalities: when we know something, we are not merely empirically familiar with it, even if that empirical familiarity plays an indispensable role in certain sciences. To have knowledge is, in this way, different from being acquainted with something; it is to have some intellectual grasp, however vague, of the essential being of a thing. In other words, knowledge – and this is the case for all science – concerns the kinds of things that there are, not things themselves in their individuality.20 And, as we saw above, the kinds of things that there are, for Schelling, are various configurations of subject−object identity, the essence of all things. All knowledge, then – insofar as it is indeed knowledge – is attentive, to some degree, to the essential being of all things. And in this way all knowledge involves an essentially philosophical insight: by attending to any particular kind of thing, one gains insight into the nature of reason or being as such. We can take the empirical study of nature as an example of science that, while not, strictly speaking, philosophical, nevertheless has something philosophical about it. On the one hand, throughout the Lectures on the Method of Academic Study – and throughout the other texts of this period – Schelling states that the empirical sciences of nature must be guided by a rationalist philosophy of nature. In this way, ‘the empirical theory of nature … must be subordinated to the absolute science of nature, grounded in Ideas’ (i.e., configurations of subject−object identity), and it must even ‘take as its model the kind of construction that characterizes an absolute science’ (SW I/5: 323; 1966: 122). That is to say, the empirical study of nature must be guided by the rational constructions that take place in the philosophy of nature; it ‘must be accelerated onwards’ by philosophy and

The Science of All Science and the Unity of the Faculties 281 the ideas presented in philosophical science (SW I/4: 92; 2020: 54). On the other hand, however, Schelling notes that the history of empirical science involves certain moments in which such construction occurs, ‘as though by instinct’ (SW I/5: 323; 1966: 122). Consequently, according to Schelling, ‘even empiricism, no matter how unclearly, views nature as a whole, every individual element in it determining and being determined by all the others’ (SW I/5: 324; 1966: 122).21 And this is not exclusively because empirical research into nature, under ideal conditions, is guided by a rationalist philosophy of nature; it is also because, in the empirical sciences of nature – which are not properly philosophical – there is an implicit orientation towards, instinctive sense for, or even vague perception of the rational structure of the universe. I take it that this is explained by the fact that even the empirical sciences are forms of scientific knowledge, and scientific knowledge always involves some attention, however imprecise, to generalities. Indeed, for Schelling, empirical knowledge only counts as science because it involves such an orientation to the universal. This is what distinguishes empirical science, in which ‘the particular has value … insofar as it implies the universal and absolute’ (SW I/5: 212; 1966: 6), from mere empirical sensation, which attends to sheer particularity and is for that reason a ‘negation of knowledge’ (SW I/6: 199; 1994: 183). As we saw earlier, all knowledge involves attention to what is – subject−object identity – and, therefore, all knowledge involves philosophical insight. Schelling’s discussion of the activity of the scholar may clarify this idea that philosophical insight can be found within non-philosophical science. A man who confines himself to studying a subordinate discipline, however praiseworthy his industry may be, cannot be expected to rise to the vision of knowledge as an organic whole. This vision can be found only in the science of all sciences, in philosophy. (SW I/5: 214; 1966: 8) Philosophical knowledge has the whole in view, and someone who studies another discipline – it seems – simply lacks that philosophical knowledge. And yet, [t]he more a scholar conceives of his particular domain as an end in itself, even making it, as far as he is concerned, the central point of all knowledge – hoping to expand it into an all-encompassing totality – the more he is striving to express universals and Ideas in it. (SW I/5: 232; 1966, 25)22 There is a difference, then, between the scholar who confines himself to the particularity of his discipline and the scholar who conceives his discipline

282  Benjamin Berger as expressive of something universal. The latter is involved in a process Tilottama Rajan calls ‘the absolutization of the particular’ (Rajan 2012: 49). And it is in this process that, it seems to me, we can again see something philosophical implicitly at work within non-philosophy. Note that the scholar who strives to express universals in his thinking does not, according to Schelling, begin with foundational philosophical principles that allow him to subsequently come to an understanding of a given subject matter, itself a step (or more) removed from philosophy. On the contrary, the scholar treats his subject matter as ‘the central point of all knowledge’. But when executed properly, attention to one’s particular field actually involves an orientation to the whole, i.e., an orientation that does not treat the particular as particular but as an expression of some more essential, universal law, namely the law of reason – A = A – which recurs throughout all being and all instances of knowledge (SW I/4: 116; 2012: 147). For this is what the particular truly is: an expression of subject−object identity. To see one’s own field of inquiry in relationship to the whole of scientific knowledge is not, therefore, to see one’s own field as incomplete or as fundamentally lacking in genuine insight; on the contrary, Schelling argues, it is to see one’s own field as ‘an end in itself, an absolute’ (SW I/5: 232; 1966: 25), since every element of the whole proves to be absolute itself once it is grasped in its essential being, subject−object identity. In this way, philosophical insight into the nature of things can be understood to be immanent to the various sciences. For, according to Schelling, the presentation of the absolute within a particular is philosophy (SW I/5: 107; 1989: 282). It should be clear from the passages cited above that Schelling thinks the scholar can be better or worse at ‘absolutizing the particular’, to use Rajan’s phrase; indeed, much scholarly activity is lacking any explicit reference to anything universal, and so it is important to see it as largely nonphilosophical. The degree to which philosophical insight is explicit will therefore vary from case to case. What I’d like to suggest, however, is that, while not all intellectual activity is properly philosophical for Schelling, it is all philosophically inspired. Since all science involves some recognition – even if this is largely unconscious – of some aspect of reason or subject−object identity, there is an essential philosophical insight within each and every form of knowledge. Even the scholar who ‘cannot be expected to rise to the vision of knowledge as an organic whole and is not, strictly speaking, a philosopher’ (SW I/5: 214; 1966: 8), nevertheless, has an oblique view on what is universally: reason or subject−object identity. And because all sciences involve some view, however obscure, of what is, they are unified by that philosophical insight. This, I take it, is one thing Schelling has in mind when he claims that ‘all the sciences are one in and through philosophy’ (SW I/5: 261; 1966: 56).

The Science of All Science and the Unity of the Faculties 283 It is not accidental that this way of describing genuine knowledge – i.e., as the method in which a person focuses on something particular and considers how that particularity is connected to and, moreover, is an expression of something universal – is precisely how Schelling describes intellectual intuition during this period of his thought: to see the whole in the part or ‘the universal in the particular’ (Schelling, SW I/4: 362; 2012: 206), a process without which philosophy would not be possible (SW I/5: 255; 1966: 49). To think philosophically is precisely to see that there is subject−object identity in a particular kind of being (SW I/5: 234; 1966: 122). As we saw earlier, reaching the epistemic standpoint of subject−object identity, or intellectual intuition, requires that one has abstracted from consciousness and sunk into the structure of reason as such. This is why, in the 1821 lecture, On the Nature of Philosophy as Science, Schelling describes intellectual intuition in terms of ecstasy and as a process of self-abandonment (SW I/9: 229; 1997: 228).23 It is from this standpoint – not that of consciousness but of reason or subject−object identity – that philosophy happens and the universal law of identity is glimpsed in the particular. According to Schelling, one cannot acquire this philosophical ability to intuit the universal; it must be within a person from the start. And since intellectual intuition is a natural talent, philosophy should not be expected to offer a ladder to non-philosophical consciousness: there is no path that will lead one from the non-philosophical, ordinary experience of subject−object difference to the philosophical, intellectual vision of subject−object identity (SW I/4: 361−362; 2012, 206).24 This idea is, of course, famously mocked by Hegel (HW 3: 31; 1977, 16). But Hegel is not alone in being critical of the apparent esotericism of Schelling’s intuitionism. For instance, in his review of the Lectures on the Method of Academic Study, Schleiermacher criticises Schelling for developing a ‘one-sided polemic’ against everything in life that is not truly scientific (Schleiermacher 2002: 484), as if the non-scientific or non-philosophical elements of life were simply undeserving of philosophical attention. There is undoubtedly something compelling about such criticisms, especially when we consider the pedagogical context in which Schelling presented his Lectures on the Method of Academic Study. Yet as a counterpoint to his critics, the very fact that we find the ‘instinct’ to philosophise in nonphilosophical science (SW I/5: 323; 1966: 122) suggests that Schelling has a relatively expansive view regarding who may prove to think philosophically and how such philosophical thinking might begin. Something philosophical gets underway whenever a person, in whatever epistemic situation, sees the universal law of reason operating in a particular instance. No ladder to philosophy is required, then, because philosophical insight is immanent to all knowledge.

284  Benjamin Berger In the fragment, On the Essence of German Science, Schelling identifies the ability to intuit the universal in the particular as the condition of metaphysics (SW I/8: 9). In Schelling’s hands, therefore, “metaphysics” cannot be reduced to “first philosophy” in the manner of Descartes and Fichte, where this discipline provides the foundational principles upon which other forms of knowledge can be secured.25 Although he does remain committed to some version of this Cartesian-Fichtean conception of philosophy, Schelling also distances himself from what he sees as the tendency of modern philosophy, as the provider of first principles, to devolve into metaphilosophy, i.e., into a reflection upon what knowledge is rather than remaining a striving for knowledge as such.26 Philosophy, for Schelling, is indeed the science of all science, and this has everything to do with the fact that philosophy unifies the other sciences. Yet philosophy is not “science of science” in the sense that it provides us with a scientific account of what science is or how all sciences must operate in order to be scientific. Philosophy is, to put it colloquially, knowledge of what matters most, not merely because it is knowledge of the necessary conditions of all other knowledge, but because it is knowledge of what is most true, namely reason or subject−object identity.27 All other sciences prove to be sciences and prove their unity with one another to the extent that there is philosophical insight at the heart of their intellectual activity. It is on this account that all sciences are indebted to philosophy, the supreme science and essence of all knowledge. Consequently, according to Schelling, the university does not become properly philosophical by merely demanding that students undergo a preparatory study of philosophical principles upon which their other studies will rely.28 An institution devoted to the cultivation of knowledge requires the ‘communal spirit’ (SW I/5: 230; 1966: 24) constitutive of a ‘free association [Verein]’ (SW I/5: 284), which only comes from the philosophical insight into the unity of all things. Since the unification of the sciences is only realised when the sciences associate with one another in a free and philosophically inspired manner – and since the structure of the university ought to mirror the structure of knowledge – it is no wonder that Schelling believes philosophy should not be set apart from the sciences as a separate faculty (or, presumably, a department within the faculty of the arts [SW I/5: 284; 1966: 79−80]). Yet we should not lose sight of the fact that while philosophy, for Schelling, should not be granted a faculty of its own – and while philosophical knowledge is an essential feature of all knowledge – philosophy is nevertheless a science of its own, precisely insofar as it is the science of all sciences. The integrity of philosophy is therefore not lost in its dispersal among the sciences; on the contrary, such dispersal is central to the distinctiveness of what philosophy is.

The Science of All Science and the Unity of the Faculties 285 12.5  Conclusion: Schelling after Kant In Section 12.2, I suggested that Schelling’s conception of philosophy as a discipline distinct from the other liberal arts and sciences can be seen as a moment in the history of modern European thought in which the term “philosophy” comes to take on a more determinate meaning than it had previously. I also suggested that, in making this move, Schelling can be understood to be following Kant’s idea of “pure philosophy”, where “purity” marks a difference between that which is properly and that which is improperly philosophical. One of the things I have sought to show here is that, while Schelling is committed in this manner to the specificity of philosophy, he does not set philosophy apart from other forms of knowing, such as empirical science. Instead, he claims that such impure forms of knowing are implicitly philosophical. This, I have argued, is central to the distinctively Schellingian conception of the way philosophy realises the unity of the sciences. Yet, despite his idea that philosophy is at work within all intellectual activity, Schelling does not simply expand the Kantian conception of pure philosophy, such that it becomes a more inclusive category – one, for instance, approaching the more traditional conception of philosophy as comprehending all liberal arts and sciences. In fact, while Schelling understands philosophical insight to animate all genuine thought, he also goes further than Kant in narrowing the scope of what counts as properly or purely philosophical. We already hit upon this in an implicit way when we saw Schelling criticise Fichte’s conception of philosophy as a scientific investigation into the necessary principles of any science. Such epistemological concerns are not, for Schelling, properly philosophical. Philosophical insight is insight into what there is, and one is led astray by lingering upon questions regarding how knowledge is possible. Unlike Kant, then, for whom pure philosophy is comprised of both (i) a critique of the faculty of reason and (ii) a systematic science of reason (itself comprised of both a metaphysics of nature and a metaphysics of morals) (A841/B869), Schelling does not see the former as being properly philosophical. One who wishes to highlight Kant’s interest in not only criticising metaphysics but also grounding it might emphasise the continuity between Kant’s broader project and Schelling’s understanding of the nature of philosophy: what Kant calls the metaphysics of nature and morals is, one might think, not far from philosophy as understood by Schelling. But, importantly, from Schelling’s perspective, critique is not philosophy, and philosophy is not dependent upon critique.29 In fact, for the Schellingian, critique inhibits the possibility of genuine philosophical thinking, because it investigates the faculty of reason and, as a result, all knowledge acquired on the basis of that critique will be limited

286  Benjamin Berger to the perspective of mind or merely subjective subject−object identity. Philosophy proper, by contrast, must abstract away from that standpoint; it must ‘forget the [merely] subjective element’ of reason (SW I/4: 116; 2012, 146), i.e., bracket the existence of reason as a faculty, in order to think from the perspective of reason as such, which is not a mere faculty of the mind but is being itself. Thus, against Kant and Fichte, the mature Schelling develops a conception of philosophy that we might identify as exclusively metaphysical – lacking any critical component and, moreover, opposed to all attempts at transcendental grounding.30 In addition to highlighting the way Schelling’s conception of philosophy ultimately works against critical philosophy in the service of a more traditionally metaphysical form of thinking, the foregoing also puts us in a position to briefly revisit his more subtle departure from Kant on the institutional practice of philosophy. Kant and Schelling share the aim of securing the autonomy of philosophical research and distinguishing philosophical activity from that of the higher faculties by identifying the source of philosophical knowledge as reason. Yet Kant ties the autonomy of philosophy (in the broad sense) to the legitimacy of conflict between philosophy and other forms of knowing. For Kant, it is a good thing that philosophy can challenge the findings of medicine, law, and theology. Such conflict is necessary if philosophical reason is to be allowed to reorient the other sciences when they go astray. The traditional notion that philosophy is the handmaid to theology is thus correct, Kant suggests, but philosophy is a handmaid that leads the way with a torch (KGS I/7: 28; 1996: 255). Philosophy does not follow the higher faculties; it redirects them when they get things wrong. The conflict of the faculties therefore establishes reason’s unique role in the university as the guide of all science. For Schelling, it is not the conflict but the unity of the faculties that guarantees the influence of reason.31 He thus seeks to secure the autonomy of philosophy by arguing that philosophy is the essence of all other forms of knowing, their point of absolute unification. And while it is certainly difficult to imagine today’s universities becoming Schellingian in this respect, there is nevertheless something that Schelling helps us to see more clearly about this fragile institution: if the university is indeed going to be a university, we must recognise that the ever-increasing compartmentalisation of the sciences is detrimental to knowledge (SW I/5: 227; 1966, 21). Schelling understood that there must be a way to grant the particular sciences the freedom to specialise without leading to their utter disconnection, and that one way of achieving the right balance between autonomous scientific pursuit and the unity of all knowledge may be to recognise that there is universal philosophical insight at the heart of every instance of genuine thinking.32

The Science of All Science and the Unity of the Faculties 287 Notes 1 F.W.J. Schelling, Sämmtliche Werke, Division I, Volume 5: 230; Schelling 1966: 24. Further references to Schelling’s Sämmtliche Werke appear as SW, followed by division number, volume number, and page number, e.g., SW I/5: 230. References to Schelling’s Historisch-kritische Ausgabe appear as HKA followed by series, volume, and page number. Further abbreviations appear as follows: Kant’s Gesammelte Schriften – KGS; Fichte’s Sämmtliche Werke – FSW; Fichte’s Gesamtausgabe – FG; Hegel’s Werke in 20 Bänden – HW. 2 On the connections between the series of Jena lectures delivered by Schiller, Fichte, and Schelling, see Ziolkowski 1990: 237−252 and Ziolkowski 2004: 16−19. 3 C.P. Snow’s discussion of the “two cultures” is, in this respect, as relevant today as it was in 1959. See Snow 1998. 4 Indeed, from an economic-materialist perspective, one might emphasise the fact that, when the philosophers of this period entered into public conversations about the university, they were doing so in the hopes of securing their livelihoods. 5 As Michael Hofstetter argues, Wolff and C.G. Heyne play important roles in the lead-up to this classical defence of the university (Hofstetter 2001: 23−24). It is also worth noting that, although the emphasis on the inherent value of knowledge is a major theme of the period, it is by no means novel; it is, rather, a reaction to what was perceived as the overly utilitarian character of the modern Baconian research programme and an attempt to return to an ancient Greek conception of knowledge and its relationship to practical activity. Cf. Jonas 2001. 6 Significantly, however, Schelling rejects the critical component of what Kant calls “pure philosophy”. This is a third point to which I will return, and it is closely related to Schelling’s transformation of Fichte’s conception of philosophy as science of all science. 7 I do not explore it here, but there is a significant connection between Schelling’s comments about the ultimate unity of the faculties, on the one hand, and his understanding of the relationship between theoretical and practical knowledge, on the other. As Derrida notes, establishing institutional unity (as opposed to making room for conflict between the faculties) can be understood in light of Schelling’s more fundamental critique of Kantian philosophy and its apparent inability to conceive reason as properly unified (Derrida 2004: 72−76). 8 See, for instance, SW I/5: 279−280; 1966: 75; SW I/3: 629; 1978: 232−233; SW I/7: 142; 1984: 245; and SW I/8: 200; 2000: xxxv. 9 See Whistler (2013) for a defence of the idea that ‘from a Schellingian point of view … there is a fundamental equality between all the sciences, including philosophy’ and that ‘there is no reason to prefer one science to another’ (Whistler 2013: 232). 10 The line of thought explored here has significant implications for Schelling’s difference from Kant and, more specifically, for the way Schellingian metaphysics must eschew the critical project that Kant understood to be required to rationally justify metaphysics. I briefly return to this in the concluding section. 11 The type of knowledge Schelling has in mind is closely related to what Josef Pieper describes in his essay on leisure and philosophy: this is not the activity of intellectual labour but, rather, a kind of vision that follows a passive moment

288  Benjamin Berger of letting go of oneself (Pieper 1998: 32), and thus a form of knowledge that, in some sense, allows the philosopher to transcend the human perspective (Pieper 1998: 12). 12 In 1804, Schelling puts the point as follows: ‘it is not me who recognises [the] identity [of subject and object], but [this identity] recognises itself, and I am merely its organ’ (SW I/6: 143; 1994: 144). 13 For an invaluable account of the method involved in this strictly rational enterprise, see Breazeale 2014. 14 To be sure, Schelling’s method of abstracting from consciousness in order to begin philosophising is significantly different from Fichte’s own method of abstraction, and this is connected to a major difference between their respective conceptions of the nature of reason and their disagreement about whether reason can be identified with the “I”. Although I do not consider these issues here, they are closely related to the discussion that follows. On the difference between Schelling and Fichte on abstraction, see Berger and Whistler 2020: 162−183. 15 One might think that Schelling has moved too quickly here, from insisting on the importance of a philosophy of reality or a philosophy of what actually is to insisting on the importance of a philosophy of nature in particular. Why, for instance, couldn’t a philosophy of what is begin with a philosophy of spirit or mind? And is this not, from a certain perspective, what Fichte does? Perhaps. But for Schelling, naturalistic explanations are more basic than explanations that refer to the mind, and they are necessary for theoretical philosophy to get off the ground (SW I/4: 76−78); a system that begins with spirit, according to Schelling, will not only mistakenly privilege practical reason, but it will also tend to prioritise epistemological and metaphilosophical concerns. 16 It should not be overlooked that, despite his expressed eagerness to be done with preparations in On the Possibility of a Form of All Philosophy, this text also shows Schelling at his most similar to Fichte. Indeed, it is here that he identifies the ‘propaedeutic of philosophy’ – of which he is admittedly growing tired – with ‘science of all science’, ‘primordial science [Urwissenschaft]’, and ‘science kat’ exochen’, which ‘is supposed to condition all the other sciences’ (SW I/1: 92; 1980: 41−42). In the years that follow, Schelling gradually disentangles the ‘science of all science’ and ‘science kat’ exochen’ from the propaedeutic function of the Wissenschaftslehre – all of which are identified in 1794. Thus, in the letter of November 19, 1800, while Schelling commits himself to the task of philosophy as such and not philosophy of philosophy, he also continues to associate the latter with ‘science kat’ exochen’ (HKA III/2.1: 280; 2012: 44). By 1821, in On the Nature of Philosophy as Science, this is no longer the case: the idealism of the Wissenschaftslehre is not, and cannot possibly be, science par excellence (SW I/9: 211; 1997: 212). 17 Emphasis modified. 18 This explains why Schelling’s own texts of the period do not typically involve lengthy introductory material but, instead, enter straightaway into discussing the nature of things. His 1804 Propaedeutic of Philosophy might be seen as an exception, but in another sense it proves the same point: Schelling is clear in the Propaedeutic that such intellectual work is not philosophy; it is a ‘purely negative preparation for philosophy’ (SW I/6: 73). 19 To be clear, what I am describing as the ‘merely’ propaedeutic function of philosophy – the importance of which, again, Schelling acknowledges, despite his worries about what he perceives as Fichte’s overemphasis on it – does not

The Science of All Science and the Unity of the Faculties 289 necessarily put philosophy into the service of sciences presumed to be superior to it. As Paul Ziche has argued, there is a significant difference between the way Schelling, following Fichte, conceives philosophy as introduction to the other sciences and the institutional conception of philosophy prevalent at the turn of the century (Ziche 2008: 148, 167−168; Ziche 2011: 13−14). According to the latter, the lower faculty of philosophy provides students with the opportunity to pursue a basic preparation for scientific learning, beginning with the most ordinary and, indeed, simple ideas in order to make possible the comprehension of more complicated ideas. By contrast, Schelling – and here, again, he is thinking along broadly Fichtean lines (Ziche 2008: 162−163) – conceives philosophy as the distinctive science that provides students with the highest principles of all knowledge and, in this way, realises the ‘systematic groundwork and foundation’ of the sciences (Ziche 2008: 150). What I argue in the following is that, in addition to conceiving philosophy as foundational in this unique sense, Schelling also seeks to develop a novel understanding of how philosophy in fact unifies the other sciences. And this unification, it seems to me, involves the idea that philosophy is something more than the foundation of knowledge – even in this special, Fichtean sense. From this perspective, although the Fichtean conception of philosophy as foundation is certainly different from the institutional conception of philosophy as merely introductory, the two similarly identify philosophy as the preparatory ground for knowing. 20 It is only in his late thought that Schelling comes to see this exclusive attention to kinds of things – and the lack of attention to individuals in their sheer existence – as a limit to rational science that must be overcome by a non-rationalist form of philosophy. 21 Emphasis modified. 22 The point about scholarship is also directly related to pedagogy for Schelling: ‘He who knows his particular profession only as a specialty, who is unable to see what is universal in it or to express it in terms of universal scientific culture, is unworthy to be a teacher and custodian of science. He may be useful in various ways – as a physicist in the erection of lightning rods, as an astronomer in composing almanacs, as a physician in utilising galvanism for therapeutic purposes, etc. – but the vocation of teacher requires more than mechanical skills’ (SW I/5: 231; 1966: 24−25). 23 ‘What man needs is not to be placed inside himself, but outside himself’ (SW I/9: 230; 1997: 229). Contrary to Fichte, for whom abstraction from experience bumps up against a fundamental limit, namely, the ‘pure I’ (FG II/3: 329; 1988: 204), Schelling argues that ‘the only thing from which we cannot abstract … is … absolute identity itself’ (SW I/6: 146−147; 1994: 147). 24 The cultivation of philosophical ability can be hindered, however, if students are encouraged to think non-philosophically (SW I/5: 236; 1966: 30). There is consequently one way in which students can be helped to think philosophically: by demonstrating to them how philosophy differs from other ways of knowing (SW I/5: 248; 1966: 42). 25 As Dalia Nassar has shown, the difference between Fichte and Schelling on the nature of philosophy can already be seen in 1794, in the otherwise quite similar On the Concept of the Wissenschaftslehre and On the Possibility of a Form of All Philosophy. Nassar argues that, for the young Schelling, the fundamental principle of science is unconditional not because it is certain, as it is for Fichte, but because it is self-causing. In this way, Schelling is – again, already in 1794 – more of a Spinozist than a Fichtean. See Nassar 2013: 167−170.

290  Benjamin Berger 26 According to Schelling, that philosophy is love of knowledge implies that, even if it is the absolute and primordial science (Urwissenschaft), it is not primordial knowledge (Urwissen): it is the ‘striving to participate in primordial knowledge’ (SW I/5: 218; 1966: 12; emphasis modified). See also SW I/5: 254−255, 280; 1966: 48, 75. Understandably, this has suggested to many readers of Schelling that even philosophy – universal as it is – is somehow limited with respect to that which it knows, as if the unity of all sciences is in some way dependent upon a form of scientific knowledge that must remain incomplete. For an account of philosophy as science of science that emphasises the idea that ‘the systematic unity of knowledge can never be reached’, see Steigerwald 2020: 39. 27 From this perspective, the law of identity is better understood as the essence of all things and all knowledge, as opposed to a first principle or foundation of knowledge. Indeed, it is because the law of identity is essential that it can operate as the ultimate explanation of everything that is and everything that is known. I take it that this slight shift in emphasis – from conceiving the principle of philosophy as a foundation to conceiving it in terms of essence – is part of Schelling’s more general project of developing a form of idealism that is not transcendental in orientation. 28 See Ziche 2008. 29 See García-Romero 2011: 144, for an excellent account of Schelling’s subsequent embrace of critique and ‘radicalization of Kantian criticism’ by way of Aristotle. 30 Prior to 1801, of course, Schelling remains interested in various forms of transcendental argumentation. Yet in my view, after the 1800 System of Transcendental Idealism – and prior to his late system – he fully abandons this continued (although already subdued) concern for critique, such that, as Nectarios Limnatis remarks, he can return unambiguously to a ‘metaphysics of a pre-Kantian type’ (Limnatis 2008: 127). 31 Derrida suggests that even this departure from Kant is, in an important sense, Kantian, since there are two sides to Kant’s understanding of the conflict of the faculties: ‘There is the Kant … who wants to bring a department of philosophy into existence and to protect it (in particular from the State). In order to protect it, one must delimit it. And then there is the Kant who grants the Faculty of Philosophy the right of critical and panoptical supervision over all the other departments, in order to intervene in them in the name of truth’ – i.e., as a science that is not set apart from the others but ‘is in fact everywhere’ and in some sense encompasses them all (Derrida 2004: 72). 32 I would like to thank the editors and the other contributors to this volume for their incredibly helpful feedback at various stages of my writing this chapter.

Bibliography Berger, Benjamin and Daniel Whistler (2020) The Schelling-Eschenmayer Controversy, 1801: Nature and Identity (Edinburgh: Edinburgh University Press) Breazeale, Daniel (2014) ‘“Exhibiting the Particular in the Universal”: Philosophical Construction and Intuition in Schelling’s Philosophy of Identity (1801–1804)’. In Interpreting Schelling: Critical Essays, edited by Lara Ostaric (Cambridge: Cambridge University Press), pp. 91–119 Derrida, Jacques (2004) Eyes of the University: Right to Philosophy 2. Translated by Jan Plug and Others (Stanford: Stanford University Press)

The Science of All Science and the Unity of the Faculties 291 di Giovanni, George (2005) “Hegel’s Anti-Spinozism: The Transition to Subjective Logic and the End of Classical Metaphysics”. In Hegel’s Theory of the Subject, edited by David Gray Garlson (Baskingstoke: Palgrave Macmillan) Fichte, J.G. (1845–1846) Sämmtliche Werke. Edited by I. H. Fichte (Berlin: Veit) Fichte, J.G. (1971–) Gesamtausgabe. Edited by the Bayerischen Akademie der Wissenschaften (Stuttgart: Frommann-Holzboog) Fichte, J.G. (1988) Fichte: Early Philosophical Writings. Translated and edited by Daniel Breazeale (Ithaca and London: Cornell University Press) Fichte, J.G. and F.W.J. Schelling (2012) The Philosophical Rupture between Fichte and Schelling. Edited and translated by Michael G. Vater and David W. Wood (Albany: SUNY Press) García-Romero, Marcela (2011) “Schelling’s Late Negative Philosophy: Crisis and Critique of Pure Reason”, Comparative and Continental Philosophy 3.2, pp. 141–164 Hegel, G.W.F. (1977) Phenomenology of Spirit. Translated by A.V. Miller (Oxford: Oxford University Press) Hegel, G.W.F. (1986) Werke in 20 Bänden. Edited by Eva Moldenhauer and Karl Markus Michel (Frankfurt am Main: Suhrkamp) Hofstetter, Michael J. (2001) The Romantic Idea of a University: England and Germany, 1770–1850 (Basingstoke: Palgrave) Jonas, Hans (2001) “The Practical Uses of Theory”. In The Phenomenon of Life: Toward a Philosophical Biology (Evanston: Northwestern University Press), pp. 188–210 Kant, Immanuel (1900–) Gesammelte Schriften. Edited by the Royal Prussian Academy of Sciences (Berlin: de Gruyter) Kant, Immanuel (1996) The Conflict of the Faculties. In Religion and Rational Theology, translated by Mary J. Gregor and Robert Anchor (Cambridge: Cambridge University Press) Kant, Immanuel (1998) Critique of Pure Reason. Translated by Paul Guyer, and Allen W. Wood (Cambridge: Cambridge University Press) Kant, Immanuel (2004) Metaphysical Foundations of Natural Science. Translated by Michael Friedman (Cambridge: Cambridge University Press) Limnatis, Nectarios G. (2008) German Idealism and the Problem of Knowledge: Kant, Fichte, Schelling, and Hegel (New York: Springer) Menand, Louis, Paul Reitter, and Chad Wellmon (2017) The Rise of the Research University: A Sourcebook (Chicago and London: University of Chicago Press) Nassar, Dalia (2013) The Romantic Absolute: Being and Knowing in Early German Romantic Philosophy, 1795–1804 (Chicago: University of Chicago) Pieper, Josef (1998) Leisure: The Basis of Culture. Translated by Gerald Malsbary (South Bend, IN: St. Augustine’s Press) Pinkard, Terry (2019) “German Idealism: The Thought of Modernity”. In The Cambridge History of Modern European Thought, Volume I: The Nineteenth Century, edited by Warren Breckman and Peter E. Gordon (Cambridge: Cambridge University Press) Purvis, Zachary (2016) Theology and the University in Nineteenth-Century Germany (Oxford: Oxford University Press)

292  Benjamin Berger Rajan, Tilottama (2012) “Excitability: The (Dis)Organization of Knolwedge from Schelling’s First Outline (1799) to Ages of the World (1815)”. In Romanticism and Modernity, edited by Thomas Pfau and Robert Mitchell (London and New York: Routledge), pp. 47–63 Schelling, F.W.J. (1856–1861) Sämmtliche Werke. Edited by K.F.A. Schelling (Stuttgart and Augsberg: Cotta) Schelling, F.W.J. (1976–) Historisch-kritische Ausgabe, edited by the Schelling Commission of the Bavarian Academy of Sciences (Stuttgart: Frommann-Holzboog) Schelling, F.W.J. (1978) System of Transcendental Idealism. Translated by Peter L. Heath (Charlottesville: University Press of Virginia) Schelling, F.W.J. (1980) On the Possibility of a Form of All Philosophy. In The Unconditional in Human Knowledge: Four Early Essays (1794–1796), translated by Fritz Marti (Lewisburg: Bucknell University Press) Schelling, F.W.J. (1984) Aphorisms as an Introduction to Naturphilosophie, translated by Fritz Marti. In Idealistic Studies 14.3, pp. 344–358 Schelling, F.W.J. (1994) System of Philosophy in General and of the Philosophy of Nature in Particular, translated by Thomas Pfau. In Idealism and the Endgame of Theory (Albany: State University of New York Press) Schelling, F.W.J. (1996) On University Studies. Translated by E.S. Morgan and edited by Norbert Guterman (Athens, OH: Ohio University Press) Schelling, F.W.J. (1997) On the Nature of Philosophy as Science, translated by Marcus Weigelt. In German Idealist Philosophy, edited by Rüdiger Bubner (Harmondsworth: Penguin), pp. 210–243 Schelling, F.W.J. (2000) The Ages of the World. Translated by Jason Wirth (Albany: State University of New York Press) Schelling, F.W.J. (2020) On the True Concept of Philosophy of Nature, translated by Judith Kahl and Daniel Whistler. In The Schelling-Eschenmayer Controversy, 1801: Nature and Identity (Edinburgh: Edinburgh University Press) Schleiermacher, F.D.E. (2002) Rezension von Friedrich Wilhelm Joseph Schelling: Vorlesungen über die Methode des akademischen Studiums. In Kritische Gesamtausgabe, Division 1, Volume 4: Schriften aus der Stolper Zeit (1802–1804), edited by Eilert Herms, Günter Meckenstock, and Michael Pietsch (Berlin and New York: de Gruyter) Snow, C.P. (1998) The Two Cultures (Cambridge: Cambridge University Press) Steigerwald, Joan (2020) “Schelling’s Romanticism: Traces of Novalis in Schelling’s Philosophy”. In Schelling’s Philosophy: Freedom, Nature, and Systematicity, edited by G. Anthony Bruno (Oxford: Oxford University Press) Whistler, Daniel (2013) Schelling’s Theory of Symbolic Language: Forming the System of Identity (Oxford: Oxford University Press) Ziche, Paul (2008) “Philosophie als Propädeutik und Grundlage akademischer Wissenschaft: Schellings Vorlesungen über die Method des akademischen Studiums im Kontext der Universität Jena um 1800”. In Philosophie und Gestalt der Europäischen Universität: Akten der Internationalen Fachtagung Budapest, vom 6–9. November 2003, edited by István M. Fehér and Peter L. Oesterreich (Stuttgart-Bad Cannstatt: Frommann-Holzboog), pp. 147–168

The Science of All Science and the Unity of the Faculties 293 Ziche, Paul (2011) “Die Welt der Wissenschaft im Innersten erschüttern”: Schellings Vorlesungen über die Methode des akademischen Studiums als philosophisches Programm zur Wissenschaftsorganisation”. In “Die bessere Richtung der Wissenschaften”: Schellings “Vorlesungen über die Methode des akademischen Studiums” als Wissenschafts- und Universitätsprogramm, edited by Paul Ziche and Gian Franco Frigo (Stuttgart-Bad Cannstatt: Frommann-Holzboog), pp. 3–21 Ziolkowski, Theodore (1990) German Romanticism and Its Institutions (Princeton, NJ: Princeton University Press) Ziolkowski, Theodore (2004) Clio, the Romantic Muse: Historicizing the Faculties in Germany (Ithaca: Cornell University Press)

13 Two Models of Critique of Metaphysics Kant and Hegel Dietmar H. Heidemann

13.1 Introduction The aim of this chapter is to assess two paradigmatic models of critique of metaphysics, i.e., as I will call them, Kant’s “capacities first” model and Hegel’s “thought determination” model. Although their models are fundamentally different, both, Kant and Hegel, nevertheless dismiss the same type of metaphysical claims for very similar reasons. Whereas Kant concludes from his criticism that metaphysics is not possible as a science, Hegel explicitly reinvents metaphysics as science, but as a new type of metaphysics. The comparison of both models reveals systematic insights into the different kinds of reasons that are given for calling the scientific status of metaphysics into question in Classical German Philosophy and whether those reasons are compatible. The issue of the compatibility of both models is particularly revealing. For in his critique of classical rationalists such as Leibniz and Wolff, Hegel praises Kant for having identified the cardinal mistake of traditional metaphysics. Kant demonstrates, as Hegel concedes, that the classical metaphysical principle, tacitly presupposed by such thinkers, according to which the principles of thought are the principles of being in general, is a dogmatic presumption and of no philosophical use whatsoever. According to Kant, the possibility of cognition and henceforth the possibility of metaphysics as science depends on the critical analysis of what human cognisers are able to cognise in general, i.e., on the prior critical analysis of cognitive capacities. Although Hegel accepts Kant’s critique of the aforementioned principle, he dismisses Kant’s transcendental account of cognitive capacities for depending on subjectivist presuppositions. He also thinks that in the end rationalist metaphysics, too, relies on such subjectivist and even psychologist assumptions. The grounds of his critique of metaphysicians such as Leibniz and Wolff must therefore be fundamentally different from Kant’s.

DOI: 10.4324/9781003187561-14

Two Models of Critique of Metaphysics 295 In Section 13.2, I shall show what Kant’s “capacities first” model consists in, and that Kant’s critique of metaphysics follows from his analysis of what human cognisers can know in general. In Section 13.3, I turn to Hegel’s “thought determination” model, i.e., the view that classical metaphysics goes wrong about its objects while it is at the same time possible to arrive at metaphysical truths on a purely conceptual basis by overcoming the Kantian model. Kant’s “capacities first” model and Hegel’s “thought determination” model represent competing strategies of criticising metaphysics. By “critique of metaphysics”, I understand the investigation of the presuppositions, methods and resulting claims that classical metaphysics makes or subscribes to with respect to its objects. The aim of such a critique is not to better understand metaphysics but to examine whether such presuppositions, methods, and claims are justified and how far they reach towards counting metaphysics as science.1 In Section 13.4, I discuss how Kant’s and Hegel’s models relate and specifically whether the “thought determination” model falls victim to the “capacities first” model. For, on the face of it, it looks as if Hegel’s model itself is in essence not much different from methods of classical metaphysics as criticised by Kant, i.e., from amplifying cognition merely on the basis of conceptual analysis. If this would be the case, as I will argue, it would be hard to see how Hegel himself could establish his own metaphysics as science.2 13.2  Kant’s “Capacities First” Model In most general terms, the “capacities first” model3 is the view that cognitive abilities define the possibility of metaphysical cognition. More specifically, this means that the possibility of (human) cognition is essentially determined through the nature and scope of (human) cognitive faculties and their corresponding cognitive capacities, such that, in order to assess the limits of cognition, the nature and range of cognitive faculties must be known first. For Kant there are no more than two cognitive faculties − sensibility and understanding − each of which exhibits its own irreducible cognitive capacity. While sensibility exhibits the capacity to have intuitive or singular representations, the understanding has the capacity to produce conceptual or universal (general) representations. As we will see, the “capacities first” model is indispensable not only for Kant’s transcendental account of cognition but also for the broader question whether metaphysics is possible as a science. The question whether metaphysics is possible as a science allows for four different answers: (1) yes, metaphysics is possible as a science; (2) no, it is not; (3) we cannot know; and (4) the question is meaningless because metaphysics is meaningless. While Hegel, as we will see in Section 13.3,

296  Dietmar H. Heidemann favours answer (1), Kant makes a case for answer (2). However, he does not concur with answer (3), i.e., the sceptic’s response that we cannot know whether metaphysics is a science because we cannot justify our beliefs, nor with answer (4), e.g., the logical positivist’s view that metaphysical claims are nonsensical. Although metaphysics is not possible as a science, metaphysics is not philosophically meaningless, illogical, or even absurd for Kant, especially not with respect to the metaphysics of morals. As such, metaphysics must be conceived, he explains, as ‘the system of pure reason (science), the whole (true as well as apparent) philosophical cognition from pure reason in systematic interconnection’ (CpR A841/ B869). Prima facie, Kant says, metaphysics seems to be a science since it engages with ideas of pure reason that transcend the possibility of experience and aims ‘to expand the domain of our judgments beyond all bounds of experience through concepts to which no corresponding object at all can be given in experience’ (CpR B6). The most prominent of these ‘unavoidable problems of pure reason itself are God, freedom and immortality’ (CpR B7), ideas of pure reason that classical metaphysics claims to be able to apprehend and grasp through mere concepts. Kant dismisses the possibility of cognitively accessing ideas of pure reason. Ideas of pure reason are uncognisable not for sceptical reasons, or because they are empty concepts, but because the critical determination of the ‘sources of metaphysics’ (Prolegomena, Kant 2004: 15, AA 4: 265; cf. CpR AXII)4 reveals that any ‘wholly isolated speculative cognition of reason that elevates itself entirely above all instruction from experience’ is questionable as long as metaphysics is not yet on ‘the secure course of a science’ (CpR BXIV). Kant’s conception of science as the purposeful, systematic “unity” of cognition(s) that is not only a priori and necessary but also certain (cf. CpR A832/B860–A851/B879) may be too narrow as to its scope and as it compares to the contemporary understanding of “science” that usually does not lay claim on apriority or necessity. Kant has, however, a clear idea of the standard based on which it can be decided whether or not metaphysics is a science. The standard of this measurement and evaluation is not the ideal of already established sciences like mathematics, physics, and logic, after which metaphysics should then be modelled. Whether or not metaphysics is a science can only be decided according to the kinds of judgements it essentially proposes and, of course, their justification. Since metaphysics is the ‘speculative cognition of reason … through mere concepts’ (CpR BXIV), its pivotal claims seem to be analytic. On the other hand, metaphysics is meant to be non-tautological since, on Kant’s understanding, metaphysical judgements are supposed to amplify our cognition (a priori). If metaphysical propositions were in fact analytic, metaphysical problems such as the immortality of the soul, the simplicity of the thinking I, or the existence of God could be decided by merely conceptual means

Two Models of Critique of Metaphysics 297 based on the principle of contradiction as the necessary and sufficient criterion of the truth of analytic propositions. This is, however, obviously not the case. For if metaphysical questions could be definitely decided by way of conceptual analysis, why are there still ‘endless controversies’ on the ‘battlefield’ of metaphysics (CpR AVIII), as Kant notes? For one should expect that after more than 2000 years of the history of metaphysics at least some of these problems would have been solved.5 Since metaphysical judgements are not meaningless but cannot be analytic either, they seem to be synthetic (a priori). Now synthetic judgements require a third thing (CpR B194/A155) for their justification which cannot be provided through concepts, for concepts as such are empty and do not provide information about the world. Conceptual relations simply cannot account for the ‘condition of the possibility of things themselves’ (CpR B289). The justifying third thing in synthetic judgements rather is “intuition” and its a priori forms (CpR B195/A155–B197/A158). At this point of the argument, one might object that even if it is true that synthetic judgements cannot be justified by means of abstract concepts, there might be special, non-abstract concepts that are as contentful as intuition and can serve as the justifying third thing for synthetic judgements (a priori). In this case metaphysical judgements (a priori) could be justified and metaphysics be made a science of pure reason on merely conceptual grounds. The “capacities first” model, however, demonstrates that human cognisers cannot have concepts that could possibly serve as a source of metaphysical cognition. The primacy of the exploration of our cognitive faculties even reveals that the crucial reason why it is that metaphysics cannot be a science is grounded in Kant’s cognitive dualism, i.e., the view that the minimum condition for cognition is the cooperation of sensibility and understanding as the two irreducible sources of human cognition. In the “Architectonic” chapter of the Critique of Pure Reason and its discussion of the place of metaphysics within the ‘system of all philosophical cognition’, Kant states explicitly that the determination of the scientific status of any cognition in general and ‘metaphysics’ in particular depends on its decomposition and specifically the isolation of its elements: It is of the utmost importance to isolate [isolieren] cognitions that differ from one another in their species and origin, and carefully to avoid mixing them together with others with which they are usually connected in their use. … One must nevertheless admit that the distinction of the two elements in our cognition, one of which is in our power completely a priori but the other of which can be derived only from experience a posteriori, has remained very indistinct, even among professional thinkers, and hence the determination of the bounds of a special kind of cognition, and thus the genuine idea of

298  Dietmar H. Heidemann a science with which human reason has so long and so intensively occupied itself, has never been accomplished. (CpR A 842/B 870–A 843/B 871) To “isolate” is the methodological keyword that Kant already introduced at the beginning of the Transcendental Aesthetic and the Transcendental Logic. There he implements the method or procedure of isolation to motivate cognitive dualism: the view that cognition is possible only through the cognitive cooperation of intuitive and conceptual representations, that is, through sensibility and understanding and intuition or concept, respectively. The “capacities first model” presupposes cognitive dualism: at the outset of the Transcendental Aesthetic Kant says that ‘we will therefore first isolate sensibility by separating off everything that the understanding thinks through its concepts, so that nothing but empirical intuition remains’ (CpR A22/B36). Corresponding to this, he explains in the introduction to the Transcendental Logic that here ‘we isolate the understanding (as we did above with sensibility in the transcendental aesthetic), and elevate from our cognition merely the part of our thought that has its origin solely in the understanding’ (CpR A62/B87). This dual isolation identifies sensibility and understanding as the sole sources or stems of cognition that must cooperate to make cognition possible. The cooperation of sensibility and understanding as the minimum condition of cognition does not yet follow from the bare fact that they are the sole sources of cognition. That they must work together in order to make cognition possible is rather a consequence from the specific cognitive capacities pertaining to them. Cognitive capacities enable cognitive faculties to exercise their unique cognitive functions. For Kant, “sensibility” and “understanding” are the “sources” or “faculties” of cognition, whereas “intuition” and “concept” are the kinds of representations they are capable of. That is, sensibility has merely the capacity of receiving intuitive, i.e., singular representations, and the understanding has merely the capacity of bringing about or forming conceptual representations. That is, sensation and understanding as such cannot figure as cognitive capacities, nor can intuition and concept count as sources or faculties of cognition. It is only by assigning unique cognitive capacities to sensibility and understanding, namely receiving intuitional representations, and forming conceptual representations, respectively, that they can fulfil their function.6 When it comes to the question of whether metaphysics can be a science, it is a precondition to specify the faculties of cognition by assigning them cognitive capacities. Kant’s strategy is to first identify these cognitive capacities and then show that both are not only indispensable for cognition but also irreducible. His objection is that because metaphysics neglects the dual-compositional structure of cognition, it fails as a science. In the

Two Models of Critique of Metaphysics 299 Transcendental Aesthetic and Transcendental Logic, Kant portrays both capacities as follows: the cognitive capacity of the faculty of sensibility is the capacity of intuition, i.e., of representations that are singular, sensible, and immediate for the cogniser. Furthermore, intuitional representations are inclusive, since the “parts” of intuition are limitations that can only contain one another, just as bigger parts of spaces contain smaller parts, and they are coordinate since parts of intuition can only coexist, rather than being logically ordered. By contrast, the cognitive capacity of the faculty of the understanding is the capacity of concepts, i.e., of representations that are general, intellectual, and indirect for the cogniser. Furthermore, conceptual representations are non-inclusive since concepts are not composed of parts but underly logical subordination with respect to the genus and logical coordination with respect to species. These features that Kant attributes to intuition and concept are irreducible and mutually exclusive such that sensibility does not have the capacity to represent concepts and the understanding does not have the capacity to represent intuitions. Since their cognitive capacities are mutually exclusive, the cognitive faculties are mutually exclusive, too.7 Kant’s general line of criticism is that metaphysics, i.e., metaphysica specialis as well as generalis, violates cognitive dualism by claiming to be able to cognise its objects independent of sensible intuition by mere concepts. Throughout his critical work he exemplifies this objection with reference to rationalist metaphysics in general and the Leibniz−Wolffian tradition in particular. His main point is that the difference between sensibility and understanding, as Leibniz and Wolff have it, is merely a formal or logical one, for it fails to attend to ‘the content of thought’. In the Transcendental Aesthetic Kant writes: ‘The Leibnizian−Wolffian philosophy has therefore directed all investigations of the nature and origin of our cognitions to an entirely unjust point of view in considering the distinction between sensibility and the intellectual as merely logical’ (CpR A43f/B61).8 According to this view, call it cognitive monism, cognition only bears on one source or faculty that allows for both, indistinct (sensible/intuitive) as well as distinct (conceptual) representations. The continuous increase of mental transparency of conscious representations up to its highest degree, i.e., the most distinct conceptual determination of an object, makes insight into the essences of things possible by merely conceptual means. This is exactly what cognitive dualism dismisses because, on Kant’s account, sensibility and understanding are categorically distinct such that it is not possible to transform (indistinct) intuitive representations into (distinct) conceptual representations in order to realise metaphysical cognition.9 Based on the “capacities first” model, Kant argues that any attempt in terms of metaphysics to establish cognition a priori about the world through concepts alone must fail such that metaphysics cannot be a

300  Dietmar H. Heidemann science. When it comes to Kant’s critique of metaphysics, the “capacities first” model does not, of course, exhaust the matter. In the antinomy of pure reason and in the Transcendental Dialectic as such he demonstrates in great detail the inconsistencies and flaws that metaphysics cannot avoid if it holds onto cognitive monism and the resulting transcendental realist picture of the word. The crucial basis for all of this is the “capacities first” model and the proof that metaphysics’ cognitive monism is illegitimate. As Kant shows, the underlying misconceived principle that misleads metaphysics is the presupposed, false claim that the principles of thought are the principles ‘of the possibility of things themselves’ (CpR B289). The principle that objectively holds instead is: ‘The conditions of the possibility of experience in general are at the same time conditions of the possibility of the objects of experience, and on this account have objective validity in a synthetic judgment a priori’ (CpR A158/B197). In what follows I shall discuss whether Hegel can, on the one hand, agree with Kant’s criticism of cognitive monism, while, on the other, reject transcendental conditions of the possibility of experience. Prior to this discussion we need to make ourselves familiar with the “thought determination” model that Hegel employs in order to show that although rationalist metaphysics is unsustainable, metaphysics is possible as a science. 13.3  Hegel’s “Thought Determination” Model The “thought determination” model is the view that conceptual analysis demarcates the possibility of metaphysical cognition. In Hegel, “conceptual analysis” may refer – in rather more contemporary terms – to the procedure of transforming pre-theoretical beliefs into theoretical concepts. This is the project of the Phenomenology of Spirit and its cognitive transition from pre-theoretical “Sense Certainty” to the scientific concept of “Absolute Knowing”, i.e., transforming the Weisen des Fürwahrhaltens (literally: ways of taking to be true) into scientific, true knowledge.10 The more specific terminological meaning of Hegelian “thought determination” is, however, to make available the essences of things themselves by exploring concepts or categories as objective determinations of what there truly (essentially) is. Hegel believes that this procedure not only generates metaphysics as the true science,11 but also simultaneously dismisses classical metaphysics. Whereas classical metaphysics relies on the finite limitations of the subject−object division of thought, he argues, the “thought determination” goes beyond the dogmatic restriction of such divisions. However, like Kant, he regards classical metaphysics’ cardinal mistake to consist in the dogmatic “presupposition” that ‘thought determinations’ are the ‘fundamental determinations of things’ such that ‘what is, by being thought, is known in itself’ (Enc. §28; cf. SL 21: 35 and 48).12 Hegel’s

Two Models of Critique of Metaphysics 301 positive reestablishment of metaphysics as science, to be more precise, as the “science of logic”, by way of thought determination must hence be fundamentally different from that dogmatic “presupposition” classical metaphysicians make. This conception of metaphysics as logic is evidently non-standard and can only be adequately grasped from the historical perspective of Hegel’s intellectual developmental. It is only from there that the “thought determination” model of critique of metaphysics becomes intelligible. In what follows, I shall therefore first briefly sketch the developmental stages of Hegel’s major metaphysical views in order to be able to show why Hegel, at the final stage, thinks that logical thought determinations necessarily replace the rationalist thought determinations of classical metaphysics. As already indicated, the success of Hegel’s critique of metaphysics crucially depends on whether logical thought determination is significantly different from rationalist thought determination.13 The first stage of this intellectual development is the metaphysics of unification that Hegel took up around 1797 as an adherent of Hölderlin (and Sinclair). As his Early Theological Writings show, the metaphysics of unification claims that the opposition and separation of related concepts within the finite presuppose a preceding union, i.e., a union that cannot be grasped through the faculty of understanding. In attempting to cognise absolute union, finite understanding rather disintegrates. The absolute union must be conceived as divine union and therefore reveals itself through religious faith. At that time the young Hegel believed the finite to be intrinsically characterised through its opposition to the divine or absolute as “pure life”. The opposition, however, between finite and infinite only exists for human consciousness that establishes it by limiting the sphere of the former. Hegel now claims that since the opposition is established by finite reflection, it must be overcome in direction of the divine: the ‘self-elevation of man … from finite life to infinite life, is religion’, not metaphysics.14 The philosophy of the one life does not oppose the infinite to the finite, for this form of opposition would be untrue since it would be product of finite thinking itself. Divine life rather is to be construed in terms of a spinozist-pantheist idea of allcomprising union. But on this first stage, philosophy is subordinated to religion in that it ‘has to disclose the finiteness in all finite things and require their completion by means of reason’.15 It is therefore crucial that in his early metaphysical conception Hegel acknowledges the irreconcilable conflict or opposition within the finite and concedes that the absolute or divine cannot be rationally cognised since it is only the object of religious faith. Around 1800, however, Hegel replaces the systematic status of religion by metaphysics, and assigns logic the function of introducing metaphysics.

302  Dietmar H. Heidemann In this new conception, logic is conceived as the introduction to metaphysics by unfolding the immanent limitations of finite cognition or reflexion. To do so, the method logic applies is the sceptical procedure of counterpositing that guides reason in producing antinomies. This position is based on a developmental conception of speculative cognition according to which cognition proceeds from the annihilation of the finite (antinomies) to the infinite absolute, such that the transition itself can be conceptually explicated. On this view the finite understanding is incapable of establishing the identity of oppositions, for, unlike rational cognition, finite reflection is not equipped with the concept of absolute identity with which to unify oppositions. In the fragment Logica et Metaphysica from 1801/02, Hegel outlines this conception of logic as “introduction to philosophy” as follows: philosophy’s true purpose is to cognise the absolute through speculation by sceptically nullifying the finite and elevating to the absolute. The claim is that a transition must be made from the logic of finite cognition to true logic, resulting in infinite cognition. This cognition is, however, not to be conceived as a positive way of cognising the absolute. It rather is a “negative cognising” since logic only prepares for metaphysics as the true science. During this period of his intellectual development Hegel thus does not claim that the infinite or absolute can be rationally grasped since logic and metaphysics are separate.16 In the subsequent years, Hegel’s conception once again changed dramatically. After 1802 he gives up the conception of logic as a preparatory introduction to metaphysics. The reasons for this renewed shift probably have to do with his collaboration with Schelling in Jena. Accommodating Spinoza’s philosophy of the one substance, Hegel at that time conceived the absolute as substance. At around 1804 he must have realised that to conceive of the absolute as substance is to leave it crucially underdetermined. He specifically became aware of the problem that the absolute cannot just be conceived as a fixed object of thought (substance), but must rather be comprehended as comprising complex logical, self-referential relations that can only be developed in an independent discipline. This independent discipline is then to become the science of logic. It is for this reason that Hegel abandons the idea of the absolute as substance and replaces it with the conception of absolute subjectivity incorporating self-referential logical structures. As a consequence, he conjoins logic and metaphysics as the new “science of the absolute”. An introduction to logic as the science of metaphysics is still indispensable, though. The Phenomenology of Spirit now takes on the function of introducing metaphysics as the science of the absolute, i.e., the science of logic.17 It is only after or in the context of the publication of the Phenomenology of Spirit that Hegel arrives at his positive conception of metaphysics and the “thought determination” model. That metaphysics is a science is

Two Models of Critique of Metaphysics 303 beyond doubt as he affirms in his almost canonical definition: ‘Logic … coincides with metaphysics, i.e., the science of things captured in thoughts that have counted as expressing the essentialities of things’ (Enc. §24). In Hegel’s account of metaphysics, “thought” means “to come up with a concept of things”, i.e., to grasp the “essentialities” or “the universal in things”. For this reason, in metaphysics, “thought” must be conceived not as mental activity but as “objective thought” (Enc. §24), that is, as conceptually determining of what things are. Hegel is fully aware that the claim that ‘objective thought constitutes the core of the world’ can be misleading since attributing “thought” or “consciousness” to “natural things” seems to implicate an overall mentalist conception of what the world is. To avoid this mentalist misconstruction of the formula ‘reason in the world’ (Enc. §24), he invents the technical term ‘thought-determination[s|’ (Gedankenbestimmungen) (Enc. §24, Addition 1). The neologism (objective) “thought determination” is thus an auxiliary term that is supposed to refer to his own logical rather than to classical or traditional (rationalist) metaphysics. Nevertheless, in his critique of metaphysics Hegel’s use of “thought determination” is ambiguous. Confronting classical (rationalist) with logical (scientific) metaphysics he juxtaposes the untrue and the true use of the term “thought determination”.18 This seems to indicate that he is willing to concede that classical (rationalist) metaphysics makes at least an effort to grasp what the world truly is by means of conceptual analysis. The untrue and true use of “thought determination” in Hegel’s critique of metaphysics may be distinguished as follows: 1 The untrue use of “thought determination”: The untrue conception of “thought-determination”, as it occurs in pre-Kantian, classical (rationalist) metaphysics, consists in ‘the belief that through thinking things over the truth comes to be known’ (Enc. §26). Reflecting on things and thereby producing thoughts is the finite understanding’s cognitive activity such that the thoughts produced by the understanding are (untrue) “finite thought determinations”. But finite thoughts are one-sided thoughts that consider the objects of metaphysics (‘the objects of reason’ (Enc. §27, cf. §36 Addition) such as soul, world, God,) by finite cognitive standards. That is to say, the perspective of the understanding alone only allows for a restricted access of what truly is. In this respect, Hegel seems to directly appropriate the central point of Kant’s critique of metaphysics according to which the way we cognise cannot count as the ‘condition of the possibility of things themselves’ (CpR B289): ‘This science regarded the thought determinations as the fundamental determinations of things’ (Enc. §28).19 For Hegel, the identification of thought determinations and ontological determinations

304  Dietmar H. Heidemann is a bare presupposition and is based on the further assumption that thought determinations as ontological determinations are ‘predicates of the true’ (Enc. §28).20 The problem Hegel sees here is that rationalist metaphysics does not examine whether finite thought determinations are themselves true nor whether in formal terms predication is at all the appropriate manner of grasping the true. This is particularly true of traditional metaphysics’ first part, ontology, Hegel claims, since ontology is foundational in that it is the ‘the doctrine of the abstract determinations of essence’ (Enc. §33). Hegel repudiates rationalist ontology because of its one-sidedness, its empirical methodology, and, again, for its unjustified belief in the predicative structure of the true. His overall diagnosis is that the use of “thought determination” in rationalist metaphysics must be regarded as untrue since it is bounded through its finitude; however, truth must not be conceived as constrained. Unlike Kant’s critique of (rationalist) metaphysics, Hegel’s objections are not immanent: they presuppose his own positive, logical account of metaphysics as ‘the science of things captured in thoughts that have counted as expressing the essentialities of things’ (Enc. §24). The true use of “thought determination” is key in this account. 2 The true use of “thought determination”: The true metaphysical conception of “thought determination”, as established in the logic, explicates the “essentialities of things” beyond the immanent opposition of reflecting understanding, i.e., beyond the finite and therefore one-sided way of predicative thought. As unbounded, thought is ‘objective thought’ (SL 12.33-34). It is “objective” or non-subjective in that it is not restricted through cognitive capacities that determine its nature and limit its scope. Objective, true thought determination is therefore ‘thought in so far as this thought is equally the thing as it is in itself; or the thing in itself in so far as this is equally pure thought’ (SL 12.33). The tripartite structure of the logic, Being – Essence – Concept, is supposed to spell out what objective thought is: objective logic, i.e., the logic of Being and Essence, replaces former (rationalist) metaphysics, namely metaphysica specialis of soul, world, and God, as well as metaphysica generalis or ontology, which Hegel conceives as ‘that part of metaphysics intended to investigate the nature of ens in general’ (SL 21.48). Objective logic recognises the ens independently of (subjective) ‘determinations of thought’. As already indicated, on Hegel’s view, traditional metaphysics suffers from considering ‘the pure forms of thought’ such as soul, world, and God through the lens of ‘substrates’ and ‘representations’, which he takes to be inadequate for the scientific explication of the true, not least because the metaphysics of finite reflexion conceives those ideas as ‘figurative representations’ (SL 21.49; cf. SL 12.28). In particular, rationalist metaphysics

Two Models of Critique of Metaphysics 305 is unable to overcome the logical separations of finite thought such as the subject−object division. Overcoming these kinds of separations is a requirement for attaining a higher conceptual unity in scientific metaphysics, specifically the concrete universal of speculative logic. By contrast, scientific metaphysics comprehends objective thought determinations not merely as forms of thought. In true metaphysics these forms are themselves what thought is. The logic of Being and Essence is, however, not only the self-standing doctrine of objective thought determination. Hegel also claims that ‘objective logic is therefore the true critique of such determinations – a critique that considers them, not according to the abstract form of the a priori as contrasted with the a posteriori, but in themselves according to their particular content’ (SL 21.49).21 Objective logic and critique of metaphysics by means of the “thought determination” model are thus two sides of the same coin. The subjective logic, or the logic of Concept, integrates being and essence and finalises true thought determination independent of the traditional substance-predicate scheme of cognition. The nature of Hegel’s critique of metaphysics based on the “thought determination” model is obviously very different from the Kantian critique based on the “capacities first” model. Hegel’s model presupposes the scientific conception of metaphysics, whereas Kant’s model seems to not rely on claims that are themselves metaphysical. In what follows, I will not discuss in detail which model is more promising and which is less. I will rather compare both models and point to some important structural commonalities as well as differences that can improve our understanding of what a critique of metaphysics should look like in more general terms. 13.4  Determinations of Intuition and Determinations of Thought As indicated earlier, from Kant’s “capacities first” model and Hegel’s “thought determination” model, basically the same conclusion follows. For both, Kant as well as Hegel, argue that classical (rationalist) metaphysics is not possible as a science. Hegel would, however, insist that Kant draws this conclusion from false (untrue), subjectivist premisses, i.e., the view that cognitive capacities come first, whereas he himself operates on the basis of thought as such independent of subjectivist suppositions. The crucial point can be put as follows: Hegel blames Kant for misconceiving categories as determinations of intuition rather than comprehending them as thought determinations proper. Scholarship on Hegel has not paid much attention to the Kantian background of Hegel’s notorious technical term “thought determination”. This background is, however, crucial with respect to the question of whether

306  Dietmar H. Heidemann metaphysics is possible as a science. For Kant, the possibility of metaphysics as a science hinges on whether logical functions in judgements or categories, respectively, can at all count as determinations of thought or must be conceived as determinations of intuition for achieving objective cognition. This is exactly what Hegel denies, for he claims that objective metaphysical cognition can in fact be achieved merely on conceptual grounds, although in a different way, to be sure, than the rationalists do. It is striking that Kant himself considers the possibility of metaphysical cognition by conceiving logical functions in judgements or categories, respectively, as “thought determinations”. Kant’s use of “Gedankenbestimmung” (CpR B149) might well have inspired Hegel to develop his own account of thought determination by contrasting it with the Kantian model. To better appreciate this point, I shall discuss two schema (Figures 13.1 and 13.2) that illustrate where exactly Kant and Hegel differ. In the Critique of Pure Reason Kant states: The same function that gives unity to the different representations in a judgment also gives unity to the mere synthesis of different representations in an intuition, which, expressed generally, is called the pure concept of understanding. (CpR A79/B104) Whereas in the first edition of the Critique of Pure Reason Kant merely points to the identical cognitive procedure of giving unity, both in judgement and intuition, two years later in the Prolegomena he specifies that in objective judgements, i.e., judgements of experience, [t]he given intuition must be subsumed under a concept that determines the form of judging in general with respect to the intuition, connects the empirical consciousness of the latter in a consciousness in general, and thereby furnishes empirical judgments with universal validity; a concept of this kind is a pure a priori concept of the understanding, which does nothing but simply determine for an intuition the mode in general in which it can serve for judging. (Kant 2004: 52; AA 4: 300; my emphasis)22 That categories are logical functions in judgement conceived as determinations of intuition is then made clear again in the second edition of the Critique of Pure Reason. The ‘explanation of the categories’ is, Kant states, that ‘[t]hey are concepts of an object in general, by means of which its intuition is regarded as determined with regard to one of the logical functions for judgments’ (CpR B128). In §20 of the Transcendental Deduction the same idea reads as follows: ‘the categories are nothing other than these

Two Models of Critique of Metaphysics 307

Figure 13.1  Kant’s “capacities first” model.

very functions for judging, insofar as the manifold of a given intuition is determined with regard to them’ (CpR B143). As Figure 13.1 shows, Kant seems to have the following division in mind: logical functions as ‘the unity of the action of ordering different representations under a common one’ (CpR A68/B93) can be conceived in two fundamentally different ways: either as determinations of thought or as determinations of intuition in general. In the first case, logical functions must be understood as forms of thought that arrange concepts such that they figure, e.g. as subject or predicate term in a categorical judgement: “S is P”. Since thought determination is, for Kant, merely formal and does not refer to intuition, it is of no metaphysical value whatsoever. This is so because in this case there are no (non-logical) criteria for codifying why a concept must be determined as subject term rather than as predicate term and vice versa: ‘as long as intuition is lacking, one does not know whether one thinks an object through the categories, and whether there can ever be any object that even fits them’ (CpR B288).23 In the second case, logical functions are forms of thought referring to intuition, i.e., to representational content. It is here that they operate as categories in that ‘the manifold of a given intuition is determined with regard to them’ (CpR B143). Now, since categories do not analytically implicate the specific kind of intuition they can refer to (cf. CpR B150), two possibilities must be distinguished. They can either refer to non-sensible intuition, i.e., a kind of intuition humans don’t have, or they refer to exactly the kind of intuition we do have, that is, sensible

308  Dietmar H. Heidemann intuition in space and time as deduced in the transcendental deduction. For Kant, the first option is the paradigm case of classical (rationalist) metaphysics. Here categories are then merely empty concepts of objects, through which we cannot even judge whether the latter are possible or not – mere forms of thought without objective reality – since we have available no intuition to which the synthetic unity of apperception, which they alone contain, could be applied, and that could thus determine an object. Our sensible and empirical intuition alone can provide them with sense and significance. (CpR B 148)24 Since with respect to metaphysical, non-sensible intuition the use of the categories is non-objective, metaphysics cannot be a science as Kant’s critique of classical rationalism reveals. Only in the second case can categories have objective meaning or reality since only with respect to spatiotemporal sensible intuition the possibility of an object can be demonstrated.25 This, however, is a case not of metaphysics but of transcendental philosophy proper. As Figure 13.2 indicates, Hegel does not allow for intuition as a possible referent of thought determination. Thought determination is self-sufficient and non-referential in that the objective meaning, or validity of the categories, is generated by thought itself. Hegel is not denying the cognitive

Figure 13.2  Hegel’s “thought determination” model.

Two Models of Critique of Metaphysics 309 relevance of intuition, e.g. in world-directed perceptual cognition. This, however, only concerns empirical circumstances and the accidental composition of cognition. On the “thought-determination” model intuition is, however, irrelevant because it does not contribute to the logical selfconstitution of thought. Hegel’s main objection is that intuition is alien to conceptual analysis and only restricts or limits the scope of thought determination. The restriction and limitation of thought is, he emphasises, the major flaw of the “capacities first” model, not least because Kant ties the categories back to logical functions of judgement instead of developing them on their own merely by thought: [T]he Kantian philosophy did not consider the categories in and for themselves, but declared them to be finite determinations unfit to hold the truth, on the only inappropriate ground that they are subjective forms of self-consciousness, still less did it subject to criticism the forms of the concepts that make up the content of ordinary logic. What it did, rather, is to pick a portion of them, namely the functions of judgments, for the determination of categories, and simply accepted them as valid presuppositions. (SL 12.28) For Hegel it is obvious that, on the Kantian model, thought determinations or categories do not stand on their own feet because of a twofold deficiency: first, Kant has categories depend on logical functions, and, secondly, as to their objective meaning, he restricts them to (sensible, spatiotemporal) intuition, which is to say that Kant transforms thought determinations (categories) into determinations of intuition. Kantian categories cannot, therefore, count as true thought determinations, so Hegel only considers rationalist metaphysics as an alternative possible candidate for metaphysical reasoning. As we have seen earlier though, by Hegelian standards the thought determinations of rationalist metaphysics are problematically subjective and do not achieve the level of objectivity of thought, since metaphysics as a true science requires “total presuppositionlessness” (Enc. § 78). Although not relying on intuition, rationalist metaphysics therefore does not reach the status of a science. As a consequence, it is only the kind of thought determination that eliminates any reference to subjective points of view (such as cognitive capacities) and equally does not hold on to intuition that establishes objectivity, i.e., scientific metaphysics or logic.26 The upshot of this reasoning is that whereas Kant denies the possibility for metaphysics to be a science no matter what, Hegel believes that true logical thought determination makes metaphysics a science. The point of contention between the Kantian and the Hegelian model is the status of intuition. For Kant thought determinations or categories cannot be

310  Dietmar H. Heidemann objectified independently of intuition, for without sensible spatiotemporal intuition the real possibility of things cannot be determined. That is to say, categories are determinations of intuition. Kant’s “capacities first” model therefore establishes the hetero-objectivity of categories because the objective meaning of categories is realised outside of themselves through the synthesis of the manifold given in sensibility. For Hegel, by contrast, the possibility of things can only be cognised if categories are kept separate from intuition and conceived as mere determinations of thought. Purified thought determination is therefore ‘objective thinking’ (SL 12.34) and opens up the domain for scientific metaphysics. Hence, Hegel’s “thought determination” model establishes the auto-objectivity of categories because their objective meaning is realised within thought itself: ‘thought is equally the fact as it is in itself; or the fact in itself in so far as this is equally pure thought’ (SL 21.33). Hegelian auto-objectivity requires that the role of Kantian intuition must be completely absorbed by thought, although Kant takes this to be impossible. Hegelian thought determinations do not depend on what is given, e.g. in sensibility. Their objectivity is produced immanently through themselves in a self-referential dialectical process that presupposes the possibility of the speculative identity of what is taken to be opposed by the finite understanding, such as concept and intuition. However, merely because logic as the theory of objective thought is not concerned with intuition, Hegel is not exempted from giving an account of the role that, for Kant, is played by intuition. This is because the logic of thought determination must also account for how the singular can be determined by merely conceptual means. If the logic of thought determination were not able to do so, it would be hard to see that it is really about the world, rather than just being a play of concepts. In fact, this point seems to be the very core of the Doctrine of the Concept in the Science of Logic. If Hegel is able to successfully demonstrate how intuition as repraesentatio singularis can be spelled out in the merely conceptual terms of thought determination, Kant’s “capacities first” model would fail. If he is not, his own “thought determination” model would be indefensible. This issue represents one of the most exciting and to my mind still open questions of Classical German Philosophy.27 Notes 1 Cf. CpR AXII: ‘Yet by this I do not understand a critique of books and systems, but a critique of the faculty of reason in general, in respect of all the cognitions after which reason might strive independently of all experience, and hence the decision about the possibility or impossibility of a metaphysics in general, and the determination of its sources, as well as its extent and boundaries, all, however, from principles.’ References to the Critique of Pure Reason (CpR) are from Kant 1998 (first edition = A, second edition = B).

Two Models of Critique of Metaphysics 311 2 To be clear, in what follows I argue that Kant is rejecting metaphysics as the science of the supersensible, which I take to be uncontroversial. Kant’s use of the term “metaphysics” is, however, ambiguous in the sense that he does allow for metaphysics as science but merely in the sense of cognition a priori about the world. This science is obviously different from classical metaphysics in that it spells out such cognition under the condition of transcendental philosophy. Kantian (future) metaphysics as science can therefore not go beyond what has antecedently been demarcated by the transcendental critique of the possibility of cognition. Within this narrow frame of the Transcendental Aesthetic and the Transcendental Analytic metaphysics as science has its place. If it goes beyond it, it doesn’t. For an alternative reading of Kantian metaphysics as science, cf. Chapter 6 by Gava in this volume. 3 Note that throughout this chapter my use of “model” is philosophically unambiguous, i.e., I do not use “model” in contemporary terms as model of scientific explanation. Here “model” rather signifies in a non-technical sense a paradigm conception or pattern of critique of metaphysics. 4 References to the Prolegomena are from Kant 2004 (AA = Kant Akademieausgabe). 5 On metaphysics and science in Kant, cf. Dryer 1966: 17−29. Here I cannot engage in detail with the literature on Kant’s critique of metaphysics broadly construed. On this point, cf. the comprehensive study Grier 2001. 6 Kant’s usage of “isolation” (also with respect to “reason”, see CpR A305/B362) relates to scientific methodology, especially chemistry (CpR A842/B870). For details, see Heidemann 2024: Ch.1. I would even argue that this does not only apply to the Kantian theory of cognitive faculties but also seems to be a general point for cognitive theories that operate on the basis of faculties of cognition. Cognitive faculties in terms of sources of cognition must be epistemically specified through capacities pertaining to them in order to be able to take up a cognitive role. Here I cannot elaborate on this point. 7 On this point, cf. Heidemann 2002. 8 Cf. the Inaugural Dissertation On the form and principles: ‘one can see that the sensitive is poorly defined as that which is more confusedly cognised, and that which belongs to the understanding as that of which there is a distinct cognition. For theses are only logical distinctions which do not touch at all the things given, which underlie every logical comparison. Thus, sensitive representations can be very distinct and representations which belong to the understanding can be extremely confused’ (Kant 2003: 387, AA 2: 394). See also the Anthropology: It ‘was a great error of the Leibniz-Wolffian school … to posit sensibility in a lack (of clarity in our partial ideas), and consequently in indistinctness, and to posit the character of ideas of understanding in distinctness’ (Kant 2007: 251, AA 7: 141−142). Cf. Prolegomena, Kant 2004: 41−42, AA 4: 290; What Real Progress Has Metaphysics Made in Germany since the Time of Leibniz and Wolff, Kant 2002: 368−369; 372−375, AA 20: 278, 281−285. 9 An example of cognitive monism, Kant might have in mind, is Leibniz. Cf. Meditationes de Cognitione, veritate et Ideis (1684), where Leibniz determines the difference between true and false ideas as follows: ‘Est ergo cognitio vel obscura vel clara, et clara rursus vel confusa vel distincta, et distincta vel inadaequta vel adaequata, item vel symbolica vel intuitiva: et quidem si simul adaequata et intuitiva sit, perfectissima est’ (Leibniz 1684: 422). At the final level (“perfectissima”) conceptual clarification is supposed to ultimately lead to metaphysical insight in ideas that are general and singular alike.

312  Dietmar H. Heidemann 10 Since this transition is not arbitrary, according to Hegel, it necessarily follows the categorical determinations of the Science of Logic. The Phenomenology therefore presupposes the validity of the Logic, which results in the problem of circularity concerning the relation between both works, as Hegel himself admits. Cf. Heidemann 2007: §§25, 36. 11 For Hegel “science” is what results from the Phenomenology’s transition of consciousness to true self-consciousness in “Absolute Knowing”. In the Science of Logic he makes this point as follows: ‘Pure science thus presupposes the liberation from the opposition of consciousness. It contains thought in so far as this thought is equally the fact as it is in itself; or the fact in itself in so far as this is equally pure thought. As science, truth is pure self-consciousness as it develops itself and has the shape of the self, so that that which exists in and for itself is the conscious concept and the concept as such is that which exists in and for itself’ (SL 21.33−34). On Hegel’s concept of science, see Heidemann 2008. 12 References from the Encyclopaedia (abbr. Enc.) are from Hegel 2010a. References from the Science of Logic (abbr. SL) are from Hegel 2010b. For the Enc., I provide the number of the §, and for the SL, I provide volume and page number of Hegel 1968. 13 As is well known, the debate on Hegel’s critique of metaphysics is rather complex because it comprises not only Hegel’s critique of metaphysics as such but also Hegel’s relation to Kant and his critique of metaphysics. Two camps can be distinguished: those who see Hegel’s critique of metaphysics as independent of his relation to the Kantian project of transcendental philosophy such as Houlgate (2000), and those who see Hegel’s critique of metaphysics and his overall philosophy as further developing and completing the Kantian project such as Pippin (1989). In this chapter, I cannot discuss the arguments for and against these two approaches. For a very helpful overview of the positions in play here, especially in the wake of Taylor’s and Hartmann’s approaches to Hegel, see Lumsden 2008. Düsing (1990) shows how Hegel also carries out his critique of metaphysics by way of discussing specific Kantian topics of critique of metaphysics such as the antinomies. Cf. also Heidemann 2021. 14 Cf. Hegel 1971: 311 and 245−255. 15 Cf. Hegel 1971: 313. 16 Cf. Hegel 1998: 272−274. 17 For details on how Hegel conjoins metaphysics and logic and the Phenomenology’s introductory role, cf. Heidemann 2007: 199−271 and 323−348. 18 Here I prefer “untrue” to “false” since for Hegel “untrue” just means “insufficiently developed” rather than being bluntly false for logical reasons. 19 Although Hegel concurs with Kant’s criticism, he judges the metaphysical principle of ‘thought-determinations as the fundamental determinations of things’ as such to be superior to ‘the later critical philosophising’ (Enc. §28). The second major criticism that he seems to directly borrow from Kant (cf. CpR A571/B600–A583/B611) is the rationalist presumption that ‘of two opposite assertions (which is what those sentences were) one had to be true while the other was false’ (Enc. §32). 20 Cf. Enc. §28, Addition: ‘Objects of reason, however, cannot be determined by means of such finite predicates, and the aspiration to do so was the defect of the old metaphysics’. 21 Hegel therefore believes that his own critique of metaphysics is superior to the Kantian one. 22 Cf. What Real Progress Has Metaphysics Made in Germany since the Time of Leibniz and Wolff, Kant 2002: 363−364, AA XX, 272.

Two Models of Critique of Metaphysics 313 23 Determinations of thought therefore merely set the norms of how concepts as possible predicates of judgements can be used in logically well-formed combinations. Logical functions as thought determinations are the formal preconditions of judging in general but do not by themselves convey information about the concepts they order. From the Kantian point of view, Hegel’s “thought determination” model would belong under this species of intuition-free thought. 24 There are, however, passages where Kant seems to be arguing that classical (rationalist) metaphysical claims are merely conceptual, i.e., without referring to intuition whatsoever. Cf. e.g., CpR B XIV, B 288−294, or B 309. The point is that prima facie metaphysical claims seem to be analytic, but turn out to be synthetic (a priori); however, they cannot be justified because the justifying third thing is missing: sensible, spatiotemporal intuition. 25 Cf. CPR B 288−294. 26 Cf. SL 21.12−13: ‘Indeed, the need to occupy oneself with pure thoughts presupposes a long road that the human spirit must have traversed; it is the need, one may say, of having already attained the satisfaction of necessary need, the need of freedom from need, of abstraction from the material of intuition, imagination, and so forth; from the material of the concrete interests of desire, impulse, will, in which the determinations of thought hide as if behind a veil. In the silent regions of thought that has come to itself and communes only with itself, the interests that move the life of peoples and individuals are hushed’. 27 Longuenesse (2007: 183−185) insightfully hints at how this research question could be tackled.

Bibliography Dryer, Douglas Poole (1966) Kant’s Solution for Verification in Metaphysics (London: Allen & Unwin) Düsing, Klaus (1990) “Hegels Metaphysikkritik, dargestellt am Beispiel seiner Auseinandersetzung mit Kants Antinomienlehre”. In Denken unterwegs. H. Kimmerle zum 60. Geburtstag, edited by Henk Oosterling and Frans de Jong (Amsterdam: Rodopi), pp.109–125 Grier, Michelle (2001)  Kant’s Doctrine of Transcendental Illusion (Cambridge: Cambridge University Press) Hegel, Georg Wilhelm Friedrich (1968ff) Gesammelte Werke. Edited by the RheinischWestfälische Akademie der Wissenschaften (Hamburg: Meiner) Hegel, Georg Wilhelm Friedrich (1998) Schriften und Entwürfe (1799–1808). In Hegel 1968 ff. Volume 5. Edited by Manfred Baum and Kurt R. Meist (Hamburg: Meiner) Hegel, Georg Wilhelm Friedrich (2010a) Encyclopedia of the Philosophical Sciences in Basic Outline Part 1: Science of Logic. Translated and edited by Klaus Brinkmann and Daniel O. Dahlstrom (Cambridge: Cambridge University Press) Hegel, Georg Wilhelm Friedrich (2010b) The Science of Logic. Translated and edited by George Di Giovanni (Cambridge: Cambridge University Press) Heidemann, Dietmar H. (2002) “Anschauung und Begriff. Ein Begründungsversuch des Stämme-Dualismus in Kants Erkenntnistheorie” In Aufklärungen. Festschrift

314  Dietmar H. Heidemann für Klaus Düsing zum 60. Geburtstag, edited by Kristina Engelhard (Berlin: Duncker & Humblot), pp. 65–90 Heidemann, Dietmar H. (2007) Der Begriff des Skeptizismus. Seine systematischen Formen, die pyrrhonische Skepsis und Hegels Herausforderung (Berlin/New York: De Gruyter) Heidemann, Dietmar H. (2008) “Substance – Subject – System: The Justification of Science in Hegel’s Phenomenology of Spirit”. In Hegel’s “Phenomenology of Spirit”. A Critical Guide, edited by Dean Moyar and Michael Quante (Cambridge: Cambridge University Press), pp. 1–20 Heidemann, Dietmar H. (2021) “Hegel’s Philosophy of Philosophy”. In The Relevance of Hegel’s Concept of Philosophy: From Classical German Philosophy to Contemporary Metaphilosophy, edited by Luca Illeterati and Giovanna Miolli (London: Bloomsbury) Heidemann, Dietmar H. (2024) Das Einzelne und das Allgemeine. Bd. 1: Kants Begründung des Dualismus der Erkenntnisstämme (Berlin/Boston: De Gruyter) Houlgate, Stephen (2000) “Substance, Causality, and the Question of Method in Hegel’s Science of Logic”. In The Reception of Kant’s Critical Philosophy. Fichte, Schelling, & Hegel, edited by Sally Sedgwick (Cambridge: Cambridge University Press), pp. 232–252 Kant, Immanuel (1998) Critique of Pure Reason. Translated and edited by Paul Guyer and Allen W. Wood (Cambridge: Cambridge University Press) Kant, Immanuel (2002) Immanuel Kant: Theoretical Philosophy after 1781. Edited by Henry E. Allison and Peter Heath (Cambridge: Cambridge University Press) Kant, Immanuel (2003) Immanuel Kant: Theoretical Philosophy 1755–1770. Edited by David Walford (Cambridge: Cambridge University Press) Kant, Immanuel (2004) Prolegomena to Any Future Metaphysics.Translated and edited by Garry Hatfield (New York: Cambridge University Press) Kant, Immanuel (2007) Immanuel Kant: Anthropology, History, and Education. Edited by Günter Zöller and Robert B. Louden (Cambridge: Cambridge University Press) Leibniz, Gottfried Wilhelm (1684) Meditationes de Cognitione, veritate et Ideis. In Die philosophischen Schriften von Gottfried Wilhelm Leibniz, edited by Carl Immanuel Gerhardt. Vol. 4 (Berlin 1880: Weidmann) Longuenesse, Béatrice (2007) Hegel’s Critique of Kant (Cambridge: Cambridge University Press) Lumsden, Simon. (2008) “The Rise of the Non-metaphysical Hegel” Philosophy Compass Volume 3, No. 1, pp. 51–65 Pippin, Robert B (1989) Hegel’s Idealism. The Satisfaction of Self-Consciousness (Cambridge: Cambridge University Press)

14 Quietism, Dialetheism, and the Three Moments of Hegel’s Logic G. Anthony Bruno

14.1 Introduction The history of philosophy can present us with false friends and false foes. We can overestimate a historical figure’s ability to answer our questions by citing their use of concepts or methods that seem familiar to us, yet whose meaning and purpose differ across epochs. We thereby miss an opportunity to learn to what extent we have moved away from our philosophical past.1 We can also underestimate a historical figure’s ability to answer our questions by citing their use of concepts or methods that are marked by controversy or disrepute, as if controversy or disrepute may not signal resistance to relinquishing perhaps still dominant prejudices. We thereby miss an opportunity to learn in what way we perpetuate our philosophical past. The history of philosophy thus risks a kind of self-opacity whereby we fail to acknowledge either our remoteness from or our proximity to prior modes of thinking. The risk associated with the history of philosophy is relevant to an assessment of Hegel’s appropriation by John McDowell and Graham Priest. McDowell enlists Hegel for a quietist answer to a problem stemming from the dualistic assumption that concepts and reality belong to different orders, viz. the problem of how concepts are answerable to the world. If we accept Hegel’s absolute idealist view that the conceptual is boundless, this dualism and its resultant problem are said to dissolve. Priest enlists Hegel for a dialetheist answer to a problem stemming from the dualistic assumption that truth and falsity are mutually exclusive, viz. the problem of how certain sentences are both true and false. If we accept Hegel’s dialectical view that certain contradictions are necessary, this dualism and its resultant problem are said to dissolve. For both McDowell and Priest, then, contemporary philosophy finds a true friend in Hegel. I will argue that McDowell’s and Priest’s appropriations of Hegel exhibit the historical self-opacity of overestimating Hegel’s affinity with quietism and dialetheism, respectively. On the one hand, McDowell reads DOI: 10.4324/9781003187561-15

316  G. Anthony Bruno Hegel as a quietist who silences sceptical questions that threaten common sense, yet he cannot account for Hegel’s adaptation of ancient scepticism for use against common sense. On the other hand, Priest reads Hegel as a dialetheist who subordinates formal logic to dialectical logic by affirming the truth of certain contradictions, yet he cannot account for Hegel’s commitment to resolving contradictions for the sake of truth as a whole. I will diagnose their misreadings in terms of what Hegel regards as the three moments of logic and argue that while McDowell jumps prematurely to its third moment, Priest stalls needlessly at its second moment. Hegel concludes the Preliminary Conception in the Encyclopaedia Logic (1830) with ‘descriptive anticipations’ of the main structure of the text, according to which logic, i.e., the science of intelligibility, is said to have three ‘moments’. The first is the abstractive moment of the understanding, which ‘stops short’ at fixed categories. The second is the negative moment of dialectic, which discovers the ‘genuine nature’ of the categories, viz. that each ‘passes over, of itself, into its opposite’. The third is the positive moment of speculation, which grasps the ‘unity’ of categories through the ‘dissolution’ of their inner opposition. Hegel warns that if these moments are ‘kept separate from each other … then they are not considered in their truth’ (Hegel GW 19: §§79−82).2 In other words, the truth of the three moments of logic lies in their unity, just as the truth of the stages of a plant lies in their unity. Any appropriation of Hegel should accordingly be measured against the standard of truth to which he holds the moments of logic. I will present the three moments of logic in Section 14.2 in order to show that quietist and dialetheist readings of Hegel fail to consider these moments ‘in their truth’. On the one hand, in his quietist critique of metaphysics, McDowell enlists Hegel to dissolve problems stemming from the assumption of the duality of concept and reality. But, as I will show in Section 14.3, McDowell helps himself directly to the third moment of logic, where the boundlessness of the conceptual would be fully articulated. Since he arrives at the third moment prematurely, ignoring its prior moments, he obscures its truth. On the other hand, in his dialetheist critique of formal logic, Priest enlists Hegel to dissolve problems stemming from the assumption of the duality of truth and falsity. But, as I will show in Section 14.4, Priest restricts himself gratuitously to the second moment of logic, where contradictions within the categories are not yet resolved. Since he stalls at the second moment, severed from its final moment, he obscures its truth. We can extricate Hegel from quietist and dialetheist misreadings only if we see that the truth of the moments of logic lies in their unity. Admittedly, McDowell and Priest offer minority readings of Hegel. Nevertheless, they appropriate Hegel in order to advance relatively influential positions, according to which, following McDowell, the philosophical questions that we ask are pseudo-questions that require a therapeutic

Quietism, Dialetheism, and the Three Moments of Hegel’s Logic 317 response (McDowell 1996: 95) and, following Priest, the concepts that we use are useful yet mostly logically inconsistent (Priest 2006: 4). If Hegel endorses neither quietism nor dialetheism, it is crucial to retrieve his metaphysics from recruitment to such positions. I will consequently treat these misreadings of Hegel as a ladder to discard once we rise to a more adequate reading. 14.2  The Three Moments of Logic For Hegel, logic has three moments, which he calls ‘moments of everything logically real; i.e., of every concept or of everything true in general’ (Hegel GW 19: §79). In the first moment of logic, the understanding posits the categories as abstract concepts, i.e., as universals that are ‘held onto in firm opposition’ to particulars. Furthermore, the understanding judges according to the ‘principle of identity’, applying the categories so that a thing may be ‘grasped in its full precision’ in contrast to other things and, hence, so that ‘nothing should remain vague and indeterminate’ in the ‘domains either of theory or of practice’ (Hegel GW 19: §80). Bringing an object or an action under a category in a judgement preserves the identity and thus the determinacy of that object or action to the exclusion of other objects or actions. In the second moment of logic, dialectic discovers that the categories do not actually preserve the identity of what is judged because a category’s own ‘nature’ is dialectical in that it ‘passes over, of itself, into its opposite’. Dialectic discovers the ‘immanent transcending’ whereby a category’s ‘one-sidedness and restrictedness … displays itself as what it is’, viz. its ‘negation’. We see this when, despite its apparent ‘one-sidedness’, being, since it is undifferentiated and indeterminate, negates itself by collapsing into nothing and vice versa, producing a contradiction. Evoking a distinction that we will observe in Section 14.3, Hegel claims that dialectic is ‘high ancient scepticism’, for it consists in ‘complete despair about everything that the understanding holds to be firm’ regarding the categories. Dialectic therefore ‘must not be confused’ with ‘modern’ scepticism, which denies knowledge of the ‘supersensible’ yet dogmatically ‘hold[s] onto’ knowledge of the ‘sensible’. By exposing the process whereby a category ‘immanent[ly] transcend[s]’ itself for its own opposite, dialectic reveals itself to be ‘[s]cepticism proper’ (Hegel GW 19: §81).3 Hegel notes that dialectic is not merely negative, but also positive, for while it is the nature of a category to sunder itself into contradictory claims, the ‘dissolution’ of these claims and their ‘transition’ towards ‘unity’ in a successor category equally belong to a category’s nature. This is why, towards the end of his description of the second moment of logic, Hegel calls dialectic ‘the principle through which alone immanent coherence and

318  G. Anthony Bruno necessity enter into the content of science’ (Hegel GW 19: §81). First, the development of logic as the science of intelligibility, i.e., the development of truth as a systematic whole, would be interrupted if a category negated itself into claims that remained in contradiction, for such claims would lack coherence.4 Second, Hegel says shortly after that ‘necessity’ consists in thinking’s ‘drive to find’ a ‘stable meaning’ for the opposing claims in a category (Hegel GW 19: §87). This is to say that a category’s opposing claims would lack necessity unless their contradiction were resolved by the unity of a successor category. As Hegel says at the end of his description of the second moment, while ‘philosophy … contains the sceptical as a moment within itself – specifically as the dialectical moment’, it ‘does not stop at the merely negative result of the dialectic’, i.e., at contradiction. This is because dialectic ‘mistakes its result’ if it ‘holds fast’ to contradiction, which thwarts the scientific goals of ‘coherence and necessity’. Instead, the negative result of dialectic must be ‘sublated’ into a ‘positive’ result, which occurs in the third, ‘speculative’ moment of logic (Hegel GW 19: §81).5 In the third moment of logic, then, the ‘positive result’ of dialectic is the speculative unity of the opposing claims in a category. This, Hegel says, is ‘not simple, formal unity, but a unity of distinct determinations’ (Hegel GW 19: §82), for it contains, not just the sameness, but also the distinctness of these claims. In other words, a speculative unity contains the identity and difference of a category’s opposing claims. Were this unity strictly an identity of opposing claims, it would merely repeat the contradiction with which dialectic negatively results. As Hegel says in Section One, Chapter Two, of the Doctrine of Being in the Science of Logic (1812/16), affirming the identity or ‘equal[ity]’ of the opposing claims in a category is only ‘another shape of the still abiding contradiction’. Resolving this contradiction requires grasping opposing claims in their ‘ideality’, i.e., as moments of a unity that maintains the difference of these claims (Hegel GW 21: 139). Thus, being and nothing are ideal moments of becoming because in it they are ‘the same’, since they have ‘passed over into’ each other, and ‘not the same’, since each is distinguished by its ‘distinct’ vanishing into the other, viz. and respectively, as ‘coming-to-be and ceasing-to-be’ (Hegel GW 21: 69−70, 94−95). In the next section, I will articulate the shortcoming of McDowell’s appropriation of Hegel by tracing the ancient sceptical character of logic’s second moment to Hegel’s early adaptation of an ancient sceptical method. 14.3  Quietism and the Third Moment of Logic In Mind and World, McDowell diagnoses an antinomy whose resolution he regards as Hegelian in spirit. The antinomy ensnares two theses about how concepts are answerable to the world such that our beliefs can count

Quietism, Dialetheism, and the Three Moments of Hegel’s Logic 319 as knowledge: the empirical world both must and cannot stand as a tribunal over our use of concepts. It must, according to the first thesis, if our concepts are to be guided by sensible interaction with the world, and cannot, according to the second thesis, if our concepts are to be guided, not only by sensibility, but also by our normative and therefore spontaneous capacity to make warranted judgements about the world (McDowell 1996: xii, 67, 69). However, contra the first thesis, unless we are normatively capable of judging what is sensibly given, what is given cannot figure in warranted judgement and so cannot be appealed to as anything but a ‘myth’. And, contra the second thesis, if the relations among warranted judgements are justificatory, such relations are confined to the ‘space of reasons’ and so, however coherent, fail to grasp the world (McDowell 1996: 5−9, 14). The theses in the antinomy share the idea that sensibility is a receptive capacity that, in its use, is distinct from the spontaneous capacity for judgement. This idea expresses the dualistic assumption that whereas concepts are actively grasped by judgement in thought, reality is passively received by sensibility in experience, i.e., the dualistic assumption that concepts and reality belong or at least cannot be ruled out as belonging to different orders. McDowell resolves the antinomy by rejecting its spoiling idea. He argues that sensibility is inseparable from judgement because, in its use, it ‘draws’ on our normative capacity for judging (McDowell 1996: 13). In other words, the empirical world is always a matter for judgement and, hence, empirical content is always implicitly conceptual content. To judge that something empirically given has some property is always to sense it in such a way as to be capable of grasping the concept of that property.6 McDowell’s resolution of the antinomy is inspired by Kant’s claim that ‘intuition without thought’ falls outside the unity of consciousness and so is ‘nothing for us’ (Kant A111).7 However, Kant holds that our forms of intuition condition how the world appears to us and thereby exclude the world as it is in itself. This leaves intact the duality of concept and reality insofar as it permits the coincidence of concept use strictly with empirical reality, not with transcendental reality. In order to avoid Kant’s transcendental idealist view of an unknowable world beyond the empirical world, McDowell invokes Hegel’s ‘Absolute idealis[t]’ view that the ‘conceptual realm’, i.e., the realm whose empirical content is always implicitly conceptual content, has no ‘outer boundary’ (McDowell 1996: 44).8 We dislodge the dualistic assumption that concepts and reality belong to different orders only if we acknowledge that a reality beyond one’s concept use neither entails nor requires a reality beyond what is empirically conceivable.9 In other words, we eliminate the duality of concept and reality only if we distinguish ‘the act of thinking’ from ‘thinkable contents’ and recognise that such contents are not bounded by an unknowable world (McDowell

320  G. Anthony Bruno 1996: 28).10 We thereby follow Hegel in grasping the sense in which concepts and reality constitute a single order.11 The Hegelian view that the conceptual is boundless accordingly serves to resolve the antinomy. But why does McDowell regard this view as quietistic? McDowell’s aim is to silence such sceptical questions as whether ‘we are open to facts’ or whether ‘states of affairs’ are ‘directly manifest’ to us, questions that ask after the unity of concepts and reality precisely because they begin by assuming their duality. We silence such questions if, following Hegel, we recognise that concepts and reality constitute a single order. However, for McDowell, this recognition must be therapeutic, not constructive. He argues that ‘constructive philosophy’ has ‘no prospect of answering’ sceptical questions because it guarantees their return by leaving their dualistic assumption intact. We must instead ‘exorcise’ such questions (McDowell 1996: xxiii-iv).12 As he says: ‘The aim here is not to answer sceptical questions, but to begin to see how it might be intellectually respectable to ignore them, to treat them as unreal, in the way that common sense has always wanted to’ (McDowell 1996: 113). McDowell’s quietistic goal is to ignore scepticism and to restore our sense of being at home in the world.13 On his reading, the Hegelian view achieves this goal by dislodging scepticism’s dualistic assumption and the sceptical questions that it provokes. Hegel’s awareness that there is ‘no ontological gap’ between thought and being, i.e., that thought and being are of the same ‘sort’, silences questions that are ‘threatening to common sense’ (McDowell 1996: 27−8, 83). In “Hegel’s Idealism as Radicalization of Kant”, McDowell puts this point by saying that, by discarding the transcendental idealist ‘frame’ whereby Kant retains the duality of concept and reality, Hegel’s ‘authentic idealism’ provides assurance for ‘common-sense realism’ (McDowell 2009: 81). While McDowell regards Hegel as entitled to a quietistic expression of the claim that ‘thinking does not stop short of the facts’ (McDowell 1996: 33), he acknowledges that his interpretive strategy is to ‘domesticate the rhetoric’ of absolute idealism (McDowell 1996: 44). In “Hegel’s Idealism”, he describes Mind and World as taking a ‘simple’ path towards absolute idealism, Hegel’s own path to which he says is ‘more complex’ (McDowell 2009: 89). And in “Responses”, a reply to his readers, McDowell describes his thinking as Hegelian ‘in spirit’, if not in letter, because his anti-sceptical ‘exorcisms’ provide ‘ways of entering into Hegel’s work’, whereas Hegel himself struggles to make absolute idealism ‘inviting to people who find it alien’ (McDowell 2006: 269, 277).14 It is difficult to reconcile McDowell’s quietism with his appropriation of Hegel. In the Preface to the Phenomenology of Spirit (1807), Hegel says that philosophy’s attainment of scientific form is ‘the prize at the end of

Quietism, Dialetheism, and the Three Moments of Hegel’s Logic 321 a complicated, tortuous path’. In the Introduction, he describes this path as a ‘pathway of doubt’, one that involves not the irresolute scepticism of ‘shillyshallying about this or that presumed truth, followed by a return to that truth again’, but rather the ‘thoroughgoing scepticism’ that is ‘directed against the whole range of phenomenal consciousness’, specifically, consciousness’ ‘so-called natural ideas’ about truth. Irresolute scepticism is exhibited precisely by what McDowell regards as the hopeless movement of dogmatically assuming the duality of concepts and reality, asking after their unity, and constructing an answer that leaves their duality intact and that thereby secures merely a ‘return’ to a ‘presumed truth’. By contrast, thoroughgoing scepticism ‘brings about a state of despair’ in consciousness’ ideas or categories of truth by discovering contradictions within them and ‘bring[ing] to pass the completion of the series’ of their negations (Hegel W 3: 19, 72−73). Hegel’s focus is therefore not the same variety of scepticism that McDowell aims to exorcise. As Michael Forster shows, Hegel adapts an ancient sceptical method of setting opposing arguments against each other in order to produce equipollence, suspend judgement, and achieve tranquillity. For Hegel, we discover opposing claims of equal force within individual categories, fall into the despair of being forced to affirm the contradiction that is produced by these opposing claims, resolve this contradiction by sublating these opposing claims into a successor category, and ultimately attain a kind of repose by dialectically exhausting the totality of such contradictions and arriving at a stable set, i.e., a system, of categories.15 Crucially, this adapted ancient method assumes no specific beliefs in order to achieve its goal. By contrast, modern scepticism is beset by disquiet precisely because it begins with dogmatically assumed beliefs, e.g., the assumption of the duality of concepts and reality (Forster 1989: 10−12). We can see, then, that whereas McDowell aims to silence and exorcise the irresolute scepticism of modernity, Hegel aims to harness and repurpose the thoroughgoing scepticism of antiquity.16 Still, perhaps Hegel’s adaptation of ancient scepticism is consistent with McDowell’s quietist method of ignoring modern scepticism insofar as Hegel might be said to arrive at McDowell’s quietist destination by taking a path that differs merely in degree of complexity. This prospect fades when we consult Hegel’s review of G.E. Schulze’s Critique of Theoretical Philosophy (1801) in “On the Relationship of Scepticism to Philosophy” (1802). Hegel claims that Schulze ‘is only acquainted with sceptical and dogmatic philosophizing’ because he fails to recognise that, ‘apart from scepticism and dogmatism, there was still a third possibility, to wit, a philosophy’ (Hegel 2000: 316, 325).17 Schulze dogmatically assumes the duality of concept and reality, first, by asserting that ‘what is given within the compass of our consciousness has

322  G. Anthony Bruno undeniable certainty’ and, second, by defining theoretical philosophy as the science of the ‘unconditioned cause of all conditioned things’, a cause that he locates ‘outside and above our consciousness’. This duality makes it both necessary and impossible that ‘existence’ is ‘discoverable’ by means of ‘concepts’, for, on the one hand, theoretical philosophy seeks to comprehend the existence of an unconditioned cause and yet, on the other, it can find ‘certainty’ only ‘within the compass of our consciousness’. Since no ‘bridge’ can be built from our concepts to existence, Schulze sceptically resigns himself to ‘a philosophy which does not go beyond consciousness’ (Hegel 2000: 317−318). Schulze’s sceptical conclusion is unavoidable given the dogmatic premise on which it rests, viz. the duality of concept and reality. This is why Hegel says that ‘Schulzian scepticism integrates the crudest dogmatism into itself’. Moreover, we saw that dogmatic assumption defines the irresolute scepticism of modernity insofar as this variety of scepticism is helplessly tethered to a ‘presumed truth’. This is why Hegel regards modern scepticism as having ‘sunk so far in company with dogmatism’ through a ‘communal degeneration of philosophy’ (Hegel 2000: 330). McDowell might appear to echo Hegel’s critique of Schulze, given that McDowell quietistically ignores sceptical questions in an apparent attempt to avoid the ‘integrat[ion]’ of scepticism and dogmatism. However, McDowell’s alliance with common sense betrays the Schulzian character of his quietism. As Hegel says: [M]odern scepticism lacks the noblest side of scepticism, its orientation against the dogmatism of ordinary consciousness … For the most recent scepticism … the ordinary consciousness with its whole infinite range of facts has an indubitable certainty … Furthermore, according to this latest scepticism, our physics and astronomy, and analytical thought, bid defiance to all rational doubtfulness; and thus it lacks also the noblest side of the later ancient scepticism, i.e., its orientation against limited cognition, against finite knowledge. (Hegel 2000: 339) McDowell’s aim is to ignore sceptical questions on behalf of common sense. However, by deferring to the ‘indubitable certainty’ of common or ‘ordinary consciousness’, even against the duality of concept and reality, McDowell lacks what Hegel regards as the ‘noblest side of scepticism’, viz. an ancient sceptical ‘orientation against the dogmatism of ordinary consciousness’. Ordinary consciousness consists of ‘finite knowledge’ whereby the understanding offers bare assurances about truth in the ‘phenomenal world’, i.e., the truth of ‘the given, the fact, the finite (whether this finite is called “appearance” or “concept”)’ (Hegel 2000: 331−332). Ordinary

Quietism, Dialetheism, and the Three Moments of Hegel’s Logic 323 consciousness exhibits dogmatism because the finitude of its knowledge entails its opposition to contrary finite knowledge, i.e., to contrary bare assurances about the truth. As Hegel says, it is the ‘essence of dogmatism’ to posit as ‘absolute’ something that is merely relative, viz. ‘something finite, something burdened with an opposition’ (Hegel 2000: 335). We cannot resolve the opposition within the finite knowledge of ordinary consciousness on the basis of still more finite knowledge on pain of regress. This is why Hegel says that ‘sundering’ the ‘absolute identity of thought and being’ into ‘finite cognition’ constitutes a ‘dogmatic scepticism’ (Hegel 2000: 354, 357). Resolving the opposition within finite knowledge instead requires adapting an ancient sceptical method of producing equipollence between opposing arguments, which Hegel does by dialectically exhausting the totality of inner oppositions in the categories. We see, then, that McDowell repeats rather than avoids Schulze’s integration of scepticism and dogmatism.18 Despite ignoring the questions of modern scepticism, McDowell perpetuates its indulgence of ordinary consciousness, whose dogmatic endorsement of endlessly opposable finite knowledge, including the bare assurance of the identity of concept and reality, demands a solution that derives from antiquity. For Hegel, modern scepticism has a dogmatic shadow that we cannot outrun by simply ignoring its questions. We must instead pursue ‘a third possibility, to wit, a philosophy’. While Hegel regards this philosophy as ‘neither scepticism nor dogmatism’, he also regards it as ‘both at once’, for he acknowledges a variety of scepticism that is ‘in its inmost heart at one with every true philosophy’, viz. a scepticism that is adapted from an ancient method. First, this scepticism observes the inner opposition of, not a mere concept, but rather a dogmatic claim about truth. Second, it observes the self-destruction to which this claim’s inner opposition leads. Third, it observes the successor claim that resolves this inner opposition. The adapted ancient sceptical method thereby teaches us how ‘philosophy may possibly be something other than the dogmatism’ with which modern scepticism is exclusively acquainted (Hegel 2000: 322−323). Crucially, Hegel’s early adaptation of this three-step ancient sceptical method prefigures the movement of his subsequent logic through its three moments of understanding, dialectic, and speculation. However, as we saw, these moments only have their truth together, i.e., as a unity. We therefore cannot, like McDowell, help ourselves directly to the third moment, as if the speculative insight into the articulated identity of concept and reality could be won simply by an appeal to common sense and hence by means other than philosophy, i.e., by means that are unqualifiedly anti-sceptical. The pathway of doubt cannot be evaded on pain of dogmatism. McDowell’s quietistic refusal to take this path and his deference to common sense abandons the spirit of Hegelianism because it abandons what Hegel

324  G. Anthony Bruno regards as the sceptical essence of ‘every true philosophy’ and what he eventually regards as the second, dialectical moment of logic. McDowell’s premature speculation about the goal of Hegel’s philosophy accordingly violates the unity of its logical moments.19 Indeed, McDowell risks what in the Preface to the Phenomenology Hegel calls ‘the rapturous enthusiasm which, like a shot from a pistol, begins straight away with absolute knowledge, and makes short work of other standpoints’ – including scepticism – ‘by declaring that it takes no notice of them’ (Hegel W 3: 31). Hegel is aware that such a ‘royal road to Science’ will find ‘no more easy-going way than to rely on sound common sense’. However, given the inherent dogmatism of common sense, he concludes that the ‘scientific insight’ of philosophy is ‘only to be won through the labour of the Concept’ (Hegel W 3: 65). 14.4  Dialetheism and the Second Moment of Logic Just before he charges Schulze with ignoring the philosophical alternative to modern scepticism and dogmatism, Hegel invokes a principle that he says is ‘explicit’ in Plato’s Parmenides yet is ‘implicit in every genuine philosophical system’, viz. ‘the principle of scepticism’, which states that ‘panti logōi logos isos antikeitai [against every argument there is an equal one on the other side]’. Applying this principle yields contradictions insofar as we are forced to affirm opposing claims of equal force within individual categories. However, Hegel denies that these contradictions are unacceptable violations of the ‘principle of [non-]contradiction’ on the grounds that ‘every proposition of reason must, in respect of concepts, contain a violation of [the principle of non-contradiction]’. Unlike propositions of the understanding, which prohibit contradiction, propositions of reason are those whose component ‘concepts’, ‘assertions’, or claims ‘contradict themselves’, e.g. the proposition “being is and is not nothing”. According to Hegel, propositions of reason must violate the principle of non-contradiction because ‘every genuine philosophy’ has as its ‘negative side’ that it ‘sublates’ this principle (Hegel 2000: 324−325).20 As the first thesis of his habilitation (1801) states: ‘contradictio est regula veri, non contradictio falsi [contradiction is the rule of truth, non-contradiction that of falsity]’ (Hegel W 2, 533). An apparent dilemma arises according to which we must either reject Hegel’s critique of the principle of non-contradiction or reject his philosophy.21 On the one hand, as John McTaggart argues, we must reject Hegel’s critique of the principle of non-contradiction because, ‘far from denying’ this principle, his philosophy is ‘based on it’, for, were it not, it would ‘reduce itself to an absurdity’ (McTaggart 2000: 15).22 In other words, the comprehensibility of Hegel’s philosophy is inconsistent with his critique

Quietism, Dialetheism, and the Three Moments of Hegel’s Logic 325 of the principle. On the other hand, as Karl Popper argues, we must instead reject Hegel’s philosophy because if philosophy affirms violations of the principle of non-contradiction, then there is ‘no rational motive for changing our theories’ and thus no ‘critique’ and ‘no intellectual progress’ (Popper 1965: 316).23 In other words, the critical value of Hegel’s philosophy is inconsistent with his critique of the principle. Priest offers a solution to this dilemma by arguing that we can accept both Hegel’s philosophy and his critique of the principle of non-contradiction so long as we read him as a dialetheist. Dialetheism is the view that there are true contradictions or ‘dialetheias’.24 Priest states that the ‘main claim’ of his book In Contradiction is that ‘Hegel was right’ to hold that our concepts ‘produce dialetheias’ (Priest 2006: 4). In “Dialectic and Dialetheic”, he claims that Hegel’s philosophy is ‘based on dialetheism’ because it affirms that dialetheias ‘occur in reality’, specifically, in what moves and what lives, a view that Priest endorses (Priest 1989/90: 388−391). As Hegel says in Section One, Chapter Two, of the Doctrine of Essence in the Science of Logic: [A]s regards the claim that there is no contradiction … experience itself testifies that there do exist at least a great many contradictory things … Something moves, not because now it is here and there at another now, but because in one and the same now it is here and not here. [S]elf-movement … is likewise nothing else than that something is, in itself, itself and the lack of itself (the negative), in one and the same respect … Something is alive, therefore, only to the extent that it contains contradiction within itself. (Hegel GW 11: 287).25 At any given moment, something moving is both arriving at and leaving a location in space and hence both is and is not so located. Likewise, at any given moment, something living is both arriving at and leaving a stage of its self-organisation and hence both is and is not so self-organised. These, for Hegel, are true contradictions, i.e., dialetheias. According to Priest, Hegel’s critique of the principle of non-contradiction follows from his affirmation of the existence of dialetheias. Given this affirmation, Priest regards Hegel’s dialectical logic as a species of dialetheic logic, whose general form consists in assigning to sentences either or both truth values (Priest 1989/90, 395). Priest remarks that since dialetheic logic permits dialetheias as well as non-contradictions, thereby avoiding trivialism, it governs a ‘more general’ domain than formal logic, which permits only non-contradictions. The wider scope of dialetheic logic allows it to accommodate the contradiction that he agrees with Hegel is essential to all ‘change’, without

326  G. Anthony Bruno which contradiction everything would always have a ‘consistent’ identity and thus be permanently ‘static’. Priest’s remark about dialetheic logic is intended to apply to its dialectical species. This, he says, poses a ‘stretch’ to the letter, but not to the spirit, of the passage above from the Science of Logic (Priest 1989/90, 395).26 To be sure, just prior to that passage, Hegel says: It is … one of the basic prejudices of previous logic and of ordinary thought that contradiction is not as essential and immanent a determination as identity. But in fact, if order of precedence were an issue, and the two determinations were to be held separate, it would be the principle of contradiction that should be taken as the more profound and the more essential. For in contrast to it, identity is only the determination of simple immediacy, of inert being, whereas contradiction is the root of all movement and life; it is only insofar as something has a contradiction within it that it moves, is possessed of instinct and activity. (Hegel GW 11: 286−287) If a thing’s ‘essential and immanent determination’ were its identity, it would always be consistent and so would involve no intrinsic contradiction. In that case, Hegel says shortly after, any apparent contradiction in a thing would be ‘an accident, an abnormality as it were, a momentary fit of sickness’ (Hegel GW 11: 287). However, such a thing would be ‘inert’, for, as we saw, ‘contradiction is the root of all movement and life’. Hegel accordingly infers that contradiction is a thing’s ‘more profound’ and ‘more essential’ determination, i.e., its true ‘principle’. On Priest’s dialetheist reading of Hegel, then, we can accept both Hegel’s philosophy and his critique of the principle of non-contradiction and thereby avoid an apparent dilemma.27 More recently, Priest claims that dialetheism sheds light on Hegel’s criticism of Kant’s resolution to the antinomies. In “The Antinomy of Pure Reason” in the Critique of Pure Reason, Kant shows that cosmology yields justified yet opposing inferences about the ‘unconditioned unity of objective conditions in appearances’, i.e., about the world. These inferences use the categories to ascribe contradictory properties to the world, e.g. that it is spatiotemporally finite and infinite. According to Kant, the resulting ‘self-contradictory concept’ of the world stems from reason indulging the ‘illusion’ that the categories apply to the world as ‘a whole existing in itself’, when in fact they apply only to the world as ‘the sum total of all appearances’ (Kant A340/B398, A406/B433, A506−507/B534−535). Kant thus resolves the antinomies by showing that their contradictory concept of the world refers only to how reason behaves, not to what exists.28

Quietism, Dialetheism, and the Three Moments of Hegel’s Logic 327 Hegel credits Kant with recognising the ‘necessity of the contradiction’ in the ‘operation of reason’ (Hegel GW 21, 40). However, in Section Two, Chapter Two, of the Doctrine of Being, he charges that Kant’s resolution exhibits ‘an excessive tenderness for the world to keep contradiction away from it, to transfer it to spirit instead, to reason, and to leave it there unresolved’. We saw that the world, by containing movement and life, cannot be shielded from contradiction. As Hegel says, ‘nowhere … does it escape contradiction’ (Hegel GW 21: 232).29 In “Kant’s Excessive Tenderness for Things in the World and Hegel’s Dialetheism”, Priest claims that Hegel’s criticism of Kant’s resolution to the antinomies rests precisely on the dialetheist view that the world contains dialetheias. Indeed, it is because Hegel observes dialetheias in the world that Priest regards him as the ‘zenith of dialetheic thinking’ since antiquity (Priest 2019: 67−71).30 In a recent rebuttal of Priest’s dialetheist reading of Hegel, Michela Bordignon makes four important observations. First, unlike Priest, Hegel endorses a developmental conception of truth according to which contradiction is the means by which truth develops into a systematic whole (Bordignon 2019: 2).31 Second, whereas Priest divides logical space into true and false regions on which dialetheias are said to overlap, Hegel regards logical space as the singular region of truth, understood as a selfdeveloping systematic whole (Bordignon 2019: 12).32 Third, unlike Priest, Hegel holds that the contradictory claims that compose a proposition of reason must be resolved (Bordignon 2019: 13). As Hegel says in the Doctrine of Being, the ‘resolution’ of the contradictory claims in the proposition “the finite is and is not the infinite” consists in, not the acknowledgment of the equal correctness, and of the equal incorrectness, of both claims – this would only be another shape of the still abiding contradiction – but the ideality of both, in the sense that in their distinction, as reciprocal negations, they are only moments … Here we have … the nature of speculative thought displayed in its determining feature: it consists solely in grasping the opposed moments in their unity. (Hegel GW 21: 139) Fourth, whereas Priest acknowledges isolated instances of the coincidence of truth and falsity yet leaves the regional duality of truth and falsity otherwise intact, Hegel regards instances of the coincidence of truth and falsity as progressive determinations of the singular region of truth (Bordignon 2019: 14). These observations demonstrate the extent to which Priest’s dialetheist reading of Hegel poses a stretch to the spirit, not just to the letter, of Hegel’s philosophy.

328  G. Anthony Bruno While I am in broad agreement with Bordignon’s four observations,33 she does not offer a diagnosis of the root cause of Priest’s abandonment of the spirit of Hegelianism. I will suggest that the root cause of Priest’s abandonment is a misreading of Hegel, one that stalls needlessly at the second moment of logic. Recall that, for Hegel, the three moments of logic cannot be ‘kept separate’ if we are to grasp them ‘in their truth’. We saw that McDowell distorts their truth by helping himself directly to the third moment, as if speculative insight into the articulated identity of concept and reality could be secured by quietistically circumventing what Hegel regards as scepticism proper, which alone affords a path to speculative unity. We can see that Priest also distorts the truth of the moments of logic, viz. by restricting himself gratuitously to the second moment, as if science were possible on the basis of contradiction alone and hence in the absence of what we saw in Section 14.2 are the scientific goals of ‘coherence and necessity’. By holding fast to contradiction, Priest’s dialetheist reading of Hegel stalls needlessly at philosophy’s sceptical moment, satisfying itself with dialectic’s negative result while forgoing its positive result, viz. the resolution of contradiction. Without the speculative unity of a category’s opposing claims, that category’s contradiction remains unresolved, halting the progression of science. Indeed, when Priest says that contradiction alone ‘produce[s] a train of conceptual development’ (Priest 2006: 4), he starves this production of its speculative purpose, without which there is no resolution of contradiction, no determination of successor categories, and thus no development of truth as a whole.34 Priest rightly notes that post-Kantians like Hegel regard reason as producing necessary contradictions (Priest 1989/90: 399). However, he neglects their commitment to systematically resolving such contradictions. As Fichte argues in Foundations of the Entire Wissenschaftslehre (1794/95), the contradiction of the I and the not-I threatens to ‘eliminate’ the unity of consciousness, which imposes the ‘task’ of ‘discover[ing] some X’ at which this contradiction is not ‘further extended’, but rather is ‘completely resolved’ (Fichte SW I: 107, 143). For Hegel, resolution involves sublating contradiction into unity. In Section One, Chapter Two, of the Doctrine of Being, he calls ‘sublation’ ‘one of the most important concepts in philosophy’, for it has the ‘speculative meaning’ that it signifies both ‘to “preserve”’ and ‘to cause to cease’. This co-signification is exhibited by becoming, in which being and nothing are preserved as ‘moments’ yet cease insofar as they come to ‘possess a different determination’ in this unity, viz. ‘coming-to-be and ceasing-to-be’ (Hegel GW 21: 94−95). Being and nothing are sublated into becoming like nourishment is incorporated into a body, their present form dissolving while being absorbed into a higher form. As Hegel says: ‘it is contradiction that moves the world, and

Quietism, Dialetheism, and the Three Moments of Hegel’s Logic 329 it is ridiculous to say that contradiction cannot be thought. What is correct in this assertion is just that contradiction is not all there is to it, and that contradiction sublates itself by its own doing’ (Hegel GW 19: §119).35 As we have seen, whereas the second moment of logic is a dialectical unity of the opposing claims in the first moment, which yields the negative result of a necessary contradiction, the third moment is a sublated unity of the opposing claims in the second moment, which yields the positive result of a resolved contradiction. This is not to say that logic simply rids itself of contradiction. Just as the plant no more rids itself of the bud when flowering than rids itself of the flower when fruiting, these being ‘reciprocally necessary moments’ of the plant’s growth (Hegel W 3, 12), so too logic is such that dialectic no more rids itself of abstraction – whose fixed claims dialectic unifies into contradiction – than speculation rids itself of dialectic – whose contradictory claims speculation unifies into successor categories and, ultimately, into the absolute idea – these being reciprocally necessary moments of logic’s truth. Moreover, although becoming provisionally resolves the contradiction of being and nothing, it does not thereby annul movement or life. As Hegel says, ‘becoming’ is ‘not just the unity of being and nothing, but it is inward unrest – a unity which in its self-relation is not simply motionless, but which, in virtue of the diversity of being and nothing which it contains, is inwardly turned against itself’. Becoming is precisely the movement of coming to be and ceasing to be, one in which being and nothing are ‘undivided’ determinations (Hegel GW 19: §88). Movement thus persists as logic discovers ever-more determinate categories and thereby reveals the coherence and necessity of truth as a whole.36 Thus, when he claims that dialetheism is what explains Hegel’s charge against Kant of an excessive tenderness for things, Priest betrays an excessive tenderness for contradiction. To be sure, Hegel regards the antinomies as exhibiting the ‘dogmatism’ of ‘adhering to one-sided determinations of the understanding while excluding their opposites’, which compels the understanding to hold, e.g. that ‘the world is either finite or infinite, but not both’. Understanding’s dogmatic aversion to contradiction provokes a ‘struggle of reason’ to instead affirm contradictions and thereby ‘overcom[e] what the understanding has made rigid’, viz. categories that are ‘opposed’ and ‘separated’ as if ‘by an infinite abyss’. Nevertheless, Hegel is clear that reason’s struggle consists in the ‘speculative’ task of sublating contradictory properties, e.g. finitude and infinitude, and affirming that the world is ‘both the one and the other, and hence neither the one nor the other’ (Hegel GW 19: §32).37 In other words, reason’s struggle is not merely to override the understanding by preserving contradictory properties in a conjunction, i.e., ‘both … and’, but rather to sublate these properties into a speculative unity in which they acquire ‘a different

330  G. Anthony Bruno determination’ than they have in their mere contradiction, i.e., ‘neither … nor’.38 It is this struggle that facilitates reason’s scientific movement toward coherence and necessity. However, if, as Priest asserts, reason does ‘not produce consistency, but rather inconsistencies’, such that ‘[e]verything’, including ‘the Absolute’, is ‘inconsistent’ (Priest 2006: 5),39 it must abandon these scientific goals. But then such reason is not Hegelian. The scientific goals of coherence and necessity reflect what Hegel regards as spirit’s need to be ‘at home with itself, and thereby free’. ‘In the Logic’, he says, the ‘content’ of thinking ‘is brought forth by thinking’ rather than being imposed by an external source. Moreover, whereas the ‘ordinary’ meaning of truth is ‘the agreement of an object with our representation of it’, the ‘philosophical’ meaning of truth is ‘the agreement of a content with itself’. But then since the content of logical thinking derives from thinking alone, philosophical truth must be none other than thinking’s agreement with itself, which is to say, spirit’s being at home with itself and being free (Hegel GW 19: §24).40 Of course, Hegel recognises that logical thinking ‘gets entangled in contradictions’ in a ‘conscious loss of its being at home with itself’, viz. when it posits categories that contain contradictory claims. However, he insists that ‘thinking will not give up, but remains faithful to itself’, viz. through ‘the resolution of its own contradictions’. If thinking ‘despairs of being able to bring about, from its own resources, the resolution of the contradiction in which it has put itself, then it returns to the solutions and appeasements in which the spirit has participated in its other modes and forms’. But Hegel denies that spirit’s momentary despair will ‘degenerate into misology’ (Hegel GW 19: §11).41 This is because ‘spirit is the one which is strong enough that it can endure contradiction’ and ‘knows how to resolve it’ (Hegel GW 21: 232). As he says in the Phenomenology, ‘the life of Spirit is not the life that shrinks from death and keeps itself untouched by devastation, but rather the life that endures it and maintains itself in it. It wins its truth only when, in utter dismemberment, it finds itself’ (Hegel W 3: 36). *** Quietist and dialetheist readings of Hegel obscure the truth of the three moments of logic. They thereby exhibit a kind of self-opacity to which the history of philosophy is vulnerable. We cannot enlist Hegel to prove such theses as that concepts are answerable to the world or that certain sentences are both true and false if such theses make assumptions that conflict with his conception of logic. This is both to overestimate his affinity for present concerns and to neglect the extent to which we may have moved away from our philosophical past. Of course, we might have good reason to move on from Hegel. However, knowing this too requires a kind of historical self-knowledge.42

Quietism, Dialetheism, and the Three Moments of Hegel’s Logic 331 Notes 1 Cf. Lewis 1961: 13: ‘The dominant sense of any word lies uppermost in our minds. Wherever we meet the word, our natural impulse will be to give it that sense. When this operation results in nonsense, of course, we see our mistake and try over again. But if it makes tolerable sense our tendency is to go merrily on. We are often deceived. In an old author the word may mean something different. I call such senses dangerous senses because they lure us into misreadings’. 2 Cf.: ‘the logical has three sides: (α) the side of abstraction or of the understanding, (β) the dialectical or negatively rational side, [and] (γ) the speculative or positively rational one’ (§79). 3 Crucially, whereas the opposing arguments that concern ancient scepticism exclude each other, Hegel’s adapted ancient skeptical method concerns the opposing claims that are internal to a category. This signals a shift of logical focus from mere contradiction to self-contradiction. On the difference between the classical principle of non-contradiction and the speculative principle of selfcontradiction, see de Boer 2010. 4 See Bordignon (forthcoming) on intermediate categories of the logic as illegitimate totalities. 5 Cf.: ‘The second negative at which we have arrived, the negative of the negative, is this sublating of contradiction’ (GW 12: 246). Cf. Hegel’s 1820s lectures on the history of philosophy: ‘the operations of skepticism are undoubtedly directed against the finite. But however much force these moments of its negative dialectic may have against the properly speaking dogmatic knowledge of the understanding, its attacks against the true infinite of the speculative idea are most feeble and unsatisfactory’ (Hegel 1995: 367). 6 Cf. Boyle’s gloss on McDowell: ‘what is needed is not just any sort of constraint on the subject’s judging; what is needed is something intelligible as a constraint from the subject’s own point of view – something she could see as a reason for judging the world to be thus-and-so, if she were to reflect on the question “Why should I believe that P?”’ (Boyle 2016: 534). 7 See McDowell 1996: 4, 87. 8 Cf.: ‘the effect of [Kant’s] philosophy is to slight the independence of the reality to which our senses give us access. What is responsible for this is precisely the aspect of Kant’s philosophy that struck some of his successors as a betrayal of idealism: namely the fact that he recognizes a reality outside the sphere of the conceptual. Those successors urged that we must discard the supersensible in order to achieve a consistent idealism’ (McDowell 1996: 44). 9 See Sedgwick 1997. 10 Cf. Hegel’s distinction between ‘thoughts’ and ‘thought-determinations’ (GW 19: §24). 11 See Hegel: ‘the Logical is to be sought in a system of thought-determinations in which the antithesis between subjective and objective (in its usual meaning) disappears. This meaning of thinking and of its determinations is more precisely expressed by the Ancients when they say that nous governs the world, or by our own saying that there is reason in the world, by which we mean that reason is the soul of the world, inhabits it, and is immanent in it, as its own, innermost nature, its universal’ (GW 19: §24). 12 McDowell acknowledges that exorcism is constructive ‘in another sense’ (xxiv) that is left implicit, but that presumably involves dislodging the spoiling idea that produces the antinomy and reorienting those who indulge its component theses.

332  G. Anthony Bruno 13 Terry Pinkard (2018: 19) suggests that McDowell’s quietism is non-classical because it treats sceptical questions as genuine rather than pseudo-questions. But this neglects McDowell’s view that quietism is meant to restore common sense, for which such questions are precisely not genuine. It also neglects his claim that the ‘vertigo’ that results from sceptical questions should not be felt ‘in the first place’ (McDowell 1998: 63). 14 Some of McDowell’s readers doubt whether his quietism is Hegelian even in spirit. Robert Stern (1999: 260–263) grants that Hegel is not a sceptic, but notes that he demands more than the repose of common sense, viz. a metaphysics that can transform the ossified categories of ordinary understanding into self-developing categories of reason. As Hegel says in the Science of Logic, common sense and modern science work ‘hand in hand to cause the downfall of metaphysics’ and to produce ‘a cultivated people without metaphysics – like a temple richly ornamented in other respects but without a holy of holies’ (GW 21: 6). Stephen Houlgate (2006: 251–255) argues that whereas McDowell seeks to show how experience grounds judgement, Hegel seeks to show how judgement grounds experience, a reversal that reflects Hegel’s view of judgement as constrained, not externally by the world, but internally by the categories. Sebastian Rödl (2007) denies that McDowell refutes Kant’s idea that the forms of intuition are imposed on the unity of consciousness on the grounds that this requires not the simple path of quietism but rather the complex path of absolute idealism, whereby the categories are derived through a deduction that is at once metaphysical and transcendental. (On the metaphysical and transcendental character of Fichte’s genetic deduction of the categories, which simultaneously answers the question quid facti and the question quid juris, see Bruno 2018.) Finally, Sebastian Gardner (2013: 135–136) holds that, even granting that quietism is non-dualistic, McDowell’s preference for quietism over constructive philosophy is simply dogmatic. 15 See Hegel W 3 (73−74): ‘the exposition of the untrue consciousness in its untruth is not a merely negative procedure. … [I]n the negation the transition is made through which the progress through the complete series of forms comes about of itself’. 16 Cf. Hegel: ‘Ancient scepticism must be differentiated from the modern form; we are only dealing with the former, for it alone is of a true, profound nature’ (W 19: 360). On the role of Agrippan, Pyrrhonian, and Platonic scepticism in Hegel’s Jena period, as well as the connection between Agrippan scepticism and nihilism, see Franks 2008. 17 Cf. Hegel’s 1820s lectures on the history of philosophy, in which he charges Schulze with the error of ‘recogniz[ing] nothing but dogmatism and [modern] scepticism, and not the third philosophy’ (W 19: 400). 18 For a related comparison of McDowell and Jacobi, see Bruno 2020. 19 See Hegel 1977a: 99: ‘Common sense cannot understand speculation; and what is more, it must come to hate speculation when it has experience of it’. Cf. Hegel 2000a: 283: ‘[philosophy] only is philosophy in virtue of being directly opposed to the understanding and hence even more opposed to healthy common sense, under which label we understand the limitedness in space and time of a race of men; in its relationship to common sense the world of philosophy is in and for itself an inverted world’. 20 For a Hegelian critique of the presupposition of the principle of non-contradiction by Aristotle, Sextus Empiricus, and Kant, and how this presupposition does not secure, but rather impedes, the determinacy of both being and meaning, see Winfield 2018.

Quietism, Dialetheism, and the Three Moments of Hegel’s Logic 333 1 See Ficara 2012. 2 22 Cf. Brandom 2002: 179; for a criticism of Brandom, see Bordignon 2012. 23 According to Redding (2007: 213–215), for Hegel, not only can the principle of non-contradiction not be assumed as mythically given, but also its significance shifts depending on whether contradictions apply to terms or to propositions. Moss (2020: 241–242) argues that since, for Hegel, presupposing the principle of non-contradiction relativises truth to that presupposition, he must hold the dialetheist view according to which the absolute exists only as a true contradiction. 24 Priest and Robert Routley coin ‘dialetheism’ and ‘dialetheia’ in 1981 (Priest et al. 1989: xx). 25 Cf. Hegel GW 19: §81. For an account of contradiction in the Doctrine of Essence, see Pippin 1996 and de Boer 2010. 26 One worry is that Priest’s remark imposes a scientific dualism onto Hegel, according to which dialetheic and formal logic are self-standing, mutually external sciences of thought that simply coincide in cases of non-contradictions. On formal logic, see Hegel GW 19: §§115, 160, 182. Bowman (2013: 255.n33) argues that dialectical logic is not comparable to non-classical logic, including dialetheic logic. 27 On the difference between the static logic of the understanding (Verstandeslogik) and the revisable logic of reason (Vernunftlogik), and for a comparison of Hegel’s threefold division between institutional logic, natural logic, and logical form and Priest’s threefold division between taught logic, used logic, and logical facts, see Ficara 2020. 28 Once reason critiques itself, however, it is no longer afflicted by contradiction. See Kant: ‘It is worrisome and depressing that there should be an antithetic of pure reason at all, and that pure reason, though it represents the supreme court of justice for all disputes, should still come into conflict with itself. We had such an apparent antithetic of reason before us above [viz. in “The Antinomy of Pure Reason”], to be sure, but it turned out that it rested on a misunderstanding, namely that of taking, in accord with common prejudice, appearances for things in themselves, and then demanding an absolute completeness in their synthesis, in one or another way (which were both equally impossible), which could hardly be expected in the case of appearances. There was thus in that case no real contradiction of reason with itself in the propositions “The series of appearances given in themselves has an absolutely first beginning” and “This series is absolutely and in itself without any beginning”; for both propositions are quite compatible, since appearances, as regards their existence (as appearances) in themselves are nothing at all, i.e., something contradictory, and thus their presupposition must naturally be followed by contradictory consequences’ (A740/B768). 29 Cf. GW 19: §48, W 20: 358−359. See Žižek 2013: 395. 30 See Hegel GW 19: ‘antinomy is found not only in the four particular objects taken from cosmology, but rather in all objects of all kinds, in all representations, concepts, and ideas. To know this, and to be cognizant of this property of objects, belongs to what is essential in philosophical study’ (§48). 31 See Hegel: ‘The true shape in which truth exists can only be the scientific system of such truth’ (W 3: 14). 32 See Hegel: ‘The more conventional opinion gets fixated on the antithesis of truth and falsity, the more it tends to expect a given philosophical system to be either accepted or contradicted; and hence it finds only acceptance or rejection. It does not comprehend the diversity of philosophical systems as the progressive unfolding of truth, but rather sees in it simple disagreements. The

334  G. Anthony Bruno bud disappears in the bursting-forth of the blossom, and one might say that the former is refuted by the latter; similarly, when the fruit appears, the blossom is shown up in its turn as a false manifestation of the plant, and the fruit now emerges as the truth of it instead. These forms are not just distinguished from one another, they also supplant one another as mutually incompatible. Yet at the same time their fluid nature makes them moments of an organic unity in which they not only do not conflict, but in which each is as necessary as the other; and this mutual necessity alone constitutes the life of the whole’ (W 3: 12). 33 A fifth observation of Bordignon’s is that whereas Hegel regards reality as the bearer of contradictions, Priest states in In Contradiction that he is concerned with sentences as bearers of contradictions (Bordignon 2019: 8; see Priest 2006: 54). However, in “Dialectic and Dialetheic”, Priest is clear that he follows Hegel when he opposes ‘non-literal’ interpretations of contradictions on the grounds that contradictions ‘occur in reality’ (Priest 1989/90: 390−391; cf. 2009: 71). 34 See Kreis 2015: 335−336. 35 Cf.: ‘Speculative thought consists only in this, in holding firm to contradiction and to itself in the contradiction, but not in the sense that, as it happens in ordinary thought, it would let itself be ruled by it and allow it to dissolve its determinations into just other determinations or into nothing’ (GW 11: 287−288). See Heidegger 1988: 28: ‘sublating or Aufhebung must, of course, be conceived, as always in Hegel, in terms of the resonance of its threefold meaning: tollere, removing and eliminating the mere, initial illusion; conservare, preserving and including in the experience; but as an elevare, a lifting up to a higher level of knowing itself and its known’. 36 Moreover, Hegel states in his 1820s lectures on aesthetics that life always resolves its own contradiction: ‘to say that opposites are to be identical is precisely contradiction itself. Yet whoever claims that nothing exists which carries in itself a contradiction in the form of an identity of opposites is at the same time requiring that nothing living shall exist. For the power of life, and still more the might of the spirit, consists precisely in positing contradiction in itself, enduring it, and overcoming it. This positing and resolving of the contradiction between the ideal unity and the real separatedness of the members constitutes the constant process of life, and life is only by being a process’ (Hegel 1975: 120). 37 On Hegel’s distinction between understanding and reason and its role in resolving Kant’s antinomies, see Winegar 2016. 38 See de Boer 2010: 368: ‘According to the speculative meaning of the principle of contradiction, a conceptual determination such as indivisibility only contradicts its ultimate principle – the concept as such – insofar as it opposes its contrary, that is, insofar as it does not establish the unity of indivisibility and divisibility’. 39 Moss (2020: 278) argues that the contradiction between being and nothing remains both true and false on the grounds that the atemporal character of Hegel’s logic entails that neither a contradiction’s truth nor its falsity passes away. This suggests a subtle difference between Moss’ and Priest’s dialetheist readings of Hegel regarding the third moment of logic. See Moss (forthcoming). 40 See Stang (manuscript). See Werner 2020 for an account of why Hegel’s logic is not modelled on the theoretical cognition of what is given.

Quietism, Dialetheism, and the Three Moments of Hegel’s Logic 335 41 Cf. Kant on sceptical despair and misology, on which see Bruno 2018a and Callanan 2019. 42 Thanks to Michela Bordignon, Charlie Cooper-Simpson, Robb Dunphy, James Kreines, Toby Lovat, Gregory S. Moss, Sebastian Stein, David Suarez, Andrew Werner, and audiences at University College Dublin, the Federal University ABC, and the Chinese Hong Kong University for helpful comments on this chapter.

Bibliography Bordignon, Michela (2012) “Contradiction or Non-Contradiction? Hegel’s Dialectic between Brandom and Priest” Verifiche 41(1−3), pp. 241–45 Bordignon, Michela (2019) “Hegel: A Dialetheist? Truth and Contradiction in Hegel’s Logic” Hegel Bulletin 40(2), pp. 1–17 Bordignon, Michela (2021) “Hegel’s Logic as a System of Illegitimate Totalities”. In Hegel’s Encyclopaedic System, edited by S. Stein and J. Wretzel (London: Routledge), pp. 116–32 Bowman, Brady (2013) Hegel and the Metaphysics of Absolute Negativity (Cambridge: Cambridge University Press) Boyle, Matthew (2016) “Additive Theories of Rationality: A Critique” European Journal of Philosophy 24(3), pp. 527–55 Brandom, Robert (2002) Tales of the Mighty Dead (Cambridge, MA: Harvard University Press) Bruno, G. Anthony (2018a) “Skepticism, Deduction, and Reason’s Maturation”. In Skepticism: Historical and ​Contemporary Inquiries, edited by G.A. Bruno and A.C. Rutherford (London: Routledge), pp. 203–19 Bruno, G. Anthony (2018) “Genealogy and Jurisprudence in Fichte’s Genetic Deduction of the Categories” History of Philosophy Quarterly 35(1), pp. 77–96 Bruno, G. Anthony (2020) “Jacobi’s Dare: McDowell, Meillassoux, and Consistent Idealism”. In Idealism, Relativism and Realism: New Essays on Objectivity beyond the Analytic-Continental Divide, edited by D. Finkelde and P. Livingstone (Berlin: de Gruyter), pp. 35–55 Callanan, John (2019) “Kant on Misology and the Natural Dialectic” Philosophers’ Imprint 19(47), pp. 1–22 Conant, James (2004) “Varieties of Skepticism”. In Wittgenstein and Skepticism, edited by D. McManus (London: Routledge), pp. 97–136 De Boer, Karin (2010) “Hegel’s Account of Contradiction in the Science of Logic Reconsidered” Journal of the History of Philosophy 48(3), pp. 345–73 Ficara, Elena (2012) “Dialectic and Dialetheism” History and Philosophy of Logic 34(1), pp. 35–52 Ficara, Elena (2020) “Hegel and Priest on Revising Logic”. In Graham Priest on Dialetheism and Paraconsistency, edited by C. Baskent and T.M. Ferguson (Dordrecht: Springer), pp. 59–71 Fichte, J.G. (1965) Sämmtliche Werke. Edited by I.H. Fichte (Berlin: de Gruyter) [SW] Fichte, J.G. (1982) Science of Knowledge. Translated by P. Heath and J. Lachs (Cambridge: Cambridge University Press)

336  G. Anthony Bruno Forster, Michael (1989) Hegel and Skepticism (Cambridge, MA: Harvard University Press) Franks, Paul (2008) “Ancient Skepticism, Modern Naturalism, and Nihilism in Hegel’s Early Jena Writings”. In The Cambridge Companion to Hegel and Nineteenth-Century Philosophy, edited by F. Beiser (Cambridge: Cambridge University Press), pp. 52–73 Gardner, Sebastian (2013) “Transcendental Philosophy and the Possibility of the Given”. In Mind, Reason, and Being-in-the-World: The McDowell-Dreyfus Debate, edited by J.K. Schear (London: Routledge), pp. 110–42 Hegel, G.W.F. (1896) Lectures on the History of Philosophy, Volume III. Translated by E.S. Haldane and F.H. Simson (London: Kegan Paul, Trench, Trübner & Co) Hegel, G.W.F. (1968–) Gesammelte Werke Deutsche Forschungsgemeinschaft (Hamburg: Meiner) [GW] Hegel, G.W.F (1970) Werke in 20 Bänden. Edited by E. Moldenhauer and K.M. Michel (Frankfurt: Suhrkamp) [W] Hegel, G.W.F. (1975) Aesthetics Volume 1: Lectures on Fine Art. Translated by T.M. Knox (Oxford: Oxford University Press) Hegel, G.W.F. (1977) Phenomenology of Spirit. Translated by A.V. Miller (Oxford: Oxford University Press) Hegel, G.W.F (1991) The Encyclopedia Logic (with the Zusätze). Translated by T.F. Geraets, W.A. Suchting, and H.S. Harris (Indianapolis: Hackett) Hegel, G.W.F. (1995) Lectures on the History of Philosophy, Volume 2. Edited by E.S. Haldane and F.H. Simson (Lincoln: University of Nebraska Press) Hegel, G.W.F. (2000) “On the Relationship of Skepticism to Philosophy, Exposition of Its Different Modifications, and Comparison of the Latest Form with the Ancient One”. In Between Kant and Hegel: Texts in the Development of Post-Kantian Idealism, edited and translated by G. di Giovanni and H.S. Harris (Indianapolis: Hackett), pp. 311–62 Hegel, G.W.F (2006) Lectures on the History of Philosophy Volume II: Greek Philosophy. Translated by R.F. Brown (Oxford: Oxford University Press) Hegel, G.W.F (2010) Science of Logic. Translated G. di Giovanni (Cambridge: Cambridge University Press) Hegel, G.W.F. (1977a) The Difference between Fichte’s and Schelling’s System of Philosophy. Translated by W. Cerf and H.S. Harris (Albany: SUNY Press) Hegel, G.W.F. (2000a) “The Critical Journal, Introduction: On the Essence of Philosophical Criticism Generally, and Its Relation to the Present State of Philosophy”. In Between Kant and Hegel: Texts in the Development of Post-Kantian Idealism, edited and translated by G. di Giovanni and H.S. Harris (Indianapolis: Hackett), pp. 272–91 Heidegger, Martin (1988) Hegel’s Phenomenology of Spirit.Translated by P. Emad and K. Maly (Indianapolis: Indiana University Press) Houlgate, Stephen (2006) “Thought and Experience in Hegel and McDowell” European Journal of Philosophy 14(2), pp. 242–61 Kant, Immanuel (1998) Critique of Pure Reason.Translated by P. Guyer and A.W. Wood (Cambridge: Cambridge University Press)

Quietism, Dialetheism, and the Three Moments of Hegel’s Logic 337 Kreis, Guido (2015) Negative Dialektik des Unendlichen: Kant, Hegel, Cantor (Berlin: Suhrkamp) Lewis, C.S. (1961) Studies in Words (Cambridge: Cambridge University Press) McDowell, John (1996) Mind and World (Cambridge, MA: Harvard University Press) McDowell, John (1998) “Virtue and Reason”. In Mind, Value, and Reality (Cambridge, MA: Harvard University Press), pp. 50–73 Routledge, John (2006) “Responses”. In Routledge McDowell: On Mind and World, edited by N.H.Routledgeondon: Routledge), pp. 269–305 McDowell, John (2009) “Hegel’s Idealism as Radicalization of Kant”. In Having the World in View: Essays on Kant, Hegel, and Sellars (Cambridge, MA: Harvard University Press), pp. 69–89 McTaggart, John (2000) Studies in the Hegelian Dialectic (Kitchener: Batoche) Moss, Gregory S. (Forthcoming) “Absolute Dialetheism”. In On Markus Gabriel’s New Realism, edited by O. Bueno and J. Voosholz (Dordrecht: Springer) Moss, Gregory S. (2020) Hegel’s Foundation-Free Metaphysics: The Logic of Singularity (London: Routledge) Pinkard, Terry (2018) “McDowell’s Hegel: Quietism versus the Dialectic” Revista Eletrônica Estudos Hegelianos 15(25), pp. 18–38 Pippin, Robert (1996) “Hegel’s Metaphysics and the Problem of Contradiction”. In The Hegel Myths and Legends, edited by J. Stewart (Evanston: Northwestern University Press), pp. 239–52 Popper, Karl (1965) Conjectures and Refutations: The Growth of Scientific Knowledge (London: Routledge) Priest, Graham (1989/1990) “Dialectic and Dialetheic” Science and Society 53(4), pp. 388–415 Priest, Graham (2006) In Contradiction: A Study of the Transconsistent (Oxford: Clarendon Press) Priest, Graham (2019) “Kant’s Excessive Tenderness for Things in the World and Hegel’s Dialetheism”. In Logic from Kant to Russell: Laying the Foundations for Analytic Philosophy, edited by S. Lapointe (London: Routledge), pp. 56–72 Priest, Graham, Robert Routley, and Jean Norman (Ed.) (1989) Paraconsistent Logic: Essays on the Inconsistent (München: Philosophia) Redding, Paul (2007) Analytic Philosophy and the Return of Hegelian Thought (Cambridge: Cambridge University Press) Rödl, Sebastian (2007) “Eliminating Externality”. International Yearbook of German Idealism 5 (Berlin: de Gruyter), pp. 176–88 Sedgwick, Sally (1997) “McDowell’s Hegelianism” European Journal of Philosophy 5(1), pp. 21–38 Stang, Nicholas (Manuscript) “Determinacy, Contradiction, Movement: Towards a Reading of Hegel’s Logic” Stern, Robert (1999) “Going Beyond the Kantian Philosophy: On McDowell’s Hegelian Critique of Kant” European Journal of Philosophy 7(2), pp. 247–69 Werner, Andrew (2020) “Hegel’s Dialectical Method: A Response to the Modification View” Canadian Journal of Philosophy 50(6), pp. 767–84

338  G. Anthony Bruno Winegar, Reed (2016) “To Suspend Finitude Itself: Hegel’s Reaction to Kant’s First Antinomy” Hegel Bulletin 37(1), pp. 81–103 Winfield, Richard Dien (2018) “On Contradiction: Hegel versus Aristotle, Sextus Empiricus, and Kant”. In Hegel and Ancient Philosophy: A Re-examination, edited G. Magee (New York: Routledge), pp. 149–60 Žižek, Slavoj (2013) Less Than Nothing: Hegel and the Shadow of Dialectical Materialism (London: Verso)

15 Metaphysics on the Model of Natural Science? A Kantian Critique of Abductivism Nicholas Stang 15.1 Introduction For several decades in the 20th century, analytic philosophy suffered under the despotic rule of logical empiricism, and metaphysics was driven into exile. But, as Kant predicted, the reign of the empiricists proved to be brief, and metaphysics was returned, if not to the throne, at least to a central place in contemporary philosophy. Work on metaphysical topics that would have been familiar to Kant’s rationalist predecessors – e.g., monism, grounding, the principle of sufficient reason – now thrives in analytic philosophy, while the once-feared principle of verification, long used to drive metaphysics into hiding, finds few adherents. In an ironic reversal, though, those who defend metaphysics today often do so by claiming that it is continuous with natural science. The very accusation that Kant thought undermined metaphysics – that ‘the birth of the purported queen [of the sciences] was traced to the rabble of common experience’ (Ax) – is now taken to vindicate it instead. This is the view of metaphysics I will call abductivism. Abductivism is a package of views, a key component of which is a thesis about the common methodology of metaphysics and science: Abductive Methodology. The basic methodology of metaphysics is the same as the basic methodology of the sciences, i.e., inference to the best explanation (IBE, or abduction for short).1 Just as the physicist infers the theory that best explains, for instance, the paths of particles in a cloud chamber, so too, according to the abductivist, does the metaphysician infer the theory that best explains some target phenomenon, such as the modal profiles of objects, or their persistence through time. Abductivism is also committed to a thesis about the epistemology of metaphysics: Abductive Epistemology. The core source of our knowledge of metaphysical theories is abduction.2 DOI: 10.4324/9781003187561-16

340  Nicholas Stang Of course, the abductivist need not think that all of our knowledge in metaphysics comes from abduction; some of our knowledge in metaphysics might be knowledge of logic, knowledge of conceptual truths, or knowledge of mathematics or physics. But that knowledge by itself is not enough to get us knowledge of metaphysical theories, and the difference is made up by abduction. The empirical, conceptual, logical, mathematical, and physical truths under-determine metaphysical theory, so we infer to the metaphysical theory that best explains the “data”. Another component of abductivism is realism about natural science:3 Scientific realism. The theoretical terms in our best scientific theories refer to entities and structures in the world that are (i) theory-independent and (ii) mind-independent. Since abductivists see metaphysics as continuous with science, they are committed to the parallel claim about metaphysics: Metaphysical realism. The theoretical terms in our best metaphysical theories refer to entities and structures in the world that are (i) theoryindependent and (ii) mind-independent. Abductivism is thus committed, in Kantian terms, to transcendental realism.4 Abductivism is an attractively unified set of views about the methodology, epistemology, scientific status, and semantics of metaphysics. It has many contemporary adherents.5 Abductivism offers a putative explanation of the possibility of metaphysics as knowledge and as a science. In Kantian terms, abductivism is a critique of metaphysical reason, an explanation of how (abductive) reason can achieve knowledge in metaphysics that simultaneously establishes the limits of reason in metaphysics.6,7 Reason can go as far in metaphysics as abduction reaches; once our inferences to the best explanation gives out, our rational warrant for metaphysical theorising gives out too, and we should refrain from speculating further.8 In fact, Kant agrees with abductivism on a few core claims. He agrees that metaphysics and natural science are about explanations, i.e., not merely knowing what is the case, but also knowing why.9 He further agrees with the abductivist about the first conjuncts of Scientific realism and Metaphysical realism, though he rejects the second conjunct: the concepts in our natural scientific and metaphysical theories refer to theory-independent entities and structures in the world which are mind-dependent.10 But this narrow band of agreement conceals a vast gulf of disagreement about the very nature of metaphysics itself. For Kant, abduction provides only comparative certainty in its conclusion: the grounds of the inference (the data) make the conclusion (the putative explanation) probable (or at least more probable than the alternatives), but they are not sufficient

Metaphysics on the Model of Natural Science? 341 guarantors of its truth; one can abduct from incomplete or misleading data to a false conclusion.11 But metaphysics, according to Kant, must have apodictic certainty: the grounds on which we base our metaphysical claims must suffice, by themselves, to ensure that those claims are true.12,13 Because he has this apodictic conception of metaphysics, Kant must reject both Abductive methodology and Abductive epistemology. Kant thus cannot accept an abductivist conception of metaphysics. But neither can he accept a purely abductivist conception of natural science. Kant thinks that all natural sciences, properly speaking, rest on metaphysical, hence apodictically certain, foundations, which therefore cannot be justified abductively.14 Once these metaphysical foundations are in place, though, inference to the best explanation can play a significant role in natural science.15 Given this vast difference in their respective conceptions of metaphysics, prospects for an informative dialogue between Kant and the abductivist seem bleak. In this chapter, I try to break this impasse by exploring Kant’s reasons for thinking that abductivist metaphysics cannot do what it claims to do, that is, it cannot explain the possibility of knowledge in metaphysics. But because the abductivist conception of metaphysics is so radically unlike Kant’s conception of metaphysics, and in fact is more similar to his conception of (the empirical part of) natural science, it will actually be Kant’s critique of natural science that will be relevant to abductive metaphysics. I will argue that Kant’s arguments about the possibility of natural science can be redeployed to show that abductive metaphysics fails. What is more, these arguments do not depend on Kant’s controversial assumption, which would in any case be rejected by any contemporary metaphysician, that metaphysics must be apodictically certain. Although abduction plays a significant role in the metaphysical theories of some of his predecessors16 and in his own pre-Critical metaphysics,17 Kant was not acquainted with any philosopher who held to the strictly abductivist conception of metaphysics.18 Consequently, my manner of proceeding in this chapter will be more reconstructive than strictly exegetical. Although this chapter is based on a reading of Kant developed elsewhere, it is not primarily a work of Kant exegesis but an attempt to take some ideas from Kant and use them to criticise contemporary metaphysics. Nonetheless, I will indicate along the way how my argument maps onto Kant’s texts. 15.2  Abductive Metaphysics How is abductive knowledge in metaphysics, knowledge by inference to the best explanation, possible? To answer this question, we need to say a bit more about what explanation is. There is a vast literature on this topic, and it is not my intention to address it comprehensively here.19 Instead, I

342  Nicholas Stang will attribute a fairly simple (possibly oversimplified) set of commitments about explanation to the abductivist. Over the course of this chapter, I will consider whether the abductivist can evade my arguments by abandoning or revising some of these commitments. Let us begin by distinguishing two senses of explanation. Firstly, we can understand explanation as an activity of epistemic agents, and derivatively as the product of that explanation (i.e., the activity of explaining produces explanations). Secondly, we can understand explanations as entities or structures in the world, the real things that “back” or “correspond to” our explanations in the first sense. For instance, causal explanation in the first sense is the activity of searching for, and finding, causes for given events; causal explanations in the second sense are the causes themselves, the causes referred to and described by explanation in the first sense. It is controversial whether the activity of explanation needs to be “backed by” or “correspond to” worldly explanatory structure, but this assumption is shared by many contemporary metaphysicians, so I will take it for granted.20 The first condition, therefore, that must be satisfied for the possibility of metaphysical knowledge, as the abductivist conceives it, is that the world have explanatory structure. Furthermore, it must have the kind of explanatory structure relevant to metaphysical explanations. If metaphysical explanation includes causal explanation, then there must be causal structure in the world (i.e., there must be causes and effects). Likewise, if metaphysical explanation involves various kinds of non-causal explanations, then it must include relevant corresponding worldly explanatory relations. For instance, the possibility of grounding explanation in metaphysics (i.e., explaining facts by citing their grounds) requires that there be grounding structure in the world. Which particular explanatory structures a given metaphysician is committed to will depend on the details of her metaphysical theory; the key idea is that abductivist metaphysicians are committed to there being ontic explanatory structure “in the world”. According to the abductivist, the core source of our knowledge of metaphysical theories is abduction. Since knowledge requires truth, the abductivist is committed to holding that abduction in metaphysical theories is truth-tracking, i.e., it generally takes us from true premises to true conclusions. But for abduction to be successful, it not only needs to be alethically successful (truth-tracking), it must also be explanatorily successful, i.e., it must take us from true premises to truths about their explanations. These are distinct requirements; a mode of inference (e.g., logical deduction) might be truth-tracking without necessarily tracking explanations of its premises. Finally, if abduction in metaphysics is to provide knowledge, then its alethic and explanatory success cannot be a matter of mere luck. “Mere luck” here is simply a stand-in for the post-Gettier idea that knowledge is incompatible with being simply

Metaphysics on the Model of Natural Science? 343 a matter of accident, of having “gotten lucky” in having true beliefs. In other words, abduction in metaphysics must provide (i) un-Gettier-able (i.e., non-lucky) (ii) true beliefs about the (iii) explanations of the premises. But the distinction drawn above between explanation-as-somethingwe-do and explanation-in-the-world commits the abductivist even further. For our practices of explanation have a certain structure, sometimes called “canons of theory choice” or “canons of explanation”: explanations must be unified, parsimonious, rich in consequences, agree with other theories, etc. These features are what constitute the bestness of inference to the best explanation. For example, to pick a feature of explanations that almost everyone agrees is a virtue, the abductivist is committed to claiming that unified explanations are ceterus paribus better than disunified ones. But in order for abduction guided by this principle to be “alethically successful” there must be unified ontic explanatory structure “in the world”. If there were, for instance, fundamentally distinct kinds of explanatory structure in different domains (or even different regions of the universe), then in abducting to the more unified explanatory theory we would not be tracking the ontic explanatory structure, i.e., we would not be making explanatorily successful inferences (even if they were alethically successful). Note that the ceterus paribus character of our canons of theory choice means that the requirements imposed on the ontic explanatory structure in the world are somewhat “loose”: the ontic explanatory structure must be to some degree unified, fruitful (rich in consequences), etc.21 Successful explanation (and thus successful IBE) requires more than there just being explanatory structure in the world. Explanations are sensitive to structure. If I explain some target phenomena by citing an explanans that gets the structure of the underlying ontic explanation wrong, I have failed to explain the phenomena, even if my beliefs about the explanans are true. This means that the success of IBE is sensitive to the language in which we conduct it.22 If our language does not “carve” ontic explanatory structure “at the joints”, then our explanations of facts stated in that language, assuming the explanations are also formulated in our language, will refer to non-joint-carving entities and structures in the world. We may systematically infer to true theories, but they will not be theories backed by the objective explanatory structure in the world. Our inferences may track the truth, but they will fail to be successful inferences to the best explanation. I will illustrate these claims with a pair of thought experiments. I will then draw a general lesson from them. Recall the famous “grue” example from Goodman 1944 (updated for 2023): there is a community whose language includes the predicate “grue”, which applies to objects if and only if they are either green and first observed before 2050 or blue and first observed after 2050. Speakers of this language will infer in 2023 from the fact that all observed emeralds

344  Nicholas Stang are grue to the conclusion that all emeralds whatsoever are grue. This example is unsuitable for my purposes, for three reasons: it is a case of induction, not abduction; it is clearly not a case of metaphysical abduction; and the conclusion is false, rather than merely non-explanatory. To make my point I need a Goodman-style case involving metaphysical abduction to a true but non-explanatory conclusion. 15.2.1  Gerrymandered Persistence

Let us assume that perdurance is the correct explanation of why objects persist through time (i.e., exist at more than one time): objects persist by having instantaneous temporal parts at the times at which they exist. Objects, let us assume, are 4D spacetime “worms”: mereological fusions of instantaneous temporal parts. Consider a community that has correct beliefs about the patterns of persistence of objects, i.e., correct beliefs about which objects exist at which times, but that has a term in their language “gr-sists”, which refers to objects that persist either by perduring (having instantaneous temporal parts) or by enduring (by being “wholly present” at each time at which they exist, or however you want to spell out endurantism). They have no term in their language that refers specifically to the property of perduring or to the property of enduring. The speakers of this community will reason abductively as follows: (P1) [A set of true beliefs about the patterns of persistence of objects.] (C1) ∴ The gr-sistence theory of the persistence of objects. By hypothesis, their abductive premise(s) and conclusion are true; the speakers of this community have true beliefs about the pattern of persistence of objects, and objects do gr-sist. But the conclusion does not explain why objects persist the way they do. They persist because they perdure. In fact, objects gr-sist because they perdure; perdurance is one of the disjuncts of gr-sistence, and disjuncts ground disjunctions (or so I will assume). The speakers of this community will infer to abductive conclusions that fail to explain their premises because their language includes the predicate “gr-sists”, rather than “perdures”. Their abductive inferences will fail because the language they speak fails to carve the world at its explanatory joints. 15.2.2  Reversed Grounding

Let us assume that there is an ontic explanatory relation in the world that answers to the contemporary notion of “grounding” and that standard assumptions about that relation are true, i.e., that it is irreflexive (nothing

Metaphysics on the Model of Natural Science? 345 grounds itself), asymmetric (no circular grounding), and transitive. Let us assume further that the actual world is grounding paradise: the Principle of Sufficient Reason holds for the grounding relation, so absolutely every fact has a ground (I will leave it open whether the PSR itself has a ground). Assume additionally that every fact has a consequence: every fact grounds some further fact (e.g., a disjunction of which it is one of the disjuncts). Consider a community whose language contains the term “gr-grounds” that refers to the inverse of the grounding relation: “p grgrounds q” is true in their language just in case the fact that q grounds the fact that p.23 Speakers of this language will infer from facts to their consequences, not to their grounds. Because every fact has a consequence (we have assumed), they may systematically infer to true conclusions, but they will systematically infer to the wrong (metaphysical) explanations; they will infer consequences, not grounds, as explanations. Consider, further, how such a community would construct a metaphysical theory of “what grgrounds what” (to borrow Schaffer’s (2009) phrase). The premises of their abductive inference would consist in the facts that constitute the target phenomena of such a theory, while their conclusion would consist in a set of claims that gr-ground those facts. For instance, their theory of what gr-grounds the existence of material parts will cite facts about the existence of wholes of which they are parts (assuming that material objects are grounded in their parts); their theory of what gr-grounds the existence of complex material objects will cite facts about their shadows (assuming that the existence of complex material objects grounds the existence of their shadows), etc. In general, their metaphysical theorising will be alethically successful but not explanatorily successful, that is, it will not track ontic explanatory structure. These examples are meant to constitute an intuitive case for the principle that theories will be explanatorily successful, i.e., track not only the truth but also the ontic explanatory structure, only if they involve terms that refer to ontic explanatory structure in the world rather than, for example, gerrymandered relations and properties. To summarise, then, the Abductivist is committed to all of the following: A1. Ontic explanatory structure. There is enough ontic explanatory structure in the world to “back” explanations in true metaphysical theories. A2. Abductive epistemology. Abduction in metaphysics according to our canons of explanation (unity, etc.) tracks the truth and the underlying ontic explanatory structure, and both in a non-Gettierable (not merely lucky) fashion. A3. Reference dependence. The terms in our best metaphysical theories carve the ontic explanatory structure at its joints.

346  Nicholas Stang If A1 is false, then abductive metaphysics is impossible because there is not enough explanatory structure in the world to “back” our explanations. If A2 is false, then abduction in metaphysics might generate true beliefs, but it would not be knowledge (it would be accidentally true belief). And if A3 is false, abduction in metaphysics might generate knowledge, but it won’t generate knowledge of underlying explanatory structure; we will systematically fail to carve the world at its explanatory joints. 15.3  Kant’s Critique of Natural Scientific Abduction Kant gives a critique of abduction in natural science, in his specific semitechnical sense of “critique”: an account of the nature and limits of a capacity for knowledge, in this case, our capacity for knowledge of Nature through inference to the best explanation. He does not offer a separate critique of abduction in metaphysics because he thinks metaphysics cannot be abductive; metaphysics, he thinks, must have apodictic, rather than merely comparative, certainty.24 Furthermore, Kant’s critique of abduction, the bulk of which occurs in the Appendix to the Transcendental Dialectic, depends on his own transcendental idealist metaphysics, which by that point in the KrV, he takes himself to have proven. Therefore, in using that Kantian critique to criticise contemporary abductive metaphysics I will have to proceed somewhat indirectly. In this section, I summarise Kant’s critique of abduction, organised around his explanations of principles in his own theory that correspond to A1–A3. Because my primary aim here is systematic rather than textual, I will simply state what I take Kant’s view to be, keeping textual details, and my reasons for reading Kant the way I do (as well as engagement with secondary literature), to a minimum.25 The aim of this section is twofold: to motivate a critical question about abductivism (what explains A1–A3?) and to give some reasons for thinking of this as a genuinely Kantian question, the kind of question Kant (or “the Kantian”) should ask when confronted with abductive metaphysics. I will formulate these three Kantian theses K1–K3 in the contemporary terms of A1–A3, in order to bring out their similarities. This will involve some translation of Kant into contemporary lingo; Kant would not formulate them this way exactly (nor would I if my task were primarily exegetical). K1. Ontic explanatory structure. There is enough ontic explanatory structure in the spatiotemporal world to back explanations in natural scientific theories. The key to Kant’s explanation of K1 is his transcendental idealism, the doctrine that the form of the spatiotemporal world is grounded in the form

Metaphysics on the Model of Natural Science? 347 of our experience of the spatiotemporal world. Transcendental idealism entails the following: Transcendental idealism. If our experience, in virtue of its very form, represents the spatiotemporal world as having enough ontic explanatory structure to back natural scientific theories, then this grounds the presence of sufficient such structure in the spatiotemporal world.26 Because the spatiotemporal world is a world of appearances, and the “being” of appearances is grounded in the content of our experience of them, if we experience the world as having ontic explanatory structure (e.g., causal structure), this grounds the presence of that structure in the world of appearances. But notice that Transcendental idealism will explain the presence of structure only in the phenomenal world, the world of appearances, not in things in themselves. It can provide no explanation of why there is any ontic explanatory structure in things in themselves (assuming there is any). Kant’s explanation of the possibility of abduction is an explanation of the possibility of abduction from (premises about) appearances to (conclusions about) appearances. It provides no explanation of how we could use abduction to acquire knowledge of things in themselves, because Kant thinks such an abduction could never provide us with any knowledge. One immediate consequence is that Kant’s transcendental idealist explanation of K1 cannot be borrowed by the contemporary abductivist, who, in Kantian terms, is a transcendental realist about metaphysics. Kant thinks that the principal kind of explanation in natural science is causal explanation, so the first thing he must account for is the presence of casual structure in the phenomenal world. He does so by arguing, in the Analogies of Experience, that: Experience. Experience, in virtue of its very form, represents spatiotemporal objects as absolutely persisting substances in law-governed causal interaction, i.e., every alteration in a substance is the effect of a substance (whose state is reciprocally altered by the prior substance), where these alterations are governed by universal and necessary laws. This, combined with Transcendental idealism, explains why the spatiotemporal world has causal and nomic structure: experience represents it as having this structure, and that grounds the presence of such structure (because the objects in questions are appearances). But Kant thinks that the phenomenal world has further “explanationbacking” structure, which is crucial for his critical reconstruction of abduction in the Appendix to the Transcendental Dialectic (henceforth, “the Appendix”): real essences. The connection with causal-nomic structure is

348  Nicholas Stang this: laws are necessary in virtue of being grounded in the real essences of the objects about which they are laws.27 This means that causal-nomic structure brings with it essence structure. Put more precisely, the connection is as follows: Law. If it is a law that φ has ψ then (a) it is true that all samples of φ have ψ, and (b) every sample of φ has ψ in virtue of the real essence of φ.28,29 Real essences are essences of (kinds) of things, unlike logical essences, which are essences of concepts.30 Logical essences contain the marks that constitute a concept, while real essences contain the properties that constitute objects themselves.31 To use one of Kant’s favourite examples, both attractive and repulsive forces are part of the real essence of matter, but only the latter is a mark contained in the concept .32 This means that it is a law that matter has both attractive and repulsive forces. Different speakers can associate the same empirical concept with different logical essences (they can give it different nominal definitions) but they refer to the same kind of object, e.g., matter.33 Logical essences of concepts are relatively trivial for natural scientific purposes; they contain only the marks that speakers happen to think in a certain concept. What is far more important are the properties contained in the real essence, for they ground the laws that are the proper topic of scientific inquiry.34,35 To use a piece of non-Kantian terminology, I will say that, when φ is as above, the φs constitute a natural kind, or, equivalently, that is a natural kind concept. Some care is required here, because Kant is a conceptualist about universals: all that exist are individuals (individual substances and their individual accidents), and any “generic” or “universal” entity (e.g., a general property, a kind) “exists” only in the content of conceptual representation.36 So natural kinds are not part of Kant’s inventory of what exists, not part of what Quine would call his “ontology”. Consequently, talk about natural kinds always has to be, in principle, paraphrasable in different terms: saying of a concept that it is a natural kind concept, or that objects fall under that concept in virtue of their sharing a common real essence.37 Empirical concepts are at least partly individuated by the natural kinds to which they refer: if C and C* are the same empirical concept, they refer to the same natural kind.38 But natural kinds are themselves individuated by their real essences (even if we can never know the complete real essence).39 The properties in the real essence make that kind the thing it is; without them, it would not be possible. For instance, nothing could be matter that did not have its essential attractive and repulsive forces. But if natural kinds themselves partly individuate the empirical concepts that refer to them, and natural kinds are themselves individuated by their

Metaphysics on the Model of Natural Science? 349 real essences, then real essences partly individuate empirical concepts. For instance, nothing could be the empirical concept if it did not refer to matter, i.e., a kind with the same real essence as matter (including attractive force).40 This means that if there were no natural kinds whatsoever (constituted by objects sharing a generically similar real essence), then there would be no empirical concepts whatsoever. What unifies the instances of an empirical concept is not possession of the marks contained in its logical essence, for possession of these marks is not by itself sufficient to fall under the empirical concept; it is possession of the marks that constitute the real essence of the corresponding kind.41 For instance, possession of repulsive force is not sufficient to fall under ; this requires attractive force as well, even though only the former, but not the latter, is contained in the logical essence of that concept. If there are no common real essences, there would be nothing to ground membership in a common empirical concept. In particular, mere perceptible similarity among objects is not by itself sufficient for them to fall under a common empirical concept; a generically similar real essence is required. One immediate consequence of Kant’s essentialist theory of empirical concepts and laws is that there is an isomorphism between the system, if there is one, of empirical concepts (a hierarchy of more and less general empirical concepts) and the system, if there is one, of natural laws (a hierarchy of more and less general laws of nature). Empirical concept C is more general than empirical concept C* if and only if the real essence of the natural kind corresponding to C is more general than the real essence of the natural kind corresponding to C*.42 But there is also an isomorphism between empirical natural kinds and natural laws: natural kind k is more general than natural kind k* if and only if the k law is more general than the k* law. By transitivity, this entails that there is an isomorphism between the system of empirical concepts and the system of natural laws: empirical concept C is more general than empirical concept C* if and only if the corresponding C law (the k law, where C is the concept of k) is more general than the corresponding C* law (the k* law, where C* is the concept of k*). We can translate unproblematically between a system (hierarchy of relative generality) of empirical concepts and a system of natural laws. K2. Abductive epistemology. Abduction in natural science according to our canons of explanation (i.e., generality, specificity, continuity) tracks the truth and the underlying ontic explanatory structure, and both in a non-Gettierable (not merely lucky) fashion. Despite the centrality of causal explanation to Kant’s conception of natural science, and his recognition that many of our causal inferences will be

350  Nicholas Stang abductive (for a single effect could have many causes, which means we must infer to the most likely cause),43 his most sustained discussion of abduction in natural science, in the Appendix, is not about causal inference, but inference to the system of empirical concepts (ordered into species and genuses). This is because Kant thinks a great deal more can be determined a priori about the causal laws that govern material substance than is given by the pure transcendental principles of the KrV alone (this project is undertaken in MAN). Only once these impure yet a priori “metaphysical” principles are in place is it possible to engage in causal inference about objects of experience.44 Consequently, my focus in discussing K2 will be on natural scientific abduction that reveals the underlying system of empirical concepts, rather than on casual abduction. Kant attributes the cognitive task of inferring from perceptible similarities and differences among objects to the underlying system of empirical species and genuses to what he calls the “hypothetical” use of reason. The hypothetical use of reason is clearly not deductive, ‘that is, not such that if one judges in all strictness the truth of the universal rule assumed as a hypothesis thereby follows’ (A647/B675).45 But it is equally clear that the hypothetical use of reason is not exclusively (or even primarily) inductive. One of the main topics of the Appendix is the hypothetical use of reason in inferring from particular empirical concepts (and their associated laws) to more general empirical concepts, i.e., from species to genus. The inference from two or more species to their common genus is not an inductive inference, i.e., an inference from a premise of the form that p has held in all observed instances (e.g., that all observed samples are F) to the conclusion that p holds in all cases whatsoever (that all samples whatsoever are F). It is an inference from two or more empirical concepts to the common underlying features that unite and (partially) explain them, as well as potentially other species of the same genus. In contemporary terms, it is abductive. Kant’s discussion of the principles that govern the hypothetical use of reason is complex, confusing, and controversial. What follows is a simplified presentation, sufficient for my purposes in this chapter. Kant argues for three regulative principles that guide our abductive search for the system of empirical concepts: generality (For any two species, seek a common genus), specification (For any genus, seek further species), and continuity (Between any two species of a genus, search for an intermediate species). Together these constitute the regulative Idea of systematicity in Nature. In each case, Kant argues that these regulative principles (which he frequently dubs “logical”) are possible only under the assumption of a corresponding transcendental one. I will focus on generality: ‘The logical principle of genera therefore presupposes a transcendental one if it is to be applied to Nature (by which I here understand only objects

Metaphysics on the Model of Natural Science? 351 that are given to us). According to that principle, sameness of kind is necessarily presupposed in the manifold of a possible experience (even though we cannot determine its degree a priori), because without it no empirical concepts and hence no experience would be possible’ (A654/B682). Corresponding claims are made about the logical/regulative principles of specification and continuity: they are applicable only on the assumption of a corresponding transcendental principle, without which experience would not be possible.46 On my reading, the transcendental status of the principle of generality is a consequence of Kant’s essentialist conception of empirical concepts: if there were no common real essences (if no empirical object were similar to any other object in virtue of their real essences), then there would be no empirical concepts.47 The presence of common, perceptually manifest, qualities in objects is not sufficient to ground their belonging to the same empirical concept. The most this will guarantee is a common logical essence, formed by abstracting from their perceptible similarities; but, as we have seen, having the marks contained in a logical essence is not enough to ground membership in a common empirical concept. For that, common real essences are required. It is a commitment of this reading that the species and genuses of which Kant speaks in the Appendix are essential species and genuses: they divide objects, not according to their accidental or merely perceptually manifest properties but according to their real essences. Without empirical concepts, experience is impossible;48 so without at least some common real essences, experience is impossible.49 This is why the principle of generality is transcendental, i.e., it makes experience possible. This is a sketch of Kant’s explanation of K2 restricted to a very general formal feature of our explanations and the ontic structure that backs it: abduction in natural science according to the principle of generality (search for common empirical genuses) must be backed by ontic explanatory structure constituted by natural kinds sharing common real essences. I say that this is general and formal because the form of experience explains why the search for common genuses in general will not be entirely in vain. But it does not explain why our actual attempts to find these common genuses will succeed. In particular, it leaves open the possibility that although Nature satisfies the principle of generality, indeed may even satisfy it ideally (for any two objects there is a common genus that subsumes both), those common genuses are cognitively inaccessible to human beings. To make this vivid, it might be that the empirical similarities in terms of which we initially classify objects, due as they are to the contingent constitution of our sense organs (e.g., we classify green things as similar because of the contingent structure of our eyes), are systematically at variance with the underlying system of empirical genuses, so that proceeding

352  Nicholas Stang from the former and trying to infer the latter, as we do, will never succeed. More generally, Nature might be ideally systematic (i.e., ideally satisfy the principles not only of generality, but also of specificity and continuity), but either so complicated or so unlike the system of perceptible similarities by which we initially categorise objects that we will never be able to achieve knowledge of the underlying system of species and genuses. This is the problem that Kant introduces in the two introductions to the KU and for which he introduces the regulative principle of purposiveness: Inquire into Nature as though it is purposive for our cognition, i.e., exists in order for us to cognise it.50 Kant takes this to be equivalent to: Inquire into Nature as though it is the sensible product of non-sensible (nonspatiotemporal) intelligent author.51 In particular, if we assume this regulative principle, we will assume that our inferences to the species-genus system underlying Nature will, at least under the right conditions, track the truth (we will infer to true conclusions) and ontic explanatory structure (because the genuses are real explanatory structures in the world), in a non-Gettierable fashion (we will not merely be getting lucky, for Nature was created, in part, to enable us to make these inferences in a reliable way). This is Kant’s explanation, such as it is, of K2. It raises difficult questions, principal among which is this: Can a regulative principle, like the principle of purposiveness, explain why K2 is true, or merely why we are rationally warranted in assuming K2 in our inquiry into Nature? But given my purposes in this chapter, I will leave that issue.52 K3. Reference dependence. The concepts in our best natural scientific theories carve the ontic explanatory structure at its joints. First, we need to distinguish, among the concepts in our natural scientific theories, between a priori formal concepts such as , , , and , and empirical concepts of particular natural kinds of substances, their real essences, and the causal laws they obey (e.g., , , , etc.). Kant’s explanation works very differently in the two cases. In the formal case, the ontic explanatory structure of the phenomenal world is grounded in the structure of our experience of that world, according to Transcendental idealism. Since these formal concepts (, , , and, I have argued, as well) are part of that a priori form, they ground that structure in the phenomenal world. For instance, since the principle of generality is transcendental, there can be no experience that does not represent there being at least some real essence shared by some spatiotemporal objects; consequently, there is at least some such real essence shared by some spatiotemporal objects. But, by the same token, formal concepts represent the very structure

Metaphysics on the Model of Natural Science? 353 they ground, or in contemporary terms, they “carve” that structure “at its joints”. Given that the formal structure of our experience grounds the formal structure of the phenomenal world, there is no possibility that concepts in which that former structure is articulated will fail to correspond to the joints in the latter structure. But this formal joint carving does not, by itself, entail that empirical specifications of these formal concepts will refer, much less that they will carve at the explanatory joints. It might be, for instance, that while our concept of carves at the joints (Nature is divided into natural kinds), none of our empirical concepts of specific natural kinds (e.g., ) refer, or, insofar as they do, they refer to gerrymandered kinds that do not carve at the joints (like ). This gap is meant to be filled by the principle of purposiveness: if Nature exists for the sake of being comprehended by us, then, at least under the right conditions, our empirical natural kind concepts (and concepts of corresponding laws) will refer and carve at the joints. It may be that, at any given time, some of our empirical natural kind concepts fail to refer or carve at the joints, but, in the fullness of time, these will be replaced by empirical concepts that do carve at the joints. We are rationally warranted in inquiring into Nature as though this is the case, so we are rationally warranted in inquiring into Nature as though K3 is true. As with K2, whether this constitutes an explanation of why K3 is true, or merely of our rational warrant for assuming it in inquiry, remains unclear. But, again, rather than address that question directly, I will instead move on to consider how to mount a Kantian critique of abductive metaphysics. 15.4  A Kantian Critique of Abductive Metaphysics The previous section established that Kant has at least the beginnings of an explanation of K1–K3. But this means that there is a natural Kantian critique of abductivism, beyond the mere insistence (rejected in any case by the abductivist) that metaphysics must have apodictic certainty, which of course abduction can never give us. Kant (or “the Kantian”) can ask, What is the abductivist explanation of A1–A3? I think we can immediately dispense with one answer, which at least some abductivists will want to give: whatever explains the possibility of abduction in general explains A1–A3.53 The reason that this answer by itself does not suffice is that, according to abductivism itself, metaphysics is more general than natural science.54 Since metaphysics is more general, it is at least logically possible that abduction is alethically and explanatorily successful in natural science, but not in metaphysics.55 It might, for instance, be the case that abduction is successful when we restrict our abductive premises and conclusions to natural scientific phenomena and

354  Nicholas Stang theories, but that it systematically misfires when we extend our inferences to the topics of metaphysics. For instance, it might be that the criteria by which we judge the “bestness” of an explanation in natural science does not apply in metaphysics, so that when we abduct by those criteria in metaphysics, our inferences systematically go wrong. In a Kantian vein, the fact that metaphysics since the time of Aristotle has been ‘a battlefield of endless controversies’ (Aviii) might even be taken as positive evidence that abduction does indeed fail in metaphysics, that when we extend our inquiry beyond the bounds of natural science we are left without any sufficient criteria for theory choice, and all we have is a ‘mock combat’ (Bxv). With respect to A1, to my knowledge, no contemporary abductivist metaphysician attempts to explain why the world includes ontic explanatory structure necessary to “back” their explanatory theories. This is no accident. For, assuming that explanations must themselves be backed by explanatory structure, any explanation we might give of one ontic explanatory structure would require another one, ad indefinitum.56 The dominant view among abductive metaphysicians is that at this point we reach explanatory “bedrock” and we cannot go any further. Matters are slightly more complicated with A2. Again, I’m not aware of any abductive metaphysician who attempts to explain why A2 is true. The most that abductive metaphysicians offer is a “good company” argument: we adopt similar canons of explanatory reasoning (e.g., Occam’s razor) in physics and the rest of natural science as we do in metaphysics, so whatever explains why these are truth- and explanation-tracking in a non-lucky (non-Gettierable) fashion in the former case presumably does so in the latter as well.57 But, as I argued in Section 15.2, A2 is false without A3: unless our language carves the world at its explanatory joints, our abductive inferences will not generate explanatory knowledge. Furthermore, A1–A3 as claims about abduction in metaphysics are logically separable from corresponding claims about abduction in natural science. Assuming that abduction in natural science is alethically and explanatorily successful, neither entails that, nor explains why, abduction in metaphysics is alethically or explanatorily successful. It might be, for all that contemporary abductivists have shown, that we come to explanatory knowledge of natural phenomena through abduction in natural science, but these abductive methods systematically fail to produce knowledge when applied to metaphysics. In the case of A3, there is a prominent philosopher, Ted Sider, who does purport to explain why the terms in our metaphysical theories carve the world at its joints. Sider (2011) gives a theory of reference in general, but, unlike many other reference theorists, he also explicitly applies his account to metaphysical reference.58 Sider argues that the world has an objectively

Metaphysics on the Model of Natural Science? 355 privileged structure and the aim of metaphysics is to uncover that structure. He further holds that explanations must be couched in terms that carve that structure at its joints; of two putative explanations, the one that carves closer to the objective structure of the world is the better explanation. “Structure” in his account is thus playing a role similar to the role played by “ontic explanatory structure” in mine: the very same structure that “backs” explanations is also the structure we aim to uncover in metaphysics. If we idealise somewhat and call the conjunction of all the sentences we hold to be platitudes or quasi-definitional of our terms our “theory”, the first part of Sider’s view is straightforward: any model of our theory (any assignment of objects to singular terms and extensions to predicates on which the theory is true) is a possible interpretation of the theory.59 However, as Putnam (1981) pointed out, for any such model with a domain larger than a single object, there is a “permuted” model, which is also an interpretation of the original theory.60 In general, these permuted interpretations will be grossly gerrymandered; they will assign intuitively bizarre objects and extensions to the singular terms and predicates of our language. In virtue of what is the intended interpretation of our theory the correct one? How do we eliminate, in a principled fashion, these gerrymandered models? Sider follows Lewis (1984) in thinking that what determines reference is that some reference candidates are intrinsically more eligible for reference. Whereas Lewis originally restricted the notion to objects and predicates (some objects are more natural than others, some predicates pick out more natural properties than others), Sider extends the notion (which he calls being ‘structural’) to any item of any syntactic category whatsoever: quantifier domains, operators, relations of any adicity, even the logical connectives themselves can be said to be more or less structural. Among, the possible interpretations of our theory, the correct interpretation maximises structuralness.61 Reference is determined by descriptive fit plus structuralness. Sider’s “official” notion of structure is an absolute one: an item is either structural or it is not.62 But reference cannot be determined by descriptive fit plus absolute structure, for that would entail the absurd result that we only ever refer to the absolutely structural, i.e., the fundamental metaphysical structure of reality. Not only would this make most of ordinary thought and talk impossible, it would also make Sider’s meta-semantics impossible, for it would make it impossible for us to refer to our own language, which, for Sider, is not itself absolutely structural. (The English language, plausibly enough, is not part of the fundamental structure of reality.) Thus, Sider’s view is that reference is determined by descriptive fit plus comparative structuralness. However, Sider never gives us more

356  Nicholas Stang than an intuitive sketch of what makes one item more structural than another. At one point he suggests an account of comparative structuralness in terms of the length of metaphysical definitions: s is more structural than s* if and only if the metaphysical definition of s in absolutely structural terms is shorter than the definition of s*.63 But Sider immediately rejects this account, for it has counter-intuitive consequences. For instance, if F and G are absolutely structural properties, then the proposed analysis entails that the disjunctive property being F or G is equally structural as the conjunctive property being F and G. But since disjunctions are less explanatory than conjunctions, and the structural is supposed to be explanatory, disjunctions should be less structural than conjunctions, even though conjunction and disjunction contribute the same amount to the length of metaphysical definitions. Sider never provides us with a replacement account of this flawed definition of comparative structuralness. Reference is a relation between terms in our language and entities and structures in the world. As we have seen, it is one among many such reference-like relations, i.e., relations between terms in our language and entities in the world that preserve the truth of our theory, but assign as the “meanings” of those terms something other than their intended meanings (their referents).64,65 Sider’s official view is that reference is the unique maximally structural reference-like relation. Alternatively, there might have been a tie, i.e., multiple reference-like relations that are equally structural. I will argue that both possibilities lead to serious problems.66 Evaluating these two possibilities is difficult because they both involve comparisons of structuralness between (reference-like) relations and, as we have seen, Sider never commits to a precise account of what comparative structuralness consists in. But as a matter of simple logic, either the structuralness of a non-absolutely structural relation partly supervenes on the structuralness of its relata or it does not. But I think we can quickly dispense with the second option. If the structuralness of a relation does not supervene on the structuralness of its relata at all, nothing prevents arbitrarily structural relations holding among arbitrarily non-structural relata. But, surely, if a relation holds exclusively among arbitrarily non-structural items, e.g., the permuted predicate extensions or arbitrary mereological fusions generated by Putnam-style arguments, then this must make that relation comparatively non-structural. So I think Sider has to admit that the structuralness of non-absolutely structural relations partly supervenes on the structuralness of their relata. However, this partial supervenience is in tension with the uniqueness of the reference relation. All of these reference-like relations agree on their first relata (words in our language), so, assuming partial supervenience, differences in their structuralness must supervene on differences in

Metaphysics on the Model of Natural Science? 357 the structuralness of their second relata, the meanings they assign to our words. What is more, on Sider’s view, they must fully supervene on these differences. Sider’s claim is that reference is more structural than other reference-like relations because it assigns more structural meanings. If he were to claim that reference is more structural than some other reference-like relation, even though they both assign equally structural meanings, it would no longer be clear what he means by the structuralness of a referencelike relation. If the structuralness of a reference-like relation supervenes on the structuralness of the items it assigns as meanings, we should expect that a slight decrease in structuralness in one item can be compensated for by a corresponding difference in structuralness in another item. But then why can’t one reference-like relation assign a slightly less structural item as the meaning of one singular term and a slightly more structural item as the meaning of another singular term, while remaining overall just as structural as reference itself? For instance, this reference-like relation might assign a slightly more structural meaning to the name “Ted Sider” and a correspondingly slightly less structural meaning to “Immanuel Kant”. If Sider wants to maintain that there is a unique maximally structural reference-like relation, this is an assumption that stands in need of explanation as much as anything does. On this view, the possibility of reference is highly sensitive to the measure of comparative structuralness over the set of possible reference-like relations. Sider never makes this measure precise, nor gives any explanation of why it has a maximum. So it is worth considering what other options Sider has. It would be overall more plausible for Sider to admit that there is no unique such maximally structural reference-like relation. “Reference” would then refer indeterminately to a family of equally structural referencelike relations. Sider has a model for terms like this; because there is no uniquely maximal reference candidate, there is no fact of the matter as to which of the equally maximal reference candidates it refers to. In this case, Sider says, it is a “merely verbal dispute” to what it refers. But this, combined with the view that the structuralness of a relation supervenes on the structuralness of its relata, potentially undermines Sider’s whole picture. Recall that we can compensate for a decrease in the structuralness of one assigned meaning with an increase in the structuralness of another. If this is the case, then for one reference-like relation R that assigns absolutely structural items as the meaning of some of our terms, we should expect there would be another reference-like relation R* that assigns less than absolutely structural items as the meaning of those terms, but slightly more structural items as the meaning of other terms. But if this is correct, then there is no fact of the matter whether we ever refer to the absolutely structural. Some reference candidates for “reference” will

358  Nicholas Stang assign absolutely structural items as the meaning of terms in fundamental metaphysics; some will assign somewhat less absolutely structural meanings to those terms, but more (but still not absolutely) structural meanings to other terms. But since metaphysics (on Sider’s view) is concerned with absolute structure, metaphysics will not be possible. It is not determinately the case that we ever succeed in referring to the absolute structure of reality. But this would mean Sider had not explained the possibility of metaphysics. It would remain a “verbal dispute” as to whether we have ever succeeded in talking about the fundamental metaphysical structure of reality. I have posed a dilemma for Sider: either reference is the uniquely maximally structural reference-like relation, or there is no such unique maximally structural relation. On the first horn, metaphysics is possible, but only if we assume that the metric of comparative structuralness over reference-like relations has a maximum. On this horn, Sider has explained the possibility of metaphysics, but by appeal to an assumption that is as much in need of explanation as anything else. On the second horn, for all Sider has shown, among the family of equally structural reference-like relations, there may be some that never assign absolutely structural meanings. In this case, Sider will have failed to explain the possibility of talk about the absolute structure of reality, i.e., metaphysics. *** Contemporary abductivists thus fail to explain A1–A3.67 From a Kantian point of view, therefore, abductivism in metaphysics appears to be a form of dogmatism: ‘the prejudice that without criticism reason can make progress in metaphysics’ (Bxxx). The abductivist offered the beginnings of an explanation of how metaphysics is possible, through abduction, but did not carry that through by offering an explanation of why abductive inference in metaphysics is knowledge-generating. Abductivism is not wholly uncritical (e.g., it admits that when inference to the best explanation gives out, so too does our justification for metaphysical theorising), but nor does it rise to level of a ‘critique of the faculty of reason in general’ (Axii). In particular, abductivism dogmatically assumes without explanation that the canons of abductive reasoning that generate knowledge in natural science will do so in metaphysics as well. What is more, our discussion up to this point shows that the charge of “dogmatism” is not simply a Kantian pejorative; it articulates an immanent critique of abductivism. Abductivism, doing metaphysics by inference to the best explanation, does not issue in a good explanation of the possibility of metaphysics. By abductivism’s own lights, then, a metaphysics that could explain its own possibility would be ceterus paribus superior to one that does not. Consequently, such a theory is ceterus paribus preferable to abductivism.

Metaphysics on the Model of Natural Science? 359 15.5 Conclusion Some abductivists will seek to defend abductive metaphysics through a “good company” objection: any reason to be sceptical about abduction in metaphysics is a reason to be, implausibly, sceptical about abduction in natural science. But this misrepresents the dialectic. Kant accepts that abduction is alethically and explanatorily successful, as long as it restricts itself to the spatiotemporal world and obeys certain regulative (and transcendental) principles (see Section 15.2). The greater generality of metaphysics leaves logical room for abduction to be successful in natural science without it being successful in metaphysics. Furthermore, the empirical success of natural science over the past several hundred years makes it overwhelmingly plausible that the natural sciences are at least alethically, and perhaps even explanatorily, successful. Metaphysics has no such string of successes to point to, for it is a battlefield of endless controversies. In the face of this, the lack of a satisfying explanation of its possibility is a powerful challenge to the science of metaphysics. Some will respond that the fact that abductive metaphysics cannot explain its own possibility, i.e., it remains dogmatic, is a compelling reason to reject it only if some other conception of metaphysics can do better. Kant would agree; he does not think that the third option, to reject metaphysics altogether (to “feign indifference” to it), is a viable option for human reason, for our moral vocation (the Highest Good) depends on certain metaphysical foundations (i.e., free will, the existence of God, the immortality of the soul). Kant also, of course, thinks he has an explanation of the possibility of metaphysics, one that relies on transcendental idealism. Whether that explanation succeeds or not, however, is outside the scope of this chapter. In conclusion, I just want to note that even if Kant’s transcendental idealist explanation of metaphysics is beset with insuperable problems, I think, it is, this does not automatically redound to the credit of the abductivist, for there is a whole separate family of metaphysical views still to be considered: the metaphysical views of the postKantian idealists, who agreed with Kant in rejecting pre-Kantian dogmatism and would have agreed with him in rejecting contemporary abductivism, but who thought we had to go beyond Kantian transcendental idealism to obtain a properly critical metaphysics, i.e., one that can explain its own possibility. But that is a story for another time.68 Abbreviations for the Work of Kant All works of Kant are cited according to volume and page number in the ‘Akademie Ausgabe’ (AA): Immanuel Kant. 1900-. Gesammelte Schriften. Ed. Vol. 1–22: Preussische Akademie der Wissenschaften, Vol. 23: Deutsche

360  Nicholas Stang Akademie der Wissenschaften zu Berlin; from Vol. 24: Akademie der Wissenschaften zu Göttingen. De Gruyter.  ritik der reinen Vernunft: A edition (1781, AA 4), K B edition (1787, AA 3) BDG Der einzig mögliche Beweisgrund zu einer Demonstration des Daseins Gottes (AA 02) EEKU Erste Einleitung in die Kritik der Urteilskraft (AA 20) KU Kritik der Urteilskraft (AA 5) Log Jäsche Logik (AA 9) MAN Metaphysische Anfangsgründe der Naturwissenschaften (AA 4) Refl. Reflexionen (AA 14–19) ÜE Über eine Entdeckung, nach der alle neue Kritik der reinen Vernunft durch eine entbehrlich gemacht werden soll (AA 8) V-Lo/Blomberg Logik Blomberg (AA 24) V-Lo/Busolt Logik Busolt (AA 24) Vo-Lo/Dohna Logik Dohna-Wundlacken (AA 24) V-Lo/Philippi Logik Philippi (AA 24) V-Lo/Pölitz Logik Pölitz (AA 24) V-Lo/Wiener Wiener Logik (AA 24) V-Met/Dohna Metaphysik Dohna (AA 28) V-Met/Herder Metaphysik Herder (AA 28) V-Met/Mron Metaphysik Mrongovius (AA 29) V-Met-L2/Pölitz Metaphysik Pölitz (AA 28) V-Met/Schön Metaphysik von Schön (AA 28) V-Met/Volckmann Metaphysik Volckmann (AA 28) A/B

Notes 1 Here, and henceforth, “sciences” refers simply to natural sciences. 2 I am going to assume that it is knowledge that the abductivist is after in metaphysics. Whether the abductivist could set their sights on a lower epistemic status (e.g., justified true belief) I will not consider here. 3 “Abductivism” refers to a set of doctrines shared by many contemporary metaphysicians, not to a single thesis. So I am not claiming that scientific realism follows logically from abductive methodology or abductive epistemology (it doesn’t). I am claiming only that abductivism, as actually endorsed by practising metaphysicians, contains this commitment. 4 A369, A491/B519. In Kantian terms, the primary topic of contemporary metaphysics is things in themselves, although there is also some contemporary work on the metaphysics of appearance as well (e.g., McDaniel 2017: 140–169). 5 For example, Biggs 2011; Hawley 2006; Paul 2012; Sider 2009: 385, 2011: 12–15; and Williamson 2016. Some of these references are drawn from Biggs and Wilson 2017: 739.

Metaphysics on the Model of Natural Science? 361 6 Technically, for Kant, a critique is a critique of a capacity (Axii), e.g., practical reason or the power of judgement. Without too much distortion, though, we can see abduction as capacity as well, although not all abductivists would agree that it is a basic capacity (as argued in Biggs and Wilson 2017). 7 This does not mean it is a successful critique, i.e., a fully adequate explanation of how metaphysics is possible. My argument in this chapter is precisely that it does not succeed in that ambition. 8 All abductive metaphysicians with whom I am familiar admit that metaphysics is less abductively secure than natural science, and that there are questions in metaphysics that are epistemically undecidable because there are not sufficiently good grounds to abduct to a single answer or set of answers, though they differ on where the “limits” of metaphysics lie (cf. Bennett 2009 and Sider 2011: 12). 9 For Kant, metaphysics is concerned with conditions (grounds), either the conditions of possible experience (immanent metaphysics) or the conditions of supersensible objects (transcendent metaphysics). See A845/B873. 10 For Kant there is also transcendent metaphysics of the mind-independent (supersensible), but that is either grounded practically (in the moral law) or is merely speculative and fails to constitute cognition. In either case, it lies outside the scope of this chapter. 11 Log 9: 84–85; V-Lo/Dohna 24: 743; V-Lo/Wiener 24: 888. 12 Axv; Log 9: 86; Refl. 2680, 16: 467; Refl. 5645, 18: 293; V-Lo/Busolt 24: 647; V-Lo/Pölitz 24: 559. Cf. Bxxii on how what is initially merely the Copernican hypothesis (that objects conform to our capacity for cognition) becomes apodictically certain through the KrV itself. 13 There are two other Kantian reasons to reject abductive metaphysics. First, metaphysics must be a priori, but abduction is empirical. However, few contemporary metaphysicians would accept that metaphysics must be a priori, and some even deny that abduction is empirical (e.g., Biggs and Wilson 2017). Second, the abductive conclusion functions initially as a hypothesis, and the hypothesis must be known apodictically to be at least really possible (A770/ B798). But this just reiterates the questionable assumption that metaphysics must be apodictic at the level of the possibility. 14 MAN, 4: 468. 15 As Kant argues in the Appendix to the Transcendental Dialectic, “On the regulative use of ideas of pure reason” (A642–668/B670–696). See Section 15.4 for discussion. 16 For example, Leibniz gives an essentially abductive justification for pre-established harmony in the New System of Nature; see Leibniz 1989: 144–145. 17 In BDG, Kant endorses an essentially abductive argument for the existence of a ‘rational Author’ from the ‘purposeful provisions’ in all things, in addition to the more famous apodictic proof from the real possibility of all things (2: 159). 18 Locke is a possible exception, but further exploring that connection lies outside the scope of this chapter. 19 The locus classicus is Lipton 2004. 20 I take it that this distinction should be acceptable to both those whom Raven (2015) calls “unionists” about metaphysical explanation (e.g., Dasgupta 2014, Fine 2012, and Rosen 2010), and those he calls “separatists” (e.g., Audi 2012, Schaffer 2012). 21 One can have a more deflationary conception of explanation (e.g., a broadly pragmatist one) but this will push one’s metaphysics in a more deflationary direction. Cf. Dasgupta 2018.

362  Nicholas Stang 22 I have put this point at the level of language, but it could also be put at the level of thought: our concepts must carve the world at its explanatory joints. 23 Assuming grounding is a relation between facts. If not, adjust the example accordingly. 24 See the texts cited in the Introduction. 25 This material is presented more fully in Stang n.d.1. 26 I include the qualification ‘in virtue of its very form’ because how, and whether, the matter of experience grounds the matter of the objects of experience is complex and controversial. For discussion, see Stang 2016a and 2018. 27 MAN 4: 468, 469. Kant there talks of nature, but in the case of causally efficacious substances (which is what laws are about), natures are real essences (see 4: 468n.). 28 For further discussion, see Stang 2016b: 234–244. 29 This raises the question of which is explanatorily prior: law or real essence. I think real essence is prior; the necessity of laws is grounded in real essences. I argue for this at greater length in Stang 2016b. 30 ÜE 8: 238, Log 9: 143, Br. 11: 37, V-Lo/Wiener 24: 839, and V-Lo/Dohna 24: 757. 31 Log 9: 61, ÜE 8: 229, Br. 11: 36, Refl. 5706, V-Lo/Blomberg 24: 116, V-Lo/ Philippi 24: 408, 456, V-Lo/Pölitz, 535–536, V-Lo/Busolt 24: 634, V-Lo/Dohna 24: 727, V-Lo/Wiener 24: 839, V-Met-L2/Pölitz, 28: 553. 32 MAN, 4: 509, 511; V-Met-L2/Pölitz, 28: 553. 33 A727–728/B755–756, Refl. 3966, V-Lo/Blomberg 24: 116–117, V-Lo/Dohna 24: 757, V-Lo/Wiener 24: 919. 34 This is only true of given empirical concepts; in the case of made empirical concepts (concepts of artefacts, concepts of invented fictions), the logical essence and the real essence coincide. For the given-made distinction, see Log 9: 93. 35 “Essentialist” readings of Kantian laws of nature similar to that presented here and in Stang 2016b have also been defended by Watkins (2005) and Kreines (2008). 36 V-Lo/Volckmann 28: 422; V-Met/Schön 28: 503; V-Met-L2/Pölitz 28: 560. 37 This of course just raises the question of the ontological statues of essences themselves. Are they also part of Kant’s “ontology”?. For reasons of space, I will not address that question here. 38 I want to remain neutral on whether the identity of the natural kind fully individuates empirical concepts, i.e., whether, as seems plausible, there could be two distinct empirical concepts of one and the same natural kind (e.g., and ). 39 Kant thinks we only ever know, at most, parts of the real essences of things; see Br 11: 37, V-Lo/Dohna 24: 728, V-Met-L2/Pölitz 28: 553. 40 This is why Kant claims that not merely the possession, but the very possibility, of an empirical concept of a force depends upon the actual existence of that force in space and time. If that force did not exist (if a force without its real essence did not exist), no concept we form would be a concept of that very force. See B252, MAN 4: 486–487. 41 This follows from a point Kant repeatedly makes in his lectures − that the marks contained in (given) empirical concepts are typically so impoverished as to be virtually uninformative; see LB, 24: 116, 117–118, 271; WL 24: 919, A727–728/B755–756. 42 By “there being” a set of empirical concepts I mean that were we to engage in the relevant acts of concept formation, we would form those concepts. So the “being” of those concepts, for the purposes of this argument, is independent of whichever acts of concept formation we actually perform.

Metaphysics on the Model of Natural Science? 363 43 A368; V-Met/Volckmann, 28: 401, 408; V-Met-L2/Pölitz 28: 548; V-Met/ Dohna 28: 624, 628; V-Met/Mron 29: 808, 818. 44 They are impure because they involve the empirical concept of matter (see B for the notion of the impure a priori). For the transcendental-metaphysical distinction, see MFNS 4: 469–470; KU 5: 181. 45 In the terms of the Introduction, it produces only comparative, not apodictic, certainty. 46 See A650–651/B678–679, where a transcendental principle is said to underlie the logical/regulative principle of systematicity as such; at A656/B684 he claims there is a transcendental principle of specification and, at A660/B688, of continuity. Pickering (2011) reads these references to a transcendental principle, implausibly, as references to transcendental illusion. 47 I agree with Geiger (2003: 275–282) that the systematicity of nature is a transcendental principle (and with his trenchant criticisms of prior readings), but I do not accept his argument for that conclusion. 48 This is because “experience” for Kant has a much richer, more determinate, and more scientific content than the relatively minimal “experience” of the empiricist tradition. See Stang 2018 for discussion. 49 A slightly stronger conclusion is warranted: since every object is governed by causal laws, and laws concern natural kinds (objects in virtue of a common real essence), every object belongs to some common natural kind. 50 An especially clear statement of the problem occurs at EEKU 20: 208–209; see also KU 5: 183, 186, 188. 51 KU 5: 180. 52 It might seem obvious to some readers that Kant’s explanation can only be an epistemic one (of our rational warrant for assuming K2) rather than a metaphysical one (of the truth of K2). But, arguably, this depends upon assuming a transcendental realist view of Nature on which the “being” of Nature can transcend what we, in principle, could have rational warrant for judging about it (i.e., assuming that the epistemology and metaphysics of Nature could come that far apart). But I will leave this issue aside. 53 In conversations with abductive metaphysicians, I have found this a common manoeuvre. 54 A more strictly naturalist position, on which metaphysics is simply identical to natural science, is not targeted by the arguments of this chapter. 55 It is even clearer that Kant himself cannot accept this answer, for natural scientific abduction is limited to objects of possible experience, while metaphysics makes claims that are either beyond the bounds of possible experience (transcendent metaphysics) or claims about the forms of experience themselves (immanent metaphysics). That abduction in the former cause is successful is no reason to think it would be successful in the latter cases. For the immanenttranscendent distinction, see A846/B874. 56 Schaffer (2003) raises the intriguing possibility of an infinite downward regress of (in my terms) ontic explanatory structure, but I will not pursue that thought any further here. 57 Notice that a Bayesian account of abduction alone will not suffice: even if we build our canons of abductive theory choice into our priors (e.g., assigning a prior probability to more unified explanations), this will not account for why reasoning according to such priors constitutes knowledge (why it non-accidentally tracks the truth). What is more, those priors must be formulated in a language that carves the world at its explanatory joints. So a Bayesian account by itself will not

364  Nicholas Stang explain A2 or A3. See Bradley 2020 for further discussion. Thanks to Trevor Teitel for helpful conversations about this issue. 58 Williams 2019 offers an interpretationist meta-semantics in a similar spirit, although it departs from Sider in key respects (56: n. 26; 67: n. 10). Unlike Sider, though, Williams does not explicitly apply it to reference in metaphysics. In Stang n.d.2, I argue that Williams’s theory does not adequately explain why metaphysics is semantically possible. 59 I follow Sider in making the simplifying assumption that our theory is firstorder. This assumption is potentially fraught, however, because, as Bayes (2001) shows, Putnam’s original argument fails in higher-order theories. But I will not pursue this issue further here. 60 The same idea underlies Quine’s argument for the inscrutability of reference; see Quine 1960. Putnam’s original argument used the upwards and downwards Löwenheim-Skolem theorems to show that if our theory has any infinite models, for any infinite cardinality it has a model of that cardinality. Lewis (1984) points out that the main result follows from much simpler considerations about the permutation of the domain. 61 In fact, it is more complicated than this; see below. 62 Sider 2011: 128. Because anything of any syntactic type whatsoever can be evaluated for its structuralness, I choose the dummy term “item”. Bear in mind that the structuralness of a term in our language is, in general, distinct from the structuralness of its referent. Even if spacetime points are absolutely structural, names for them are not, for names are not (according to Sider) part of the fundamental structure of reality. 63 Sider 2011: 129–133. 64 Following Frege, it is standard in analytic philosophy to distinguish between two notions of meaning, sense and reference. In line with Sider’s approach, I am ignoring sense and focusing only on meaning as reference. 65 I will speak of reference-like relations as assigning “meanings” to terms, because it would be needlessly confusing to speak of them as assigning “referents” to them (only reference assigns their referents) and needlessly prolix to talk about “their semantic value assigned them by that reference-like relation” (or something of that ilk). But just to be clear: the meaning of some term according to a reference-like relation is, in general, not its intended or actual meaning (its referent), unless that reference-like relation is reference itself. 66 Talk about reference-like relations is just a way of talking about interpretations of the language different from the intended interpretation. But for ease of exposition I focus on the (reference-like) relation between terms and the meanings assigned to it by different interpretations, rather than on the interpretations themselves. 67 This does not mean that abductivism cannot explain A1–A3. However, I think my discussion so far shows that the prospects for such an abductivist explanation are bleak. 68 I would like to thank audiences at the University of Kansas and the participants in a “block seminar” at the University of Bonn for their helpful feedback on earlier presentations of this material. Brad Cokelet, Catharine Diehl, Alex Englander, Tyler Hildebrandt, Karl Schafer, and Trevor Teitel gave me invaluable feedback and comments as well. Special thanks go to Robb Dunphy for extensive comments on the penultimate draft, which vastly improved the (still imperfect) final chapter.

Metaphysics on the Model of Natural Science? 365 Bibliography Audi, P. (2012) “Grounding: Toward a Theory of the In-Virtue-Of Relation” Journal of Philosophy Volume 109, pp. 685–711 Bayes, T. (2001) “On Putnam and His Models” Journal of Philosophy Volume 98(7), pp. 331–350 Bennett, K. (2009) “Composition, Colocation, and Metaontology”. In Chalmers, D., Manley, D., and Wasserman, R. (Eds.) (2009) Metametaphysics: New Essays on the Foundations of Ontology (Oxford: Oxford University Press) Biggs, S. (2011) “Abduction and Modality” Philosophy and Phenomenological Research Volume 83, pp. 283–326 Biggs, S. and Wilson, J. (2017) “The A Priority of Abduction” Philosophical Studies Volume 174, pp. 735–758 Bradley, D. (2020) “Naturalness as a Constraint on Priors” Mind Volume 129(513), pp. 179–203 Chalmers, D., Manley, D., and Wasserman, R. (Eds.) (2009) Metametaphysics: New Essays on the Foundations of Ontology (Oxford: Oxford University Press) Correia, F. and Schnieder, B. (2012) Metaphysical Grounding: Understanding the Structure of Reality (Cambridge: Cambridge University Press) Dasgupta, S. (2014) “On the Plurality of Grounds” Philosophers’ Imprint Volume 14, pp.1–28 Dasgupta, S. (2018) “Realism and the Absence of Value” Philosophical Review Volume 127(3), pp. 279–322 Fine, K. (2012) “Guide to Ground”. In Correia, F. and Schnieder, B. (2012) Metaphysical Grounding: Understanding the Structure of Reality (Cambridge: Cambridge University Press). Geiger, I. (2003) “Is the Assumption of a Systematic Whole of Empirical Concepts a Necessary Condition of Knowledge?” Kant Studien Volume 94(3), pp. 273–298 Goodman, N. (1944) “The New Riddle of Induction”. In Fact, Fiction, and Forecast (Cambridge, MA: Harvard University Press) Hawley, K. (2006) “Science as a Guide to Metaphysics?” Synthèse Volume 149, pp. 451–470 Kreines, J. (2008) “Kant on the Laws of Nature: Laws, Necessitation, and the Limitation of Our Knowledge” European Journal of Philosophy Volume 17(4), pp. 527–558 Leibniz, G.W. (1989) Philosophical Essays R. Ariew and D. Garber (Trans. and Eds.) (Hackett) Lewis, D. (1984) “Putnam’s Paradox” Australasian Journal of Philosophy Volume 62(3), pp. 221–236 Lipton, P. (2004) Inference to the Best Explanation, 2nd ed. (London: Routledge) McDaniel, K. (2017) The Fragmentation of Being (Oxford: Oxford University Press) Paul, L. (2012) “Metaphysics as Modeling: The Handmaid’s Tale” Philosophical Studies Volume 160, pp. 1–29 Pickering, M. (2011) “The Idea of the Systematic Unity of Nature as a Transcendental Illusion” Kantian Review Volume 16(3), pp. 429–448

366  Nicholas Stang Putnam, H. (1981) Reason, Truth, and History (Cambridge: Cambridge University Press) Quine, W.V.O. (1960) Word and Object (Cambridge, MA: Harvard University Press) Raven, M. (2015) “Ground” Philosophers’ Compass Volume 10(5), pp. 322–333 Rosen, G. (2010) “Metaphysical Dependence: Grounding and Reduction”. In B. Hale and A. Hoffmann (Eds.) Modality: Metaphysics, Logic, and Epistemology (Oxford: Oxford University Press) Schaffer J. (2003) “Is There a Fundamental Level?” Noûs Volume 37(3), pp. 498–517 Schaffer, J. (2009) “On What Grounds What”. In Chalmers, D., Manley, D., and Wasserman, R. (Eds.) (2009) Metametaphysics: New Essays on the Foundations of Ontology (Oxford: Oxford University Press) Schaffer, J. (2012) “Grounding, Transitivity, and Contrastivity”. In Correia, F. and Schnieder, B. (2012) Metaphysical Grounding: Understanding the Structure of Reality (Cambridge: Cambridge University Press) Sider, T. (2009) “Ontological antirealism”. In Chalmers, D., Manley, D., and Wasserman, R. (Eds.) (2009) Metametaphysics: New Essays on the Foundations of Ontology (Oxford: Oxford University Press) Sider, T. (2011) Writing the Book of the World (Oxford: Oxford University Press) Stang, N. (2015) “Who’s Afraid of Double Affection?” Philosophers’ Imprint Volume 15(18), pp. 1–28 Stang, N. (2016a) “Appearances and Things in Themselves: Actuality and Identity” Kantian Review Volume 21(2), pp. 283–292 Stang, N. (2016b) Kant’s Modal Metaphysics (Oxford: Oxford University Press) Stang, N. (2018) “Kant, Cohen, and the Concept of Experience”. In C. Damböck (Ed.) Philosophie und Wissenschaft bei Hermann Cohen (Cham: Springer), pp. 13–40 Stang, N. (n.d.1) “Kant on Natural Kinds”. Unpublished manuscript Stang, N. (n.d.2) “Is Kant’s Critique of Metaphysics Obsolete?” https://philpapers. org/rec/STAIKC Watkins, E. (2005) Kant and the Metaphysics of Causality (Cambridge: Cambridge University Press) Williams, J.R.G. (2019) The Metaphysics of Intentionality (Oxford: Oxford University Press) Williamson, T. (2016) “Abductive Philosophy” The Philosophical Forum Volume 47(3–4), pp. 263–280

Index

a posteriori 37, 100n21, 168, 182n10, 228, 260, 297, 305 a priori 2, 8–9, 12, 14, 29–33, 36–37, 100n21, 105–106, 111–112, 115–122, 126, 129, 134, 139–140, 143, 150–155, 159–160, 163, 167–176, 179–180, 182, 185–191, 193, 196–197, 199n11, 200n17, 206–208, 210, 212, 215, 217, 220n1, 222, 223n21, 229, 244, 253, 260–262, 296–300, 305, 307, 311n2, 313n24, 350–352, 361n13, 363n44 abductivism 16–17, 339–347, 349–351, 353–354, 358–361, 363–364 actuality 21–22, 30, 37, 63, 100–101, 110–111, 127n.2, 219, 229, 265, 288n15 Allais, Lucy 115, 222n16 Allison, Henry 7, 106, 113–115, 121–125, 128–129, 221 Ameriks, Karl 115, 128n7, 199, 201n25, 222n16, 246n10 antinomy 117, 119, 121, 123–125, 137, 140, 300, 302, 312n13, 318–320, 326–327, 329, 331n12, 333–334 appearances 5, 7, 75, 85, 99n3, 100n18, 106, 109, 111–121, 125–130, 137–138, 169, 187, 191, 200n17, 202n41, 221n12, 229, 319, 322, 326, 333n28, 347, 360n4 apperception 80, 109, 128n8, 209, 212, 221n8, 222n17, 308

arbitrariness 12, 17, 65, 152, 159–160, 235, 251, 264, 267, 273, 311n10, 356 aspirational 6, 242–243, 247n23, 248n37 atheism 7–8, 137–138 Baum, Manfred 199n5, 200n19, 222n15 being 4, 20, 28, 35–36, 41, 126, 134, 259, 264, 270, 276–283, 286, 294, 320, 323, 332n20, 347, 362n42, 363n52 Beiser, Frederick 45n103, 198n4, 222n20, 229, 236, 239, 246–247, 259–260 Berry, Jessica 11–12, 227–228, 230–241, 243–244, 246–247 Breazeale, Daniel 199n9, 202, 245–246, 267–268, 288n13 capacities 14–16, 36, 58, 108–109, 150, 154, 184, 187, 193–195, 197–198, 201, 202n44, 221n4, 294–300, 304–306, 309–310, 311n6, 319, 346, 361 categories 10–12, 14, 89, 101n26, 107, 109–110, 116–119, 122–126, 128, 129n15, 130n17, 168, 171–176, 179–180, 187, 189, 191, 193, 199n11, 200n20, 201n23, 206–213, 217–221, 223n29, 229–230, 253, 259–261, 267n6, 300, 305–310, 311n10, 316–318, 321, 323–324, 326, 328–332; deduction of 10–12,

368 Index 117, 119, 128, 168, 191, 201n23, 206–224, 229, 232, 235, 247n26, 253, 259, 261–263, 267n6, 268n11, 307–308, 332n14 causality 29, 32–33, 35, 55, 133–134, 170, 178, 229, 238, 260–261, 265, 322 342, 350, 352 circularity 65, 223n27, 242; vicious, 11, 62, 112, 173–175, 201n26, 207, 209, 211–212, 215, 217, 220, 221n9, 222n15, 247n26, 311n10, 345 cognitive dualism 14, 297–299, 315, 319–320 consciousness 43n64, 102n30, 109, 193–194, 202–203, 212–213, 215, 219, 222n17, 223, 228, 261, 263, 268n11, 283, 288n14, 301, 303, 307, 309, 312n11, 319, 321–323, 328, 332; principle of 10–11, 193–194, 197, 200n16, 212–213, 217, 219–220, 221n5, 222n18, 228, 244n1, 245, 246n13 cosmology 4, 19, 26, 28, 34–36, 38, 40n6, 42n50, 56, 58, 61, 125, 133, 139–140, 143, 326, 333n30 critique 6–7, 10, 13, 16, 77, 99n6, 105, 122, 137, 168, 185–188, 193–199, 200n12, 202n46, 206, 236, 285, 290, 324–326, 332n20, 333n28, 341, 346, 353, 358, 361; of metaphysics 14–15, 163n3, 191, 294–313, 340 demonstration 1–2, 4, 6, 11, 15–16, 20, 28, 30–39, 44, 45n95, 51, 61, 63, 65–66, 71n17, 88, 242–243, 252, 276 Descartes, René 1, 4, 17n3, 21, 24–28, 42–44, 75, 83, 105, 132, 134, 136, 141, 252, 284 determinism 137–138, 141 dialetheism 15, 315–317, 324–330, 333–334 discursive cognition 7, 61, 105–111, 113–119, 125–126, 128, 253, 261–262 doctrine of method 9, 162, 163n5

dogmatism 7, 14–15, 17, 38–39, 105, 117, 126–127, 129n12, 132, 136, 184, 227, 231–245, 246n19, 247n36, 253, 294, 300–301, 317, 321–324, 329, 331n5, 332, 358–359 empiricism 7, 37, 120, 129n12, 132, 135, 137, 140, 145, 184, 281, 339, 363n48 epistemology 12, 111, 114, 126, 141, 211, 232–233, 238, 241, 251–252, 255, 284, 288n15, 339–341, 345, 349, 360, 363n52 ethics 27, 38, 43n51, 44n75, 66, 70n7, 138, 244 faculties 6, 13, 14, 37, 68–69, 105, 107, 110, 113, 121–123, 150, 155, 180, 186–187, 190, 192–197, 201, 208, 214, 218, 228–229, 235–236, 245n8, 247n34, 285–286, 295, 297–301, 310n1, 311n6, 358 faith 7, 29, 51–52, 55, 58, 64–65, 69, 136–137, 141, 301 fatalism 7, 137–138 foundationalism 10, 79, 185, 191–192, 197–198, 227–228, 233, 242–244, 255, 262, 282, 284, 289n19, 304 Franks, Paul 145n4, 199n5, 246n13, 259, 332n16 freedom 7, 136, 138, 144, 167, 263, 273, 296 geometry 1–2, 5–6, 31, 35–37, 41n51, 73–98, 100n18, 101n28, 102n36 German Idealism 3, 10–12, 18n7, 199n9, 213, 226–227, 230–234, 241–244, 246n17, 247n23, 248n36 di Giovanni, George 199, 201n25, 246n13, 279–280 God 4–5, 23–27, 35, 43n50, 45n95, 53, 55–63, 66–70, 105–106, 110–11, 114, 123, 127, 134, 136–145, 150, 161, 191, 264, 276, 278, 296, 303–304, 359 grounding 62, 112, 114, 175, 339, 342–345, 362n23

Index  369 highest good 9, 156–162, 165n17, 359 human being 4, 54, 57–60, 65–69, 136–145, 158–161, 211, 259, 265–266, 271–272, 351 the “I” 12, 252–253, 260–264, 267, 328 ideas of reason 169, 187, 193, 206, 220n1, 296; constitutive use of 9, 123, 168, 174–175, 179; regulative use of 9, 66, 70n11, 123, 168–169, 172, 174–176, 179–181, 182n10, 350–352, 359, 361n15, 363n46 imagination 44n74, 190, 260, 313n26 intellectual intuition 6–7, 105–115, 118–128, 242–243, 248n38, 283 Jacobi, Friedrich Heinrich 3, 267n1, 332n18 Leduc, Christian 44n75, 85, 100n18 Leibniz, Gottfried Wilhelm 4, 19, 21, 25, 28, 30, 33, 40n7, 41n13, 50, 69, 75, 79–80, 85, 120, 129n15, 134, 137, 198n2, 201n22, 210, 221n11, 294, 299, 311, 361n16 logic 15–16, 26, 34, 37, 39, 43n51, 45, 62, 78, 80–82, 88, 100, 102n36, 116–117, 128n5, 203n48, 206, 220n1, 236, 296, 301–305, 309–310, 312n17, 316–318, 323–330, 331n4, 333–334, 340; transcendental 107, 116, 298–299 Maimon, Salomon 3, 12, 200n19, 201n26, 212, 235, 247n26, 251–253, 259–263, 266–268 materialism 7–8, 135, 137–141, 287n4 McDowell, John 15–16, 315–324, 328, 331–332 Mendelssohn, Moses 3, 73, 98n1, 137, 145n1 metametaphysics 1–3, 15–16, metaphilosophy 10, 13, 185, 189, 194, 197–198, 199n8, 231, 233, 239, 241–243, 279, 284, 288n15 metaphysical explanation 17, 342, 345, 361n20

metaphysics of morals 133, 144, 168, 244, 285, 296 metaphysics of nature 7–9, 132–133, 139, 141–144, 145n5, 168, 172, 175, 180, 182n10, 285, 363n52 natural science 1, 8, 15–17, 138, 141, 167–168, 171–176, 179–182, 271, 273–274, 339–341, 346–354, 358–359 necessity 22, 69, 110–111, 122, 133, 175, 192, 219, 296, 318, 327–330 ontology 3–4, 19–45, 53, 56, 58, 76–77, 99n4, 126, 128n2, 133, 139, 141–142, 304 Pinkard, Terry 199n5, 279, 332n13 possibility 6, 22, 29, 54, 74, 78, 87, 90–94, 100n21, 101n28, 110, 127n2, 133, 191, 219, 265 postulate 5–6, 31, 36, 52, 74, 77–78, 82–94, 98, 100–102, Priest, Graham 15–16, 315–317, 325–330, 333–334 principle: first, 4–6, 12, 23–24, 27–30, 35–36, 43n52, 52–54, 58, 60–69, 71n16, 78–80, 83–84, 99n11, 189–195, 198n5, 199n15, 203n48, 206, 209, 212, 220, 221n5, 222n23, 226, 233, 242, 245n6, 254–258, 261–262, 265, 267n5, 284, 290n27; of non–contradiction, 5, 27, 29–36, 43n64, 44, 45n79, 61–66, 70n14, 71, 191, 201n22, 203n48, 245n6, 297, 324–334; of sufficient reason, 27, 29–32, 36, 43n58, 44n67, 59–62, 191, 201n22, 339, 345 proof 6, 31–35, 39, 44, 111, 133, 136–139, 207–210, 213, 217, 220, 236, 241, 243, 254 psychology 26–28, 34–35, 56, 58, 133–135, 140 quietism 15, 315–317, 320–323, 328, 330, 332

370 Index rationalism 2, 4, 7, 14–15, 120, 129n12, 184, 223n27, 226, 241, 253, 267, 275, 280–281, 294, 299–309, 312n19, 313n24, 339 reason: practical, 129n10, 162, 244, 262, 265, 288n15; speculative, 107, 129n10, 133, 140, 171, 296, 331n2; theoretical, 7, 115, 129n10, 262, 265 scepticism 7, 11–12, 18n7, 43n57, 127, 136, 226–261, 317, 320–324, 328, 331n3, 332; Humean 229, 231–232, 246; Pyrrhonian 11–12, 226–227, 231–244, 245n4, 246n16, 247n25 scholasticism 23, 26, 33 sensibilit 14, 88, 106–126, 129n12, 153, 187, 193, 199n5, 201n24, 202n44, 209, 213–216, 222n13, 229, 235, 264, 295–299, 310, 311n8, 319 Sider, Theodore 134, 354–358, 360n5, 361n8, 364 Spinoza, Baruch 21, 134, 137, 141 substance 134, 210, 219, 260, 302, 305, 350, 352 systematicity 2, 9, 105, 163n5, 164n13, 168–175, 179–182, 200–201, 249–254, 262, 288, 292, 350 theology 23, 26–28, 34, 35–38, 56–58, 62–63, 72, 133–134, 139, 271–272, 277, 286, 291 thing in itself 7, 36, 106, 111–129, 130n17, 115, 187, 199n11,

209–212, 215, 217, 220–224, 229–230, 233, 237, 247n12, 263, 304n2, 333n28, 347, 360n4 transcendent 58, 70n11, 86, 126, 129, 169, 170, 361, 363n55 Transcendental Aesthetic 113, 221n12, 298–299, 311 Transcendental Analytic 10, 185, 187, 189–191, 311n2 Transcendental Dialectic 115, 119, 127, 163n5, 202n41, 300, 346, 361n 15 transcendental philosophy 139, 172, 207, 230, 308 unconditioned 119, 123–124, 127, 129n10, 150, 155, 163, 252, 261–262, 289n25, 322 understanding 11, 14–15, 38–39, 62, 86, 98n3, 100n13, 105–113, 116–129, 137, 140, 170, 180, 186–191, 193, 199n11, 200n17, 201, 202n44, 207–218, 220, 221n10, 222n15, 229, 235, 253, 259, 266, 295–304, 306, 310, 311n8, 316–317, 322–324, 329, 331–333, 334n37; intuitive 106, 108–110, 113, 117, 124, 128n3, 253 will 54, 86, 138, 141, 264, 313n26, 359 wisdom 8, 151, 155–162, 164–165 Wissenschaftslehre 12–13, 202n46, 245n6, 247n23, 252, 271, 279, 288n16, 289n25, 328