System and Freedom in Kant and Fichte: Festschrift in honor of Günter Zöller (Routledge Studies in Eighteenth-Century Philosophy) 9780367480585, 9781032288765, 9781003044376, 0367480581

Citation preview

System and Freedom in Kant and Fichte

In both Kant’s and Fichte’s thought, system and freedom are intimately correlated and can be said to define the whole of their respective philosophies. Therefore, speaking of system and freedom with regard to Kant and Fichte is, on the one hand, to consider the methodically developed entirety of their philosophies and, on the other, to go to the very heart of their ways of thinking. The essays in this volume explore both Kant’s and Fichte’s proposals how to integrate freedom into a system of thought. System and Freedom in Kant and Fichte will be of interest to scholars and advanced students working on Kant, modern philosophy, and German Studies. Giovanni Pietro Basile is currently Associate Professor in the Philosophy Department of Boston College. After completing studies in physics, theology, and philosophy, he earned his PhD in Philosophy at the LMU Munich, where he also received the German Habilitation. He is a member of the Reviewers Panel of the journals Gregorianum and KantStudien. Among his main publications are two books—Transcendance et finitude. La synthèse transcendantale dans la Critique de la raison pure de Kant (Paris: L’Harmattan, 2005) and Kants Opus postumum und seine Rezeption (Berlin/Boston: De Gruyter, 2013)—as well as several articles on Kant, Karl Jaspers and Paul Ricœur. Ansgar Lyssy is currently Researcher at the University of Heidelberg, working on a project on causality in Hegel, funded by a grant from the Thyssen Foundation. In 2020, he finished his Habilitationsschrift at LMU Munich, a yet unpublished monograph titled Humankind and Humanity in Kant. This research was funded by a research grant from the German Research Foundation. Notable publications include Kausalität und Teleologie bei G. W. Leibniz, Stuttgart: Franz-Steiner (Studia Leibnitiana, Special Issue No. 48), 2016, three anthologies on Kant and the philosophy of the eighteenth century, and several papers on Leibniz, Kant, Hegel, and other related thinkers.

Routledge Studies in Eighteenth-Century Philosophy

Hume on Art, Emotions, and Superstition A Critical Study of the Four Dissertations Amyas Merivale A Guide to Kant’s Psychologism via Locke, Berkeley, Hume, and Wittgenstein Wayne Waxman Kant and the Continental Tradition Sensibility, Nature, and Religion Edited by Sorin Baiasu and Alberto Vanzo 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 For more information about this series, please visit: www.routledge. com/Routledge-Studies-in-Eighteenth-Century-Philosophy/book-series/ SE0391

System and Freedom in Kant and Fichte Edited by Giovanni Pietro Basile and Ansgar Lyssy Festschrift in honor of Günter Zöller

First published 2022 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 © 2022 Taylor & Francis The right of Giovanni Pietro Basile and Ansgar Lyssy 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. Library of Congress Cataloging-in-Publication Data A catalog record for this book has been requested ISBN: 978-0-367-48058-5 (hbk) ISBN: 978-1-032-28876-5 (pbk) ISBN: 978-1-003-04437-6 (ebk) DOI: 10.4324/9781003044376 Typeset in Sabon by Apex CoVantage, LLC

Contents

Prefacevii List of Figuresix List of Citations and Abbreviationsx List of Contributorsxiii Introduction

1

  1 The Identity of Reason

9

STEPHEN ENGSTROM

  2 Lichtenberg’s ‘Es denkt’ versus Kant’s ‘Ich denke’

31

PATRICIA KITCHER

  3 Modal Concepts in Kant’s Transcendental Discourse

50

CLAUDE PICHÉ

  4 Can Practical Reason Be Artificial?

71

DIETER SCHÖNECKER

  5 “The Eye of True Philosophy”: On the Relationship Between Kant’s Anthropology and His Critical Philosophy

86

ROBERT B. LOUDEN

  6 Kant am Pregelflusse: Site and Systemicity in the Preface to the Anthropology

103

SUSAN M. SHELL

  7 Kant’s Philosophy of Religion—A Provocation to the Historical Religions BERND DÖRFLINGER

135

vi  Contents   8 Hume and Kant on Utility, Freedom, and Justice

147

PAUL GUYER

  9 Reading Fichte Today. The Prospect of a Transcendental Philosophy

166

MARCO IVALDO

10 Fichte’s Original Presentation of the Foundational Principles of the Wissenschaftslehre: The Question of Method

178

MARINA F. BYKOVA

11 The Idea of Universal Monarchy in Fichte’s Practical Philosophy

206

DAVID JAMES

Index227

Preface

This book is designed to be a Festschrift for Günter Zöller, who we, the editors as well as the contributors, would like to honor for his considerable philosophical work, and especially for his numerous and fundamental contributions to the study of Kant and Fichte. As the notions of ‘system’ and ‘freedom’ take up center stage of his research,1 they have been chosen as the threads that guide this volume. Günter Zöller worked first at the University of Iowa, and then for over 20  years at the Ludwig-Maximilians-Universität München, from which he retired in 2020. He has furthered the cause of philosophy not only with his numerous and influential writings but also with his helpful, patient, and insightful contributions to countless live discussions in which he has participated, with his collegiality, and with the support that he has offered to his students and friends.2 This book contains 11 papers, which all engage, directly or indirectly, with the work of Zöller. Originally, this book was designed to contain significantly more papers and was aimed for a much earlier publication, but in 2020 the Covid pandemic began, and the various lockdowns disrupted our work as editors and presumably also that of the contributors. Several authors had to drop out and could not fully develop their papers in time. We are grateful to all contributors who have powered through these difficult times with first-class research. We would also like to thank our editor at Routledge, Andrew Weckenmann, and his assistant, Alexandra Simmons, for their support and their patience. Ansgar Lyssy would also like to thank the German Research Foundation (DFG) and the Thyssen Foundation, which have both funded his research while this book was developed, and hence have both contributed to making this volume possible. Editors’ thanks also go to Dillon Reihill for his assistance in proofreading and Robert McQueen for the creation of the index.

Notes . See Zöller (1998, 2001, 2017), for example. 1 2. A comprehensive bibliography of Zöller’s writings up to the year 2019 as well as a short overview over relevant biographical dates can be found in Kisner/ Basile/Lyssy/Weiss (2019).

viii  Preface

References Kisner, Manja; Basile, Giovanni Pietro; Lyssy, Ansgar; Weiss, Michael B. (eds.) (2019). Das Selbst und die Welt—Denken, Handeln und Hoffen in der Klassischen Deutschen Philosophie. Würzburg: Königshausen & Neumann. Zöller, Günter (1998). Fichte’s Transcendental Philosophy: The Original Duplicity of Intelligence and Will. Cambridge: Cambridge University Press. Zöller, Günter (2001). “Die Seele des Systems: Systembegriff und Begriffssystem in Kants Transzendentalphilosophie.” In Hans Friedrich Fulda und Jürgen Stolzenberg (eds.), Architektonik und System in der Philosophie Kants. Hamburg: Meiner, 41–72. Zöller, Günter (2017). “Weder Methode noch Disziplin. Zum historisch systematischen Ort transzendentalen Philosophierens.” In Marco Ivaldo, Hans Georg von Manz and Ives Radrizzani (eds.), Vergegenwärtigung der Transzendentalphilosophie. Das philosophische Vermächtnis Reinhard Lauths. Würzburg: Königshausen & Neumann, 167–183.

Figures

6.1 Map of Königsberg 6.2 Detail from 1808 map of Königsberg

133 134

Citations and Abbreviations

Kant References to Kant’s works other than the Critique of Pure Reason will be given in the text by the volume and page of Kant (1900–), the Akademie edition, cited with volume and page number. English citations are, as far as possible, from The Cambridge Edition of the Works of Immanuel Kant, Cambridge: Cambridge University Press, 1992. Significant deviations are noted and explained. The Critique of Pure Reason will be cited by the page numbers of the first and second editions. Anth CF CPJ CPracR CPR Dreams FS G HL JL L MF MM NDMR Notes OP Ped PhilEnc Physical Geography Prol TPP

Anthropology from a Pragmatic Point of View Conflict of the Faculties Critique of the Power of Judgment Critique of Practical Reason Critique of Pure Reason Dreams of a Spirit Seer The False Subtlety of the Four Syllogistic Figures, Groundwork for the Metaphysics of Morals Lectures on Logic (Hechsel) Lectures on Logic (Jäsche) Letters Metaphysical Foundations of Natural Science Metaphysics of Morals New doctrine of motion and rest Notes on the Observations on the Feeling of the Beautiful and Sublime Opus Postumum Lectures on Pedagogy Philosophical Encyclopedia Physical Geography Prolegomena to Any Future Metaphysics Toward Perpetual Peace

Citations and Abbreviations Refl Rel TP UG UNH V-Anth/Collins V-Anth/Parow V-Anth/Mensch V-Anth/Mron V-Anth/Fried V-Anth Pillau V-Met/L1/Pölitz V-Met/L2/Pölitz V-Met/Dohna V-Met/Schön V-Met/Mron V-Met/Volckmann

xi

“Reflections” from Kant’s Handschriftlicher Nachlass Religion within the Boundaries of Mere Reason On the Common Saying: That May Be Correct in Theory, but it is of No Use in Practice On the Ultimate Ground of the Differentiation of Regions Universal Natural History and Theory of the Heavens Lectures on Anthropology (Collins) Lectures on Anthropology (Parow) Lectures on Anthropology (Menschenkunde) Lectures on Anthropology (Mrongovius) Lectures on Anthropology (Friedländer) Lectures on Anthropology (Pillau) Lectures on Metaphysics 1 (Pölitz, Original) Lectures on Metaphysics L2 (Pölitz, Original) Lectures on Metaphysics (Dohna) Lectures on Metaphysics (von Schön) Lectures on Metaphysics (Dohna) Lectures on Metaphysics (Volckmann)

Fichte AGN CCS CCP CPA EPW FEW GA

Addresses to the German Nation. Trans. and ed. by Gregory Moore. Cambridge: Cambridge University Press, 2008. The Closed Commercial State. Trans. and ed. by Anthony Curtis Adler. Albany, NY: State University of New York Press, 2012. Concerning the Concept of the Wissenschaftslehre. In EPW, 87–136. The Characteristics of the Present Age. In The Popular Works of Johann Gottlieb Fichte. Trans. and ed. by William Smith, vol. 2. London: Trübner & Co., 1889, 1–288. Fichte: Early Philosophical Writings. Trans. and ed. by Daniel Breazeale. Ithaca: Cornell University Press, 1988. 2nd ed., 1993. Foundation of the Entire Wissenschaftslehre and Related Writings, 1794–5. Trans. and ed. by Daniel Breazeale. Oxford: Oxford University Press, 2021. Gesamtausgabe der Bayerischen Akademie der Wissenschaften. Ed. by Reinhard Lauth, Erich Fuchs and Hans Gliwitzky. Stuttgart and Bad Canstatt: Frommann-Holzboog. Cited by series, volume and page number.

xii IWL RPP SE

Citations and Abbreviations Introductions to the Wissenschaftslehre and Other Writings (1797–1800). Trans. and ed. Daniel Breazeale. Indianapolis, In: Hackett, 1994. Review of Immanuel Kant, Perpetual Peace: A  Philosophical Sketch. Trans. by Daniel Breazeale. The Philosophical Forum 32.4, 2001, 311–21. The System of Ethics. Trans. and ed. by Daniel Breazeale and Günter Zöller. Cambridge: Cambridge University Press, 2005.

Contributors

Marina F. Bykova is Professor of Philosophy at North Carolina State University. Her area of specialization is the history of the nineteenthcentury continental philosophy, with a special focus on German idealism and emphasis on Fichte and Hegel. Her most recent books include The German Idealism Reader: Ideas, Responses and Legacies (Bloomsbury Academic, 2019), Hegel’s Philosophy of Spirit: A Critical Guide (ed., Cambridge University Press, 2019), The Bloomsbury Handbook of Fichte (ed., Bloomsbury Academic, 2020), and The Palgrave Hegel Handbook (co-ed. with K. Westphal; Palgrave Macmillan, 2020). Bernd Dörflinger is Professor at the University of Trier. He worked on history of philosophy (Kant, Kant’s legacy, and German Idealism) and on systematic philosophy as well (theoretical philosophy, moral, esthetic, and philosophy of religion). His publications include the following monographies: Die Realität des Schönen in Kants Theorie rein ästhetischer Urteilskraft (1988) and Das Leben theoretischer Vernunft. Teleologische und praktische Aspekte der Erfahrungstheorie Kants (2000). He is President of the Kant-Gesellschaft. Stephen Engstrom is Professor of Philosophy at the University of Pittsburgh. His areas of interest include ethics, metaphysics, modern philosophy (especially Kant), and ancient philosophy. He is the coeditor of the book Aristotle, Kant, and the Stoics: Rethinking Happiness and Duty (1996) and the author of the monograph The Form of Practical Knowledge: A Study of the Categorical Imperative (2006). Paul Guyer is Jonathan Nelson Professor of Philosophy at Brown University. He is the author of nine books on Kant, including  Kant and the Claims of Taste  (1979),  Kant and the Claims of Knowledge (1987), Kant (2006), Kant’s Groundwork for the Metaphysics of Morals (2007), and Knowledge, Reason, and Taste: Kant’s Response to Hume (2008). He is the editor of six anthologies of work on Kant, including three Cambridge Companions.

xiv  Contributors Marco Ivaldo is Professor of Moral and Practical Philosophy at the University “Federico II” of Naples. His research interests lie mainly in modern German philosophy (Leibniz, Humboldt, Jacobi, Kant, Fichte, Reinhold, Hegel, Schelling). He is the coeditor of Fichte-Studien. He is the author of numerous books and papers. His most recent monographs are Ragione pratica: Kant, Reinhold, Fichte (2012), Fichte (2016), and Filosofia e religione. Attraversando Fichte (2016). David James is Associate Professor at the University of Warwick. His research interests lie mainly in German Idealism together with its legacy (e.g., the Frankfurt School of Critical Theory) and social and political philosophy, especially the tradition that can be traced from Rousseau through Kant, Fichte, and Hegel to Marx. He is the author of three books, Fichte’s Social and Political Philosophy (2011), Rousseau and German Idealism (2013), Fichte’s Republic (2015), coeditor (with Günter Zöller) of The Cambridge Companion to Fichte (2016), and editor of the Book Hegel’s Elements of the Philosophy of Right: A Critical Guide (2017). Patricia Kitcher is Professor of Philosophy at Columbia University. She works mainly on Kant, philosophy of psychology, and Freud. In recent years she has focused on the cognitive psychology of the Critique of Pure Reason. She has written numerous articles and three books: Kant’s Transcendental Psychology (1990), Freud’s Dream (1992), Kant’s Thinker (2011). She has served as President of the Society for Philosophy and Psychology and as President of the North American Kant Society. Robert B. Louden is Professor of Philosophy at the University of Southern Maine. He works on Ethical Theory, History of Ethics (Ancient, eighteenth–twentieth centuries), Kant’s Ethics, History of Modern Philosophy, Kant, the Enlightenment and Modernity. Among his many publications are the following monographs: Morality and Moral Theory (1992), Kant’s Impure Ethics (2000), The World We Want (2007), and Kant’s Human Being (2014). He is also coeditor and translator of two volumes in the Cambridge Edition of the Works of Immanuel Kant and former President of the North American Kant Society. Claude Piché is Professor Emeritus at the University of Montréal. His research focuses on Kant, Jacobi, Reinhold, Fichte, Rickert, and Simmel. He is the author of two books: Das Ideal. Ein Problem der Kantischen Ideenlehre (1984) and Kant et ses épigones (1995). He also published numerous volumes and acts of congress as an editor and several papers and articles. Dieter Schönecker is Professor of Practical Philosophy at the University of Siegen. He works especially on ethics (meta-ethics and medical ethics)

Contributors  xv and on Kant. He is the author of the monograph Kant: Grundlegung III. Die Deduktion des kategorischen Imperativs (1999) and coauthor with Allen W. Wood of Kant’s Groundwork for the Metaphysics of Morals—A Commentary (2011). He has also published numerous volumes as an editor and several papers. Susan M. Shell is Professor and Chair of the Department of Political Science at Boston College. She is the author of several monographs on Kant’s practical philosophy: The Rights of Reason: A Study of Kant’s Philosophy and Politics (1980), The Embodiment of Reason: Kant on Spirit, Generation and Community (1996), Kant and the Limits of Autonomy (2009). She is also the coeditor of Kant’s ‘Observations’ and ‘Remarks’: A Critical Guide (2012). She has been Visiting Professor at Harvard University, and received fellowships from The National Endowment for the Humanities, The American Council of Learned Societies, The Bradley Foundation, the Deutsche Akademi­ sche Austauschdienst, and the Radcliffe Institute.

Introduction

In both Kant’s and Fichte’s thought, system and freedom are intimately correlated and can be said to define the whole of their respective philosophies. Therefore, speaking of system and freedom with regard to Kant and Fichte is, on the one hand, to consider the methodically developed entirety of their philosophies and, on the other, to go to the very heart of their different ways of thinking. For Kant, philosophical thinking means thinking systematically. If philosophy is to produce proper philosophical knowledge, philosophizing can never be severed from system building. A system is an organized body of knowledge which derives its unity and structure from a foundational principle. An unsystematic body of knowledge would be a mere aggregate of separate bits of insight, derived from experience, conjecture, or common sense. In this sense, Kant is a builder of philosophical systems, as evidenced by the system consisting of the three critiques. This serves as propaedeutics to the building of the system of transcendental philosophy, which, in turn, was designed to comprise a system of the philosophy of nature, which was never finished, and a system of freedom, which comprises the metaphysics of morals, as both a book and a project. Here, the system of reason translates into metaphysics as science. Nevertheless, none of the realized systems fully satisfies Kant’s idea of a comprehensive system of reason and he never ceased working on the extension and fulfillment of the system of philosophy. Indeed, new problems regarding the unity of his philosophy still loom large until the late Opus Postumum.1 However, Kant’s philosophy is systematic not only because it builds doctrinal systems. It is also systematic in an other sense, because Kant considers metaphysics as the result of a disposition of human reason which leads us inevitably into thinking systematically. He maintains that the ideas of soul, God, and freedom necessarily play a role within the systematic outline of our thinking. Yet freedom is arguably the ultimate focus of Kant’s philosophy, and practical reason claims a central place in this philosophy. Precisely for this reason, freedom can only be thought of in a systematic relationship with the foundation of knowledge and within a critique of the limits of speculative metaphysics. That is why freedom DOI: 10.4324/9781003044376-1

2  Introduction is already at stake in the foundational reflections of the first Critique, namely, the problem of the systematic unity and functional diversity of theoretical and practical reason. Freedom is again at stake in Kant’s applied philosophy. Where his philosophy concerns anthropology, pedagogy, history, religion, and politics, the notions of spontaneous freedom of the cognizing subject and practical freedom for embodied agents are made fruitful in regard to contingent social or historical circumstances. Here, agents are not only governed by principles alone, but they also act within their social horizon, they are driven by contingent incentives, and they are limited through external forces. This is where the ‘pure’ reflection of critical philosophy finds a concrete application, but the importance of the concept of freedom in the field of applied philosophy also brings a significant complement to the purely philosophical concept. That is, doing philosophy for its own sake and without having its application for human purposes and self-knowledge in sight, theoretical knowledge remains ‘myopic’ and will ultimately fail to make sense of itself (cf. 25:682). This raises the question of the complex relationship between the critical system and the philosophical fields that lie at the margin of it. Fichte’s Philosophy could hardly be understood without reference to Kant’s Philosophy. The correlation between system and freedom is therefore as important to Fichte as it is to Kant, although it receives for the former a new meaning and a new function. Fichte understands his own thinking as the revision and fulfillment of Kant’s system of transcendental philosophy, aiming to make it less susceptible against skeptical attacks. Therefore, he strives to find a certain and apodictic foundation for the unity of theoretical and moral philosophy, namely, freedom as the original and irreducible ground. This transformation of critical philosophy in the sense of a radical conjunction of freedom and system under the auspices of transcendental idealism also has consequences for his applied philosophy, especially in the realm of political thinking. Here, Fichte develops a much stronger understanding of universal monarchy, compared to Kant, and he fully realizes the relevance of the French Revolution for the course of world history. This inquiry into the role of the notions of system and freedom in Kant and Fichte is relevant for at least two reasons. First, it is of historical significance, as the relationship between system and freedom plays a foundational role both in Kant’s critical and applied philosophy and in Fichte’s transcendental philosophy. Moreover, a reflection on this topic opens new perspectives for the ongoing philosophical debate. In contemporary philosophy, systematicity of thinking and metaphysics of freedom face a radical criticism. While systematicity still serves as a criterion for distinguishing good philosophy from bad philosophy, few thinkers still strive to establish their own philosophy as a system. This criticism seems often motivated by the alleged excesses of doctrinal systems, but it also

Introduction  3 interprets systematicity as nothing but coherently developed arguments. The foundational notion of system that is developed by Fichte and Kant points to the systematic character of reason beyond any given system, i.e. to metaphysics as disposition (Metaphysik als Anlage, as Kant puts it) that remains the ultimate horizon of every philosophical thinking.

Note 1. Cf. Basile, Lyssy (2022).

Bibliography Basile, Giovanni Pietro and Ansgar Lyssy (eds.) (2022). The Unfinished System. New Essays on Kant’s Opus Postumum. London: Routledge. Kant, Immanuel (1900–). Kants gesammelte Schriften (Akademie-Ausgabe). Ed. by the Königlichen Preußischen Akademie der Wissenschaften and successors, 29 vols. Berlin: Reimer/De Gruyter.

Chapter Summaries The contributions in this volume are divided into three parts. In the first part are collected the texts that deal with the concepts of system and freedom in relation to Kant’s critical system. The second part groups together contributions that reflect on the significance of these two concepts in those sections of Kantian philosophy that hold on the edge of or are beyond the critical system, with particular reference to anthropology, philosophy of religion, and political philosophy. Finally, the third part includes works that deal with system and freedom in Fichte, as a continuation and overcoming of the Kantian perspective. The first part includes four texts. In the first chapter, “The Identity of Reason,” Stephen Engstrom argues that Kant’s distinction between theoretical and practical reason gives rise to what is sometimes called the problem of the unity of reason, the problem of showing that theoretical and practical reasons are in harmony or at least consistent with one another, despite appearances to the contrary (e.g., the seeming conflict between determinism and freedom). But this problem seems to depend on an acknowledgment of the identity of reason, some acceptance of the idea that, as Kant puts it, “in the end there can be only one and the same reason, which must be distinguished merely in the application.” (G, 4: 391) So before addressing questions relating to the unity of reason, we should first ensure that we have sufficient comprehension of reason’s identity in both theoretical and practical regard. Famously, Kant uses the pure subject as grounding for his derivation of the categories and the unity of thought. But how should we understand this subject? In the second chapter, Patricia Kitcher takes up this question by means of responding to a challenge issued by Lichtenberg in her

4  Introduction chapter “Lichtenberg’s ‘Es denkt’ versus Kant’s ‘Ich denke’.” Here, she picks up Lichtenberg’s famous dictum that we should not say ‘I think,’ but ‘it thinks,’ which is usually taken to be a criticism of Descartes. However, Günter Zöller has recently argued that Kant is an equally plausible target. Zöller situates his account of Lichtenberg’s attack on the Kantian ‘I think’ in a broader discussion of Lichtenberg’s views of the shortcomings of the system of transcendental idealism, that is, the alleged ‘hole.’ As Zöller notes, Kant comes close to this formulation himself when he refers to “this I or he or it (the thing) that thinks” (CPR, A346/B404), yet he also thinks that Kant has resources for avoiding the problem of the absent thinker that Lichtenberg lacked. Kitcher situates the exact problem that Lichtenberg raises for Kant and then develops in greater detail the Kantian resources for solving it than Zöller indicates but does not fully exploit. Clearly, this defense of the subject as source of all thinking is relevant for all transcendental conceptions of freedom. Next, Claude Piché discusses Kant’s use of modal concepts in his transcendental discourse in his chapter entitled “Modal concepts in Kant’s transcendental discourse.” Kant makes use of modal concepts like possibility, existence, necessity, and contingency, and these modal concepts are part of the table of the categories of the Critique of Pure Reason. The rules for their application to objects of experience are exposited in the chapter of the Critique entitled “Postulates of the Empirical Thought in General.” The question is then: does the critical discourse which depicts the restrictions accompanying the application of the modal categories transcend these same restrictions to the extent that it would use for itself the modal concepts in their ‘absolute’ meaning? This might appear plausible at first sight since transcendental philosophy describes the conditions of experience from the outside, that is, from a higher standpoint. But Piché argues to the contrary that the recourse to modal concepts in the writing of the first Critique complies mutatis mutandis with the limitations imposed upon the modal categories at play within experience, leaving room for the freedom of the thinker to apply the modal concepts in accordance with their epistemic interest. In other words, all modal concepts, when they are not taken in the sense of their merely logical meaning, must be considered as ‘conditioned’ and not as absolute. This is relevant for the task of the transcendental philosopher, who tries to integrate the philosophical conception of freedom into a rigid system, as much as it is of importance to the empirical scientist, who tries to make sense of things. In the fourth chapter, Dieter Schönecker deals with the question: Can practical reason be artificial? Practical reason is defined as the power to cognize and to will the good. Can it be part of an artificial intelligence? From a Kantian point of view, the answer is clearly negative: Practical reason cannot be artificial. After presenting a preliminary remark on the possibility of Kantian moral machine, Schönecker argues that in a

Introduction  5 Kantian model of moral obligation, the typical (human) moral subject must have not only practical reason but also moral feelings in order to cognize the validity of the moral law. Using the knowledge argument against physicalism and functionalism, he continues to argue that computers have no feelings and, a fortiori, no moral feelings; therefore, computers are no moral subjects. This conclusion is based on a Kantian ‘I feel’ rather than the ‘I think.’ Schönecker discusses two problems related to this argument and concludes with an analogy: Just as planets do not fly, computers do not feel. The second part of this book is called “Freedom and the Human Condition: Religion, Anthropology and Politics.” It includes four contributions as well. Robert B. Louden presents freedom from an anthropological point of view, as he examines Kant’s views about freedom not from the well-worn metaphysical perspective of the Critique of Pure Reason and other works in theoretical philosophy, but rather from the impure perspective of his writings on anthropology, history, and education. Approaching Kantian freedom from this alternative perspective yields numerous advantages that are not to be found elsewhere: it allows us to better understand the relationship between transcendental philosophy and Kant’s anthropology, as only anthropology enables us to put philosophy to good use for human purposes. In “Anthropology and Knowledge of the World in Kant’s Late Thought,” Susan M. Shell addresses a crucial moment in an ongoing dispute concerning the diverging assessments of the specific character of Kant’s anthropology and its role in his system of philosophy more generally. In some passages, Kant conceives of anthropology as a purely ‘empirical’ study, only episodically related to critical philosophy; in others, as an Urdisziplin that summarizes and contains the entire critical project. He also defines ‘empirical anthropology’ as a ‘theoretical’ science that considers man insofar as he is subject to the ‘laws of nature’ rather than (as with ‘practical science’) to the ‘laws of freedom.’ Then again, ‘moral anthropology’ is understood as a necessary counterpart to a ‘metaphysics of morals’ within the larger field of practical philosophy. Physiological anthropology, by way of contrast, makes up no proper part of ‘knowledge of the world’ at all. To understand this dichotomy, it is helpful to acknowledge the dual worldly roles of the human being, where knowledge of the world is concerned: humans are both the observer and the observed. To avoid the science of anthropology resulting in a vicious circle, we must already have acquired knowledge of what a human being is; and we can do so by understanding the humans at our local home, through Umgang with one’s neighbors and countrymen. Local knowledge presupposes general knowledge; and both, if they are to give rise to science, must be guided by a plan—one that bears notable relation to Kant’s own worldly Umgang, both geographical and professional.

6  Introduction The seventh chapter is Bernd Dörflinger’s “What Kant’s Philosophy of Religion Exacts from the Historical Religions of Revelation.” Here, Dörflinger opposes a traditional line of interpretation, according to which Kant’s enlightened philosophy of religion supposes that harmony is possible between his conception of religion within the limits of reason and the historical religions of revelation, at least coexistence without conflict. Kant holds up the idea that moral rational religion marks the completion of his systematic reflection of morals. Without such a completion, practical philosophy would not have a systematic unity. However, the historical religions of revelation stand in opposition to the concept of a moral rational religion. They cannot claim a position within the system of pure practical reason. According to Kant, human beings don’t have a sensorium for the reception of supernatural messages. Inversely, Kant suspects it to be impossible that human beings affect God. Hence for him, all phenomena of an explicit worship must be irrational phenomena. Faced with pretended Holy Scriptures, he contends the superiority of reason over these scriptures. He rejects specific religious statutory laws because they necessarily include the potential of violent conflicts among different historical religions. In the long run, there is no other reasonable goal in history than the end of this type of religion. By means of increasing enlightenment, which is not itself violent but intellectually offensive, this goal can be reached. Then, Paul Guyer compares Hume and Kant on utility, freedom, and justice. Guyer argues that the normative differences between Hume’s presumably proto-utilitarian approach to justice and Kant’s approach based on the innate right to freedom of every human being is not as great as might be expected. A more important difference is that Kant founds his conception of justice on the demand for the greatest possible freedom for all, but places such a value on the coherence and continuity of government that he rejects a right to rebellion in favor of a right to criticize and petition, while Hume grounds our allegiance to government on the utility of security but concedes a much more liberal right to resistance. Although their normative and meta-ethical foundations are different, Hume’s utility- or happiness-based rules of justice do make room for freedom of action, and Kant’s freedom-based laws of justice do make room for happiness or utility. The third part of this book includes three chapters and is titled: “Beyond Kant: Freedom as System in Fichte.” First, Marco Ivaldo discusses a recent consideration made by Günter Zöller, according to which transcendental philosophy modifies the ancient question of metaphysics on the forms and principles of being and poses the question about the conditions of possibility of our knowledge of the forms and the principles of being. Ivaldo’s chapter, titled “Transcendental Philosophy as a System of Freedom” continues this line of argument and emphasizes how Fichte connects the forms of knowledge that Kant had treated separately

Introduction  7 (theoretical philosophy and moral philosophy). The Wissenschaftslehre, understood as fundamental and unitary philosophical science, becomes in Fichte the ‘system of freedom.’ There is no fundamental opposition between the system and freedom, but freedom is a principle of the system: freedom is not only an object of the system but also a ‘subject’ of the system. As such, freedom is not identified with freedom of choice, but is a practical and theoretical principle. As such, a unitary philosophy of freedom the Wissenschaftslehre thus becomes the theory of the self-legislation of freedom both in will (practical knowledge) and in knowledge (theoretical knowledge). It is not identified either with an ontology or with a gnoseology (in the sense of the university philosophical tradition), but instead conceived as a science of the transcendental unity of being and of knowledge. It presents itself as an understanding (Verstehen) of the sense (Sinn) of experience. In the Foundation of the Entire Wissenschaftslehre (1794–95), Fichte openly associates the methodology of his systematic presentation of philosophical derivations with what he calls “the synthetic method.” However, there is no consensus among scholars on how to construe Fichte’s method employed in his first Jena presentation. Focusing on the “First Installment” of the Foundation, Marina F. Bykova’s contribution offers a discussion of Fichte’s style of reasoning and the methodical procedure he uses by the presentation of his first principles and their foundational demonstration. The author argues that Fichte’s conception of ‘synthesis’ is not identical to Kant’s and that the way he enacts the ‘synthetic’ procedure in his first presentation of the Wissenschaftslehre points to his departure from Kant’s usage of the ‘synthetic method.’ Bykova also suggests that the ‘way of proceeding’ Fichte uses to demonstrate his foundational principles can be properly comprehended only as a combination of methods construed dialectically and that the ‘synthesis’ he has in mind is not a simple ‘unity’ of the opposing principles but rather a complex dialectical resolution of the arisen contradictions. Finally, David James discusses an aspect of Fichte’s political philosophy in his chapter “The Idea of Universal Monarchy in Fichte’s Practical Philosophy.” After a brief reconstruction of Fichte’s explicit references to the idea of universal monarchy, James explains Fichte’s interest herein in terms of the nature of the historical event that he is seeking to defend, namely, the French Revolution. Herein, two questions emerge. The first question concerns whether Fichte subsequently provides a systematic transcendental explanation of the tendency toward territorial expansion that he associates with the idea of universal monarchy. The second question concerns whether he provides an account of how the same tendency can be countered. James argues that answers to both questions can be found in Fichte’s theory of the drive to absolute self-sufficiency developed in the 1798 Sittenlehre. The desire for territorial expansion is explained in terms of the lawless form of this drive, whereas the lawful form of

8  Introduction this drive limits this desire. I shall then discuss the implications of these claims of transcendental psychology in relation to some of Fichte’s later remarks on the state in The Closed Commercial State, which imply that the modern state as such, and not only the monarchical state, exhibits expansionary tendencies that must be curbed.

1 The Identity of Reason Stephen Engstrom

At his point of entry into practical philosophy, Kant remarked that just as theoretical philosophy must be grounded in a critical investigation of theoretical reason, practical philosophy must be grounded in a critical investigation of practical reason. He added, however, that the latter investigation must also exhibit practical reason’s “unity” with theoretical “in a common principle,” because “in the end there can be only one and the same reason, which must be distinguished merely in the application” (G, 4: 391).1 Soon thereafter Kant published his critical investigation of practical reason, and in that work he drew comparisons between theoretical and practical reason and considered their relation to one another. He also dealt with associated metaphysical issues, such as the reconciliation of freedom and nature and the conditions of the possibility of the highest good. These questions are of great interest in their own right, and they are bound up with the highest aspirations of critical philosophy. But they lie downstream from the topic I  will be mainly addressing here. The metaphysical questions are connected with what is sometimes called the problem of the unity of reason, the problem of showing that, despite appearances to the contrary, theoretical and practical reason are in harmony, or at least consistent with one another.2 But this problem presupposes the identity of reason. Only under the supposition that one and the same reason is at work in two applications could an apparent conflict between theoretical and practical reason pose a threat to unity. So before pressing forward to questions relating to unity, it may serve us well to ensure that we sufficiently comprehend reason’s identity.3 Kant’s fruitful insight that consciousness of the identity of the I think can yield recognition of the unity of the understanding encourages the anticipation that reflection on reason’s identity may similarly illuminate the unity of reason.

1  The Divisions of Philosophy 1. Kant’s statement that in the end there can be only one and the same reason occurs, as I said, at his point of entry into practical philosophy, and it will be helpful here to recall the survey of the divisions of philosophy DOI: 10.4324/9781003044376-2

10  Stephen Engstrom offered at that juncture, for it affords an overview of human reason’s most basic scientific employment, which lies not in any art or special science but in the systematic elaboration of the principles of knowledge that reason legislates for the sake of its final end (G, 4: 387–8).4 Taking up the traditional Hellenistic division of philosophy into the sciences of logic, physics, and ethics, Kant adds only the principle of the division, with the aim, he says, of ensuring the completeness of the division and the correctness of the subdivisions. This aim is served by way of two distinctions, one dividing philosophy into formal and material, and the other dividing material philosophy into the laws of nature and the laws of freedom, or, as Kant will also put it, into theoretical and practical philosophy. Logic is then identified with formal philosophy, and physics and ethics make up the two species of material. Before considering the two divisions, we should note that Kant speaks of philosophy as rational knowledge (Vernunfterkenntnis), intimating thereby that reason is a capacity for knowledge (Erkenntnis), and that the distinction between theoretical and practical philosophy is a distinction between two types of rational knowledge. This point bears underscoring. For we also find in Kant’s writings statements to the effect that the aim in matters practical is not knowledge but the determination of the will and action, and these statements, despite their Aristotelian ring, have encouraged readers to suppose that the distinction, as Kant understands it, is between a cognitive and an other-than-cognitive application of reason.5 But while such remarks do imply that reason’s practical application is not directed to theoretical knowledge, they do not imply that it is other than cognitive. Reason can be at bottom a cognitive capacity, whose acts are acts of cognition, even though the aim in matters practical is the determination of the will and action; for that aim can be the aim of practical reason’s own knowledge. 2. The first division, between formal and material, is fundamental for philosophy. Rational knowledge takes a first step toward becoming systematic by gaining explicit consciousness of itself as such, distinguishing what belongs to its form from what belongs to its matter, and in doing so, it recognizes itself to lie in the exercise of discursive—that is, conceptual, rather than intuitive—understanding. The resulting knowledge of the form of discursive knowledge constitutes the formal knowledge in which logic consists. It depends on an act of reflection that can figure in the act of knowing. In this reflection, the act of knowing attends to itself and to the element in its consciousness that is absolutely necessary and as such common to all such acts, and it can regard this element—its form— in abstraction from all contingent or particular elements that make for diversity in acts of knowing. Logic’s formal knowledge accordingly lies in rational cognition’s explicit consciousness of its own form. And once this consciousness of its form has been gained, rational knowledge can itself receive the name material knowledge. So logical knowledge, as formal,

The Identity of Reason  11 lies nowhere outside of material knowledge yet is nevertheless distinct from all such knowledge. To the extent that material knowledge is distinguished from formal by the diversity of the acts of knowing that it comprises, it can also be described as differing from the latter in that, as Kant says, it “considers some object” (G, 4: 387). That this is so can be seen by further reflection on the difference just noted between what is contingent and what is necessary in the consciousness figuring in rational knowledge. This difference lies at bottom in a difference between modifications of consciousness and the consciousness modified, or the consciousness itself, which is identical across the diversity of modifications that marks them as contingent and hence as modifications. These modifications, however, cannot be arbitrary, given that the knowledge in question is philosophical and therefore not reliant on any arbitrary acts of choice. The comprehension of them as contingent depends rather on understanding them as affections of consciousness or as resting on such affections and hence as indicating the presence of a material condition outside consciousness that enables knowledge of an object, or something whose actuality is distinct from the actuality of the consciousness and its modification. Thus the division between formal and material knowledge can also be characterized in terms of a distinction between self, or subject, and other, or object: material knowledge “considers some object,” and formal knowledge, as cognition’s knowledge of its own identical form, is “a self-knowledge of the understanding and of reason” (JL, 9: 14),6 in that this form, considered as potentiality, is just the cognitive capacity itself. As I hinted at the outset, Kant identifies this form of rational knowledge that logical reflection brings to explicit attention with a certain unity: according to its own basic self-understanding, knowledge is constituted by self-agreement. Knowledge must agree with knowledge. I will return to this unity of knowledge in due course. Let us turn now to the second division, between the two sorts of material knowledge. This division may at first appear to rest merely on a distinction between the objects known, in that Kant speaks of two objects, nature and freedom, the former comprising what is, or the real, the latter what ought to be, or the good. Elsewhere, however, he explains the division by distinguishing the relations the two sorts of knowledge bear to the objects (CPR, Bix—x, CPracR, 5: 46). According to that explanation, the two sorts of knowledge are alike in that each constitutes itself as material through understanding itself to be of an object, of which it has a concept, and also as rational through determining that concept and thereby (in accordance with the Copernican way of thinking) the object as well. But they differ in their understanding of the relation they bear to their respective objects: the object of theoretical knowledge must be “given from elsewhere” in order to be known, whereas practical knowledge works to make its object actual. In short, since both sorts of

12  Stephen Engstrom knowledge determine their objects, they do not differ in the formal aspect of their relation to their objects: the object depends on the knowledge. But they do differ in the material aspect, or in point of actuality or existence: theoretical knowledge depends on its object, while in the practical case the object depends on the knowledge. 3. In the division we have been considering, philosophy’s three parts are represented statically and symmetrically. But the relation among them also involves a certain dynamic character and a certain asymmetry between the two species of material knowledge. The asymmetry I have in ­ ractical mind is not the often-noted primacy that Kant ascribes to pure p reason, but another, quite different asymmetry, which he also recognizes. It might be expressed by saying that reason’s primordial use is theoretical,7 and it follows directly from our basic understanding of theoretical and practical knowledge. There is nothing in the notion of theoretical knowledge—as knowledge that determines an object that must be given from elsewhere—that implies that such knowledge depends for its possibility on the possibility of practical knowledge. Kant registers his recognition of the at least notional possibility of a merely theoretical knower when at one point for the sake of argument he conceives of a rational creature in whom nature allotted to its reason neither the power to frame its end nor even the power to determine by what means it would pursue it, assigning both to instinct, leaving reason with no further occupation than the contemplation of nature (G, 4: 395).8 In the case of practical knowledge, on the other hand, there is no corresponding notional possibility. According to its own self-understanding, practical knowledge is efficacious, working to make its object—what it knows—actual. But to make an object actual is to bring it under the heading of what is, making it an object of theoretical knowledge. Indeed, practical knowledge perfects itself practically through attaining theoretical knowledge of that object. Though in its cognitive moment practical knowledge rests in itself, in its practical moment it comes to rest in theoretical knowledge. The practical perfection of practical knowledge therefore includes theoretical knowledge of its object. Practical knowledge thus presupposes the possibility of theoretical. So theoretical knowledge is the primordial species, and practical knowledge, according to its very idea, is possible only as a second application of the very same reason. No inconsistency blocks us from thinking of a merely theoretical knower, but we cannot even conceive of a merely practical knower. If theoretical knowledge is primordial, then the distinction between form and matter on which philosophy rests must first be drawn in respect of knowledge conceived as theoretical, and the first division of philosophy must accordingly be between the formal science of logic and the material science of nature. This consequence has a clear bearing on our question about reason’s identity. If reason is first recognized in its theoretical use, then our task of comprehending reason’s identity across its

The Identity of Reason  13 two applications becomes the task of comprehending how reason, as explicated in logical reflection on theoretical knowledge, can also be at work in a practical application.

2  Theoretical Reason 1. Let us first focus, then, on the theoretical use, beginning with a wellknown passage, the opening lines of the Critique of Pure Reason’s Analytic of Principles: Universal logic is built on a ground plan that quite exactly coincides with the division of the higher faculties of knowledge. These are: understanding, the power of judgment, and reason. That doctrine deals therefore in its analytic with concepts, judgments, and conclusions, in direct conformity with the functions and the order of those mental powers, comprehended under the broad designation of the understanding in general. (CPR, A130–1/B169) It might seem that Kant is here using ‘understanding’ (Verstand) in different senses, a narrow sense, signifying one of the three higher faculties of knowledge, and a broad sense, encompassing all of them together. But if there are two senses, they are intimately related. Kant first introduced the understanding in general as “a capacity to think,” or to know “through concepts” (fixing it that the understanding under consideration is discursive, not intuitive), and then declared in the transcendental deduction that the I think is “in all consciousness one and the same,” implying thereby that this I of apperception is the consciousness on the part of the understanding of its own identity, or analytic unity, in all consciousness, an identity that he also described as an identity of function (CPR, A69/ B94).9 Similarly, in the deduction’s account of the original synthetic unity of apperception, Kant characterized this unity as “unity of concept” and then explicitly identified it with the understanding, describing it as “the highest point to which all use of the understanding, even the whole of logic and, after it, transcendental philosophy, must be attached.”10 Kant is evidently thinking, then, that there must be an identity of the understanding in general across its three different functions, just as there must be an identity of reason across its different applications. And in giving one of those functions the same name as the generic function, he indicates that it is basic, figuring constitutively in the others, even though they are set over against it in the division. The understanding in general is thus the unity of the very thing—­ concept—that universal logic deals with in its treatment of the function of the understanding in the narrow sense. As for the other two functions, universal logic’s treatment of these subordinates and traces them to the

14  Stephen Engstrom function of concepts, subsuming judgments under concepts and conclusions under judgments. A judgment is said to be “the manner of bringing given cognitions to the objective unity of apperception” (CPR, B141)11 and is thus subsumed under the unity that was characterized as “unity of concept”; and a conclusion of reason is said to be “nothing other than a judgment by means of the subsumption of its condition under a universal rule (major premise)” (CPR, A307/B364).12 From these passages, we can see that the understanding consists in unity of concept, the power of judgment in a specific type of unity of concept, and reason in a specific type of the type of unity in which the power of judgment consists. To mark this order, we might speak of three grades of unity: unity of concept, or of thought and understanding; unity of judgment, or of knowledge; and unity of conclusion, or of a priori knowledge, of reason and science. Kant notes that this order in unity can also be captured in modal terms, as problematic, assertoric, and apodeictic, or as thought of the possible, the actual, and the necessary (CPR, A74–5/B100).13 These grades of unity, then, constitute the three higher faculties of knowledge. Their division and order is not one of external opposition but an internal, logical progression. The capacity to judge is in the understanding, the capacity to think; and reason, the capacity to conclude, is in the capacity to judge and thereby also in the understanding. The three powers are thus just moments of a single capacity to know, the understanding, distinguished by the different grades of unity the exercise of the latter can achieve. To forestall a possible misunderstanding, it bears noting that although the three faculties are ordered, the understanding is not primordial in relation to the other two in the way that the theoretical application of reason is primordial in relation to the practical. For although we can at least conceive of a merely theoretical knower, we can form no conception at all of a subject whose faculty of knowledge includes the capacity to use concepts but lacks the capacity to judge. This impossibility is registered in Kant’s well-known argument that the understanding, conceived as the capacity to think, or to know through concepts, is a capacity to judge, in that the only use it can make of its concepts in knowing is to judge by means of them (CPR, A68–9/B93–4).14 2. When Kant eventually turns his attention to reason, he introduces it as the “supreme cognitive power” (CPR, A299/B355).15 At first sight, this characterization may seem to conflict with the passages just considered, according to which reason stands under the understanding in general, which as the synthetic unity of apperception is “the highest point” to which all use of the understanding must be attached. But no difficulty will confront us here if we keep in mind that while the understanding is just synthetic unity, or the form of thought, reason is this same unity in its highest grade: “the highest unity of thought” (CPR, A298–9/B355). To understand reason, then, we need to understand this highest unity.

The Identity of Reason  15 As discursive, theoretical knowledge consists in judgment, an at least implicitly self-conscious act in which, in the basic case, the concept of an object is determined in an act of predication, as in the judgment this stone is warm. As an act of determination, such judgment constitutes itself as synthetic, distinguishing itself from analytic judgment by implicitly recognizing a thought set over against itself as its logical contradictory, whereas nothing contradicts an analytic judgment. Although the thought opposite to a synthetic judgment is excluded by that judgment from being a judgment itself, it is still a thought and as such no mere aggregate of representations. It has a minimal unity, unity of thought. The judgment to which it stands opposed, in contrast, while also a thought, is constituted by a higher unity, unity of knowledge, which can be characterized as self-agreement. Reflection on these self-conscious acts of thought and judgment can bring their different grades of unity into sharper focus. In a bare thought, diverse concepts are jointly used in a single act, and on account of this joint employment, each use depends on every other. In thinking or conceiving of this stone as not warm, I employ the two concepts stone and warm, therein bringing into play the capacity to use the concept stone and the capacity to use the concept warm. Since these capacities are distinct, in that the use and the possession of each are possible without the use or even the possession of the other, any use of the one is distinct from any use of the other. Yet in the thinking of this stone as not warm, both capacities are used in a single act, a single joint use, so here my use of stone depends on my use of warm, and my use of warm depends on my use of stone. On account of this interdependence, this act is not a process or coming to be, any more than it is a mere aggregate of concurrent acts; its unity, in traditional Aristotelian terms, is that of an activity (energeia). Judgment secures a higher unity. Judging is thinking, so it too lies in a joint use of concepts. In the judgment that the stone is warm, the use of stone depends on the use of warm and the use of warm depends on the use of stone. But here the component acts are not only interdependent but also in agreement, reciprocally furthering one another. The whole act is thus in agreement with itself, sustaining itself in the understanding; it is therein conscious of its own validity, a validity that constitutes it as knowledge and excludes the opposing thought from any unity higher than that of mere thought. Still higher than the unity of judgment and knowledge is the third and highest grade. To understand it properly, we need first to focus attention on the materially conditioned character of theoretical knowledge, its dependence on a condition external to itself, a sensible and empirical condition, lying in affections of receptivity, through which its object is given from elsewhere. Being external to the act, these affections are contingent and diverse, and as they differ, theoretical knowledge falls into a diversity of independent acts of judgment—this stone is warm, that one is cold, the

16  Stephen Engstrom sun is shining, the ice is melting, and so on. Yet despite their independence, the self-conscious self-agreement that constitutes these judgments individually, or distributively, also informs them collectively. For in each of these judgments, it is recognized that even though the judgment itself depends on an external condition, the self-agreement constituting it as knowledge is not fortuitous but belongs to it in virtue of its being an exercise of the capacity to know. And in understanding itself to lie in the exercise of this capacity, a judgment presupposes, through its consciousness of the self-sustaining character of such exercise, that it can never (barring influence from without16) be in conflict with that same capacity’s exercise in other conditions, but is always suited to be in agreement with it. For that which is the very source of the self-sustaining character of an activity cannot itself be conceived as the source of any conflict that would undermine it, and diverse judgments could never be recognizable as one and all exercises of a single selfsame capacity that constitutes such a source (as they must be to be judgments at all) did not the concepts they employ all rest on the same original concept of an object that, through its universality, makes it possible for those judgments to be in agreement. In every judgment, therefore, it is recognized that (barring influence) conflict among judgments is never possible, whereas agreement always is. As exercises of the capacity to know, then, judgments are constituted by the recognition that knowledge must agree with knowledge. So to the extent that these judgments stand in relations in which agreement is possible, they will be in agreement with one another. Two relations of agreement can be distinguished, in accordance with the two respects—objective and subjective—in which concepts are general. In respect of concepts’ objective generality, there will be coherence of judgments, or what we might call objective agreement, where different judgments assert the same thing even though they are of different objects (albeit objects alike in kind and in like conditions). In respect of concepts’ subjective generality, there will be communicability of judgments, or subjective agreement, where different judgments assert the same thing of the same object even though they are made by different subjects (or, if by the same subject, in different contexts of use). Theoretical cognition’s recognition of its collective self-agreement introduces the idea of a higher grade of collective unity that can figure as an end for theoretical reason, namely, unity of system, or the unity of reason and science. So far as the collective unity among the judgments belonging to a body of actually acquired knowledge includes agreement—coherence—but not interdependence, it recognizably lacks completeness. Recognition of this lack presupposes the idea of complete unity in knowledge, from which may arise a representation of such unity as an end.17 Progress toward complete unity requires the discovery of material principles, or judgments constituted by a universal joint use of the component concepts, in which the interdependence and agreement

The Identity of Reason  17 among the uses of those concepts are conceived universally. By furnishing common grounds for the initially independent judgments standing in relations of agreement, such universal cognitions add to that agreement a unity that goes beyond mere coherence, in that the diverse judgments derivable from them are at once recognized as interdependently necessary and interdependently dependent on a single cognition and therein constituted as components of a single act, a system. Thus Kant speaks of the unity of reason as a unity “under principles” (CPR, A302/B359) and calls reason “the faculty of principles,” the capacity through which it is possible to “know the particular in the universal through concepts” (CPR, A299–300/B356–57).18 We can now see that the unity of reason informs all theoretical knowledge in an at least implicit or potential way. For the self-conscious selfagreement of each judgment is a consciousness of implicit necessity, and the anticipation of inter-judgmental agreement that this consciousness implicates is an implicit awareness of the judgment’s validity as universal both in respect of objects and in respect of judging subjects and hence is an implicit understanding of itself as suited to be known through derivation from universal cognition. As common to every judgment, this two-sided universality is independent of a judgment’s logical quantity. It figures implicitly even in singular judgments of experience. In judging that this stone is warm, I implicitly suppose both that any stone constituted as this stone is and in such conditions as it is in would likewise be warm, and that any subject constituted as I am and in such conditions as I am in would likewise cognize that this stone is warm. The necessity and universality of its validity give discursive knowledge the character of law, what we may call legislative form. On the one side, in respect of subjects, it constitutes such cognition’s legitimacy. On the other, in respect of objects, it constitutes its legislation, whereby its determination of the latter amounts to a kind of rulership or governance over them. Reason’s function, then, is to legislate. Legislative form is the mark of rational knowledge, or knowledge from principles.

3  Practical Reason 1. Having now outlined the understanding of reason explicated in logical reflection on theoretical knowledge, we can turn to the idea that the very same reason is also at work in a practical application. Before proceeding, however, we may pause to recall a familiar challenge that might be raised at this point. Because reason’s theoretical application is primordial, practical philosophy is perennially exposed to the question whether it is even possible for reason, on its own, to have a second application that differs so strikingly from the first. Many philosophers, as we know, have doubted or denied that pure reason, reason by itself, can be practical. Hume, the most influential member of this company, declares that

18  Stephen Engstrom “reason is perfectly inert,” resting his skepticism on the supposition that reason is a mere power of comparison, which enables us to analyze representations, but not to unite them.19 On his account, the basic power of the mind to unite its representations is vested solely in the imagination, operating in conjunction with the passions. For present purposes, we need not examine this view in detail. It will suffice to observe that it cannot accommodate the unity of thought and knowledge that we have been articulating, and to that extent it blocks comprehension of how understanding and reason could have any synthetic, materially conditioned use at all, even theoretical.20 So long as we recognize that synthetic judgment lies in a self-consciously constituted self-sustaining unity, we can recognize understanding and reason to be spontaneous powers of synthetic knowledge. This recognition preserves our comprehension of the possibility of reason’s material use, and by thus undercutting the supposition that reason’s power is confined to mere comparison, it removes the obstacle that blocked Hume from comprehending the possibility of its practical application in particular. Even with this obstacle cleared away, however, comprehending the possibility of a practical application of reason will depend on recognizing a difference in idea between reason and theoretical reason, between reason and reason-in-its-theoretical-application. For even though the theoretical application is primordial, the practical application is not to be understood as an application of the theoretical application. To conceive of practical reason along such lines would be to assimilate it to technical practical reason; but practical knowledge in the sense of interest is not technical knowledge.21 The possibility of a practical application of reason can be comprehended, then, only if reason is not conceived as originally so thoroughly sunk in theoretical knowing as to make it impossible to understand such knowing to be the application it is. Hence the distinction noted earlier between formal and material rational knowledge is of no small moment. We need therefore to consider more closely what Kant means in speaking of reason’s application (Anwendung). Application evidently falls under the general heading of use, or employment (Gebrauch). Yet Kant speaks of different sorts of use. He speaks, for instance, of a merely formal, logical use of reason, and he would clearly deem it improper to call such use an application; so application belongs under the more specific heading of material use.22 A distinction in application, then, is a distinction in such use. Now as we have noted, material rational knowledge “considers some object” and in doing so relies on affection. It can therefore be said that so far as any consciousness of diversity among the objects that understanding or reason considers in its material use constitutes a distinction in such use, that distinction will depend on some consciousness of diversity in the sensible conditions of affection. Consciousness of such diversity is for

The Identity of Reason  19 instance required, in theoretical knowledge, in order to divide the use of the concept metal into such materially distinct concepts as gold, silver, and copper. It seems clear, then, that just as the understanding depends on a difference in sensible conditions of use in making distinct concepts through which it can know distinct objects, so reason must depend on some difference in sensible conditions of application in establishing a practical application distinct from a theoretical one. Yet this distinction in reason’s application involves more than would be captured by describing it merely as a distinction in material use. The difference in sensible conditions that enables reason to establish a distinction in its application is not, as in the foregoing example, a difference that makes possible a distinction among the objects that can be known.23 As we noted, the practical perfection of practical knowledge includes theoretical knowledge of its object and so presupposes that its object can, through being practically known, also be known theoretically. Here the difference in sensible conditions must make possible a difference between ways in which objects can be known. A difference between ways in which objects can be known should not be confused with a difference between ways in which objects can come to be known. Objects of theoretical knowledge may come to be known in different ways—directly, for instance, in perception, or indirectly, through inference. But they are all as such known in the same way, namely, as actual. Objects of practical knowledge, in contrast, are not as such known in this way, even though they are also known in this way to the extent that practical knowledge perfects itself practically. A difference between ways in which one and the same object can be known by one and the same reason plainly cannot lie either simply in the object or simply in reason; it must lie rather in the relation that reason’s knowledge bears to the object. And since as we noted theoretical and practical knowledge do not differ in the formal aspect of their relation to their objects, the difference must lie on the material side, the side of actuality or existence, where theoretical knowledge depends on its object and where in the practical case the object depends on the knowledge, as an effect depends on its cause. So it is here, in the material conditions that enable these opposing dependence relations, that the difference in sensible conditions of reason’s application must lie. And it is clear from the just-described difference between theoretical and practical knowledge what the difference in those conditions must be. Insofar as the one species of knowledge depends on its object’s being given from elsewhere and the other works to make its object actual, the former must rely on the subject’s sensibly grounded awareness of a cognition-enabling capacity to be affected by the object whereas the latter must rely on the subject’s sensibly grounded awareness of a cognition-enabling capacity to be causally productive of the object. This difference in awareness must constitute a division in the subject’s sensible constitution, yielding within the subject’s

20  Stephen Engstrom susceptibility to affection two sorts of liability to be affected in a way that enables material knowledge. In particular, the sensible condition of theoretical knowledge must be a power of sense perception, a power of receptive representation (intuition) whose operation is occasioned by sensation, or sensible consciousness of affection from without; and the sensible condition of practical knowledge must be a power of sensible desire, a power of productive representation whose operation is occasioned by feeling (pleasure or displeasure), or sensible consciousness of affection from an inward source in perception or experience.24 Without the former, there could be no appearances, no material for experience; without the latter, no wish for happiness, nor indeed any material for the power of choice. A distinction in the application of reason depends, then, on the subject’s capacity to discriminate between the two modes of affection, sensation and feeling. We may accordingly say that for one and the same reason to be distinguished in the application is for a certain function universally operative in knowledge that consciously relies on sensation to be at least implicitly conscious of itself as also universally operative in knowledge that consciously relies on feeling. And that function, we saw, is cognitive unity in its highest grade. 2. As we noted, the practical application of reason contains two moments (§1.3). The first is a cognitive moment, in which reason determines its object, establishing a distinct species of knowledge, different in kind from theoretical. Here reason’s function of unity applies immediately within its practical use, beginning in subjects’ determination of their wish for happiness, prior to exercising choice. Just as in its basic determination of what is reason does not rely on knowledge of what ought to be, so in its basic determination of what ought to be it does not rely on knowledge of what is. The second moment is one of practicality, or self-conscious efficacy, in which practical knowledge works to make its object actual, or an object of theoretical knowledge as well. Here reason’s function of unity unites practical and theoretical knowledge. This moment is no less essential than the first. Insofar as reason is nothing but cognition’s highest function of unity, it cannot have two simply unrelated applications. To the extent that reason has a second application at all, that application must be practical, perfecting itself in the highest unity between the two applications. So in its first moment reason’s practical application relates only to itself, requiring agreement among practical judgments, and in its second moment it relates itself to the theoretical, requiring agreement between practical and theoretical judgments.25 Since this second moment brings us to the ideal coincidence of theoretical and practical knowledge in complementary representations of what would be at once a single reality and a single good, namely, the realization of the highest good, it points toward the metaphysical questions relating to the unity of reason that, as I said at the outset, lie beyond my main concern here. I will comment on these only briefly at the conclusion.

The Identity of Reason  21 To explicate reason’s function in the first moment, we need first to spell out more fully how the practical application of reason constitutes a distinct species of knowledge. Theoretical knowledge, as we noted, relies on sensible affection from a source outside consciousness. It therefore conceives of its object as existing independently of it and understands itself to be dependent in respect of actuality on that object. Practical knowledge, in contrast, even though it too relies on sensible affection (albeit from an inward source), is efficacious in respect of its object, indeed selfconsciously so, conceiving of its object as its own effect and hence as dependent for its actuality on the knowledge. This basic difference entails that the two species of knowledge also differ in how they originate in the subject. Theoretical knowledge begins in sense perception of particular effects in experience and proceeds, through concepts of the objects that are their causes, to material principles, which furnish universal knowledge of the latter; practical knowledge, in contrast, begins in a formal principle, a law of causality implicated in its consciousness of itself as efficacious, and proceeds, through its concept of its object, its effect, to the pleasing sense perception of the latter as it appears in experience. A  difference in respect of the constitution of their objects is also entailed. Theoretical knowledge is of phenomena, or things knowable solely through their existence, or actuality, as things that can appear, or be perceived, by affecting consciousness; while practical knowledge is of noumena, or things that can exist and appear solely through the knowledge of how they are to be.26 In representing its object as its own effect, practical knowledge is knowledge of its own action—its own causality—and so knowledge of itself as practical. It is therefore a kind of self-knowledge, representing as its own the good conduct that it works to make actual, though its representation is practical, not theoretical, and so not of what is done, but of what is to be done. If, in keeping with Kant’s traditional usage, we designate as persons the subjects that are understood in practical knowledge to be the bearers of the capacity for such knowledge, we can say, in short, that practical knowledge is persons’ self-consciously efficacious knowledge of how they themselves—at once the subjects of that knowledge and the objects known in it—should live and act. An immediate implication of this identity of subject and object is that the community of subjects who share the capacity for practical knowledge necessarily coincides with the community of objects capable of being known through it. The two communities are but two sides of a single community, the community of persons. No such coincidence is to be found in theoretical knowledge, which does not require that its objects be members of the community of subjects capable of knowing them. Yet despite this difference, practical knowledge exhibits reason’s legislative function. Its necessarily coinciding communities of subjects and of objects are constituted by the same two-sided universality that,

22  Stephen Engstrom as we noted, is already recognizable in reason’s theoretical application as integral to the form of rational knowledge. One and the same legislative form, then, is operative in both applications. The practical application is distinguished in that, as self-knowledge, its legislation is self-legislation, or the autonomy of practical reason that Kant’s analysis of common moral rational knowledge identifies as the principle of morality. So if that analysis is correct,27 it reveals morality to be the work of reason, the practical application of the very same activity of pursuing the highest unity of knowledge that logical reflection has already revealed reason to be pursuing in its theoretical application. Through the identity of its subject and its object, practical knowledge bears a certain affinity to the formal knowledge with which theoretical knowledge was originally contrasted. Yet though practical knowledge in a way overcomes the distinction between subject and object, it also preserves it and so does not collapse into logic’s merely formal self-knowledge of the understanding and of reason. Practical knowledge is material self-knowledge, a self-knowledge that is also knowledge of objects. For being reliant on the material condition of feeling, practical knowledge can be of a diversity of selves, known to one another as others, namely, other selves. In and through such knowledge, persons determine how they should live, determining themselves first in themselves yet thereby always also in interaction with others through universally self-legislated laws of conduct, which constitute them as equal members of a single community grounded in a single form or power of reason to which they all belong. So notwithstanding their diversity, persons are all united in a single original cognition of a single ultimate object, the highest good, which comprises, first, practical cognition’s own causality, or the lawful conduct that constitutes this object’s form, and second, the informed material to be realized through that causality: in a word, universal happiness collectively consequent upon universal virtuous conduct. 3. Finally, we turn to the second moment of reason’s practical application. As we saw, reason must secure not only unity within this application, but also unity between it and the theoretical. This is not a third application of reason. Reason’s legislative function secures agreement among judgments that belong to a single way of knowing, whereas here agreement is to be secured between heterogeneous ways of knowing: this agreement is not objective, not an identity of legislative form among judgments differing in their specific content, but subjective, an identity of content between judgments differing in their specific legislative form.28 This second moment, as was said, is the moment of practicality, and its perfection lies in the complete realization of the ultimate object that practical knowledge works to make actual, the realization wherein theoretical and practical knowledge necessarily coincide as complementary representations of what would be at once a single reality and a single good. Such coincidence presupposes, of course, that concepts first

The Identity of Reason  23 constituted in reason’s practical use—beginning with person, conduct, and community—are available for a practically informed theoretical use, in a development of reason’s capacity to know what is that enables practically known objects to become recognizable in experience, in the pleasing apprehension of the good.29 We noted that in its first, cognitive moment, practical knowledge differs from theoretical in that it begins not in the sensible conditions of experience but in a formal principle of self-legislation. Since this principle is independent of experience, its necessity is not sensibly conditioned. Unlike theoretical knowledge, therefore, practical knowledge does not here advance stepwise through the three grades of unity, from possibility through actuality to necessity,30 but begins in a necessary cognition. Through its consciousness of its own necessity this cognition represents its object, what it is to effect, as also necessary—that is, as practically necessary, as necessarily to be done, or to be made actual. The practical perfection of this cognition therefore implicates a coinciding necessity in the practically informed theoretical knowledge of the object, whereby what is practically known as necessarily to be done is theoretically known as necessarily done. But since theoretical knowledge advances from possibility through actuality to necessity, this perfection must be approached in stepwise fashion. In the first step, then, the step to actuality, reason requires that the object be possible, in accordance with discursive cognition’s constitutive understanding that possibility is a necessary condition of actuality. Now in the sensibly conditioned use of practical reason, this possibility condition has a merely negative, limiting use. Since an object of sensible desire can be made actual only so far as it conforms to the conditions under which the experience of it is possible, practical cognition that works to make such objects actual must conform to theoretical cognition’s judgments of possibility.31 Such use of the possibility condition is utterly familiar, having long received the bulk of attention in practical philosophy’s study of practical reason. Here practical reason limits its object to what theoretical knowledge allows as possible (practicable) before adding to it some theoretically cognizable specification of how this object, thus limited, can be made actual. Such reliance on its theoretical application is constitutive of reason in its sensibly conditioned practical application, and here reason introduces into the second moment of its practical application subalternate, theoretically conditioned modes, commonly known as prudential and technical reason. But where, as in the case of the highest good, the object is cognized as necessarily to be made actual, a positive use of the possibility condition is required. Since reason’s theoretical application, being reliant on sensible affection from a source outside consciousness, is limited to phenomena, pure reason is able, through its idea of the intelligible ground of the latter, to extend its practically informed theoretical use beyond the limits of

24  Stephen Engstrom purely theoretical knowledge in order to represent as theoretically possible what it practically represents as necessarily to be made actual. That is to say, it is warranted in this extension on the ground that if only what is possible can be made actual, then what is necessarily to be made actual must be possible. This positive use of the possibility condition is grounded, Kant holds, in the primacy of pure practical reason’s interest in its unity with theoretical (in the realization of the highest good) over theoretical reason’s interest in its own unity alone, and it receives articulation in his doctrine of the postulates of pure practical reason. In contrast to the subalternate modes of application just noted, it is the highest expression of the necessary unity of discursive reason, in that it alone preserves that unity in its full perfection, without limitation. The resulting positive representation of possibility can be briefly outlined as follows. The possibility of the object begins where the representation of it begins, in the representation of its determining form, lawful conduct. Since practical knowledge represents this conduct as necessarily to be made actual, pure reason is entitled to extend its practically informed theoretical use beyond the limits of purely theoretical knowledge in order to represent this conduct as possible. This representation is thus theoretical, though it differs from purely theoretical knowledge in that it depends on reason’s practical application. On account of this dependence, it is a postulate of pure practical reason, and as the first such postulate, it is the starting point of practically informed theoretical knowledge. Pure reason is accordingly conscious of its own power as an intelligible cause inwardly to determine a person’s power of choice to choose conduct that it, reason, represents as necessary, and it therein represents the determin­ ability of that power of choice by pure reason and hence the freedom of that power to choose in accordance with reason’s practical knowledge. With respect to the matter of practical cognition’s ultimate object, two further postulates are entailed, in accordance with the two elements that this necessary object comprises. The first element pertains to the constitution, or character, of persons, or the beings represented as subjects in practical knowledge; this is the condition of virtue, a disposition of the free power of choice that in its perfection is in complete conformity with practical cognition’s determining form, notwithstanding the natural susceptibility of this power to affection by sensible desire. The second element pertains to persons in respect of their state; this is the condition of happiness that practical knowledge, through the sensible material that figures in its predicate, includes in its representation of virtuous conduct’s complete effect. Because it is only as virtue’s effect that happiness is practically cognizable, happiness is posterior to virtue in practical cognition’s representation of its ultimate object; yet because happiness is practically cognized as virtue’s effect, it is still necessarily represented in that object. Now since virtue is included in this necessary object, pure reason represents virtue as possible, namely, by representing persons as intelligible beings originally so constituted in their finitude as to be subject to

The Identity of Reason  25 sensible conditions in a way that agrees with the potential for virtue’s attainment through freely chosen lawful conduct. And as for the second element, since on account of their finitude individual persons necessarily lack the power fully to ensure their own happiness on their own, or distributively, the ensuring of happiness is represented as possible collectively through conceiving reason itself—the intelligible form inwardly grounding the noumenal community of all persons—as also the ground of the phenomenal world in which that community is empirically manifested. Pure reason thus establishes the possibility of attaining the unity of its theoretical and practical applications in the realization of the highest good through complementing its postulation of an intelligible ground for its practical concept of conduct with the postulation of intelligible grounds for its practical concepts of person and of community. A final word relating to the step from actuality to necessity: in the highest good the ideal necessary agreement between the theoretical and the practical applications involves a relation of reciprocal furtherance between theoretical knowledge and practical cognition’s efficacy. Such reciprocity also figures in the approximation to this ideal. Through making actual the conduct it represents, practical knowledge finds its completion in practically informed theoretical knowledge, and the resulting theoretical knowledge, through its pleasing consciousness of the realization of the good, in turn reinforces the efficacy of the practical knowledge.32 This happens not only in the individual but also interpersonally, when the exemplary actions of one person elicit admiration in others, inspiring them to similar conduct. Such progress takes place not blindly, moreover, but through the virtuous cultivation of mental and physical talents in individuals and of cooperative social relations in political and ethical communities. It is sometimes thought that in advancing his doctrine of the primacy of pure practical reason to license the introduction of pure practical reason’s postulates, Kant lapses back into the transcendent metaphysics that his critical investigation of theoretical reason had proscribed. But if the foregoing outline account of pure reason’s identity is correct, that doctrine’s true purpose lies in the fullest possible self-development of the unity of one and the same reason in its two applications. For as we have seen, that identity is the identity of the unity that constitutes reason itself, and that unity, in its perfection, is its own final end. “Pure reason is in fact occupied with nothing but itself” (CPR, A680/B708).33

Notes 1. Cf. CPracR, 5: 121. Translations of quotations from Kant’s writings are my own. 2. The appearance of inconsistency also arises in reason’s theoretical application. It was the scandal of an apparent contradiction of reason with itself, Kant said, that drove him to undertake a critique of pure reason (letter to Christian Garve, September 21, 1798).

26  Stephen Engstrom 3. Not that this topic has received no consideration in recent discussions of the unity of reason. But focusing attention on it may complement what can be found in studies offered so far, for example, O’Neill (1989, Part I), Guyer (1989), Neiman (1994), and Kleingeld (1998). 4. In the Critique of Pure Reason, philosophical knowledge is called rational knowledge merely from concepts (A713–4/B741–2). Since such knowledge does not depend on any arbitrary acts of thought, it does not include rational knowledge from the construction of concepts, which embraces mathematical knowledge and in general all technical rational knowledge; “the philosopher is not an artist of reason, but the legislator of human reason” (A839/B867). 5. See, for example, CPracR,  5: 15, 89; Nicomachean Ethics I.3, 1095a5–6. Some readers who interpret Kant’s distinction along such lines register a sound opposition to approaches that seek to understand ethics as an application of theoretical knowledge to practice. See for instance Rawls (1980, 554–72) and Korsgaard (2008). But in Kant’s division of philosophy, ethics is conceived as knowledge that is sui generis, different in kind from theoretical; it is an application of reason, but not an application of theoretical reason or of theoretical knowledge (cf. §3.1 below). 6. The relation between understanding and reason will be considered in §2.1. 7. Kant’s recognition of this primordiality is reflected in his limitation of transcendental philosophy to theoretical knowledge; such philosophy must exclude everything practical, Kant says, since it must be free of everything empirical. “Transcendental philosophy is a philosophy [Weltweisheit] of pure merely speculative reason. For everything practical, so far as it contains springs [Triebfedern], relates to feelings, which belong to empirical sources of knowledge” (CPR, B29). This exclusion may seem puzzling, given Kant’s repeated statements that pure reason is practical and that it is not by means of any feeling that pure practical reason’s moral law determines the will (CPracR, 5: 3, 71–2). But Kant’s reason for this exclusion can be seen if we bear in mind that transcendental philosophy is confined to the absolutely first principles of knowledge, being the part within the critical philosophy that supplants traditional ontology, or first philosophy (CPR, A845/B873; cf. MF, 4: 469–70). The practical application of reason is possible only in a being with a faculty of desire, which, as a capacity to be, through one’s representations, the cause of their objects (MM,  6: 211), presupposes both a faculty of representation and a capacity to feel (pleasure and displeasure) (see §3.1 below). The faculty of representation, in contrast, being originally employed in bare perception, does not conversely presuppose the faculty of desire; and unlike the capacity to feel, it is integral to everything theoretical and to all knowledge whatsoever. Since the practical use of reason thus presupposes the capacity to feel, which is not a condition under which knowledge is possible at all, whereas transcendental philosophy rests solely on such conditions, practical philosophy lies outside of transcendental (cf. CPR, A801n/B829n, CPJ, 5: 181–2, MM, 6: 211–2n). So even though Kant’s practical philosophy is concerned fundamentally with the question whether pure reason is practical, it must advert to certain concepts drawn from psychology—namely, life, faculty of desire, and pleasure (CPracR,  5: 9n)—and none of its various a priori arguments, deductions, and proofs are said to be transcendental (with one apparent exception, the deduction of the concept of the highest good [CPracR, 5: 113]). 8. Another description of a notional subject in whom reason’s knowledge is only theoretical is provided at Rel, 6: 26n (though here the subject is able to apply this knowledge in deliberation).

The Identity of Reason  27 9. Cf. CPR, A15/B29, A50/B74, Anth, 7: 140–1; CPR, B132–3 (cf. A107–8). 10. This synthetic unity of apperception is identified with the understanding at CPR, B134n; it is characterized as “qualitative” at CPR, B131, with a reference back to §12, where ‘qualitative unity’ is offered as an alternative for ‘unity of concept.’ 11. Cf. JL, §17, where judgment is defined as “the representation of the unity of consciousness of different representations or the representation of the relation of the same, so far as they constitute a concept.” 12. Cf. A321–2/B378 and JL, §56: “A conclusion of reason is the knowledge of the necessity of a proposition [that is, an assertoric judgment] through the subsumption of its condition under a given universal rule.” See also A330/ B386: “Reason, regarded as the faculty of a certain logical form of knowledge, is the faculty to conclude, that is, to judge mediately (through the subsumption of the condition of a possible judgment under the condition of a given one).” 13. Kant also notes that this order grounds a modal progress in discursive cognition. In traditional Aristotelian terms, the progress could be described as from potentiality through second actuality to the perfection of first actuality (cf. CPR, A76/B101). 14. The cogency of Kant’s argument can be confirmed by considering his account of how the generality of representation that constitutes a concept is made (CPR, B133–4n, JL,  §§5–6). According to that account, the act that first constitutes a concept lies in the consciousness of a representation as containing a representation that it can share in common with other at least possible representations distinct from it, and such a consciousness is nothing other than the recognition of one representation as contained under another, or the act of subsumption that constitutes judgment (cf. CPR, A132/B171, A137/ B176). In short, concepts are born in acts of judgment. So although it is possible for a subject to think without judging, it is not possible for there to be understanding without the capacity to judge.   It is a further question whether it is possible to conceive of a capacity to judge that lacks reason. Here there is no prospect of answering in the negative by arguing in parallel fashion that judgments are born in acts of conclusion. There seem, on the contrary, to be grounds for thinking that the capacity to judge is primordial in relation to reason in a way that is analogous to the way that theoretical reason is primordial in relation to practical, but I shall not explore this question here. 15. Cf. A702/B730. 16. Namely, such influence as would constitute error (cf. CPR, A294/B350–1). 17. The act of representing the complete unity of knowledge as an end may at first seem to belong to reason’s practical application. But though a kind of wish, this act is not practical. It works, not to make actual an object it already knows (practically), but rather to know (theoretically) an object that is already actual but not yet known. It is thus peculiarly liable to prompt Meno’s famous question for philosophy: How can we look for what we seek to know without knowing what we seek? (Meno, 80d–e). Indeed, this formal wish—or Wiß­ begierde (CPR, Bxv, A856/B884; cf. A708–9/B736–7)—is philosophy itself, springing from pure reason, the natural disposition to metaphysics (B21–2), what Aristotle saw as the natural human desire to know (Metaphysics I.1, 980a21). Pure reason, then, is the formal-desiderative aspect of the cognitive faculty, the seat of its wish to actualize not its object but itself. As we shall see, such a wish also figures in reason’s practical application, but in the guise of a formal principle or law.

28  Stephen Engstrom 8. Cf. CPR, A299–300/B356–7. 1 19. A Treatise of Human Nature III.i.1. Like Kant, Hume regards human understanding as a discursive cognitive capacity, denying to it any power of intuition (beyond what figures in the mere comparison of representations), but unlike Kant he takes it to be merely analytically rather than synthetically discursive. 20. Thus, Hume remarks, “if reason be inactive in itself, it must remain so in all its shapes and appearances, whether it exerts itself in natural or moral subjects” (Treatise III.i.1). 21. As Kant makes clear: CPracR,  5: 25n, CPJ,  5: 172–3, MM,  6: 217–8; cf. JL, 9: 86–7. 22. Kant never, to my knowledge, speaks of the formal or logical use of the understanding and of reason as an application of these powers. Since he contrasts logical use with real use (e.g. at CPR, A303–9/B359–66), his notion of application might seem close to his notion of real use. The two are not to be identified, however. For Kant’s criticism of traditional speculative metaphysics can be succinctly expressed in the proposition that all real use of the discursive cognitive capacity must (so far as it is objectively valid) be material use, or application, that is, use that relies on something given. So while logical use is use of the capacity, material use is use of the capacity that is also use of given material, material available for use. It is only in relation to given material, then, that we can speak of the application of a capacity; application is such use of the capacity as is also use of given material. 23. This means that (as constructivist interpreters maintain) the distinction in application assumes no special moral sense or rational intuition of values, moral facts, or relations of fitness. 24. Here ‘sensation’ (Empfindung) is used in the narrow sense employed in the first Critique (A19–20/B34; cf. CPJ, §3). That Kant thinks of reason’s distinction in application as resting on this division in our sensible constitution is reflected in the fact that, just as he begins his critique of pure theoretical reason with a consideration of sensibility and its forms of outer and inner sense (after excluding feeling and desire and all cognition that implicates them), so he presupposes in his practical philosophy the psychological concepts life, faculty of desire, and pleasure (CPR, A14–6/B28–30, CPracR, 5: 9n; cf. MM, 6: 211–2n). 25. The order of these moments is expressed in Kant’s remark that philosophy, “the legislation of human reason,” “contains the law of nature as well as the law of morals, initially in two separate [besonderen] but ultimately in a single philosophical system” (CPR, A840/B868). If we bear in mind that reason’s primordial use is theoretical (§1.3), we can see here three stages in the progress of reason’s material use—theoretical, practical, the unity of the practical with the theoretical—which are reflected in the order of the three Critiques. 26. Cf. CPracR, 5: 16, 89–90; 43, 54–7. 27. As I shall here assume it to be, having examined it elsewhere, in Engstrom (2009). 28. It is to be noted, however, that recognition of this identity of content depends on an abstraction that was not in play earlier (in §2.2), when we were considering only subjective agreement among judgments belonging to the same way of knowing. In subjective agreement among such judgments, the identity of content is the sameness of such content as is shared for instance by your judgment and mine when we agree that this stone is warm, asserting the same thing of the same thing. In the second moment of reason’s practical application, in contrast, the identity to be secured is an identity across different ways of knowing; it is the sameness of such content as is shared for instance by a practical judgment that a certain person ought to live justly and a theoretical judgment

The Identity of Reason  29 that this person does live justly. In one sense, these two judgments do not assert the same thing of the same thing, but in another sense they do, though in different ways, in accordance with their difference in legislative form. 29. See CPJ,  §§4–5. More precisely, the apprehension is of the good in the appearance (in the experience of its “ectype”) (CPracR, 5: 43). It bears noting that the three concepts mentioned here already reflect reason’s identity, in that they are practical applications of the categories of relation (cf. CPR, A80/B106, CPracR, 5: 66), concepts that are primordially applied in theoretical knowledge but which, free of the schemata that at once enable and limit such theoretical use, can be used practically (and so also used in pure reason’s practical postulates, which, in a manner to be described below, enable that practical use by representing conduct as free, the person as immortal, and community as grounded in infinite reason, or God). 30. Cf. §2.1. Theoretical knowledge proceeds from the understanding’s sensibly conditioned (schematized) concept of its object (articulated a priori in the formal principles of pure understanding), wherein the object is represented as possible, through the exercise of the power of judgment in experience, wherein the object is represented as actual, to the exercise of reason in empirical science, wherein the object is represented as necessary (through being cognized from a material principle). 31. See CPracR, 5: 57, where Kant says that an object of practical reason must be physically possible. An object of pure practical reason, he adds, must in contrast be morally possible. 32. This reciprocity of reason’s two applications reflects and informs the homeo­ stasis of their material conditions in human life, where perception leads (inwardly, through feeling) to desire, and desire leads (outwardly, through action) to perception. I discuss the sensible side of this reciprocity in §2 of Engstrom (2010). My description of the reciprocity of reason’s applications is similar in some respects to the description of the teleological unity of theoretical and practical reason in §3 of Zöller (2016). 33. This paper was presented in 2016 at the Universities of Tel Aviv and Frankfurt and in later versions at a Berlin workshop on hylomorphism in Kant and German Idealism and at the Universities of Illinois at Urbana–Champaign and North Carolina at Chapel Hill. I thank the participants for the fruitful discussion and also the editors of this volume for their helpful comments.

References Engstrom, Stephen (2009). The Form of Practical Knowledge. Cambridge: Harvard University Press. Engstrom, Stephen (2010). “Reason, Desire, and the Will.” In Lara Denis (ed.), Kant’s Metaphysics of Morals: A Critical Guide. Cambridge: Cambridge University Press, 28–50. Guyer, Paul (1989). “The Unity of Reason.” Monist 72, 139–67. Reprinted in Paul Guyer (2000). Kant on Freedom, Law, and Happiness. Cambridge: Cambridge University Press, 60–95. Kleingeld, Pauline (1998). “Kant on the Unity of Theoretical and Practical Reason.” Review of Metaphysics 52, 500–28. Korsgaard, Christine M. (2008). “Realism and Constructivism in TwentiethCentury Moral Philosophy.” In Christine M. Korsgaard, The Constitution of Agency. Oxford: Oxford University Press, 302–26. Neiman, Susan (1994). The Unity of Reason: Rereading Kant. Oxford: Oxford University Press.

30  Stephen Engstrom O’Neill, Onora (1989). Constructions of Reason. Cambridge: Cambridge University Press. Rawls, John (1980). “Kantian Constructivism in Moral Theory.” The Journal of Philosophy 77, 515–72. Zöller, Günter (2016). “ ‘The supersensible . . . in us, above us, and after us’: The Critical Conception of the Highest Good in Kant’s Practico-Dogmatic Metaphysics.” In Thomas Höwing (ed.), The Highest Good in Kant’s Philosophy. Berlin, Boston: De Gruyter, 263–79.

2 Lichtenberg’s ‘Es denkt’ versus Kant’s ‘Ich denke’ Patricia Kitcher

1.  Zöller on Lichtenberg’s Aphorism One should say, It thinks, just as one says It lightnings. (WB, K76)1

Lichtenberg’s aphorism has often been understood as an attack on Descartes’ cogito claim in the second of his Meditations on First Philosophy (Descartes 1640/1981–94, II.17). But in 1992, Günther Zöller argued that Kant’s theory of the subject of thought was a more enlightening context within which to appreciate the strengths—and weaknesses— of Lichtenberg’s cryptic remarks about ‘I think.’ In the intervening 30 years, several other scholars, including Tyler Burge (1998) and Tobias Rosefeldt (2000), have also suggested that we can learn more from Lichtenberg’s challenge by seeing how it can be turned back by appealing to Kantian rather than Cartesian resources. Although I will appeal to their discussions later, I outline the contours of the Kantian contexts of Lichtenberg’s remarks by following Zöller’s more scholarly discussion. Zöller argues that Kant’s complex theory of self-consciousness can answer some, but not all of Lichtenberg’s objections to maintaining that there is an ‘I’ that thinks. I try to expand on Zöller’s work by arguing that Kant has resources that would enable him to provide a fuller ‘reply’ to Lichtenberg. Zöller begins his discussion by citing the entire Waste Book entry that introduces ‘Es denkt’: We become conscious of certain representations that are not dependent upon us; others, at least we believe, are dependent upon us; where is the boundary? We know only the existence of our sensations, representations, and thoughts. It thinks, we should say, just as one says, it lightnings. To say cogito is already too much if we translate it as I think. To assume the I, to postulate it, is a practical necessity. (WB, K76, cited in Zöller 1992, 417)2 DOI: 10.4324/9781003044376-3

32  Patricia Kitcher As Zöller notes, the crucial setup to the assertion that it is too much to say ‘I think’ is the (rhetorical) question about a boundary line between dependent and independent representations (Zöller 1992, 423–4). Unless that line can be drawn, how can ‘I’ be understood as that which creates representations or thoughts? This question has its heritage not just in Kant, but in the widely accepted work of Christian Wolff (another of Lichtenberg’s targets3). For Wolff and his many followers, consciousness requires differentiation.4 To be conscious of an object, one must differentiate that thing from others. For should we be conscious of that which we cognize through the senses, we must recognize the difference between that thing and others. (Meta §730, 455–6)5 Further, it is through differentiating things from one another that one is able to differentiate the self: [D]ifferentiation is an effect of the soul, and we cognize therefore through it the difference between the soul and the things that are represented. . . . This difference [between ourselves and other things] appears directly as we are conscious of other things. (Meta §730, 455–6) That is, subjects can discern themselves only because the act of differentiating among other things enables them to separate themselves from other things. They are the differentiators, the creators of differentiating representations; the objects are merely those that are differentiated and represented. Against the background of the Wolffian doctrine that consciousness requires differentiation, Transcendental Idealism (henceforth TI) poses several threats to self-consciousness. Zöller offers many texts to support his claim that Lichtenberg was fascinated by TI and that he saw it as inconsistent with the doctrine of an I  that thinks. As Zöller notes, the fundamental difficulty for Lichtenberg is that Kant holds that the representations humans form are a result of an interaction between their minds and the world (Zöller 1992, 423). Lichtenberg sees problems arising from both directions. First, the fact that representations are a mixture of what our minds contribute and what the world contributes implies that it is very hard to separate these contributions (a fact Kant concedes at CPR, B1–2). [W]henever anything affects us, the effect depends not only on the thing producing the effect but also on that upon which the effect is produced. As in collisions, both bodies are at once active and passive,

Lichtenberg’s ‘Es denkt’  33 for it is impossible for one thing to be acted upon by another without the overall effect appearing compound. (WB, K74) Note that this Waste Book entry almost immediately precedes the one containing the aphorism. Since by Kant’s own theory, the representation of an ‘external’ object is a compound of what is contributed by the world and what is contributed by the mind, how can the sources (and so objects) of such representations be understood as completely different from the mind? Hence, the difficulty of finding a demarcation line between the mind and the world. The problem with separating the world from the self is underscored by Lichtenberg’s skepticism concerning Kant’s claims about spatiality. As Zöller notes, Lichtenberg objects to what he sees as Kant’s inference from the claim that objects of representations are other than us (praeter nos) to his assertion that they are outside of us (extra nos) (Zöller 1992, 423–4). Further, Lichtenberg’s general skepticism about inferring from ‘involuntary’ representations to their cause leads him to suggest that, for all we know, the mind is the unconscious cause of involuntary representations as well as the conscious cause of voluntary ones (and as well as the unconscious cause of forms of representation) (Zöller 1992, 426–7). Thus, any attempt to separate the world from the self along either the spatial or the voluntary dimensions is going to fail. The separation does not go any better in the other direction, as we try to separate the self from the world. As Zöller notes, for the I to be a separate thing, it must be self-sufficient and so not dependent on the world. He thinks that Lichtenberg has an argument that it cannot be self-sufficient, because it depends on an ‘impersonal I’ that presents representations to it (Zöller 1992, 431). For reasons I give later, I am not convinced about that reading of Lichtenberg, but we should recall that Kant also thinks that the human ability to be self-conscious is dependent on receiving data of sense. One aspect of the Transcendental Deduction that scholars have found very puzzling is Kant’s insistence that the [s]ynthetic unity of the manifold of intuitions, as given a priori, is thus the ground of the identity of apperception which precedes a priori all my determinate thinking. (CPR, B134) That is, the unity of consciousness depends on the materials of intuition being combined according to categorial principles. Given a sufficiently skeptical bent, TI furnishes plenty of support for the impossibility of differentiating a subject from the world. Zöller notes that Lichtenberg agrees with Kant that other than things in themselves (which Lichtenberg thinks cannot be posited) all objects are nothing but

34  Patricia Kitcher representations (Zöller 1992, 432). Hence idealism seems inevitable and the distinction between what belongs to the self and what belongs to the world vanishes. Still, Zöller thinks that the principal reason for Lichtenberg’s skepticism about the ‘I’ is his puzzlement about how a self can be both a subject and an object. Zöller cites several texts that indicate the centrality of this issue in the Waste Books, for example:

and

The animal is always a subject to itself, the human being is also an object to itself. (WB, H142) I and me. I feel myself—these are two objects.

(WB, H146)

But how can the subject be an object for itself if the self as subject and the self as object are two different objects? Zöller takes this puzzle to be the key to understanding the aphorism: The reason for Lichtenberg’s puzzlement becomes clear. The “me” observed in introspection appears as foreign to the observer-I as anything else that originates in the mind without consciousness. It is in the context of these attempts at articulating the self’s internal duality that the aphorism on the “it thinks” has to be understood. The use of the third-person neuter pronoun “it” [es] in the phrase “it thinks” [es denkt] might be taken to refer to some anonymous agency behind all thinking, to some thing that thinks—thus replacing the self’s own agency in thinking with that of some other entity. Yet the proposed analogy to phrases such as “it lightnings” suggests that the identification of a determined underlying agent is given up altogether and replaced by the reference to an impersonal, anonymous agency—an agency without an agent. There simply is no person, neuter or otherwise, to which the “it” in the phrase “it lightnings” refers. Zöller (1992, 428) This puzzle leads Zöller to his interpretation of the aphorism: Lichtenberg’s phrase “it thinks” is a suggested linguistic improvement upon the earlier distinction between “I” and “me,” between the self as the hidden source of representations and the self as the subject of consciousness. The phrase “it thinks” seems to eliminate totally any reference to the I. Yet a more complete phrasing might include a second clause stating that “I become conscious of it thinking.” The I is no longer

Lichtenberg’s ‘Es denkt’  35 the agent of thinking, but the I still functions as the observer of the thinking. Lichtenberg’s point is not the outright elimination of the I, but rather its demotion. Zöller (1992, 429) Zöller’s reading of the aphorism has a number of virtues, especially in relation to his overall argument for the Kantian context. Kant also expressed wonder at how I can say: I, as intelligence and thinking subject, cognize myself as an object that is thought, insofar as I am also given to myself in intuition . . . This is no more and no less difficult than how I can be an object to myself in general and indeed one of intuition and of inner perceptions. (CPR, B155) Still, Lichtenberg and Kant address the problem from opposite ends. Kant’s puzzlement concerns how the spontaneous actively thinking ‘I’ can be the same subject as the subject whose states passively ‘appear’ in inner sense, whereas, on Zöller’s reading, Lichtenberg doubts that the subject who observes thinking can be the same as the agent who produces it. Given this (plausible) understanding of Lichtenberg’s aphorism, Zöller argues that Kant can weaken the argument for rejecting an I that thinks. The key resource available to Kant, but apparently not to Lichtenberg, is the recognition that cognition of objects involves a self-consciousness that is neither a consciousness of the states of an empirical self nor the (impossible) consciousness of a substratum underlying thought—the thing in itself that thinks. Rather, it is a consciousness of thinking (Zöller 1992, 437–8, CPR, B413). Given this central claim of the Transcendental Deduction, Kant can resist Lichtenberg’s claim that thinking is just something that happens of which the subject is then conscious in inner sense. On the other hand, Zöller thinks that the Kantian reply is only partial. It avoids the Lichtenbergian paradox of an ‘I’ that is not an I (because it is separated from the activity of thinking [Zöller 1992, 439]), but does not answer the underlying problem of differentiating a thinking subject. The problem, he argues, is that the unity of thinking required for cognition does not indicate any object that could be a subject of thought, but only the form of thinking: On Kant’s psychological theory, then, that of which I am conscious when being conscious of myself is neither the determinable, transcendental subject of thinking nor the determined, empirical self, but rather the activity of thinking as such. For Kant, self-consciousness is not a consciousness that refers to some particular and peculiar object; it does not refer to an object that is empirically given nor to

36  Patricia Kitcher one that remains elusive “behind” the appearances. The consciousness of oneself emerging in transcendental apperception is therefore “not a representation distinguishing a particular object, but a form of representation in general” (CPR, B404/A346; my emphasis)—a form or structure exhibited by my thoughts insofar as I am able to attribute them to myself. As Kant puts it most succinctly: “. . . the ‘I’ is merely the consciousness of my thinking” (B413; translation amended). Zöller (1992, 438) In Section 3 I will suggest that Kant can get further with this problem than Zöller believes. But first I  offer an alternative reading of Lichtenberg’s aphorism that brings out a different Kantian connection.

2.  Thinking and Pre-established Harmony Zöller notes that Lichtenberg was interested in the problem of pre-established harmony, but that observation does not play a role in his reading of the aphorism (Zöller 1992, 424–5). Although ‘pre-established harmony’ is usually taken to refer to Leibniz’s hypothesis (e.g., in the Monadology § 81, cf. Leibniz 1885) about how windowless monads could appear to be in causal interaction (even when they are not), both Kant and Lichtenberg mean something different by the phrase—and they mean the same thing by it. At the end of the B Deduction, Kant reflects on the system he has (thus far) presented by raising the issue of the ‘necessary agreement’ of experience with the concepts of its objects: either experience makes the concepts possible or the concepts make experience possible (CPR, B166–7). Kant’s solution to this conundrum has been the latter, since the first relation could not produce necessity, but he also considers a third possibility: The categories were neither self-thought a priori first principles of our cognition nor drawn from experience, but were rather subjective predispositions for thinking implanted in us along with our existence by our author in such a way that their use would agree exactly with the laws of nature along which experience runs (a kind of preformation system of pure reason). (CPR, B167) Having raised the possibility of a pre-established harmony between human concepts and the objects humans need to cognize, Kant’s quickly dismisses it on the ground that, again, the applicability of the categories would not be necessary, but contingent. For Kant, the problem to be addressed by pre-established harmony is how human concepts can be in agreement with objects so that cognition

Lichtenberg’s ‘Es denkt’  37 and thought are possible. Oddly, his insistence that the categories are necessary for thought in general is somewhat undermined by an earlier concession in a transitional section of the B Deduction, § 21. Again, Kant is looking back at what he has shown when he remarks: But for the peculiarity of our understanding, that it is able to bring about the unity of apperception a priori only by means of the categories and only through precisely this kind and number of them, a further ground may be offered just as little as one can be offered for why we have precisely these and no other functions for judgment or for why space and time are the sole forms of our possible intuition. (CPR, B145–6) Here he seems to be conceding that although he can show that the categories are necessary for human unity of apperception, and so for human thinking, he cannot show that they are necessary for any thinking whatsoever. When we recall that the unity of apperception (and so human thinking) is also conditioned on the combination of the manifold of intuition by the categories, it is clear that the possibility of human thought depends on the ‘peculiar’ constitution of the human mind (it combines via the categories) and on the suitability of the materials of sense to be combined by a mind with that constitution. Lichtenberg considers this issue of pre-established harmony—how human sensibility and concepts fit together in a way that permits thought—in several Waste Book passages. One occurs at K64, where he also presents a criticism of the praeter nos to extra nos inference, 12 entries before the ‘es denkt’ aphorism. Later in K64 he writes: Here then arises another question, and I cannot say if it has yet been answered: if bodies are granted objective reality, and properties are attributed to them, then given a countless number of cases, one would be possible in which the object has those properties that we, according to our nature, must attribute to it. This is so not because it has these properties, but because given the countless possible forms of intuition also this agreement would be possible. This would also be a harmonia praestabilita. (WB, K64) Lichtenberg is convinced on Kantian grounds that it makes no sense to ask whether bodies actually exist independently of human sensibility. What he is interested in is whether the properties that the mind must attribute to objects (i.e., necessary categories of the understanding) can enable humans to think. He concludes that this is likely, because there could be such a myriad of forms of intuition that some creatures would have forms that yielded representations in agreement with the categories.

38  Patricia Kitcher Although this sort of probabilistic reasoning would hardly appeal to Kant, there is a sense in which their positions agree. The possibility of human thought requires agreement between the materials of sensibility and the concepts of the understanding. Lichtenberg thinks this happens, because there are sufficient chances that it could happen. Although this is somewhat controversial,6 many scholars take Kant to start with the fact of cognition—the fact that humans have cognition—which implies that human sensibility and understanding have the harmony required for thought. As we see just in subsequent section, Kant maintained that this needed harmony could only be explained by his theory of cognition. Lichtenberg also raised the issue of the needed coordination between sensibility and thinking in an earlier entry, H147: However we imagine representing to ourselves the things outside of us, these representations will and must invariably carry some trace of the subject in them. It seems to me a very unphilosophical idea to regard our soul as merely passive; no, it also contributes something to the objects. Thus there can be no being in the world that recognizes the world as it really is. I would like to call this the affinities of the mental and physical worlds, and I can very well imagine there might be beings for whom the order of the universe would be music to which they dance while heaven plays accompaniment. I cite this entry at some length to give a sense of the Kantian tenor of the discussion. Again, Lichtenberg’s point is that while there is no chance of knowledge of things as they are independent of human forms of cognition, there must still be an affinity between human representations of objects outside us and our concepts, hence an affinity of the physical and the mental. In the A Deduction, Kant uses the same term—“affinity”—to address the question of how the manifold of intuition can be necessarily subject to constant laws. That affinity or harmony between the materials of intuition and the laws of nature can be explained only by his account of the ways in which the materials are combined by the activities of the understanding (CPR, A113). As noted, the setup to the ‘es denkt’ claim is a discussion of representations and the difficulty of drawing a line between those representations that do and those that do not depend on the subject as opposed to the ‘external world.’ Given this problem, Lichtenberg might be suggesting that Descartes or Kant cannot assert ‘I think,’ but only ‘there is thinking.’ The world contains thoughts or thinking, just as ‘es blitzt’ could naturally be read as claiming that the world includes (episodes of) lightning— ‘there is lightning.’ The former is not a trivial claim, because as the ­concern that leads to the hypothesis of pre-established harmony implies, thinking requires special circumstances of agreement between the sensory materials and the concepts that can or must be used in relation to them.

Lichtenberg’s ‘Es denkt’  39 This reading could be seen as complementary to Zöller’s, because it offers a further account of one part of his interpretation, ‘I become conscious of it thinking.’ Still, it seems that the difficulties about the boundary between subject and object raised at the beginning of the entry make it hard to sustain the ‘I’ in this reading. Further, the focus of the aphorism is not on consciousness, but on something more complicated, thinking. Against the background of Lichtenberg’s worries about pre-established harmony, there is a certain poignancy about a world that includes this very special activity, thinking. Further, if the world is one in which thinking occurs, then it seems very natural to make the move that Lichtenberg warns against: There is thinking, because there is a thinker (the ‘I’) and something that is thought about, the ‘external world.’ Since an ‘I’ cannot be differentiated from other objects, however, then despite its grammar, cogito cannot be translated as ‘I think,’ but only as ‘there is thought.’

3.  Kantian Replies to Lichtenberg’s Aphorism Although Zöller highlights Lichtenberg’s concerns with the subject of thought also being an object of thought, he addresses the difficulty of separating out an ‘I’ who is the thinker. This problem could be handled qualitatively or quantitatively. I’s are different because they have different characteristics than the rest of nature or they are different from the rest of nature and each other, because they have some sort of identity conditions that permit their individuation. Zöller argues, in effect, that Kant’s claims for the combining activities of thinking enable him to solve the first problem, but not the second. Two more recent treatments, those of Tyler Burge and Tobias Rosefeldt, aim to show that the Kantian position has the resources to deal with the individuation problem posed by Lichtenberg’s objection. They cast the issue in a somewhat different light, however, by assuming that Lichtenberg’s challenge requires a defense of the necessity of using the first person concept. Drawing in part on John Perry’s 1979 seminal discussion of the ineliminability of the I, Burge (1998) argues that Lichtenberg’s suggested linguistic innovation—use ‘think’ or, more specifically, use ‘reason,’ even though it is not possible to use ‘I’—cannot work. To make a Kantian case against Lichtenberg, Burge presents what is frequently characterized as a ‘transcendental argument.’7 In asserting that there is thought while claiming that it is illegitimate to use the concept of an ‘I,’ Lichtenberg is assuming that it is possible to have a sufficiently adequate grasp of ‘thinking’ or ‘reasoning’ without also having a grasp of the idea of individual or separate thinkers (Burge 1998, 247, 249). But this assumption behind Lichtenberg’s linguistic reform program is mistaken. As Burge argues, having the concept of a reason (or thought) does not just require having standards for judging thinking or reasoning;

40  Patricia Kitcher it also requires understanding that reasons motivate a change in attitude or action, and that requires the concept of an individual thinker: A fully explicit understanding of reason must be capable of marking conceptually the cases in reasoning where evaluating or appraising attitudes or activity under rational norms rationally motivates immediate implementation of the evaluations in shaping the attitudes or activity being evaluated. One can evaluate a system of attitudes (in another person or in the abstract) as unreasonable without its being immediately rational for one to change those particular attitudes . . . To understand reason one must distinguish conceptually from such cases those cases where particular evaluations immediately rationally require being moved to affect the attitudes or activities being evaluated in accord with the evaluation. Burge (1998, 252) In Burge’s view: The first-person concept fills this function. Its association with a thought (‘I think. . .’, ‘I judge. . .’, ‘I infer. . .’) marks, makes explicit, the immediate rational relevance of invocation of reasons to rational application, or implementation, and motivation. Burge (1998, 253) As Perry (1979) argued, knowing that someone is spilling sugar all over the grocery store offers no reason to change one’s behavior—unless one knows that ‘I am that someone.’ Similarly, recognizing a fallacy in a chain of thought would not motivate one to give up the inconsistent belief unless one understands that ‘I am the one holding all the beliefs.’ Burge’s point is not that having the I-concept is required in order to give up belief A upon recognizing the truth of not-A. That mental move is immediate. His point is, rather, that no one could have the concept of ‘thought’ or ‘reason’ who did not also understand that reasons can immediately change the attitudes of the person who has them—but not those of anyone else. As Burge notes, there is a sense in which he is agreeing with Lichtenberg that ‘I’ is required only because of practical contexts, only where there is a question of motivation (Burge 1998, 257). Still, I think Burge’s transcendental argument reveals that Lichtenberg did not think adequately about some of the demands of using the expression ‘thought’ or ‘reason.’ On the other hand, a Lichtenbergian might find Burge’s argument frustrating. Lichtenberg raises a problem about separating subjects of thought from the world. Burge does not solve that problem, but argues that humans cannot use the concept of thinking and reasoning that they have unless they can also use ‘I think’ and ‘I reason.’ They seem to be arguing passed each other.

Lichtenberg’s ‘Es denkt’  41 Tobias Rosefeldt defends Kantian self-consciousness (as he understands it) against a Lichtenbergian challenge offered by Quassim Cassam: Kant’s position lacks the resources to disarm a Lichtenbergian objection . . . the objection is that by rejecting the idea that transcendental apperception involves being able to think of one’s representations as belonging to a subject who is an object among others in the world, one is left with no clear sense in which transcendental apperception is a genuine form of self-consciousness, rather than consciousness of subjectless mental episodes Cassam (1997, 165, cited by Rosefeldt 2000, 208). Cassam elaborates on the difficulty: It ultimately makes no sense to suppose that self-consciousness, properly so-called, can be anything other than consciousness of oneself as a concrete individual subject. Since neither self-consciousness itself nor the form of ‘I think’ instances is, in the relevant sense, an individual subject, it is an illusion to suppose that Kant has the resources to reply to [Lichtenberg]. Cassam (1997, 165) This objection should seem familiar. It is an abbreviated version of Zöller’s point that since transcendental apperception is not empirical apperception (which could be tied to a body moving through the world), but only a form of thought, then it cannot form the basis of a reply to the problem of differentiation. Rosefeldt argues that Kantian self-consciousness can be defended against the ‘discrimination challenge,’ by employing conceptual rather than intuitive means. Although this might seem to be the same move that Zöller made, a move that enables the subject to differentiate herself qualitatively from all non-thinking objects, but not numerically from other subjects, Rosefeldt argues that the contents of the I-concept are different in just this respect (Rosefeldt 2000, 210). Mastery of the I-concept requires the ability to apply such logical rules as She who affirms the truth of the judgment ‘I have the thought that P and I have the thought that Q’ thereby also affirms the truth of the judgment, “I have the thought that P and the thought that Q.” Rosefeldt (2000, 210, my translation) With this ability, humans can differentiate themselves from all other things and also know that they are the object that they are differentiating from all others. Rosefeldt takes Kant to have a more refined account of concepts than of consciousness and so appeals to the Kantian idea of

42  Patricia Kitcher concepts as governed by rules to account for self-consciousness (Rosefeldt 2000, 212). Humans can be self-conscious by having the ability to apply the I-concept and applying it to themselves. Although I agree with Rosefeldt that this is an adequate Kantian reply to the Lichtenbergian objection raised by Cassam, I also agree with him that he is offering a reconstruction of a Kantian argument for the possibility of self-consciousness rather than the argument that Kant gave for a single self (Rosefeldt 2000, 212). I think that Kant’s argument is better, because it provides some understanding of a question left open by Rosefeldt’s and Burge’s accounts. How does a human being know that she falls under the I-concept and so that the rules governing it apply to her? Kant’s argument begins with the advance noted by Zöller: the recognition that cognition requires consciousness of thinking, specifically consciousness of combining (CPR, B134). Zöller thinks that this does not suffice to solve the differentiation problem, but if we examine several details of Kant’s argument for the necessity of the ‘I-think’ representation in the B Deduction, then we can see that he has resources for distinguishing I’s, at least in principle. In §15, Kant explicates the concept of ‘combination.’ The concept of combination also carries with it the concept of the unity of the manifold. Combination is the representation of the synthetic unity of the manifold. The representation of this unity cannot, therefore, arise from the combination; rather, by being added to the representation of the manifold, it first makes the concept of combination possible. This unity, which precedes all concepts of combination a priori, is not the former category of unity (§ 10); for all categories are grounded on logical functions in judgments, but in these combination, thus the unity of given concepts, is already thought. The category therefore already presupposes combination. We must therefore seek this unity (as qualitative, § 12) someplace higher, namely in that which itself contains the ground of the unity of different concepts in judgments, and hence of the possibility of the understanding, even in its logical use. (CPR, B131) In the crucial §16, we learn that the requisite ‘higher unity’ is the unity of apperception or self-consciousness. We also learn that the unity or identity of different representations in a single apperception “contains a synthesis of representations” and is possible only through “the consciousness of this synthesis” (CPR, B133). Kant then explains how the reference of different representations to an identical subject comes about: Only because I  can combine a manifold of given representation in one consciousness that is it possible for me to represent the identity

Lichtenberg’s ‘Es denkt’  43 itself of the consciousness in these representations . . . The thought that these representations given in intuition all together belong to me means, accordingly, the same as that I  unite them in one selfconsciousness, or at least can unite them therein. (CPR, B133–4) Although Kant’s focus in §16 is on intuitions, it will be easier to see his point if we start with the case he highlights in §15, that of judgments. Suppose I judge that bodies are heavy. On Kant’s account, I must be conscious in combining the representation ‘bodies’ with the representation ‘heavy.’ ‘Conscious combination’ involves a number of features. First, I must be conscious that the rules governing the use of the two concepts permit their combination in this sort of judgment. So, for example, except as a joke I cannot combine ‘bodies’ and ‘angry’ in a judgment, ‘bodies are angry.’ Second, I can predicate ‘heavy’ of ‘bodies’ only if I have reason to do so. Because ‘bodies are heavy’ is not analytic, my reason must involve some sort of sensory evidence, either firsthand or through testimony. Again, this is a matter of the conscious application of ‘heavy’ to ‘bodies,’ of following the rules for using the concept. Although it would be possible to combine ‘bodies’ and ‘green’ in the judgment ‘(all) bodies are green,’ knowing the rules for ‘(all) bodies’ and ‘green,’ I cannot do so except, again, as a joke. As Kant makes clear in §19, even though ‘bodies are heavy’ is an empirical judgment, it is not for that reason either subjective or contingent. Anyone who knows the concepts, has the relevant evidence, and chooses to opine on the subject, must judge that bodies are heavy. That is why Kant claims that these two representations are combined “in the object” and why the judgment has “objective validity” (CPR, B142). Kant also claims that these two concepts belong to each other (in the judgment) “by virtue of the necessary unity of apperception in the synthesis of intuitions” (CPR, B142). Here, it might be tempting to say that what Kant means by the ‘necessary unity of apperception’ is simply ‘the unity of the act of combining representations according to rules.’ Since the rules governing concepts are special cases of the categorial principles that would imply that the ‘necessary unity of apperception’ is equivalent to ‘the combination of representations according to rules associated with the categories.’ There are, however, two difficulties with this reading. First, it would make Kant’s thesis that use of the categories is required for the unity of apperception a tautology.8 Second, it gives an unsatisfactory reply to the question he is trying to answer at B133, viz., by virtue of what am I able to assert the identity of the apperception across different representations? The answer could only be: by uniting representations according to categorial rules, because what it is for different representations to be united in a single apperception is for them to be united according to categorial rules.

44  Patricia Kitcher I think that we can see how Kant actually answers his question by considering a third feature of ‘conscious combination.’ In judging ‘bodies are heavy,’ I am conscious not only in my use of the concepts but also in the act of combining them in the judgment. Because the contents of these representations are governed by certain rules, I am able to combine the representations in a unified thought, ‘bodies are heavy.’ And through consciously combining the representations, ‘bodies’ and ‘heavy,’ in the judgment ‘bodies are heavy,’ I am conscious of the judgment as dependent upon these representations and so as necessarily belonging together with them. Alternatively, in forging the connection between the representations ‘bodies’ and ‘heavy’ in the thought, ‘bodies are heavy,’ I have Cartesian certainty that these representations must belong together. Kant would add—“in the same subject.” I  think it is legitimate for him to do so, but we need some explicit scaffolding to see why. A  subject (of thought) is a thinker, a thing that produces thought. In consciously combining representations, a subject understands what thinking is, that she is thinking and so that she falls under the concept of a subject of thought, the I-concept. Because she produces the judgment ‘bodies are heavy,’ she is conscious of that thought as authored by her, as belonging to her; since she is conscious of the representations combined in the thought as necessarily belonging together with it, she is conscious of them as all belonging to a single subject, herself. She is also conscious of the judgment as a combination. The unity of the combination is made possible by two factors, the rules that permit the combination and the act of combining them which produces both a unified thought and a unified subject, a subject who is conscious of diverse representations as necessarily belonging together to her. The former unity would vary by the type of concepts (and so categories) involved; the latter would be constant across all thought. It is the higher unity that makes (all) judgments possible. So far, we have considered just concepts and judgments, but it is the synthetic unity of the manifold of intuitions that Kant maintains is the ground of the unity of apperception (see CPR, B134 cited earlier, p. 33). Judgment is easier to deal with, because it is clear that in making a judgment, a subject is conscious of combining different concepts. No such activity is apparent in intuition, but Kant’s discussion of the concept of ‘combination as such’ in §15 helps us to see what he has in mind. To make matters more perspicuous, consider an example of spatiotemporal perception, for example, a dog running through a park. That representation is complex in two respects. It has different spatial parts, the representations of the trees, the path and the dog, and it has different temporal parts, the representations of the beginning of the run, the middle portions, and the end. Still, most people would deny that they combine their representations of the trees and the dog, not to mention their representations of different stretches of him running; they assume that they just take in the information through their senses.

Lichtenberg’s ‘Es denkt’  45 Kant can allow this common sense view, because he is trying to highlight something different. His claim is that insofar as a cognizer understands a visual scene as complex, she must take herself to have a representation of the entire scene and also distinct representations of its parts. Since we are dealing with intuition, none of these elements need to be brought under concepts. Nevertheless, Kant maintains that insofar as the intuition is represented as complex, then it is possible only through the understanding (CPR, B130). He does not make a contrast with animals here, but it would be in keeping with his usual way of arguing to note that although animals have representations that humans regard as complex, the animals are incapable of representing visual scenes as complex (cf. FS, 2: 60). If we grant Kant the plausible enthymematic premise that humans represent all actual intuitions (i.e., p ­ erceptions) as combinations of elements, then he can conclude that they must take perceptions to be dependent on their representational e­lements and so as necessarily belonging together with them to a common subject of perception. Here the unity of the perceiver— the n ­ ecessary togetherness of her perceptual representations—is not ­something that she is ­conscious in producing, but only something that she understands, i­nsofar as she takes her perceptions to be complex. Still, that representation of c­ ombination comes from her active faculty of understanding (CPR, B130). As we have seen, Kant concludes the discussion of conscious combination in §16 with the claim that, if a subject thinks of representations given in intuition as one and all belonging to her, then that is “tantamount” to thinking that she unites them or at least can unite them in one selfconsciousness, even if she is not conscious of acts of combination (CPR, B134). But which is it? Does a representation belong to a subject because she does combine it with others in one self-consciousness or because she can do so? Kant explains that although the thought that I could unite different representations in one self-consciousness is not the consciousness of the synthesis of representations, it still presupposes the possibility of the latter, i.e., only because I  can comprehend their manifold in a consciousness do I call them all together my representations. (CPR, B134) Any use of a concept must be in a judgment, and so be part of a combination, and any (human) perception must be understood as a combination. So any possible representation must have an ‘I think,’ a unified thinker, attached to it. But now we seem in danger of the very situation that Kant is trying to avoid in the passage, that I would have a self as “manycolored and varied” as my representations (judgments and perceptions) (CPR, B134).

46  Patricia Kitcher Kant precludes that possibility by answering the question that Zöller thinks he cannot answer: How do subjects of thought differentiate themselves quantitatively? Kant highlights the conscious combination or consciousness of combination that is involved in thinking or perceiving, because that is what enables cognizers to understand what thinking is (CPR, A346/B404), what the necessary and indubitable unity of thought and of the thinker are. With the nature of thought and its unity laid out, he can use his explication to characterize the range of representations that belong to the self: I call all representations mine that I could combine with a present representation in a unified thought, because if they were so combined, I would have Cartesian certainty that the representations must belong together to a common subject. Kant’s own ‘reply’ to Lichtenberg is more satisfactory than either Burge’s ‘transcendental’ or Rosefeldt’s ‘Kantian’ arguments, because it explains in some detail how subjects understand thinking and how they are able to understand themselves as thinkers. They understand themselves as thinkers, because their representation of thinking comes from their own consciousness in thinking (see CPR, A346–7/B404–5). For purposes of Kantian interpretation, it also shows the ways in which the unity of apperception and categorial principles work together to make cognition possible; in particular, it explains what combination by categorial rules is necessary for. Although it only answers the question of the range of representations belonging to the self in principle, that seems to me to be good enough. Once it is clear that a representation belongs to a single self iff it can be combined with a current representation in a unified thought, then we can consider sources of evidence for identifying such representations. In a well-known passage, Kant remarks that during life there is no problem with identifying thinkers, because it can be done through the body (CPR, B415). The representations that have been stored in the memory of a human body would be understood as representations that could be combined in cognitions and so as belonging to the same I-think as a present cognition. Appealing to the human body to identify the representations that belong to a single apperception does not imply that Cassam is right that without a body interacting with other bodies, Kant has no clear sense in which transcendental apperception involves self-consciousness as opposed to consciousness of subjectless mental episodes. The unity of apperception is the relation that representations have when they are combined in a thought, understood as so combined and so as necessarily belonging together to a single subject. Bodily continuity provides evidence about which representations might be available to be combined with a current cognition, but it is not what the unity of a thinker consists in. Kant is careful to avoid this possible

Lichtenberg’s ‘Es denkt’  47 misinterpretation in §16. He notes the relation between the I  think and the manifold of intuitions: The I think must be able to accompany all my representations . . . Thus all manifold of intuition has a necessary relation to the I think in the same subject in which this manifold is to be encountered. (CPR, B131–2) But he immediately rejects the implication that different representations belong to a common subject simply by being taken in by the same sensory system. The manifold representations that are given in a certain intuition would not all together be my representations if they did not all together belong to a self-consciousness. (CPR, B132) That is, the representations found in a single sensory system are not thereby the representations of a single cognizer; to have that status they must belong to a single self-consciousness, which is possible only through conscious combination or consciousness of combination. Although I believe that Kant has more resources for answering Lichtenberg’s objection than Zöller supposes, I agree with him that the crucial difference between their views is that Lichtenberg lacks Kant’s sophisticated account of the requirements for cognition. In particular he lacks the distinction between apperception and inner sense. When Kant pauses in the B Deduction to draw attention to this distinction, he tries to characterize the peculiar situation of apperception: the synthetic or combining act of understanding is an act of which it [the understanding] is conscious as such even without sensibility. (CPR, B153) Six pages later, when explaining the impossibility of a cognitive subject cognizing itself as some kind of substance through intuition, he notes that the subject does have some consciousness of herself: I exist as an intelligence that is merely conscious of its faculty for combination. (CPR, B158–9) Because the self can be differentiated as the source of the combination of representations that are understood as combinations or unities, it does

48  Patricia Kitcher not matter for Kant whether some of the materials for thinking come from the external world or not (see A98). Whatever their original source, representations belong to the self, are part of the self, insofar as they can be combined with others in the unity of thought. Kant gives an enlightening and adequate ‘reply’ to Lichtenberg or, putting the matter in the proper historical order, Lichtenberg’s objection to ‘I think’ is misdirected if aimed at Kant’s ‘I think.’ The existence of thoughts requires not just a harmony between concepts and the materials of intuition, but also a conscious combination of concepts in judgments that produces both the indissoluble unity of thoughts and the indissoluble togetherness of diverse representations in a single self-consciousness. For that reason, Lichtenberg’s suggestion that we reject ‘I think’ and just maintain ‘thought happens’ cannot be made to work.

Notes 1. All citations from the Waste Books are from Lichtenberg (1764–99) and will be indicated in the text by WB and the standard letter and number designators. Because of the pandemic I have had to rely on Tester’s English edition of selections from the Waste Books, since the original was not available to me online. 2. Here I follow Tester rather than Zöller in translating the second clause. Here and below I translate “es blitzt” as “it lightnings” rather than “it lightens.” 3. See Tester (2013, 339). 4. Rosefeldt (2000, 211) also connects Lichtenberg’s concern to Wolff on differentiation. 5. I refer to Wolff (1781/1983) as he did, by the abbreviated book title and the section number. I add the page number in the Olms edition. 6. Carl (1989, 9) argues that Kant does not presuppose empirical cognition. 7. The literature on this argument form is enormous. For a recent discussion, see the Introduction to Smith and Sullivan (2011). 8. The significance of this issue was highlighted for me by Martin Muñoz’s 2020 undergraduate thesis, Apperception, ‘I Think,’ and their Relationship in the Transcendental Deduction of the Critique of Pure Reason.

References Burge, Tyler (1998). “Reason and the First Person.” In Crispin Wright, Barry C. Smith and Cynthia Macdonald (eds.), Knowing Our Own Minds. Oxford: Clarendon Press, 243–70. Carl, Wolfgang (1989). “Kant’s First Drafts of the Deduction of the Categories.” In Eckart Förster (ed.), Kant’s Transcendental Deductions. Stanford: Stanford University Press, 3–20. Cassam, Quassim (1997). Self and World. Oxford: Oxford University Press. Descartes, René (1640/1984–91). The Philosophical Writings of Descartes. Trans. and ed. by John Cottingham, Robert Stoothoff, Dugald Murdoch, and Anthony Kenny. 3 vols. Cambridge: Cambridge University Press. Leibniz, Gottfried Wilhelm (1885). Monadology (1714). In Carl I. Gerhardt (ed.), Die philosophischen Schriften von Leibniz. Vol. VI. Berlin: Weidmann, 607–23. Reprint, Hildesheim: Olms (1978).

Lichtenberg’s ‘Es denkt’  49 Lichtenberg, Georg Christoph (1764–99/ 2012). Selections from the Waste Books. In Philosophical Writings. Trans. and ed. by Steven Tester. Albany: SUNY Press, 27–182. Muñoz, Martin (2020). Apperception, ‘I Think,’ and Their Relationship in the Transcendental Deduction of the Critique of Pure Reason. Thesis, Columbia College. Perry, John (1979). “The Problem of the Essential Indexical.” Noûs 13, 3–21. Rosefeldt, Tobias (2000). Das logische Ich. Berlin: Philo Verlagsgesellschaft. Smith, Joel and Peter Sullivan (eds.) (2011). Transcendental Philosophy and Naturalism. Oxford: Oxford University Press. Tester, Steven (2013). “G. C. Lichtenberg on Self-Consciousness and Personal Identity.” Archiv für Geschichte der Philosophie 95.3, 336–59. Wolff, Christian (1751/1983). Vernünfftige Gedancken von Gott, der Welt und der Seele des Menschen, auch allen Dingen überhaupt. 4. Reprint. Hildesheim: Georg Olms Vertrag. Zöller, Günter (1992). “Lichtenberg and Kant on the Subject of Thinking.” Journal of the History of Philosophy 30.3, 417–41.

3 Modal Concepts in Kant’s Transcendental Discourse Claude Piché

In his 2017 Possibiliser l’expérience, Günter Zöller examines, with the stunning perspicacity and knowledgeability to which we have grown accustomed, Kant’s relation to modern empiricism. Although Kant has often been critical of empiricism and, more specifically, of its skeptical consequences, he nevertheless did justice to some of its intuitions by retaining in his own philosophy a ‘material element’ that, together with the ‘formal element,’ contributes to the possibility of experience. Kant may thus be said to take rigorously into account the empirical component, the a posteriori element of cognition, even though the originality of the Critique of Pure Reason consists first and foremost in bringing to light the a priori conditions that inform experience as such. As Zöller rightly points out, however, Kant mobilizes these a priori elements exclusively for their contribution to experience. They are interesting only in relation to “possible experience,” to how experience is “made possible” (Zöller 2017, 102–3).1 Accordingly, Zöller stresses this dimension of “possibilization” that impacts the status of experience’s empirical as well as transcendental conditions. Concerning the latter, Zöller writes that “it is not just that experience is made possible by the recourse to the transcendental functions of the a priori. The a priori itself receives its validation from its function of making experience possible” (Zöller 2017, 105–6). This last remark has far-reaching consequences. It tells us that the epistemic status of the a priori in transcendental discourse is dependent on its function in the possibilization of experience. In order to get a clearer view of the epistemic status of the conditions of experience, I  would like to focus on the concept of possibility used in the expression “possibility of experience.” It is, of course, a modal concept, and modality is precisely the angle that I  will take to tackle Kant’s transcendental discourse. The concepts of modality at work in his discourse are highly indicative of this discourse’s nature and especially of the limits within which it operates. My goal, in a nutshell, is to address the modal status of the conditions of the possibility of experience. For DOI: 10.4324/9781003044376-4

Modal Concepts in Kant  51 although ‘possibility’ is a modal concept, we know that to Kant it is, properly speaking, a category. In other words, the modal concepts belong to the Kantian table of the pure notions of the understanding that lies at the basis of his transcendental logic. Consequently, these concepts, bearers of a long tradition, have been the object of a critical reassessment by Kant, something Jessica Leech points out in her review of Nicholas Stang’s book on Kant’s theory of modality. This modal theory, she argues, is part and parcel of the critical revolution, one of its main components.2 Having said that, my aim does not dovetail the work of Jessica Leech, Nicholas Stang, or Giuseppe Motta,3 to name but a few. While they have been interested in Kant’s modal theory for its own sake, I instead want to think through the role of the modal categories as we see them at work in Kant’s own transcendental discourse. Which raises the following question: how could the modal categories, which the Postulates of Empirical Thought tells us are centered on objects and events of experience in their relation to the knowing subject, have meaning relative to a philosophical discourse that precisely rises over these objects and events and defines the rules of their application? In the context of transcendental discourse—which must be considered, we might recall, as real “cognition” (CPR, A11–2/B25)—are these modal concepts condemned to a merely logical meaning and to be regulated exclusively by the principle of contradiction? In other words: are they threatened to remain, for all practical purposes, mere rules of discourse that have nothing to do with its content, that is, with the description of the conditions for the possibility of experience? These legitimate questions would call for extensive and detailed answers. For the sake of brevity, I  will proceed by way of a concrete demonstration and argue that the modal categories that structure Kant’s discourse can be heard in their full meaning only when they exceed their merely logical signification. However, in exceeding the logical sphere, they must nevertheless continue to operate under certain rules. Those rules, as far as I can see, will be those that are assigned to them when they apply to experience itself. In other words, I  intend to defend the thesis that the modal concepts at work in Kant’s transcendental discourse have a cognitive meaning only if they comply, mutatis mutandis, with the rules that this discourse assigns them in their application to experience.4 I however insist here on the clause mutatis mutandis, that is, the idea that “the necessary changes must be made.” Far be it from me to bring transcendental philosophy down to the empirical level of experience. On the contrary, what must be shown is that Kant’s philosophical discourse—and the modal concepts therein—is removed from experience yet never loses touch with it (as is the case with dogmatic metaphysics). I thus propose to examine the role played by four modal concepts in the

52  Claude Piché Critique of Pure Reason: possibility, necessity, contingency, and existence. This will be achieved by way of the following points: 1. Experience in its possibility 2. The conditions of experience from the standpoint of their necessity, but also of their contingency 3. The conditions of experience from the point of view of their existence

1.  Experience in Its Possibility First, then, let us look at the concept of possibility. In order to learn more about it, we must turn to the Postulates of Empirical Thought in General where the modal categories have their natural site. In this chapter, we find a remark that will be of particular interest to us. Asking whether one can envisage other possible worlds beside our own (CPR, A232/B285), Kant mentions in the course of his discussion “absolute possibility.” But although the theme certainly has its place in that context, as soon as absolute possibility is evoked Kant chooses to postpone its discussion given that the ‘absolute’ falls within the competence of reason, and not that of understanding. Later, at the beginning of the Transcendental Dialectic, more specifically in a passage where he stresses the need to protect the meaning of the adjective ‘absolute’ against all abuses of language, Kant makes good on his word and comes back to it. There, in a halfveiled critical remark, Kant blames Baumgarten for having trivialized the word ‘absolute’ by reducing its meaning to the mere “intrinsic” possibility of a concept (CPR, A324–6/B381–2).5 Thus understood, ‘absolute possibility’ would mean nothing more than the absence of contradiction among a concept’s essential properties—nothing more, that is, than conceptual logical consistency. According to Kant, however, this internal possibility is the “least” that can be said of the concept of a thing, and this definition therefore misses the real purport of an absolute possibility. As we can see, Kant tries to safeguard the original stronger meaning of the concept, much as he had done a few pages earlier with the Platonic ‘idea.’ For in truth, absolute possibility is the ‘most’ that can be said of an object. An object is absolutely possible when its possibility is valid “in every relation,” without any “restriction” (CPR, A324–6/B381–2). How should we understand this? In order to help the reader better grasp what he means, Kant draws a parallel in the same chapter of the Dialectic with the concept of ‘absolute necessity.’ That expression is probably more familiar to the reader, as it is involved in the proofs of God’s existence, notably in the so-called cosmological argument based on the ens necessarium: God is that being, necessary in itself, which maintains itself in existence without owing to anyone or anything. Arguably, the notion is indispensable for speculative reason. Yet, despite its greater notoriety, ‘absolute necessity’ nonetheless poses

Modal Concepts in Kant  53 a serious challenge to human intelligence, a challenge similar to the one posed by absolute possibility. Kant tells us that unconditioned necessity, which is synonymous with absolute necessity, represents a “true abyss” for a finite reason: The unconditioned necessity, which we need so indispensably as the ultimate sustainer of all things, is for human reason the true abyss. Even eternity—however awful the sublimity with which a Haller might portray it—does not make such a dizzying impression on the mind. (CPR, A613/B641) Somewhat anticipating the Dialectic, Kant explains in the Postulates of Empirical Thought the reason why unconditioned or absolute necessity causes a problem for the understanding when he tells of the only manner in which the human understanding can grasp necessity. Understanding, he tells us, can only grasp necessity as “hypothetical,” viz. as the necessity implied in an ‘if, then’ proposition. Necessity ties the ground and its consequence: if the cause occurs, then the effect necessarily follows. Thus, the only necessity that is accessible to a finite understanding is a “conditioned” necessity (CPR, A228/B280). We attribute an effect to a cause according to a rule. And it is on the basis of this rule, Kant claims, that necessity becomes “intelligible” to us. The same can also be said of the intelligibility of possibility. Absolute possibility escapes human understanding because it is valid in every respect, “in all relations,” whereas possibility at a human scale is essentially subjected to “restrictions” (CPR, A326/B382),6 Kant notes. It is relative, not absolute. This allows him to claim, not in the Critique but in Reflection 4005, that “through reason we can only cognize conditioned possibility” (Refl, 17: 382), as was the case with necessity. Possibility can be understood only in certain respects, under certain conditions. In another Reflection, Kant goes so far as to argue that ‘absolute possibility’ is equivalent to existence (Refl 4297, 17: 499). What is possible without restrictions would be, by the same token, effectively real. This thesis, at first surprising, loses some of its strangeness when we recall the famous remark added to §76 of the third Critique where we learn that, to an intuitive understanding, possibility and existence are one and the same (CPJ, 5: 401–2). What can we conclude from all this for our project of elucidating the nature of transcendental discourse with the help of modal concepts? The result could be formulated in the following way: the possibility of experience, if it can be understood, will be accessible to critical philosophy only if it abides by the broad prescriptions found in the Postulates and sees it as dependent upon conditions. Whence, as Kant puts it, the expression “conditions of the possibility of experience” (CPR, A158/B197, emphasis

54  Claude Piché added). The entirety of transcendental discourse indeed revolves around this possibility and aims to bring these conditions to the fore. Having established that experience is possible only conditionally, not absolutely, we can now continue our questioning and ask: what of the modal status of these conditions of possibility? Our preliminary task, however, consists in identifying the conditions to a possible experience. As it happens, they are presented in a clear and systematic manner in the summary enumerating the three Postulates of Empirical Thought in General. Pay special attention to the three occurrences of the word ‘condition’: 1. Whatever agrees with the formal conditions of experience (in accordance with intuition and concepts) is possible. 2. That which is coherent with the material conditions of experience (of sensation) is actual. 3. That whose connection with the actual is determined in accordance with general conditions of experience is (exists) necessarily (CPR, A218/B265–6). The first Postulate states the formal “conditions” of experience, that is, the requirement that the object be in accordance with the pure forms of intuition and submitted to the pure concepts of the understanding. The second Postulate addresses the material “conditions,” which alone may anchor the object in reality. Indeed, “sensation” or “perception” is the ultimate guarantee of the object’s existence (CPR, A225/B273). These material conditions are therefore entirely empirical. Regarding the third Postulate, it manifestly constitutes the synthesis of the first two and indicates how to proceed in order to introduce necessity in experience. We would do well here to insist on what distinguishes the first two Postulates. The first deals with the pure elements of cognition and, ­ accordingly, puts forward conditions that are entirely a priori. The second deals with the actuality of the object, which can only be ascertained empirically, a posteriori, through sensation. Obviously, these two sets of conditions together make experience ‘possible.’ But it must be noted that the formal conditions bear more specifically on the object in its mere possibility, and not as it actually exists. In this regard, ‘possible experience’ obtains a more specific meaning. For even though the formal a priori conditions eventually contribute to actual experience, their specific role is confined to the determination of experience in view of its mere possibility, not its actuality. To be clear: the formal conditions are those we find in the Transcendental Aesthetic and the Transcendental Analytic; they are the pure forms of time and space and the pure concepts of the understanding. They are transcendental conditions of experience and their role is strictly a priori in those sections. As Kant states in the Metaphysical Foundations of Natural Science, a priori knowledge pertains to no more

Modal Concepts in Kant  55 than the possibility of its object, not its actuality: “to cognize something a priori means to cognize it from its mere possibility” (MF, 4: 470). That is why the transcendental principles established in the first half of the Critique of Pure Reason only have “possible” experience as their correlate. Meanwhile, this correlate proves entirely sufficient, seeing that it is able to warrant the objective validity of these a priori principles, both those of the Transcendental Aesthetic and of the Transcendental Analytic.

2. The Conditions of Experience from the Standpoint of Their Necessity, but Also of Their Contingency With these considerations in mind, we may now come back to the question of the formal conditions’ modal status, leaving aside for now the material conditions. For practical reasons, I  will consider the formal intuitive conditions (space–time) and the formal conceptual conditions (categories) as they are united in the principles of the understanding. We will proceed in two steps since we can say in advance that the formal conditions are (a) necessary but that, from another point of view, some of them are also (b) contingent. (a) The Critique establishes very clearly that the conditions of the possibility of experience in general are “necessary.” To be sure, they are not necessary in and of themselves, but only relative to experience. The formal conditions mentioned in the first Postulate are conditions sine qua non for experience. However, the meaning of “necessity” here is merely neither logical nor absolute (in the real sense). We should instead understand it as heeding Kant’s warning that the human mind can only understand ‘necessity’ when it is ‘hypothetical,’ viz. ‘conditioned.’ This means that these formal conditions are necessary only relative to their aim: the possibility of experience. Certainly, the best example that illustrates this particular intelligibility of the modal category of necessity is the principle of causality, which posits the necessity of the cause only as it relates to the effect. In the context of Kant’s philosophical discourse, the a priori conditions, necessary for the possibility of experience, of course do not operate at the level of experience itself as ‘physical’ causes, so to speak. In fact, the transcendental conditions are not, properly speaking, causes at all, although Kant unhesitantly considers them more generally as grounds (Grund: CPR, A94/B126, A95, A237/B296). This leads me to think that the very discourse of the transcendental critique complies mutatis mutandis with the Postulates’ instructions concerning the use of the modal categories within experience. If the only necessity accessible to finite intelligence is hypothetical, the transcendental discourse must consequently, in order to depict the modal status of the a priori conditions of experience, refer to an analogous form of necessity. Thus, the necessity of the a priori formal conditions cannot be but hypothetical.

56  Claude Piché But there is still more to say about the necessity of these conditions. That is why I want to look more closely at the modal status of the principles gathered in the Table of the Principles we find in the Transcendental Analytic. It ought to be recalled that Kant introduces a division in this table: on the one side, we have the so-called mathematical principles, viz., the Axioms of Intuition and the Anticipations of Perception; on the other, we have the dynamic principles, namely, the Analogies of Experience and the Postulates of Empirical Thought in General. Well, it turns out that this division has a significant impact on the kind of necessity that attaches to these different principles as formal conditions of experience. Let us first consider the mathematical principles. Their role as conditions of possibility is achieved, Kant claims, with “apodictic” necessity. That is because these principles readily and unconditionally apply to experience, dealing exclusively with the pure intuition of the object, which is an a priori form of intuition that is entirely at the disposal of the transcendental subject. That is to say that we always already know that the object of experience will have an extensive magnitude in time and space, and that its perception will be manifest to a certain degree, that is, given intensive magnitude. Even before the object presents itself to sensibility, we thus know that its different magnitudes will be measurable. As Kant says: the mathematical principles are intended to ground a priori the legitimacy of the application of mathematics to appearances. But how can we be apodictically certain of the immediate applicability of such principles? The answer, in short, is that the mathematical synthesis that unites the homogenous manifold of time (arithmetic) and of space (geometry) is the “same synthesis” (CPR, B203)7 that is at work in the apprehension of the manifold in a concrete perception. This simultaneously means that we know in advance, and with apodictic certainty, that all the results of arithmetic and of Euclidean geometry—even those of infinitesimal calculus, if one is to follow Hermann Cohen8—will automatically be valid for objects of experience. There is therefore something ‘intuitive’ to the evidence of these two principles given that it is the form of intuition that comes into play here. That is why these principles, whose necessary application is obvious, are said to be constitutive of the object of experience, even before it presents itself. The dynamic principles, for their part and the Analogies of experience in particular, are declared to be ‘regulative,’ which affects their modal status. They are defined as simply regulative because they relate to existence, which is always already beyond the control of the transcendental subject. Indeed, we have seen that the empirical conditions of experience pertain to sensation and perception, which in turn belong to the other stratum of conditions: the material conditions of possibility. We can thus understand why the dynamic principles are regulative: they are intended to simply regulate our perceptions, establish order among them. But

Modal Concepts in Kant  57 for a dynamic sequence of experience to be established, these principles depend on the impressions that give access to the object in its actuality. (b) It is at that point that Kant, in the presentation of his Table of the Principles, makes a surprising statement about the dynamic principles: “The a priori conditions . . . of the existence of the objects of a possible empirical intuition are themselves only contingent” (CPR, A160/B199, emphasis added). In other words, these a priori conditions, by definition said to be necessary and universal, are simultaneously declared to be ‘contingent’ in their application. As a result, a new modal category appears that, as we know, is the exact counterpart of necessity. If the mathematical principles intervene outright and necessarily in the constitution of the object of experience, the implementation of the dynamic principles is however contingent. This is somewhat surprising given that the dynamic principle of causality, which lies at the heart of the Table of the Principles, is thus given an inferior modal status. Yet, we know full well the strategic importance of this principle for Kant, who uses it as an example more often than any other throughout the first Critique. We also know how central it was, together with the two other Analogies, to Kant’s thinking during the elaboration of the first Critique. If we refer, for example, to Paul Guyer’s research on the Duisburg Manuscript, Kant added the mathematical principles to his table only at a later stage of the preparation of his magnum opus.9 The contingency of the dynamic principles therefore does not seem to detract from their central function in Kant’s overall project. We consequently get the following result: the dynamic principles are, without a doubt, conditions sine qua non, that is, necessary, for the possibility of experience, but their implementation is contingent. What, then, is the meaning of the modal concept of contingency? The logical definition of the word may serve as a useful starting point. Logically, contingency is defined as follows: “that whose contradictory opposite is possible” (CPR, A459/B487). There is no contradiction in thinking that a contingent thing that exists might as well not be. While this definition is relevant as a point of departure, it nonetheless leaves the reader unsatisfied: what is the point of saying that something that is may also not be, without entailing a contradiction? Thankfully, Kant sheds some light on the way he conceives the ‘real’ use of this concept, not in the chapter on the Postulates but, rather, in the General Note on the system of principles that comes immediately thereafter. There, we learn that it is impossible to go beyond the logical meaning of the concept of contingency without mobilizing the categories of relation, namely, that of cause and effect. Otherwise, the modal concept of contingency remains a mystery. Contingency therefore only becomes comprehensible if we relate it to the concept of change. Drawing from Kant’s example—with slight modifications—we can imagine the following case: a body is in movement at t1 and comes to rest at t2, which amounts to say that a change has taken place, viz., an alternation in the body’s accidents. But this is not yet

58  Claude Piché contingency. For contingency to obtain, the following also needs to be true: the same body could have continued to be in movement at t2, that is, in a state opposite to its actual state (CPR, B290n, see B48). And this is where causality comes into play, because the understanding cannot be satisfied by the acknowledgment of the mere logical possibility of the body’s movement at t2 in place of its actual rest. The body’s rest, which is a contingent state in which it finds itself at t2, calls for a causal explanation, to such an extent that contingency may be said to be in solidarity with causality, an association that provides the former with a meaning that goes beyond the merely logical. Causality is, in truth, a prerequisite for contingency that saves it from unintelligibility. Going back to our example, the contingency of the body’s new state triggers the search for the cause of this state, a cause that could explain to what extent ‘rest,’ as a contingent accident of the body, is in fact the necessary effect of the cause. Think, for instance, of a stone in free fall that suddenly comes to rest after encountering an obstacle such as the ground, for example. This makes the collusion that exists between contingency and causality obvious in Kant’s eyes: “one cognizes contingency from the fact that something can exist only as the effect of a cause” (CPR, B291). The takeaway from all this is that contingency, as a modal category, only gains its intelligibility when related to causality. We know that the stone at t2 could still have been in movement because we now understand why it is not in that state anymore: it met an obstacle. Contingency is, generally speaking, understandable only if it is related to a condition: “Everything that is contingent is possible only in a conditioned manner” (Kant, V-Met/Dohna, 28: 647). It is true that Kant sometimes mentions, in his notes and lectures, a contingency without conditions, an ‘absolute contingency,’ just as he did with absolute possibility and absolute necessity. But even in those instances he claims that the very concept is so ambitious (ein sehr grosser Begriff) that it escapes intelligibility (Kant, V-Met/Schön, 28: 499).10 These remarks on the critical concept of contingency having been made, we may now come back to Kant’s claim that the dynamic principles, in their application to experience, are merely contingent. We have indeed seen that contingency is part of a causal context that makes its real meaning intelligible only as ‘conditioned contingency.’ Now, if Kant is to be consistent, there must be a reference to some kind of “condition” when he writes of the contingent implementation of the dynamic principles. As it happens, we find such a reference in the following lines to his presentation of the Table of the Principles. The dynamic principles certainly are necessary for experience; however, as they are dependent on a condition, they are simultaneously declared contingent since they depend for their application upon a condition: [T]he principles of the dynamic use, to be sure, also carry with them the character of an a priori necessity, but only under the condition

Modal Concepts in Kant  59 of empirical thinking in an experience, thus only mediately and indirectly. (CPR, A160/B199–200)11 Kant could not be more explicit. The implementation of these principles is ‘conditioned,’ as opposed to the mathematical principles whose evidence is intuitive and immediate, that is, not submitted to any restriction in their application. Geometry, for example, simply anticipates the form of the object. The Analogies’ intervention, on the other hand, is ‘mediate and indirect.’ In this case—and this is an important point to stress— the condition to which the principles are subjected is obviously that of “empirical thinking in an experience,” that is, the dependence upon a perception’s occurrence as the ultimate index of actuality. The dynamical principles, elaborated entirely a priori for the sake of the possibility of experience, are conditioned and hence contingent. They are, in essence, related to the material conditions of experience discussed in the second Postulate. There is an interaction between the two strata of conditions insofar as both cooperate in making experience possible. It is rather clear that the two strata of conditioning do not operate separately in order to produce experience; they are not parallel conditions. But how, then, should we define their interaction? In Kant’s lectures on metaphysics, we find remarks on §§314 and 315 of Baumgarten’s Metaphysica that might indicate the way to a possible solution. There, Kant discusses the traditional theme of concaussae or, in German, Mitursachen, which we could translate by ‘contributory causes’ (Kant, V-Met/Dohna, 28: 648).12 Such causes can be either subordinated or coordinated. Kant illustrates subordinated causes with the example of the father, who is the cause of the son, but who is himself the effect of the grandfather. As we can see, these causes form a series and therefore cannot be of help to us. Given that the formal and material conditions together contribute to produce the effect, it rather looks like we are dealing with coordinated conditions, much like in Kant’s other example where the father and the mother both produce the child. In any event, I would hesitate to use the concept of an “occasional cause” (CPR, A88/ B118) to describe how perception triggers the application of the dynamical principles. The concept of ‘occasional cause’ is employed by Kant in his discussion of Locke’s physiological deduction, hence my hesitation. The context of the dynamical principles’ application is completely different, if only because the material conditions, far from simply providing the occasion for the intervention of the Analogies, are in and of themselves a constitutive part of experience, providing the object’s content. So far we have found, first, that the possibility of experience cannot be an absolute possibility but, rather, a conditioned one that is accessible to finite reason; second, that the conditions in question are indeed necessary but hypothetically so, that is to say: necessary relative to possible experience; third, that the implementation of some of these transcendental

60  Claude Piché conditions, namely, the dynamical ones—the most important ones—is contingent, that is, dependent upon material conditions. We can now press our inquiry further and move up the chain of transcendental conditions in order to explore their respective modal status. For the principles contained in Kant’s table, as we already noted, are the junction points where the pure sensible data and the pure concepts of the understanding meet. This means that the principles combine heterogeneous elements stemming from different sources, elements that remain independent despite the fact that they interact. The forms of sensibility, as we know, fall within the jurisdiction of the Transcendental Aesthetic, and the categories within that of the Analytic, whose focal point is transcendental apperception. Kant may then characterize these two ultimate sources of cognition as “principles” in the sense that transcendental apperception, for example, is the principle of the principles of the understanding: in virtue of its originality, it achieves the function of synthesis for each of the pure concepts of the understanding. For similar reasons, the pure forms of intuition can also be seen as “principles” (CPR, A22/B36, B40, A26/ B42). Our task now is to inquire into the status of these general principles, especially given that Kant calls them ‘supreme.’ The supreme principle of the possibility of all intuition in relation to sensibility was, according to the Transcendental Aesthetic, that all the manifold of sensibility stands under the formal conditions of space and time. The supreme principle of all intuition in relation to the understanding is that all the manifold of intuition stands under conditions of the original synthetic unity of apperception. (CPR, B136, emphasis added) What does ‘supreme’ mean when it is applied to these two ultimate principles of human knowledge? Should that lead us to consider them as absolute? Given that apperception is characterized elsewhere as the “highest point” (CPR, B134) of the conceptual conditions of experience, does that mean that we have reached a condition that is not only necessary but also unconditioned? We could also ask a similar question about time and space taken as ultimate principles. But before we go any further, the mere mention of the categories and the pure forms of intuition invites us to address a preliminary question concerning the constitutive elements of these two supreme principles. Could Kant acknowledge a certain kind of contingency here despite the a priori nature of the 12 categories and space and time? Let us recall Kant’s question about the precise number of categories and the two forms specific to sensibility. On that occasion, he makes no secret of their contingency, which can easily be inferred from the following excerpt of the Deduction: But for the peculiarity of our understanding, that it is able to bring about the unity of apperception a priori only by means of the

Modal Concepts in Kant  61 categories and only through precisely this kind and number of them, a further ground may be offered just as little as one can be offered for why we have precisely these and no other functions for judgment or for why space and time are the sole forms of our possible intuition. (CPR, B145–6) Obviously, the concept of contingency is not mentioned here by name, but we easily detect its presence. There is an unmistakable hint in that excerpt when Kant claims that no “ground” may be found for the pure factuality of these a priori conditions. For even if one is unable to give a reason for this state of affairs, the very search for an explanation is quite telling. There is contingency here because things could very well have been otherwise, and without contradiction, concerning the pure concepts of the understanding and the pure forms of sensibility. That is why the normal reaction is to ask ‘why’ and to seek a justification. The contingency we are facing here is specific to a finite understanding, including that of the critical philosopher, who only has access to a conditioned contingency, though in this case the search for its condition proves an impossible task. Kant therefore cannot fully account for the specific number and origin of these a priori elements of knowledge that ground experience. He in fact faces this problem as soon as he finds himself forced to juxtapose sensibility and understanding even though he cannot identify their common origin. The question has since become a classic: is there a “common root” to these two powers of knowledge? This question, which has had long and rich posterity (think, e.g., of Fichte or Heidegger13), is raised in the Critique’s Introduction where Kant is quick to dismiss it, claiming that if such a common root exists between sensibility and understanding, it is and must remain “unknown to us” (CPR, A15/B29). That answer could be read as somewhat dismissive, leading one to suspect that Kant, on the threshold of the gigantic task of presenting the results of his ‘decade of silence,’ may have chosen to ignore the problem. However the reason that explains his haste to dismiss the question is not his refusal to inquire into their common root, but the very nature of the problem itself. If Kant forgoes the goal of identifying their common origin, it is simply because such an inquiry goes beyond the scope of human intelligence. On this topic, the Anthropology is clearer and more explicit than the Critique: “[This common root is] inconceivable to us, who cannot understand that the dissimilar could stem from one and the same root.” (Anth, 7: 177) The contingency of the juxtaposition of sensibility and the understanding is due to their insurmountable heterogeneity. Kant may say in the Critique that the hypothetical root is “unknown to us,” but this passage of the Anthropology is more precise as to the reason of its unintelligibility. And it is this fundamental lack of comprehensibility that puts an end to the search. Nevertheless, the very will to find an explanation is a clear sign that we are faced with a conditioned contingency. The

62  Claude Piché human mind cannot be satisfied with blind luck, with the mere casus, viz. with brute unconditioned contingency. And this brings us back to the two ultimate a priori principles of experience, namely, pure ‘space/time’ and transcendental apperception. As a priori principles that ground all possible experiences, these are said to be ‘supreme.’ The German word used by Kant here is “oberst.” To the discerning reader, this adjective is reminiscent of the conceptual triad introduced by Kant in the second Critique to expose his conception of the highest good: Höchste, Oberste, and Vollendete. Oberste, which denotes one of the components of the highest good, is precisely the translation for the Latin supremum. We must however be careful not to carry the etymological analysis too far here, as that would lead us astray. Kant uses this concept in the context of his practical philosophy where the access to a real unconditioned is not only possible, but an undeniable ‘fact.’ The moral law, which is the supreme condition for the highest good, is described as a condition that is, properly speaking, “unconditioned” in the sense that it is genuinely original, originarium says Kant; it “is subordinated to no other” (CPracR, 5: 110). We will do well to keep this in mind: “subordinated to no other.” It stresses the radicality of the unconditioned when it is taken in the full sense of the word. As Cassirer remarked in his review of Heidegger’s Kantbuch, the practical unconditioned marks a departure from the phenomenal world, a breakthrough toward the “supersensible,” the only realm where the unconditioned has its place.14 It must be remarked, however, that never do we find in the Transcendental Analytic a description of the ultimate conditions for the possibility of experience as being unconditioned. Were it the case, Kant would be in flagrant contradiction with the teachings of his own examination of the excesses of dogmatic reason found in the Transcendental Dialectic. One could say that he was immunized from the outset against that temptation, having so vigorously sought to prevent it. Furthermore: Kant knows all too well that the wanderings of dialectic reason stem from the delusional attempt to reach the “unconditioned.” This modal concept is in fact the common denominator to the second part of the Transcendental Logic. Consequently, Kant’s recourse to the adjective “supreme” to characterize the two ultimate pure principles of knowledge in the Analytic can only be understood in the following way: the Critique undoubtedly succeeds in tracking down the conditions of the possibility of experience in their last entrenchments, but these remain ultimate conditions only for the possibility of experience. Their necessity rests entirely on their function as foundations to possible experience. Moreover, their status as ultimate is only true for a finite understanding. “Supreme principle” certainly means the highest point that philosophical intelligence can reach in the chain of conditions, but nothing forbids us from thinking that these conditions could have a further ground that remains inaccessible to our finite

Modal Concepts in Kant  63 understanding. There is a relevant passage on this topic in a remark Kant makes in his Lectures on Metaphysics (Volckmann). It deals with the idea of a cause that, though first in a series, cannot strictly speaking be said to be unconditioned. He gives the example of the myth of Adam, who may be the first individual in the chain of human generations, but nevertheless isn’t an entirely independent being. Notice how Kant uses the adjective “oberster,” viz., supreme, to designate the first link of the chain: One can call something the supreme ground [oberster Grund], ratio prima, secundum quid, for example the first man was the first ground of a series of men, namely through God. [This man] was not therefore [first ground] simpliciter talis or Independens [i.e.,] what does not contain the consequence of something else. Seen from the standpoint of everything in general, [this first man] was not the ground [of the series] without being at the same time a consequence. (28: 410) This leads us to conclude that the two supreme principles (the forms of sensibility and transcendental apperception) are, without a doubt, the highest conditions transcendental discourse can reach. But even as ultimate conditions sine qua non, they themselves remain principles secundum quid, that is: first principles, but only with respect to their principiata. Having said this of the transcendental conditions to a possible experience, that is, of the a priori conditions, what must we say of the empirical conditions discussed in the second Postulate? Those are somewhat overshadowed in the Analytic by the exposition of the pure conditions of cognition. Yet, this is where Kant may be at his most vulnerable, most notably when he holds that the affection of sensibility ultimately comes, not from the appearances, but from the thing in itself. Appearance being in truth a mere representation, one may be authorized at first to attribute to it—as the “indeterminate object of an empirical intuition” (CPR, A20/B34)—the affection of sensibility. But if Kant is to take seriously the dimension of passivity inherent to sensibility, he is compelled to recognize that sensible affection requires its own ‘dynamic,’ a relation to something else outside representation. Indeed, the appearance, which is nothing more than a representation, cannot of its own affect receptivity. The appearance is nothing but the manifestation of this ‘something else,’ the thing in itself.15 But since Kant did not pay as much attention to the material conditions as he did to the formal conditions, he very rapidly drew criticism. Recall, for instance, the reaction of Jacobi, who saw the thing in itself as a foreign body in the critical system, and came to the conclusion that if Kant’s philosophy is to be consistent, that is, unburdened by this incongruous thing, it will show itself to be an outrageous form of idealism.16 We may also consider Aenesidemus-Schulze, who

64  Claude Piché thought illegitimate Kant’s recourse to the category of causality to depict the relation of the thing in itself to sensibility since it runs counter to his explicit prescriptions concerning the use of the categories.17 Yet the most damaging reproach one might address Kant is that he may have proved unable to scrupulously comply with the teachings of the Transcendental Dialectic and allowed himself, albeit just in this case, to derive the material conditions of experience from the unconditioned. The suspicion that such is the case is raised by the clear reference to the thing in itself, which looks like an inexcusable leap into the supersensible.

3. The Conditions of Experience from the Point of View of Their Existence Let us therefore briefly consider the question of the thing in itself. I will not try to justify the use of the category of causality castigated by Aenesidemus-Schulze. I will rather attempt to draw a parallel between transcendental apperception, as the ultimate condition for a priori synthesis, and the thing in itself. In other words, I  mean to compare the transcendental subject and the thing in itself, which Kant also sometimes calls the ‘transcendental object,’ to show how they are conceived in a symmetrical manner, as though they mirror one another. This seems especially true and helpful with regard to the two following questions: (a) When dealing with the I think and the thing in itself, does Kant truly leap into the realm of the unconditioned in the strong sense of the word—as is the case with the dialectical ideas? (b) Is his recourse to the modal category of existence, to designate the ‘I think’ and the thing in itself, legitimate? (a) Regarding the first question, we must prevent the misunderstanding that might arise in the reader’s mind when Kant explicitly uses the concept “unconditioned” to describe transcendental apperception, not in the Transcendental Deduction, where the word does not appear, but in the first edition of the Paralogisms. We must keep in mind that the description he provides there does not pertain to the dialectic illusion, but to the way he retrospectively considers the ‘I think’ and its role in cognition. In this instance, he does not hesitate to employ the words “absolute” (CPR, A356) and “unconditioned” to qualify the transcendental subject; these modal qualifications in fact define the unity of the ‘I think.’ The following passage dispels any possible misunderstanding: Because, further, the only condition accompanying all thinking is the I, in the universal proposition “I think”, reason has to do with this condition insofar as it is itself unconditioned. But it is only the formal condition, namely the logical unity of every thought, in which I abstract from every object. (CPR, A398, emphasis added)

Modal Concepts in Kant  65 Clearly, the ‘I think’ is defined here as an absolute and unconditioned condition. But this absolute unity is restricted to the status of the ‘I think’ as a condition to every thought and, by extension, to all cognitions. As I have already said, neither terms were explicitly used in the Transcendental Deduction, yet the absolute and the unconditioned were nonetheless implicitly present insofar as the unconditioned unity discussed in this passage does not have an ontological bearing on the ‘I think,’ but only a logical one. Philosophical thought cannot give up the temptation to go back to this ultimate focal point, but its only value is logical. It expresses a pure identity. As for giving in to the further temptation to consider the unconditioned ‘I think’ as an object, and to determine it moreover with the help of the predicates of unity, simplicity, and substance, it goes without saying that this is out of the question for Kant. Once again, the apperception as a ‘supreme principle’ of all thought and cognition is supreme only secundum quid. Also in the Paralogisms, just a few pages earlier in the chapter, Kant manifests his intention not to give the thing in itself the status of “unconditioned” or, at least, not to speak of a modal status that is beyond the reach of human knowledge. If the thing in itself proves necessary, it is only as an inescapable condition for the possibility of experience. As for its nature as a cause, nothing can be known of it. Only its effect matters. “For whoever he may be, he knows just as little as I or anyone else about the absolute and inner cause of external and corporeal appearances” (CPR, A394). Clearly, we know nothing of the thing in itself when it comes to its status as an absolute cause or its intrinsic nature. The reader has already been warned in the Transcendental Aesthetic: its essence remains entirely “problematic” (CPR, A38/B55). Its existence, however, and its action on sensibility are certain to Kant’s eyes. Never would he consider the thing in itself as a mere object of thought. Undoubtedly its inner nature or essence is ‘problematic,’ but not its relation to possible experience, to which it contributes as the ultimate material condition. (b) This brings us to the second question: how can the modal concept of existence be legitimately applied to the thing in itself? It is a fact that Kant often asserts the reality (Wirklichkeit) of the thing in itself—in the Critique and in the Prolegomena for instance.18 But does that not contradict the rules contained in the second Postulate regarding the use of the category of existence? Once again, the parallel with the status of the ‘I think’ might help us resolve the issue. For if we know that Kant forbids in the Paralogisms any ‘real’ use of the categories of simplicity, unity, and substance to describe the essence of the ‘I think,’ we also know that he does not hesitate to mobilize the concept of existence and to characterize the ‘I think’ as an “existential proposition” [Existentialsatz]. (CPR, B423) It is certainly true that nothing is known of the essence of the ‘I think’ as an object, but its existence is attested by the very act of thinking. ‘I think’ and ‘I am’ are, for Kant, identical propositions. Kant further tells

66  Claude Piché us that this act of thinking cannot but leave a trace in the internal sense (as opposed to what we find in Hume, for instance). The ‘I think’ does not provide access to a manifold that would make possible its determination as an object. It is, on the contrary, a simple representation, a pure identity. Nevertheless, Kant says that the ‘I think’ manifests itself in sensibility by way of a perception, albeit a totally indeterminate one. It is this sensation or perception that warrants Kant’s application of the concept of existence to the ‘I think.’ Its application is however clearly exceptional as it does not take place within experience but rather relates to the necessary conditions for the possibility of this experience. [The proposition ‘I think’] expresses an indeterminate empirical intuition, i.e., a perception (hence it proves that sensation, which consequently belongs to sensibility, grounds this existential proposition), but it precedes the experience that is to determine the object of perception through the category in regard to time; and here existence is not yet a category, which is not related to an indeterminately given object, but rather to an object of which one has a concept, and about which one wants to know whether or not it is posited outside this concept. An indeterminate perception here signifies only something real, which is given. (CPR, B422–3, emphasis added) Here, existence is characterized as a pure concept of the understanding, and not as a category, in which case it would have to be schematized. The ‘I think’ therefore truly leaves a mark in inner sense, but it is not itself an object that takes place within experience. Rather, it makes experience possible. Now, it must be pointed out that Kant does not thereby derogate from the requirements of the second Postulate: sensation remains, as it was with experience, the inescapable and ultimate warrant of existence. In that respect, its mobilization has nothing in common with a transcendental use of the concept, as is the case with the ontological argument, for example, where the judgment of existence does not draw on perception. We may therefore conclude that Kant adheres to the rule of the second Postulate, albeit on a different register. One could possibly object that the parallel I  draw here between the discourse on the existence of the ‘I think’ and that on the existence of the thing in itself suffers from a serious imbalance. On the one hand, the certainty of the existence of the ‘I think’ is immediate and undeniable. This certainty accompanies, if only implicitly, every thought and cognition. The case is entirely different with the thing in itself. Nothing is more alien to the knowing subject than it. It is something intangible that Kant is forced to posit, but whose presence remains somewhat ghostly. To this line of criticism, one could reply as follows: Would it not be at the very least possible to apply the concept of existence to the thing in itself in the same specific way Kant applies it to the ‘I think,’ that is to

Modal Concepts in Kant  67 say, according to the same minimalist rules? We know that Kant meets the criterion of the second Postulate according to which perception is required to attest existence even when the category is not schematized in an experience. May we not therefore argue that the mere affection of the outer sense, on which the transcendental subject exerts no control, viz., the simple occurrence of a sensation indicating an existence through the appearance, could be transposed in another register, namely, that of the discourse on the conditions of the possibility of experience, and thus bear witness to a non-phenomenal reality underlying the appearances? Such was likely Kant’s view, even if he did not trouble himself to show that the recourse to the non-schematized concept of existence has nothing to do with a dialectical application of the concept that would aim at determining and cognizing the unconditioned. The thing in itself is not known and consequently not determined in a way that would let us know if it is conditioned or unconditioned, contingent or necessary. However, for the purpose of the possibility of experience, it must be posited assertorically as the ultimate ground of the material conditions.19 The results of our investigation are thus the following. Their merely logical sense aside, the concepts of possibility, necessity, and contingency acquire real meaning only when they are conditioned. They are tailored to the measures of a finite understanding, including that of the transcendental philosopher. And as we have just seen, the same holds for the modal concept of existence, which is also related to a condition: sensation, or perception.20 This reminds us of the importance of the material conditions for the Kantian discourse, even though they are not at the forefront of the system’s exposition. As we have seen, they have an impact on the formal conditions of experience, most notably on the dynamic principles. Their implementation is contingent because they are dependent upon perception, which shows once more the crucial role empirical conditioning plays in the possibility of experience in general. As a result, these material conditions are revealed to be the fully fledged counterpart of the transcendental principles and, in consequence, obtain the status of principle. The material conditions speak to the pure, irreducible otherness that the ‘thing in itself’ is meant to attest. Of course, we will recall that Kant, in the first edition of the Paralogisms, evokes—not without polemical intent—the possibility that the transcendental object affecting outer sense could be the same thing that is at the origin of thoughts and volitions (cf. CPR, A358). This is evidently a mere conjecture: the claim is unverifiable and therefore belongs to the realm of logical possibilities. It is assuredly conceivable, but given what we have seen, we must come to the conclusion that Kant could never have seen it as an acceptable option. That is because a transcendental subject that could also affect external sense would be, in a certain way, analogous to the intuitive understanding that autonomously produces the manifold of intuition (cf. CPR, B138–9). However, a truly dynamic relation requires that the affection of the knowing subject must be construed

68  Claude Piché as an external contingent input. It is well known that Kant is not drawn to monism. Like Crusius, whose lessons he had learned, Kant is a philosopher that advocates the plurality of principles. Moreover, very early in his career, Kant had given his preference to real influence over ideal influence. He is deeply convinced of the fecundity of a plurality of principles that condition themselves mutually. This is true, for instance, of the relations between the pure forms of intuition and the pure concepts of the understanding that together open the possibility of a priori synthesis. And it has proven equally if not more true when the reciprocal relations are dynamic in a higher sense, as with the formal conditions interacting with the material ones to make for the possibility of experience.

Notes 1. Cf. Zöller (2017, 102–3). I would like to thank Olivier Mathieu for his precious advice, philosophical as well as linguistic, in the final preparation of this text. 2. Leech (2017, 341). 3. Leech (2012); Leech (2014); Stang (2016); Motta (2012). 4. On this topic, see Piché (1995, 259–67). 5. All translations of the Critique of Pure Reason are taken from the Cambridge edition, the translations of other texts are mine; A. G. Baumgarten, Metaphysica (41757), §15, 17:29; Kant, V-Met/Volckmann, 28: 406–407. 6. See also V-Met/Schön, 28: 488. 7. See also A165–6, A167–8/B206, A224/B271. 8. See Cohen (1918, 544–5). 9. Guyer (1987, 27, 33, 62–7). 10. See also Randbemerkungen of the Preisschrift, 20: 329; Refl 3717, 17: 260. 11. The same pattern according to which a condition for the possibility of experience is itself conditioned applies to time and space in the Transcendental Aesthetic. These are obviously necessary “conditions” for the apprehension of an object, but only as long as this object is submitted to such restrictions: space, just like time, therefore entails a limitation (Einschränkung, CPR, A27/B43) of the realm of objects that may be cognized, limiting those to the sphere of appearances. The pure forms of sensibility are thus conditions for the possibility of objective knowledge, but only insofar as the objects are submitted to the subjective form of sensibility. See the following passage concerning time where the word “condition” bears both meanings (emphasis added): “Its [i.e., of time] empirical reality therefore remains a condition for all our experiences. . . . It is nothing except the form of our inner intuition. If one removes the special condition of our sensibility from it, then the concept of time also disappears, and it does not adhere to the objects themselves, rather merely to the subject that intuits them.” (CPR, A37/B54. See also A35/B51–2, emphasis added, and for “space,” A27/B43)   Time, then, conditions the form of the object (as an appearance), but this simultaneously implies that “our” mode of access to the object is itself conditioned. 12. See V-Met/L2/Pölitz, 28: 572; V-Met/Mron, 29: 844. On this subject matter, see Ameriks (2003, 155n).

Modal Concepts in Kant  69 13. See Fichte’s Wissenschaftslehre 1804 (1971, 104); Heidegger (1973, 35, 132, 155, 196). 14. Cassirer (1931, 13–14). 15. See the following passages in which Kant asserts that the thing in itself “affects” the knowing subject: CPR, A190/B225, A387, A494/B522. See also Prol, 4: 289, 315. 16. Jacobi (2004, 112). 17. Schulze (1996, 183–4). 18. CPR, A190, A387, A494; Prol, 4: 289, 314, 318. 19. See Piché (2004). 20. Since we have seen that Kant at times considers the three categories of possibility, necessity, and contingency in their absolute meaning, it is legitimate to ask if he mentions something like an absolute existence in the Critique when he writes of the fourth modal category under consideration. It is indeed the case. In the Transcendental Aesthetic chapter he uses the expression “absolute reality” for space and time in order to designate their transcendental reality. Which is to say that space and time would exist beyond the “restriction” (CPR, A27/B43) of their sphere of application and validity to human sensibility that only apprehends appearances. Yet, because absolute reality per definitionem does not submit itself to this condition, nor to any other for that matter, the hypostasis of space and time that would make them “real things” ultimately leads only to non-entities (Unding, CPR, B71).

References Ameriks, Karl (2003). Interpreting Kant’s Critiques. Oxford: Oxford University Press. Cassirer, Ernst (1931). “Kant und das Problem der Metaphysik. Bemerkungen zu M. Heideggers Kant-Interpretation.” Kant-Studien 36, 1–26. Cohen, Hermann (1918). Kants Theorie der Erfahrung. Berlin: Bruno Cassirer. Fichte, Johann G. (1971). Nachgelassenes zur theoretischen Philosophie II, Werke Vol. 10. Ed. by Immanuel Hermann Fichte. Reprint, Berlin: DeGruyter. Guyer, Paul (1987). Kant and the Claims of Knowledge. Cambridge: Cambridge University Press. Heidegger, Martin (1973). Kant und das Problem der Metaphysik. Frankfurt am Main: Klostermann. Jacobi, Friedrich H. (2004). “Beilage: Über den transzendentalen Idealismus.” In Walter Jaeschke (ed.), Werke 2.1. Hamburg: Meiner. Leech, Jessica (2012). “Kant’s Modalities of Judgment.” European Journal of Philosophy 20, 260–84. Leech, Jessica (2014). “Making Modal Distinctions: Kant on the Possible, the Actual, and the Intuitive Understanding.” Kantian Review 19, 339–65. Leech, Jessica (2017). “Review of: Nicholas F. Stang, Kant’s Modal Metaphysics.” Kantian Review 22, 341–6. Motta, Giuseppe (2012). Die Postulate des empirischen Denkens überhaupt. Berlin, Boston: De Gruyter. Piché, Claude (1995). “Self-Referentiality in Kant’s Transcendental Philosophy.” In Hoke Robinson (ed.), Proceedings of the Eighth International Kant Congress, Vol. 1. Milwaukee: Marquette University Press.

70  Claude Piché Piché, Claude (2004). “Kant and the Problem of Affection.” Symposium. Canadian Journal of Continental Philosophy 8, 275–97. Schulze, Gottlob (1996). Aenesidemus. Hamburg: Meiner. Stang, Nicholas F. (2016). Kant’s Modal Metaphysics. Oxford: Oxford University Press. Zöller, Günter (2017). “Possibiliser l’expérience. Kant sur la relation entre le transcendantal et l’empirique.” In Antoine Grandjean (ed.), Kant et les empirismes. Paris: Classiques Garnier, 99–112.

4 Can Practical Reason Be Artificial? Dieter Schönecker

Artificial Intelligence (AI) has given rise to a variety of moral, juridical, economical, and political or for short: practical questions that need to be answered soon—from how to deal with self-driving cars to the possibility of AI leading to the end of the human race due to some kind of technological singularity.1 To name just a recent incident, recall the dispute concerning the Korea Advanced Institute of Science and Technology which was accused in an open letter by 50 academics of participating in a program to develop so-called killer-machines. Since AI is, at least by philosophical standards, a fairly recent phenomenon, both these practical questions and problems as well as the possible answers and solutions to them are fairly new. Note, however, that these answers in turn will depend on foundations that are a far cry from virgin or innocent; in applied ethics and political philosophy, we get very quickly to traditional questions and positions that we need to discuss both on metaethical and normative levels in order to provide sustainable answers. Hence it is no surprise that in papers about moral machines, arguments are put forward concerning good old-fashioned utilitarianism, for instance.2 However, practical questions or the applied ethics of AI are not my concern here. Rather, the question that I shall address belongs essentially to the philosophy of mind: Can practical reason be artificial? Practical reason is best understood, I submit, as a genuine power to cognize and will the good. From a Kantian point of view, the answer to that question is clearly negative: Practical reason cannot be artificial. It is tempting to think that this answer has a foundation already in Kant’s epistemological, or as Kant would put it: theoretical thought that reason is always someone’s reason such that there is no thinking without someone who thinks or always can think I think. I will briefly look into this, but my focus will be on Kant’s practical philosophy. From this practical point of view, too, the conclusion that practical reason cannot be artificial is quick, solid, and inevitable; for practical reason is free and computers are not. However, my approach is different; it is based on the idea that moral reason comes along with moral feelings that computers cannot have. After a preliminary remark on the possibility of Kantian moral DOI: 10.4324/9781003044376-5

72  Dieter Schönecker machines (1.1) and some basics on the concept of practical reason (1.2) and Kant’s intuitionism (1.3), I  will argue that in a Kantian model of moral obligation, the typical (human) moral subject has moral feelings and must have them in order to cognize the validity of the moral law as a categorical imperative (1.3). Using the knowledge argument against physicalism and functionalism, I shall argue that computers have no feelings and, a fortiori, no moral feelings; therefore, computers are no moral subjects (1.4). This conclusion is based on a Kantian I feel rather than I think (1.5). I will then tackle two problems with this argument (2). I will conclude with an analogy (3): Just as planets do not fly, computers do not feel.

1.  The Argument from Moral Feelings It is easy to understand a Kantian argument that, if true, clearly rules out the possibility that computers have practical reason. It is the argument from transcendental practical freedom. Here is a very brief sketch: Moral obligation presupposes transcendental practical freedom of practical reason. Such freedom is, negatively speaking, independence from natural causality or physical determinism. Computers, however, are determined by the laws of physics; therefore, they cannot be free. But practical reason—and thus the human being—is free, and it needs to be free for morality to make sense (Kant is an incompatibilist).3 Therefore, no computer can have practical reason. Note that even on a non-deterministic understanding of physics, and even with regard to quantum computers, this argument from freedom will hold. For freedom is not only independence from natural causes; it is also a positive faculty, namely, the faculty of determining oneself in an autonomous act of absolute spontaneity, and such spontaneity, unlike chance, is not lawless. Thus, one can easily see that at least from a Kantian point of view, it is quite obvious that a computer cannot have practical reason. Since this is so obvious, I would like to address or rather develop an argument from Kant’s practical philosophy that often goes unnoticed; I shall call it the argument from moral feelings. 1.1  A Preliminary Remark: Kantian Moral Machines Alan Turing once listed a number of things that people think computers will never be able to do; among these was the ability to “tell right from wrong.” Of course, it depends on what one means by ‘telling right from wrong,’ but at least with regard to the output of such ‘telling,’ that assumption might very well turn out not to be true; there has been serious talk of “moral machines” for quite some time already,4 and the development of robots raises moral questions that are not only

Can Practical Reason Be Artificial?  73 of theoretical (or philosophical) interest, so to speak, but are being addressed quite practically. Moral algorithms seem possible, and such algorithms may not only help, for instance, judges to make moral decisions, but will soon make, in some very limited sense, moral decisions on their own; just think of so-called autonomous vehicles and trolley cases. It is maybe tempting to assume that such moral machines must be based on some kind of utilitarian reasoning, given the mathematical character (and prima facie easiness) of a utilitarian or hedonistic calculus. Given the formal aspect of Kant’s famous categorical imperative and the idea of universalization, however, this could be a prejudice; a computer might be able to perform a moral algorithm on Kantian grounds as well. Recall the basic idea of the so-called natural law formula: Suppose someone has a maxim, for instance, that she will commit suicide when her life irreversibly brings about or will bring about more suffering than agreeableness. The categorical imperative obligates her to ask herself whether such a maxim could be a universal law such that everyone who experiences more suffering than agreeableness will actually kill himself or herself; then she might realize that this leads to some kind of contradiction. There has been a long and ongoing debate on how to understand the contradiction Kant has in mind; but at least on a somewhat formal (logical) interpretation of the contradiction involved, a Kantian moral machine that runs a test of universalization seems possible.5 1.2  Kant’s Concept of Practical Reason6 ‘Practical reason’ is (pure) good volition: Everything in nature works in accordance with laws. Only a rational being has the faculty to act in accordance with the representation of laws, i.e. in accordance with principles, or a will. Since for the derivation of actions from laws reason is required, the will is nothing other than practical reason. (G, 4: 412).7 It is important, however, to differentiate three aspects of Kant’s concept of practical reason or a good will: The noumenally good will, the practically good will, and the holy will. The noumenally good will is the autonomous will that as such wills the good. As a moral faculty, it gives the moral law (the categorical imperative) for imperfect beings and, by means of moral feelings, it is also a motivating force. Every human being has such a will, even if he or she acts immorally.8 The noumenally good will is the basis both for the practically good will and the holy will. The practically good will is the will that finite beings have when their volition is indeed moral; it is the noumenally good will

74  Dieter Schönecker considered as a will that manifests itself successfully in a finite being against the influence of inclinations and desires. For imperfect beings, to act morally (i.e., to act with a practically good will) means to act from duty. The noumenally good will that is manifest in a person without (active) sensual hindrances is what Kant calls the ‘holy will’; it only belongs to God and other holy beings. These beings have no inclinations and desires contrary to the good; the “will whose maxims necessarily harmonize with the laws of autonomy is a holy, absolutely good will” (G, 4: 439). The noumenally good will as such (regardless of its being incorporated in a finite being) cannot be differentiated from the holy will (regardless of its being incorporated in an infinite being). It is a noumenal causality: “The rational being counts himself as intelligence in the world of understanding, and merely as an efficient cause belonging to this world does it call its causality a will” (G, 4: 453, first emphasis mine). This will is then identified with the will that is autonomous, that is, with autonomy itself: “if we think of ourselves as free, then we transport ourselves as members into the world of understanding and cognize the autonomy of the will, together with its consequence, morality” (G, 4: 453, my emphasis). Note how Kant continues: “. . . but if we think of ourselves as obligated by duty, then we consider ourselves as belonging to the world of sense and yet at the same time to the world of understanding” (G 4: 453, my emphasis). Thus the free will is the noumenal will (pure practical reason), and autonomy is its property; in some contexts, this will is considered not as the will of a human being that is also part of the sensible world, but as a noumenal will only: as “a mere member of the world of understanding, all my actions would be perfectly in accord with the principle of the autonomy of the pure will” (G, 4: 453, my emphasis). It is important to keep in mind that the noumenally good will as such is not only a mere capacity to act morally; for this will as such wills the good. Nonetheless, it is the noumenally good will that enables the human being to act morally; thus, for the human being—who is a member both of the noumenal and the sensible world—the noumenally good will is indeed a capacity. Also, unless autonomy and having a practically good will are not the same, a scoundrel would not be autonomous—which he actually is insofar as even he, to some extent, wants to be morally good, that is, insofar he has a noumenally good will. We will return to this once again. 1.3  Kant’s Moral Intuitionism It is often striking to see how to defenders of strong AI9 it seems obvious that computers “can do many things as well as or better than humans” (Russell & Norvig 2016, 1022, my emphasis). But such an assumption, of course, is begging the question; for the question really is whether they can do anything a human being can do when it comes to acting or thinking. On Kant’s account, there is no ‘doing’ in any narrow sense here.

Can Practical Reason Be Artificial?  75 Human actions, strictly speaking, are not only free actions; if they are guided by the moral law, they are embedded in moral feelings. Computers have no feelings; therefore, they do not act morally even if they in some sense make decisions in accordance with duty. Let us take a closer look at this argument from moral feelings. To the present day, Kant is considered to defend, as Edmund Husserl put it, “an extreme and almost absurd rationalism” (Husserl 1988, 412), an “extreme intellectualism” (ibid.) that leaves no room for feelings. Now every beginner philosopher in an introductory class on Kant’s ethics will learn that Kant consistently argues that reason can only bring about actions by means of feelings; hence feelings necessarily come into play here already as a determining (motivating) ground. However, it is important to realize that on Kant’s account, feelings serve a much more important function. As we have already seen, Kant draws a very strict line between holy and non-holy beings. Whereas holy beings always will what a good will wants, this is not true for non-holy, sensuous-rational beings. For them, the moral law is always a categorical imperative that necessitates them. Let me quote Kant in more detail here: If reason determines the will without exception, then the actions of such a being, which are recognized as objectively necessary, are also subjectively necessary, i. e. the will is a faculty of choosing only that which reason, independently of inclination, recognizes as practically necessary, i. e. as good. But if reason for itself alone does not sufficiently determine the will, if the will is still subject to subjective conditions (to certain incentives) which do not always agree with the objective conditions, in a word, if the will is not in itself fully in accord with reason (as it actually is with human beings), then the actions which are objectively recognized as necessary are subjectively contingent, and the determination of such a will, in accord with objective laws, is necessitation, i. e. the relation of objective laws to a will which is not thoroughly good is represented as the determination of the will of a rational being through grounds of reason to which, however, this will in accordance with its nature is not necessarily obedient. The representation of an objective principle, insofar as it is necessitating for a will, is called a ‘command’ (of reason) and the formula of the command is called an imperative. All imperatives are expressed through an ought and thereby indicate the relation of an objective law of reason to a will which in its subjective constitution is not necessarily determined by that law (a necessitation). (G, 4: 412f) The crucial step is to see that this necessitation is experienced by the feeling of respect (which in turn has negative and positive aspects that I cannot deal with here).10 But this feeling is not just a side effect, as it

76  Dieter Schönecker were. Since by ‘necessitation’ Kant means nothing but the fact that for non-holy, sensuous-rational beings like us, the moral law is an imperative, that is, a duty, the obligation involved in this is experienced in the feeling of respect. As a matter of fact, it is not only somehow experienced, but cognized by this feeling: “What I immediately recognize as a law for me, I cognize with respect” (G, 4: 402n). And it is important to see that Kant’s famous theory of the ‘fact of reason’ is directly related to this thought.11 In § 7 of the Critique of Practical Reason, Kant formulates the categorical imperative; a bit later, he says that one could call the “consciousness of this fundamental law a fact of reason” (CPracR, 5: 31). The so-called factum theory explains our insight into the binding character of the moral law; it is, among other things, a theory of justification. The basic idea is that there can be no deduction of the categorical imperative in any normal (deductive) sense,12 and yet the objective reality of the moral law is “nevertheless firmly established of itself” (CPracR, 5: 47, my emphasis). In our consciousness of the categorical imperative, the moral law is immediately given in its unconditional and binding validity; in this sense, the factum theory is a theory of moral self-evidence. This consciousness of the categorical imperative, however, is determined by the feeling of respect, that is, the unconditional validity of the categorical imperative is given in the feeling of respect. Hence it is through the feeling of respect that we cognize the validity or binding character of the moral law.13 Thus, Kant is by no means the pure rationalist that Husserl and others represent him as being. Rather, Kant is a moral intuitionist. A moral intuitionist is someone who holds the view that we cognize the validity of the moral law, the moral You ought, not by some kind of deductive (or inductive or abductive) reasoning, but by means of a certain kind of self-evidence, by a feeling. It is important not to misunderstand these claims: On Kant’s account, it is not the content of the categorical imperative that is understood by the feeling of respect; what we ought to do or omit we cognize by reason and some kind of universalization. Kant is not a moral sentimentalist. Also, the moral law itself does not depend on the moral feeling of respect for its validity; it is not that the moral law is valid because we have that feeling. Yet what we do cognize by the feeling of respect is that we ought to act morally, that the moral law is categorically binding. 1.4  A Kantian Knowledge Argument From this, however, it follows that a computer can have no practical reason. To see this, we have to take a brief look at the so-called knowledge argument or the argument from an explanatory gap put forward in one version or another by Frank Jackson, Joseph Levine, and Thomas Nagel.14 This is quite an intricate story, and here we can only sketch the main idea. For our purposes, recall Jackson’s thought-experiment concerning

Can Practical Reason Be Artificial?  77 an individual named Mary: Think of her as a scientist who knows everything there is to know about colors and their perception from the point of view of the natural sciences; but Mary, being locked in a room with only black and white books, TVs, etc., has never seen any objects that are not black or white. One day, however, she leaves her room and actually sees something that is, say, red. Now according to physicalism, all that exists are natural (physical) objects that are fully described and explained by physics (and possibly by chemistry, biology, or neuroscience). If this were true, then Mary would not gain any new knowledge of a quality—it being only a non-representational quality of her perception or of a sensedatum—she had not already known because everything there is to know about colors from an scientific, objective, third-person perspective, she does know; but there is something she did not know before she left the room, namely, how it is or what it feels like to see something red, to experience a certain quale; therefore, there is something in the world that is not physical, i.e. not fully describable by physics. This something is consciousness making phenomenal experiences. Thus we could know everything there is to know about the physical or functional facts concerning a mental state (such as having a perception of something being red) and still we would not know everything about the mental state; therefore, this mental state cannot be identical with or be reduced to those physical or functional facts. From this it follows, some at least have argued, that physicalism is false. Another implication is this: Although there is some dispute as to which mental states are qualia or are accompanied by qualia, it is obvious that feelings are qualia. But then the argument is apparent: We can know everything a computer is made of and how it works. There is no what it is like to be a computer, and therefore, unlike beings for whom there is a certain phenomenal inner life, to be in a computational state is not to be in a mental state (and vice versa). But unless we believe that computers experience qualia, they cannot have practical reason. For practical reason comes along with practical necessitation through the feeling of respect; the categorical imperative cannot be understood without this feeling; since computers have no feelings, and a fortiori no feeling of respect, they cannot understand the categorical imperative. That is the basic argument. From a Kantian point of view, three more points are important: First and only in passing, I  should note that in his later work (The Metaphysics of Morals), Kant further developed his theory of moral feelings by distinguishing four kinds of moral predispositions and, consequently, four moral feelings: the moral feeling proper, conscience, love of human beings as amor complacentiae, and self-respect.15 With regard to each of these feelings, Kant stresses that there is no obligation to have them; for to have these feelings is already a necessary presupposition to make sense of the very concept of duty in the first place. Second, to Kant, practical reason is the noumenal will that

78  Dieter Schönecker both cognizes and wills the good; it is autonomous and therefore a noumenal causality. As I have indicated already, this is a complicated story, but moral feelings cannot be naturalized as they are brought about by reason which cannot be naturalized; so even if computers did have feelings, they could not have the feeling of respect, because this feeling has its source in reason which is not a natural (physical) entity. Third, Kant, too, understands feelings as qualia. Of course, this term was not used by Kant. But he has a clear understanding of the fact that feelings have a phenomenal side that cannot be grasped by physical knowledge but must be experienced. The phenomenal side of feelings is particularly emphasized by Kant in his theory of beauty.16 Feelings as such, says Kant in the so-called First Introduction to the Critique of the Power of Judgment, “cannot be explained at all”; rather, they “must be felt, not understood [eingesehen]” (20: 232).17 In a similar vein, Kant writes in the Metaphysics of Morals that “pleasure and displeasure cannot be explained for themselves” (MM, 6: 212). In any case, the purity in pure practical reason by no means suggests that in the process of moral self-determination, no feelings are involved. The purity of practical reason consists in its being free from considerations of happiness and self-love; in human beings, it is pure reason that becomes practical on the strength of moral feelings. So even if a computer ‘makes a decision’ (as it were) on the basis of a moral algorithm, it has no idea of what it is ‘doing’ (as it were): it has no understanding whatsoever of what the moral law as a categorical imperative really is. In Kant’s terminology, a computer can perform actions (as it were) according to duty. But it certainly cannot perform actions from duty. And it certainly has no conscience or self-respect; as I  see it, such a claim is not even apprehensible. Before I move on to two problems for this Kantian position, let us have a quick look at a possible further argument, the argument from the power of judgment; given the latitude of many ethical duties, this obviously is an important aspect of practical reason. The argument could run like this: Following Kant, the power of judgment is the “faculty of thinking the particular as contained under the universal.” (CPJ, 5: 179). If there is a rule already, then the faculty of the power of judgment is the “faculty of subsuming” something particular under this rule; Kant calls this the determining power of judgment (bestimmende Urteilskraft). If there is still a rule to be found for something particular that cannot be subsumed under an already acquainted rule, then Kant calls it the reflective power of judgment (reflektierende Urteilskraft). At least for the determining power of judgment, Kant argues, there can be no further rule. For if one wanted to show generally how one ought to subsume under these rules, i. e., distinguish whether something stands under them or not, this could not happen except once again through a rule. But

Can Practical Reason Be Artificial?  79 just because this is a rule, it would demand another instruction for the power of judgment, and so it becomes clear that although the understanding is certainly capable of being instructed and equipped through rules, the power of judgment is a special talent that cannot be taught but only practiced. (CPR, A133/B172) Put another way: There can be meta-rules concerning how and when to apply rules; but on pain of a vicious circle or an infinite number of rules, there must be a point at which the power of judgment takes action without applying a rule. Computers, however, have nothing but rules to work with, that is, nothing but algorithms (and data, of course, in regard to which they are applied). If the power of judgment is a faculty that does not follow rules, then this faculty cannot be something a computer could have. The “lack of the power of judgment,” says Kant, “is that which is properly called stupidity” (CPR, A133/B172); computers, therefore, are stupid. —But it is hard to see whether this argument actually follows. The necessity of something like a power of judgment is due to the fact that there is no complete ascertainment or definition of all possible concepts and cases a priori or in advance.18 But if a decision based on the power of judgment is not based on a rule, on what is it based? It had better not be based on chance; for that is something a computer could do (following the rule to choose randomly). One might think that the power of judgment has to do with something like intuitions; but intuitions the way G. E. Moore, for instance, understands them, are different and have nothing to do with the power of judgment. Intuitions broadly understood as somewhat unconscious, strong, quick seeing (judging)-as-true states of mind (as understood in moral psychology), however, might very well not be random, but based on some (unconscious, strong, quick) weighing of goods, and such a weighing could follow rules. Hence, I would not seek to defend the argument from the power of judgment. 1.5  I Think versus I Feel Obviously, moral subjects will something, and they act on the basis of their volitions. But they also think, and even if it is disputed how thinking, and how much thinking, is actually involved in their moral decisions as such, it cannot be disputed that moral subjects must also think, at least with regard to the cognition of the surrounding world in which they will and act; moral knowledge involves non-moral knowledge about the internal and external world. Now Kant famously argues (in the second edition of the Critique of Pure Reason) that there is no thinking and hence no knowledge of the internal and external world without self-consciousness; the I think, he says, “must be able to accompany all my representations” (CPR, B131). What exactly this means has been the object of painful and

80  Dieter Schönecker long discussions among Kant-scholars;19 the basic idea appears to be that all thinking involves the synthesis of representations as someone’s representations in a judgment such that these representations as well as the act of synthesizing them belong, and must belong,20 to a self-conscious I that can always say I think (these representations). In any case, if Kant is right, and if it is true that a computer has no thinking I, then a computer does not and cannot think, and it is not intelligent the way human beings are. At best, then, a computer (AI) can imitate intelligence, and only at best imitate moral thinking. I’m not quite convinced of this argument. It seems true to me that (developed) human beings always must be able to think I think when they think (not that they always do think I think, of course), and if this is true and if it is true that computers have no I (which I think is true indeed), then computers do not think the way we do. Still it might be justified to say that they think: If theoretical (not practical) thinking is essentially the act of synthesizing content that requires a center or unit by which this act is performed, then one could possibly understand this center or unit as the control unit of a computer. My point is that the act of combining content (synthesis) might neither require Meinigkeit nor self-consciousness (no apperception, as Kant calls it) as a higher form of Meinigkeit, but just a control unit; this unit must not necessarily be aware of itself because the content of (something like) perceptions and (something like) thoughts or propositions does not require an I that thinks it and not even a form of Meinigkeit. Kant seems to have held that “everything that thinks is constituted as the claim of self-consciousness [the I think] asserts of me” (CPR, B405); “I cannot,” he claims, “have the least representation of a thinking being through an external experience, but only through self-consciousness” (B405). In other words, Kant claims that the only way to understand the possible thinking of non-human beings is based on our own self-understanding; if this is true, I cannot understand the possible thinking of a computer without thinking it as the thinking of an I. But why should this be true? This thought might simply be a consequence of the fact that computers were not around yet.21 So maybe there is thinking without a self-conscious I that thinks and also without Meinigkeit. When it comes to feelings, however, we necessarily enter into a different, namely, an inner world. There can be no feeling without someone who feels. Regardless of whether we have to distinguish between feelings in a narrow sense and emotions (as intentional), to say that there is a state which we could preliminary describe as “there is a feeling,” requires that there is an instance for whom it is to be in that state (of feeling something). Again, there might be a possible understanding of cognitive content that does not require someone for whom or whose content it is; but that is incomprehensible for feelings.

Can Practical Reason Be Artificial?  81

2.  Two Problems for the Argument from Moral Feelings The argument from moral feelings is strong, I  submit; still, it has two problems. First, what about holy wills? As we have already seen, it is a very important element of Kant’s ethics to distinguish between holy and non-holy beings. For the latter, the moral law is a categorical imperative and thus duty and obligation; for the former, there are no hindrances for morality to overcome, they have a perfectly good will. So are not holy beings in some sense moral machines? And would it not be true for a computer that always follows moral algorithms that it has such a perfectly good will? Well, no. It is certainly correct that a holy being cannot act from duty.22 But unlike computers, holy beings have a will; computers have no will; a fortiori, they do not will anything for the sake of the moral law. The claim that computers have no will is underpinned by the claim that volitions are intentional; computational states, however, are not intentional states.23 Therefore, computers have no will. Still the question of what it means for a holy will to will and to act for the sake of the moral law without intermediary moral feelings is not easy to answer. The second problem with the argument from moral feelings simply is this: Can computers really have no feelings? This, too, is a long and complicated story, and I can only sketch the problem and the possible solution. So here is the argument that maybe computers can or even do, after all, have feelings: We know that we have consciousness and feelings. At the end of the day, we have no clear, let alone comprehensive and convincing story to tell how this happens, how it really can be that we have such an inner life. However, provided that some kind of naturalistic evolutionary theory is correct, we do know that our ability to have mental events has developed out of unconscious matter. But if it is possible that the mind and its mental states somehow evolve out of matter, that is, out of the brain and its embodiment—and possible it is on the assumption that mental states are real, regardless of whether they can be reduced to brain states or not—then it could very well be possible that the mind and its mental states somehow could evolve out of a computer as yet another complicated assembly of matter. Fair enough, I  would say. So yes, this is possible if it is possible for mind to evolve out of matter. But from all that we know it is also very unlikely: Just a single biological cell already is extremely complex, let alone the brain, and complexity, as far as we know, is a precondition for, as well as a sign of, consciousness and feelings. By comparison, a computer is a very primitive object; there is no more reason to think that it has a mind than to think a sewing machine has one.24

3.  Conclusion: Swimming Submarines, Flying Planets To conclude, let me come back to Kant’s I think. The computer scientist Edsger Dijkstra famously argued that “the question of whether Machines

82  Dieter Schönecker Can Think . . . is about as relevant as the question of whether Submarines Can Swim,”25 or, one might add, whether airplanes can fly. His point, I take it, is that submarines can swim, of course, that is, move through water despite the fact that they do not swim like fish, and that airplanes can fly despite the fact that they move through air without moving wings up and down or whatever. Following Dijkstra’s analogy, it seems sensible to hold that computers think despite the fact that they do not think the way we think; there is no I, and still they think. But this analogy between computers thinking and submarines swimming or airplanes flying is misleading. As always, it all depends on how to define such words as “swimming,” “flying,” or “thinking.” If one defines “swimming” as “moving through water using limbs, fins, or tails,” then submarines do not swim; but why define “swimming” like this? To define a word—or to explicate what it is for a thing to be what it is—somehow presupposes a recognition of what is essential to that thing; for this, however, one needs paradigmatic cases.26 Now one could argue—maybe along the lines of functionalism—that computers on the basis of certain inputs perform certain operations that bring about certain outputs; and taking into account just the output—calculating, playing chess, driving a car, writing music—one is tempted to believe that computers think (and also that human beings think the way computers think). The question of how and by what means swimming is performed is not crucial to the concept of swimming. But the difference between a being that thinks I think, or at least between one that feels I feel, or that experiences I want, and a machine that has no such self-consciousness, is so enormous that those terms (to think, to feel, to want) should not be used for beings that have no I. To say that a computer thinks or feels is like saying that a planet flies just because it moves through space.

Notes 1. This paper is a slightly revised version of a paper published in Journal of AI Humanities, 2018, 2, 67–91.—I would like to thank the organizers of The First International Conference on Artificial Intelligence Humanities held at Chung-Ang University, Seoul on August 16, 2018; my special thanks go to Prof. Chan Kyu Lee and Prof. Hyeongjoo Kim. I would also like to express my gratitude to Larissa Berger, Prof. Dr. Markus Lohrey, Christian Prust, Elke Schmidt, and Dr. Thomas Sukopp for our discussions about AI. 2. Cf., for instance, Bonnefon et al. (2015). 3. Kant is a compatibilist only in that sense that freedom and determinism are compatible on the assumption of the difference between the noumenal and sensible world; but this is not the common assumption used in the debate about compatibilism and incompatibilism; cf. Schönecker (2005). 4. Cf. for instance, Wallach and Allen (2009). 5. Cf. Lindner and Bentzen (2018). 6. Here I draw from Schmidt and Schönecker (2018) and Schönecker and Wood (2015).

Can Practical Reason Be Artificial?  83 7. Cf. G, 4: 427: “The will is thought as a faculty of determining itself to action in accord with the representation of certain laws.” 8. Cf. G, 4: 400, 412, 440, 449, 455. 9. By “strong AI” I  mean for my purposes that a computer could have consciousness, an inner life (qualia), and really think the way we do; such a computer would not just imitate thinking, and would not just imitate moral thinking, but really think and also think morally. 10. For a detailed analysis, cf. Schadow (2012) and Schmidt (2013). 11. Cf. Schönecker (2013a) and (2013b). 12. “Hence the objective reality of the moral law cannot be proved by any deduction” (CPracR, 5: 47). 13. Colin et al. (2000, 260) see that emotions have more than just a motivational function; they do not, however, recognize the cognitive function as regards the validity of the moral law. 14. Cf. the pertinent texts by Jackson (1982), Levine (1983), and Nagel (1974). There are some differences in these authors, but I take it that the bottom line is the same (except maybe when it comes to Levine). 15. Cf. Schönecker (2010). 16. Cf. Berger (2021). 17. In German: “Man sieht hier leicht, daß Lust oder Unlust, weil sie keine Erkenntnisarten sind, für sich selbst gar nicht können erklärt werden, und gefühlt, nicht eingesehen werden wollen.” 18. Cf. CPR, B756: “One makes use of certain marks only as long as they are sufficient for making distinctions; new observations, however, take some away and add some, and therefore the concept never remains within secure boundaries.” 19. Cf. Klemme (1996) and Rosefeldt (2000); one example of how difficult it can be to properly understand Kant’s position is Kim’s book on the I think as an empirical proposition (Kim 2017); for a very brief overview, cf. Kitcher (2015). 20. Friebe (2005, 53) seems to think that such a reading is too strong. 21. There remains, of course, the question of in which way computers have representations. Friebe (2005, 61) is quite right in claiming that representations (Vorstellungen) as such are always someone’s representations; and it might also be true that the property of being my representations (Meinigkeit) does not necessarily imply an I in a strict sense. However, the question remains whether thinking must be understood as an operation that involves representations as something mental. 22. It also cannot have the four moral predispositions. 23. All of this is highly disputed, of course; recall the long and ongoing debate about Searle’s Chinese room argument. For a brief summary and critique cf. Gabriel (2018, 95 ff.). 24. One could also, by the way, turn the tables and argue that the existence of mental states proves that something is wrong with Darwinist evolutionary theories. Thomas Nagel, among others, has done so recently; cf. Nagel (2012). For a critical view of Nagel’s moral realism, cf. Schmidt (2018). 25. Cf. Russell and Norvig (2016, 1021). 26. Cf. Damschen and Schönecker (2013).

References Berger, Larissa (2022). Kants Philosophie des Schönen. Eine kommentarische Interpretation zu den §§ 1–22 der Kritik der Urteilskraft. Freiburg: Alber.

84  Dieter Schönecker Bonnefon, Jean- François, Azim Shariff and Iyad Rahwan (2015). “Autonomous Vehicles Need Experimental Ethics: Are We Ready for Utilitarian Cars?” arXiv:1510.03346 [cs.CY], 12 Oct. 2015. Colin, Allen, Gary Varner and Jason Zinser (2000). “Prolegomena to Any Future Artificial Moral Agent.” Journal of Experimental  & Theoretical Artificial Intelligence 12.3, 251–61. Damschen, Gregor and Dieter Schönecker (2013). Selbst philosophieren. Ein Methodenbuch. Berlin, Boston: De Gruyter. Friebe, Cord (2005). Theorie des Unbewußten: Eine Deutung der Metapsychologie Freuds aus transzendentalphilosophischer Perspektive. Würzburg: Königshausen u. Neumann. Gabriel, Markus (2018). Der Sinn des Denkens. Berlin: Ullstein. Husserl, Edmund (1988). “Kritik der Kantischen Ethik.” In Ulrich Melle (ed.), E. Husserl, Vorlesungen über Ethik und Wertlehre 1908–1914. Dordrecht, Boston: Springer, 402–18 (Husserliana XXXVIII). Jackson, Frank (1982). “Epiphenomenal Qualia.” Philosophical Quarterly 32, 127–36. Kim, Hyeongjoo (2017). Zur Empirizität des “Ich denke” in Kants Kritik der reinen Vernunft. Würzburg: Königshausen & Neumann. Kitcher, Patricia (2015). “Ich denke.” In Marcus Willaschek, Jürgen Stolzenberg, Georg Mohr and Stefano Bacin (eds.), Kant-Lexikon. Vol. 2. Berlin, Boston: De Gruyter, 1074–79. Klemme, Heiner F. (1996). Kants Philosophie des Subjekts. Systematische und entwicklungsgeschichtliche Untersuchungen zum Verhältnis von Selbstbewßtsein und Selbsterkenntnis. Hamburg: Meiner. Levine, Joseph (1983). “Materialism and Qualia: The Explanatory Gap.” Pacific Philosophical Quarterly 64, 354–61. Lindner, Felix and Martin Mose Bentzen (2018). “A  Formalization of Kant’s Second Formulation of the Categorical Imperative.” arXiv preprint, arXiv: 1801.03160. Nagel, Thomas (1974). “What It Is Like to Be a Bat.” Philosophical Review 83, 435–50. Nagel, Thomas (2012). Mind and Cosmos: Why the Materialist Neo-Darwinian Conception of Nature Is Almost Certainly False. Oxford: Oxford University Press. Rosefeldt, Tobias (2000). Das logische Ich. Kant über den Gehalt des Begriffes von sich selbst. Berlin: Philo. Russell, Stuart J. and Peter Norvig (eds.) (2016). Artificial Intelligence. A Modern Approach. Hoboken, NJ: Pearson Education Limited. Schadow, Steffi (2012). Achtung für das Gesetz: Moral und Motivation bei Kant. Berlin, Boston: De Gruyter. Schmidt, Elke E. (2018). “The Dilemma of Moral Naturalism in Nagel’s Mind and Cosmos.” Ethical Perspectives 25.2, 203–31. Schmidt, Elke E. (2013). Kants Begriff der Demütigung in der “Kritik der praktischen Vernunft.” Master-Thesis, University of Siegen. Schmidt, Elke E. and Dieter Schönecker (2018). “Kant’s Ground-Thesis. On Dignity and Value in the Groundwork.” The Journal of Value Inquiry 52.1, 81–95. Schönecker, Dieter (2005). Kants Begriff transzendentaler und praktischer Freiheit. Eine entwicklungsgeschichtliche Studie. With the collaboration of Stefanie Buchenau and Desmond Hogan. Berlin: De Gruyter.

Can Practical Reason Be Artificial?  85 Schönecker, Dieter (2010). “Kant über Menschenliebe als moralische Gemütsanlage.” With the Collaboration of Alexander Cotter, Magdalena Eckes, Sebastian Maly. Archiv für Geschichte der Philosophie 92.2, 133–75. Schönecker, Dieter (2013a). “Das gefühlte Faktum der Vernunft. Skizze einer Interpretation und Verteidigung.” Deutsche Zeitschrift für Philosophie 61.1, 91–107. Schönecker, Dieter (2013b). “Kant’s Moral Intuitionism. The Fact of Reason and Moral Predispositions.” Kant Studies Online, Feb. 2013, 1–38. Schönecker, Dieter and Allen W. Wood (2015). Kant’s Groundwork for the Metaphysics of Morals. A Commentary. Cambridge, MA: Harvard University Press Wallach, Wendell and Colin Allen (2009). Moral Machines: Teaching Robots Right from Wrong. New York, Oxford: Oxford University Press.

5 “The Eye of True Philosophy” On the Relationship Between Kant’s Anthropology and His Critical Philosophy1 Robert B. Louden Interest in Kant’s anthropological writings, as one commentator notes, “has soared in recent years” (Wilson 2015, 1). At present, however, there are still a number of contested issues regarding the proper interpretation of these writings. Two such issues that I and others have discussed previously are whether we find a distinctively moral anthropology within Kantian anthropology (see Louden 2011, 49–77; Brandt & Stark 1997, xlvi; Brandt 2003, 92), and whether Kant’s investigations of human nature also include a transcendental anthropology in addition to a merely empirical one (see Louden 2018a; Frierson 2013, 11–45). In the present essay, I wish to probe a third contested issue: how do the anthropological Kant and the critical Kant relate to one another? What is the relationship between Kant’s anthropological writings and his critical philosophy? Do his anthropological works constitute simply “a sideline to his main work in critical philosophy” and “a mere diversion from it” (Zöller 2011, 136)? Or, should we say rather that there exists a “mutually supplementary relation” (Zöller 2011, 131, 132) between the anthropological Kant and the critical Kant? (And if we choose the latter option, what is the precise nature of the mutually supplementary relation?) Or is neither of these two options satisfactory, and is a third strategy therefore needed to correctly articulate the relationship between these two very different sides of Kant’s work? In what follows I  begin by surveying two core features of Kantian anthropology (viz., its empirical and popular orientation) that together pose a significant challenge to any attempt to relate Kant’s anthropology to his critical philosophy. In Section 2 I turn to several recent defenses of the “sideline and mere diversion” option, showing why I think they fail. In Section  3 I  examine the “mutually supplementary relation” option defended by Günter Zöller, explaining what I believe are its shortcomings. Finally, in Section 4, I present and defend a third bolder option for explaining the relationship between Kant’s anthropology and his critical philosophy. Anthropology, I  shall argue, is the “true eye” of Kantian philosophy. DOI: 10.4324/9781003044376-6

“The Eye of True Philosophy”  87

1.  Empirical and Popular The two core features of Kantian anthropology that together pose a substantial hurdle for any attempt to relate it to his critical philosophy are its largely empirical nature and its popular orientation. Kant repeatedly describes his anthropology as empirical, and when he does so he is explicitly contrasting it to his work in critical philosophy. For instance, in the opening sentence of the Collins transcription Kant states that anthropology is a “science” in which “the grounds of cognition are taken from observation and experience” (V-Anth/Collins, 25: 7), and at the beginning of the lectures recorded by Parow, he notes anthropology “deserves a special set of lectures, in part because it does not at all belong to metaphysics” (V-Anth/Parow, 25: 244). Kantian anthropology, Kant notes in a frequently cited letter to former student Marcus Herz written toward the end of 1773 in which he describes the nature of his new course, is a “Beobachtungslehre [observation-based doctrine]”—a “very pleasant” one that is “distinct from all other instruction” (L, 10: 146; cf. V-Anth/Fried, 25: 472, V-Anth/Mensch, 25: 856, V-Anth/Mron, 25: 1210, Anth, 7: 121). Kant’s critical philosophy, on the other hand, is concerned with nonempirical (a priori, pure) principles of reason—reason which “has been carefully cleansed of everything empirical” (G, 4: 388). A priori cognition, Kant reminds readers in a Reflexion, “is opposed to empirical cognition: philosophy concerning this is transcendental philosophy” (Refl 4851, 18: 10), and “from empirical principles one cannot attain transcendental ones” (Refl 4899, 18: 20). Given this strong contrast between the empirical and the a priori, and given Kant’s well-known alignment of his critical philosophy with the latter, how can there be a possibility of establishing a relationship of any sort between the two? How can two such radically different enterprises be related to one another in any manner whatsoever? Part of the answer consists simply in pointing out that Kant’s anthropology (some of his own remarks to the contrary) is not in fact strictly empirical. As many commentators have argued (see, e.g., Louden 2000, 71–4; Frierson 2003, 48–67; Wilson 2006, 93–6; Cohen 2009, 93–6), Kantian anthropology contains an abundance of normative, teleological, and conceptual commitments that take it far beyond the confines of a mere Beobachtungslehre. Key examples here include both the strong focus on humanity’s Bestimmung (destiny, vocation) that pervades the anthropological writings and which also serves as “the leading center of Kantian philosophy” (Brandt 2007, 7) as a whole, and the multiple moral messages that are clearly present throughout the lectures on anthropology. An additional part of the answer consists in showing that Kant’s philosophy does not in fact eschew the empirical. Here too, however, it

88  Robert B. Louden must be admitted that some of his own remarks do run in the opposite direction. For instance, toward the end of the first Critique we read: “metaphysics  .  .  . alone [allein] properly constitutes what we can call philosophy, in the genuine sense of the term [im echten Verstande]” (CPR, A 850/B 878).2 However, only a few pages earlier he also writes: “All philosophy . . . is either knowledge from pure reason, or rational cognition from empirical principles. The former is called pure, the latter empirical philosophy” (CPR, A 840/B 868). Similarly, in the Preface to the Groundwork Kant states: One can call all philosophy, insofar as it is based on grounds of experience, empirical, but that which puts forth its doctrines solely from principles a priori, pure philosophy. Physics will thus have its empirical but also a rational part; and ethics likewise. (G, 4: 388; cf. PhilEnc, 29: 7) Another relevant text is the following passage from The Conflict of the Faculties (1798), where Kant discusses the make-up of philosophy faculties in German universities of his era: The philosophy faculty consists of two departments: a department of historical knowledge (to which history, geography, philology, and the humanities all belong, along with all the empirical knowledge contained in the natural sciences), and a department of pure rational knowledge (pure mathematics and pure philosophy, metaphysics of nature and morals. (CF, 7: 28) So Kant’s considered view, as well as the institutional reality of the time, was that philosophy consists of both empirical (a posteriori) and pure (a priori) parts. And in fact a great deal of his own work—both before and after the “critical turn” of 1770—has extensive empirical content. Kant’s philosophy is by no means limited to the a priori, and those who are obsessed with “keeping philosophy pure” (Rorty 1982, 19) have done a great disservice to him in claiming otherwise. The unabashedly popular orientation of Kant’s anthropological writings is the second major hurdle that must be cleared in order to establish a meaningful relationship between Kant the anthropologist and Kant the critical philosopher. In sharp contrast to Ernst Platner’s “scholastic anthropology”3 (V-Anth/Mensch, 25: 856), from which one cannot “obtain any enlightenment for common life” (V-Anth/Mensch, 25: 853), Kantian “anthropology can be read by everyone, even by women at the dressing table” (V-Anth/Mensch, 25: 856). Kant regarded it as a distinct virtue that audiences found his anthropology lectures “entertaining and never dry” (L, 10: 146), and his earliest biographers concur that he was

“The Eye of True Philosophy”  89 quite successful on this point. Jachmann, for instance, reports that Kant’s anthropology lectures “afforded an extremely pleasant instruction, which were also attended the most frequently” (Jachmann 1804, 31), and Rink describes them as “lively” and filled with “keen observations” (Rink 1805, 46). At the same time, however, criticism of the popular nature of Kantian anthropology goes back at least as far as 1799, when Friedrich Schleiermacher published his polemical review of Anthropology From a Pragmatic Point of View. In the opening sentence, Schleiermacher ridicules Kant’s book for being nothing but “a collection of trivialities” and “a clear portrayal of the strangest confusion” (reprinted in Kant 1980, 339). The Critique of Pure Reason, on the other hand—as its author knew all too well, and as anyone who has tried to read it will attest—is anything but a popular treatise designed for entertainment and relaxation. In the Preface to the Prolegomena, for instance, Kant himself describes the Critique as “dry, . . . obscure, . . . and . . . long-winded as well” (Prol, 4: 261), and he insists that “popularity, entertainment, and ease” are entirely out of place “when the matter concerns the existence of highly prized knowledge that is indispensable to humanity, knowledge that cannot be constituted except according to the strictest rules of scholarly exactitude” (Prol, 4: 261). How can two such radically different styles of writing be meaningfully linked to one another? One strategy, favored by Patrick Frierson, is to offer an apology for the popular dimension of Kantian anthropology and to in effect beg readers to overlook (or at least downplay) this particular feature of the text, thereby paving the way for a more scholarly, sanitized version of Kantian anthropology. As he puts it, the “desire to make his lectures popular infects even Kant’s published Anthropology, and so one frequently finds tidbits of anthropological information that holds one’s attention but that do not really deserve the attention they receive” (Frierson 2003, 180n). I myself prefer a more direct and less evasive strategy. Yes, one of the key features of Kantian anthropology is its popular orientation, but Kant himself was proud of this fact. And nothing in principle prevents popular texts and those written “according to the strictest rules of scholarly exactitude” from working together toward the common goal of enlightenment. Indeed, Kant himself advocated that they do so. Contrary to the frequently voiced criticism that “Kant turned away from the path of a popular philosopher” (Zammito 2002, 6; cf. 106, 217, 257, 292, 347), popularity remained a goal throughout his life, though after the critical turn it is true that he did maintain that popularity should “never be there at the start” (Prol, 4: 261). “The extremely rare merit of a true philosophical popularity, Kant held,” is something to be highly prized, but it should come after “the elevation to principles of pure reason has already been achieved to full satisfaction” (G, 4: 409). And Kant did of

90  Robert B. Louden course believe that the “elevation to principles of pure reason” had been “achieved to full satisfaction” with the 1781 publication of the Critique of Pure Reason (or at least with the publication of the second edition in 1787). But the exoteric, popular style of Kant’s anthropology in no way prevents it from working in tandem with his esoteric critical philosophy. There is room for both approaches.

2.  A Sideline and Mere Diversion? While all commentators agree that Kant’s anthropological writings are quite different from his critical works, at present there is little agreement concerning how these two different aspects of his corpus relate to one another. At one end of the spectrum are those who claim that the relationship is minimal or nonexistent—because they are so different from one another, the two cannot stand in any sort of systematic or mutually supportive relationship. The anthropological writings, on this view, constitute a mere sideline to or entertaining diversion from the more serious and important investigations of the critical philosophy—“a temporary and contingent relief” from pure philosophy’s “rigorous requirements of universality and necessity” (Zöller 2011, 136). Reinhard Brandt has defended one version of this view in several essays: “Empirical, pragmatic anthropology is not part of philosophy in the strict sense” (Brandt 1994, 17). “Pragmatic anthropology . . . neither belongs to philosophy in the strict sense, nor is it articulated as a system based upon an idea of reason. It is an empirical discipline” (Brandt 2003, 85). Similarly, Brandt and Stark, in their “Introduction” to volume 25 of the Academy Edition of Kant’s Schriften, point out that Kant “does not describe his anthropology as philosophical” (Brandt & Stark 1997, 25: xii n).4 John Zammito also defends a version of the “sideline and mere diversion” position, arguing that, on Kant’s view, anthropology stands in a subordinate relationship to the critical philosophy: “the critical Kant systematically subordinated anthropology to metaphysics” (Zammito 2002, 3). “The critical Kant showed no interest in displacing philosophy with anthropology, but insisted on its proper subordination to philosophy” (Zammito 2002, 182; cf. 299). “Kant created the ‘critical philosophy’ at the cost of forsaking the ‘science’ of anthropology. He sought to relegate it—not promote it—to the ‘pragmatic’ ” (Zammito 2002, 348). Because the chief arguments offered by both Brandt and Zammito in defense of this position were described and evaluated earlier in Section 1, I shall be brief here. First, they argue that Kant’s anthropology is empirical and that his critical philosophy is an a priori venture, and that “never the twain shall meet.”5 But as I  pointed out earlier, Kant’s anthropology is not merely empirical. It includes normative, teleological, and conceptual dimensions that take it far beyond the confines of a

“The Eye of True Philosophy”  91 mere Beobachtungslehre. Additionally, Kant’s critical philosophy does not block out the empirical in the draconian manner claimed by Brandt, Zammito, and others. His considered view, as well as the institutional reality of his era, was that philosophy consists of both pure (nonempirical, a priori) and impure (empirical, a posteriori) parts. Additionally, empirical considerations often do manage to work their way into Kant’s critical philosophy. Second, Brandt and Zammito argue that Kantian anthropology is merely a popular, exoteric project; one that at best offers an amusing diversion or secondary sideline and supplement to Kant’s metaphysics, pure moral philosophy, and transcendental philosophy. My reply here starts with an acknowledgment of the popular orientation of Kantian anthropology, but then follows with a re-examination and refutation of the common claim that the sage of Königsberg rejected the goal of popularity. “True philosophical popularity” remained a fundamental ambition throughout Kant’s writing career. While it is true that after the critical turn in 1770 he maintained that popularity should be sought only after the philosophical groundwork had been properly laid “according to the strictest rules of scholarly exactitude”; this in no way implies that he abandoned the authorial goal of popularity. I believe that Kant viewed the relationship between his anthropological and critical writings as being quite a bit stronger than the weak relationships that we have surveyed thus far. In the final two sections of this essay, I shall compare and contrast two different models of this stronger relationship, and explain why I find the second model superior to the first.

3.  A Mutually Supplementary Relation? Günter Zöller, in his 2011 essay “Kant’s Political Anthropology,” offers a very different account of the relationship between the anthropological and critical Kant. On his reading, these two sides of Kant’s work stand in “a mutually supplementary relation” (Zöller 2011, 131, 132) to one another. In articulating this relationship, he emphasizes that his goal is not to integrate or merge anthropology into critical philosophy, but rather to maintain a distinction between “two closely related but differently organized and oriented philosophical projects” (Zöller 2011, 138n). He aims to offer a joint consideration of each project; one that in effect shows how each side helps the other out by providing something that is missing in the other part. One key illustration of this “two-closely-related-but-differentlyorganized- and-oriented-philosophical-projects” aspect of the model is contained in the following remark: Kant’s work in anthropology counterbalances the intentionally narrow methodological and systematical orientation of the critique of

92  Robert B. Louden reason, geared at a theory of a priori principles, with a broadly conceived, empirically informed and practically geared account of the human condition. Zöller (2011, 137) This “counterbalance” claim is, I  believe, well taken, and unlikely to elicit objections—even from the “diversion and sideline” camp examined earlier. For it marks out a different task for each of the two sides, and does not hint at any encroachment into the other’s territory. However, several key components of Zöller’s mutually supplementary relation model are in my view more questionable. Consider the following three examples: (1) The critical philosophy adopts an “essentially individualist conception of the human being,” which stands in sharp contrast to “the speciesbased approach of Kant’s anthropology” (Zöller 2011, 138). But what is “individualist” about the attempt to offer “an exhaustive analysis of all human cognition a priori” (CPR, A 13/B 27), or to locate the “two stems of human cognition” (CPR, A 15/B 29)? Similarly, when Kant in the Transcendental Aesthetic describes spatial and temporal intuition as a way of perceiving objects “which is peculiar to us” (CPR, A 42/B 59), the “us” is a species-wide “us” [or perhaps even a trans-species “us:” wouldn’t the “rational creatures that inhabit Jupiter or Saturn” (UNH, 1: 359) also perceive objects as spatially and temporally located?]. In short, when one looks at what Kant actually says in the Critique of Pure Reason, it is not at all clear that this text operates with the individualist conception of the human being that Zöller attributes to it. Additionally, Zöller’s own example of the allegedly individualist assumption of freedom in the critical philosophy (and its corresponding absence in the anthropology) casts further doubt on his point: The difference between the critique of reason’s essentially individualist conception of the human being and the species-based approach of Kant’s anthropology is especially striking when it comes to the contrast between a critico-rationally geared account of interhuman relations around non-empirical “laws of freedom” and an anthropologically oriented account of human social life in terms of natural regularities. (Zöller 2011, 138) But in the Preface to Anthropology from a Pragmatic Point of View, where Kant famously contrasts his own pragmatic anthropology to the physiological anthropology of Platner and others, he says that the former “concerns the investigation of what he [viz., the human being] as a freeacting [freihandelndes] being makes of himself, or can and should make of himself” (Anth, 7: 119). Freedom, in other words, is integral to Kantian pragmatic anthropology,6 not foreign to it.

“The Eye of True Philosophy”  93 (2) The “top-down account of reason in the critical project” (Zöller 2011, 138) enable readers to “differentiate between alleged anthropological facts and their normative interpretation and significance” (Zöller 2011, 138). Or, as he remarks more directly on the same page, “the guiding role of reason” (Zöller 2011, 138) in the critical philosophy serves to unmask the “racism, sexism, [and] classism” (Zöller 2011, 138) of Kantian anthropology. Alas, this seems to me to be merely wishful thinking on Zöller’s part. Nowhere does Kant explicitly argue for this kind of relationship between anthropology and critical philosophy. (3) Kant’s anthropological writings provide “an exoteric extension of his core project of the critique of reason” (Zöller 2011, 133; cf. 134). But if the anthropological writings are merely an “extension” of the critical writings, what becomes of the earlier-cited claim that we are dealing with “two closely related but differently organized and oriented philosophical projects” (Zöller 2011, 138n)? In order for A to extend B, don’t A and B need to be similarly organized and oriented to begin with? The extension claim, in other words, conflicts with the two distinct projects claim. Overall, Zöller’s account of the relationship between Kantian anthropology and critical philosophy appears to be closer to an idealized portrait or projection of what he and many other commentators (the present author included) wish the relationship actually was. But this idealized portrait lacks textual support. Unfortunately, Kant himself does not explicitly advocate it in his own writings. My own position on the relationship between the anthropological and critical Kant, which I will present and defend in the next section, is closer to the “mutually supplementary relation” model than the “diversion and sideline” paradigm. For instance, both Zöller and I hold that Kant’s anthropology constitutes “an essential part of his philosophy” (Zöller 2011, 133). The relationship between anthropology and critical philosophy, on both of our views, is much stronger than the sideline and diversion model allows. But unlike Zöller, I do favor a more integrative model than he espouses (cf. Zöller 2011, 138n). Anthropology, in my view, is ultimately what grounds Kantian philosophy—“grounds” not only in an epistemological sense but in a moral sense as well. Anthropology “gives philosophy dignity” ( JL, 9: 23). And while I am aware that some readers may find the relationship I am about to describe as being unduly quixotic and quaint, I believe it also has more textual warrant than either of the two models discussed thus far.

94  Robert B. Louden

4.  The Eye of True Philosophy Part of the story behind Kant’s views on the relationship between anthropology and philosophy lies in the marked anthropological turn that is evident in many earlier Enlightenment writings. As Ernst Cassirer remarks in his classic study, Die Philosophie der Aufklärung, during the Enlightenment there was a tendency “to dissolve [auflösen] logic, ethics, [and] theology into mere anthropology” (Cassirer 1932, 155). Scores of Enlightenment authors argued that the new “science of man” should either replace philosophy entirely or at least become the central research project for an enlightened era, and they competed with each other in multiple efforts to shape the novel discipline in accordance with their own personal visions and concerns. Hume, for instance, in the Introduction to his Treatise of Human Nature (1739–40) writes: ‘Tis evident, that all of the sciences have a relation, greater or less, to human nature; and that however wide any of them seem to run from it, they still return back by one passage or another. Even Mathematics, Natural Philosophy, and Natural Religion, are in some measure dependent on the science of MAN; since they lie under the cognizance of men, and are judged by their powers and faculties. T Intro, Hume (1978, xv) Similarly, Hutcheson, in the first sentence of the Preface to his Inquiry into the Original of Our Ideas of Beauty and Virtue (1725), declares: “There is no part of philosophy of more importance than a just knowledge of human nature and its various powers and dispositions” (Hutcheson 1994, 3). And Rousseau, in the opening sentence of the Preface to his Discourse on the Origin of Inequality (1755), states: “The most useful and least developed of all the sciences seems to me to be that of man” (Rousseau 1984, 67). Kant, particularly in some of the earlier versions of his anthropology course, clearly echoes these Enlightenment sentiments. For instance, in the Prolegomena to Collins (1772–73), he declares: “Nothing indeed appears to be more interesting for the human being than this science [of anthropology], and yet none is more neglected than precisely this one” (V-Anth/Collins, 25: 7; cf. V-Anth/Fried, 25: 470). And in the opening sentence of Pillau (1777–78) he asserts: “There is no greater or more important investigation than the cognition of the human being” (V-Anth/ Pillau, 25: 733; cf. Notes, 20: 45). Commentators often claim that the critical Kant abandons this early infatuation with anthropology. Brandt and Stark, for instance, in their Introduction to the Lectures on Anthropology, write: [A]nthropology can and should peacefully supplement metaphysics and critical philosophy, but not reductionistically absorb and

“The Eye of True Philosophy”  95 dissolve it [reduktionistisch einziehen und auflösen]. His “Copernican Turn” is accordingly not carried out through anthropology. Brandt & Stark (1997, 25: xiii) Similarly, John Zammito, as we saw earlier, claims that Kant developed his critical philosophy, at least in part, in order to demote anthropology to a non-philosophical sideline: “Kant created the ‘critical philosophy’ at the cost of forsaking the ‘science’ of anthropology. He sought to relegate it—not promote it—to the ‘pragmatic’ ” (Zammito 2002, 348). In my view, however, these are all inaccurate descriptions of Kant’s considered position on the relationship between anthropology and critical philosophy. As I  will argue later, he continues in his later years to hold that anthropology, far from being a pleasant sideline and diversion to ‘real’ philosophy, is ultimately what grounds it. Anthropology gives dignity and inner worth to philosophy. As is well known, Kant’s most famous remark concerning the importance of anthropology is the following “Was ist der Mensch?” passage, versions of which appear in three different Kantian texts: The field of philosophy in this cosmopolitan sense can be brought down to the following questions: 1. 2. 3. 4.

What can I know? What ought I to do? What may I hope? What is the human being? [Was ist der Mensch?]

Metaphysics answers the first question, morals the second, religion the third, and anthropology the fourth. Fundamentally, however, one could reckon all of this [alles dieses] as anthropology, because the first three questions relate to the last one. ( JL, 9: 25; cf. V-Met/L2/Pölitz, 28: 533–4, and Kant’s letter to Stäudlin of May 4, 1793, L, 11: 429) In this passage, Kant is clearly putting his own spin on the anthropological turn of Enlightenment thought described earlier. He is claiming that the question, “What is the human being?” should be philosophy’s most fundamental concern, on the ground that all of philosophy’s questions are ultimately answered through anthropology. However, there are several basic problems with Kant’s “Was ist der Mensch?” argument. The first is a textual issue. Unfortunately, the “Was ist der Mensch?” question simply does not link up with the anthropology lectures in any discernible manner. Nowhere in the anthropology lectures does Kant explicitly answer or even seriously address the question, “What is the human being?” Nowhere in these lectures does he proclaim that all of philosophy’s questions are somehow answered by the information about

96  Robert B. Louden human beings presented in the lectures. The anthropology texts do not pretend to operate on such an extravagantly high level. As we saw earlier, for the most part they offer readers merely a “very pleasant observationbased doctrine” (L, 11: 146). As Martin Buber remarked, in a lecture first presented in 1938: The question, what man is, is simply not raised [in the anthropology lectures] . . . . It is as if Kant in his actual philosophizing had had qualms about setting the question which he formulated as the fundamental one. . . . Certainly Kant in his anthropology has neither answered nor undertaken to answer the question which he put to anthropology—What is man? He lectured on another anthropology than the one he asked for. Buber (1965, 120–1) And more recently, Brandt, in his extensive Kommentar on Kant’s Anthropologie, echoes Buber’s observation when he writes: [P]ragmatic anthropology . . . does not answer the question, “What is the human being?”  .  .  . Neither the Lectures on Anthropology nor the Anthropology of 1798 refers to the question, “What is the human being?” as its central problem: they do not mention it once. Brandt (1999, 16) But there are also philosophical problems with Kant’s “Was ist der Mensch?” argument. Philosophy is bigger than the merely human, and the claim that an answer to the question, “What is the human being?” will answer all of philosophy’s questions is grossly reductionist. And it also conflicts with Kant’s own extraterrestrial commitments. Kant subscribed to an infinity of possible worlds hypothesis—he believed that humans inhabit a universe “in which worlds or systems are only specks of dust in the sunlight compared with the whole of creation” (UNH, 1: 352). And he also believed that humans are not alone: “most of the planets are certainly inhabited [gewiβ bewohnt]” (UNH, 1: 354). As Lewis White Beck remarks, Kant, like “many eminent philosophers—among others, Aristotle, Nicholas of Cusa, Giordano Bruno, Gassendi, Locke, Lambert, . . . and William Whewell—believed that there is extraterrestrial life” (Beck 1985, 3). But a definitive answer to the question, “What is the human being?” is not going to help the “rational creatures that inhabit Jupiter or Saturn” (UNH, 1: 359) or any other non-terrestrial place in the infinite universe in their own philosophical investigations. Even if the rational creatures that allegedly inhabit Jupiter were to have a correct answer to the question, “Was ist der Mensch?” this would not necessarily enable them to answer their own metaphysical, epistemological, and normative queries.

“The Eye of True Philosophy”  97 Furthermore, even if we remain earthbound and bracket the intelligent extraterrestrials hypothesis, it is still not true that all of philosophy can be reckoned as anthropology. To assert this is to diminish philosophy’s mission. When philosophers ask questions about the nature of God or the nature of reality (or even the nature of morality or aesthetics), they are not necessarily asking questions about human beings or what humans can, should, or may know, do, or hope. Rather, they may simply be inquiring into the nature of these things—period. For these reasons, Kant’s radical move to dissolve all of philosophy into anthropology via the question, “What is the human being?” is unsuccessful. Philosophy is more than the merely human. However, he also has a second (less well known and more modest) argument concerning the fundamental importance of anthropology for philosophy, one that I believe is more promising. Let us turn to it now. In one of his Reflexionen zur Anthropologie, Kant criticizes an increasingly familiar kind of modern scholar who lacks humanity and “trusts his own powers too much” (Refl 903, 15: 395). Max Weber will later famously describe this scholar as a “specialist without spirit” (Weber 1958, 182), but Kant says: “I call such a person a Cyclops” (Refl 903, 15: 395). This one-eyed giant (Kant is alluding to a famous passage in Homer’s Odyssey),7 he adds, “needs another eye, so that he can consider his object from the point of view of other human beings” (Refl 903, 15: 395). This necessary second eye, which is precisely what the specialist without spirit lacks, is “what grounds the humanity of the sciences, that is, gives them the affability of judgment through which one submits to the judgment of others” (Refl 903, 15: 395). The scholarly Cyclops thus needs to acquire some humanity by cultivating a broadened way of thinking, one that will enable him “to think into the place of the other” (Anth, 7: 200). In the Anthropology, Kant also notes that despite the “gigantic erudition” of the Cyclops, he “lacks one eye, the eye of true philosophy [das Auge der wahren Philosophie], by means of which human reason appropriately uses this mass of historical knowledge, the load of a hundred camels” (Anth, 7: 227). Similarly, in his criticisms of “cyclopic learnedness” in the Lectures on Logic edited by Gottlob Benjamin Jäsche, Kant remarks that this objectionably narrow scholarship lacks one eye, the eye of philosophy [das Auge der Philosophie], and a Cyclops among mathematicians, historians, natural historians, philologists, and linguists is a scholar who is great in all these matters, but who for all that holds philosophy to be dispensable. ( JL, 9: 45) The underlying message in these passages is that one-eyed scholars (whose ranks within philosophy and other areas of academia have

98  Robert B. Louden unfortunately swelled since Kant’s day) need to acquire a broader, more humanistic way of thinking, one which Kant elsewhere calls “philosophy in the cosmopolitan sense (in sensu cosmopolitico)” ( JL, 9: 24), “philosophy according to the world concept [nach dem Weltbegriffe] (in sensu cosmico)”( JL, 9: 23), and “the science of the ultimate ends of human reason” (V-Met-L2/Pölitz, 28: 532; cf. JL, 9: 23). Only after acquiring a second eye does one see the indispensability of philosophy. Additionally, this “high concept” of philosophy in the cosmopolitan sense, he adds, “gives philosophy dignity; i.e., an absolute worth [Würde, d.i. einen absoluten Werth]” (JL, 9: 23), and it also “gives a worth to all other sciences” (V-Met-L2/Pölitz, 28: 532; cf. JL, 9: 24). But philosophy in the cosmopolitan sense and according to the world concept is only made possible by anthropology. For we are talking now about a broad-based reflection on the human condition. Philosophy, at least for humans—but not for “the rational creatures that inhabit Jupiter or Saturn” (UNH, 1: 359)—requires reflection on what it means to be a human being and on humanity’s place in the universe. Again, philosophy in the cosmopolitan sense is not an anthropology that reckons all of philosophy as anthropology (cf. JL, 9: 25)—it does not seek to replace philosophy with anthropology or dissolve it into anthropology. It is a more modest venture, but one that nevertheless holds that philosophy without an anthropological ground lacks one eye and “trusts its own powers too much” (cf. Refl 903, 15: 395). Without this necessary anthropological ground, philosophy devolves into a mere intellectual parlor game or what Kant calls philosophy in the “scholastic” sense (see CPR, A 838/B 866)—a style of philosophy from which one cannot “obtain any enlightenment of common life” (V-Anth/Mensch, 25: 853; cf. V-Anth/Mron, 25: 1209–10) and which is consistently running into the danger of becoming irrelevant to human concerns. Philosophy without an anthropological ground lacks dignity and inner worth, for anthropology is the true eye of philosophy. Granted, at first glance it may seem like a stretch to claim that, in Kant’s view, anthropology should ground philosophy for human beings—particularly abstract areas of philosophy such as the foundations of physics or mathematics that seem far removed from anthropological concerns. But I  do think that Kant and many other Enlightenment thinkers did ascribe precisely this foundational role to anthropology. Recall, for instance, the remark of Hume’s cited earlier: Tis evident, that all of the sciences have a relation, greater or less, to human nature; and that however wide any of them seem to run from it, they still return back by one passage or another. Even Mathematics, Natural Philosophy, and Natural Religion, are in some measure

“The Eye of True Philosophy”  99 dependent on the science of MAN; since they lie under the cognizance of men, and are judged by their powers and faculties. T Intro, Hume (1978, xv) On Kant’s view, however, it is not so much that mathematics and natural philosophy “lie under the cognizance of men” that gives anthropology a grounding role, for he is definitely open to the possibility that the rational creatures that inhabit Jupiter or Saturn may also engage in mathematics and natural philosophy. And if so, then mathematics and natural philosophy will also lie under their cognizance. Rather, as noted earlier, Kant’s position is that the second, anthropological eye “grounds the humanity of the sciences” (Refl 903, 15: 395). That is, mathematicians and physicists who have acquired the necessary second eye are in a much better cognitive position to appreciate the necessary conceptual and philosophical dimensions of their discipline—as well as to see that “however wide” mathematics or physics may seem to run from human nature, they are nevertheless fundamentally important and necessary activities for humanity. They will see, in other words, that philosophy is indispensable for human beings (cf. JL, 9: 45), and that mathematics and physics are themselves made possible by philosophical thinking. Kant’s frequent references to the cosmopolitan dimension of the second eye provide further support for his claim that anthropology is the eye of true philosophy. For anthropology is only properly Kantian “when it contains knowledge of the human being as a citizen of the world [als Weltbürgers]” (Anth, 7: 120). Anthropology from a Kantian point of view is not “a local, but a general anthropology. In it one comes to know the nature of humanity, not the state of human beings, for the particular properties of human beings always change, but the nature of humanity does not” (V-Anth-Fried, 25: 471; cf. V-Anth Pillau, 25: 733–4). By means of this particular normative perspective humans are able “to think into the place of the other” (Anth, 7: 120) and take into account “the point of view of other human beings” (Refl 903, 15: 395). This perspective also brings greater objectivity and impartiality to our philosophical investigations, and it is thus much better suited for the important task of unmasking the “racism, sexism, [and] classism” (Zöller 2011, 138) that Zöller assigns to Kant’s critical philosophy. And critical philosophy is itself properly grounded in this cosmopolitan dimension of the second eye in the sense that it needs cosmopolitanism in order to accurately carry out its own mission. Finally, because the second eye “gives philosophy dignity; i.e., an absolute worth” ( JL, 9: 23) as well as “a worth to all of the other sciences” (V-Met-L2/Pölitz, 28: 532), it serves to protect both philosophy and science from dishonor and discredit. This last point perhaps suggests more of a normative grounding than a strictly epistemological one, but the former is no less essential than the latter.

100  Robert B. Louden

Notes . This essay builds on Louden (2021); particularly the final chapter. 1 2. Identifying philosophy with nonempirical inquiry can also be used as a means to determine where and when one finds examples of philosophical activity among human beings. Kant, for instance, in the Jäsche logic lectures, states: “Among all peoples . . . the Greeks first began to philosophize. For they first attempted to cultivate cognitions of reason, not with images as the guiding thread, but in abstracto, while other peoples always sought to make concepts understandable only through images [nur durch Bilder] in concreto.” ( JL, 9: 27)   Park argues that the Kantian “method of a priori construction” (Park 2013, 7; cf. 23, 24, 29) was itself used as a key tool in “the campaign to exclude Africa and Asia from the history of philosophy” (Park 2013, 150). However, as he himself concedes elsewhere in his book, “a priori construction does not in principle entail the exclusion of the Orient” (Park 2013, 98)—nor, I would add, does it in principle entail the exclusion of Africa or of any other region inhabited by human beings. And whether or not the Greeks were indeed the first to “philosophize” (assuming for the moment that to philosophize means to think nonempirically) is itself an empirical question. Although I  do think that the extent to which Kant equates philosophy with the a priori has been greatly exaggerated, I  also believe that the race-based Eurocentrism in his work—which, alas, is clearly present (see, e.g., Louden [2000, 87–100])— bears no necessary connection to the “method of a priori construction.” These are two separate issues. 3. In 1772 Platner published Anthropologie für Ärzte und Weltweise (Anthropology for Physicians and Philosophers), and Kant’s former student Marcus Herz, himself a physician, soon reviewed the book for the Allgemeine deutsche Bibliothek. Kant, in the earlier-cited letter to Herz written toward the end of 1773, emphasizes that his own conception of anthropology is “quite different” (L, 10: 145) from Platner’s. 4. However, on the next page Brandt and Stark also write: “anthropology can and should peacefully supplement [friedlich ergänzen] metaphysics and critical philosophy, but not reductionistically absorb and dissolve it” (Brandt & Stark 1997, 25: xiii). Although they don’t elaborate on what they mean by “peacefully supplement,” this phrase does suggest some sort of subordinate relationship between anthropology and critical philosophy. However, this is still a much weaker relationship than the one that I shall defend later in Section 4. 5. “Oh, East is East, and West is West, and never the twain shall meet, till Earth and Sky stand presently at God’s great judgment seat” (Rudyard Kipling, “The Ballad of East and West,” 1892—see Kipling [2018]). 6. For further discussion of the place of freedom in Kantian anthropology, see Frierson (2003) and Louden (2018b). 7. In the Odyssey, Odysseus escapes from the Cyclops by seizing “a fire-pointhardened timber” and twirling it in the giant’s eye socket until “the roots of his eye crackle” (Homer 1965, 147).

References Beck, Lewis White (1985). “Extraterrestrial Intelligent Life.” In Edward Regis Jr. (ed.), Extraterrestrials: Science and Human Intelligence. Cambridge: Cambridge University Press, 3–18.

“The Eye of True Philosophy”  101 Brandt, Reinhard (1994). “Ausgewählte Probleme der kantischen Philosophie.” In Hans-Jürgen Schings (ed.), Der ganze Mensch: Anthropologie und Literatur im 18. Jahrhundert. Stuttgart: J. B. Metzler, 14–32. Brandt, Reinhard (1999). Kritischer Kommentar zu Kants Anthropologie in pragmatischer Hinsicht (1798). Hamburg: Meiner. Brandt, Reinhard (2003). “The Guiding Idea of Kant’s Anthropology and the Vocation of the Human Being.” In Brian Jacobs and Patrick Kain (eds.), Essays on Kant’s Anthropology. Cambridge: Cambridge University Press, 85–104. Brandt, Reinhard (2007). Die Bestimmung des Menschen bei Kant. Hamburg: Meiner. Brandt, Reinhard and Werner Stark (1997). “Einleitung.” In Reinhard Brandt and Werner Stark (eds.), Vorlesungen über Anthropologie. Berlin, Boston: De Gruyter, vii–cli. Buber, Martin (1965). Between Man and Man. New York: Macmillan. Cassirer, Ernst (1932). Die Philosophie der Aufklärung. Tübingen: J. C. B. Mohr. Cohen, Alix (2009). Kant and the Human Sciences: Biology, Anthropology and History. London: Palgrave Macmillan. Frierson, Patrick R. (2003). Freedom and Anthropology in Kant’s Moral Philosophy. Cambridge: Cambridge University Press. Frierson, Patrick R. (2013). What is the Human Being? London: Routledge. Homer (1965). The Odyssey. Trans. by Richmond Lattimore. New York: Harper & Row. Hume, David (1978). A Treatise of Human Nature. Ed. by L. A. Selby-Bigge, 2nd ed. Oxford: Clarendon Press. Hutcheson, Francis (1994). Philosophical Writings. Ed. by R. S. Downie. London: Dent. Jachmann, Reinhold Bernhard (1804). Immanuel Kant geschildert in Briefen an einen Freund. Königsberg: Friedrich Nicolovius. Kant, Immanuel (1980). Anthropologie in pragmatischer Hinsicht. Ed. by Karl Vorländer, with an Introduction by Joachim Kopper and a Supplementary Appendix by Rudolf Malter. Hamburg: Meiner. Kipling, Rudyard (2018). https://en.wikipedia.org/wiki/The_Ballad_of_East_ and_West. Louden, Robert B. (2000). Kant’s Impure Ethics: From Rational Beings to Human Beings. Oxford: Oxford University Press. Louden, Robert B. (2011). Kant’s Human Being: Essays on His Theory of Human Nature. Oxford: Oxford University Press. Louden, Robert B. (2018a). “Kant’s Anthropology: (Mostly) Empirical Not Transcendental.” In Francesco Valerio Tommasi (ed.), Der Zyklop in der Wissenschaft: Kant und die anthropologia transcendentalis. Hamburg: Meiner, 19–33. Louden, Robert B. (2018b). “Freedom from an Anthropological Point of View.” In Violette L. Waibel, Margit Ruffing and David Wagner (eds.), Natur und Freiheit: Akten des XII. Internationalen Kant-Kongresses. Berlin, Boston: De Gruyter, 457–72. Louden, Robert B. (2021). Anthropology from a Kantian Point of View. Cambridge: Cambridge University Press. Park, Peter K. J. (2013). Africa, Asia, and the History of Philosophy: Racism in the Formation of the Philosophical Canon, 1780–1830. Albany: State University of New York Press.

102  Robert B. Louden Rink, Friedrich Theodor (1805). Ansichten aus Immanuel Kant’s Leben. Königsberg: Göbbels und Unzer. Rorty, Richard (1982). “Keeping Philosophy Pure: An Essay on Wittgenstein.” In Richard Rorty (ed.), Consequences of Pragmatism. Minneapolis: University of Minnesota Press, 19–36. Rousseau, Jean-Jacques (1984). A Discourse on Inequality. Trans. and ed. by Maurice Cranston. London: Penguin. Weber, Max (1958). The Protestant Ethic and the Spirit of Capitalism. Trans. and ed. by Talcott Parsons. New York: Charles Scribner’s Sons. Wilson, Holly L. (2015). “Review of Alix Cohen (ed.), Kant’s Lectures on Anthropology: A Critical Guide.” British Journal for the History of Philosophy 23, 589–92. Wilson, Holly L. (2006). Kant’s Pragmatic Anthropology: Its Origin, Meaning, and Critical Significance. Albany: State University of New York Press. Zammito, John (2002). Kant, Herder, and the Birth of Anthropology. Chicago: University of Chicago Press. Zöller, Günter (2011). “Kant’s Political Anthropology.” Kant Yearbook 3, 131–61.

6 Kant am Pregelflusse Site and Systemicity in the Preface to the Anthropology1 Susan M. Shell

The present study of the Preface to Kant’s Anthropology takes its primary bearings from a common insight or assumption of his later writings, namely, that the relation between theory and practice cannot be fully specified but only exhibited, as it were, in situ. Understood in this light, the short Preface to Kant’s final published work on anthropology has a philosophic significance that has gone largely unremarked. To be sure, such neglect is understandable, given both the Preface’s brevity and its unusually personal character. Nevertheless, in these largely neglected preliminary paragraphs and autobiographical footnotes Kant establishes both that and how “knowledge of the world” can be “raised” to a “genuine [förmliche] science” (Anth, 7: 121), perhaps the only science to succeed in being both “systematic” and also fully “popular” (Anth, 7: 121). In short, in these few lines, Kant schematizes, as it were, through the inclusion of facts pertaining to his own personal biography and its setting, the union of general and local knowledge absent which the possibility of Anthropology as a true science is open, as he frankly admits, to insuperable objections.

1.  Pragmatic Anthropology: Its Scope and Meaning Kant’s Anthropology has been the subject of new and growing interest in recent years. Michael Foucault’s 1961 thesis on Kant’s Anthropologie in pragmatischer Hinsicht (Anthropology in a pragmatic regard),2 had the field virtually to itself. Today there are many excellent studies devoted to anthropological questions and themes in Kant’s thought.3 And the publication of the Akademie edition (vol. 25) of his lectures on anthropology spanning three decades promises to further enrich the field. This new attention is a welcome addition in its own right and a useful counterweight to the excessive formalism of much earlier work on Kant’s practical philosophy, especially in the English-speaking world. But it is also marked by an ongoing dispute as to the specific character of Kant’s anthropology, along with its systematic role more generally. Since these disputes bear upon some of the deepest and most vexing questions in DOI: 10.4324/9781003044376-7

104  Susan M. Shell Kant’s thought, including but not limited to the relation between nature and freedom, and between the a priori and the empirical, the perplexity they mark is no small matter. This perplexity is exacerbated by Kant’s own shifting usage and otherwise divergent indications over the many years in which he explicitly addressed himself to anthropological matters. If one limits oneself to the critical period, anthropology is presented as a purely empirical study, only episodically related to critical philosophy, or, alternatively, as what Robert B. Louden has helpfully called an Urdisziplin that summarizes and contains the entire critical project.4 Such textual divergence is no less evident when one attempts to pin down the systematic role of anthropology in Kant’s practical thought. In the Critique of Judgment, for example, Kant defines “empirical anthropology” as a “theoretical” science that considers man insofar as he is subject to the “laws of nature” rather than (as with “practical science”) to the “laws of freedom” (CPJ, 5: 277, 461). Anthropology, so understood, seems to fall outside the realm of “practical science” entirely. But Kant does not consistently exclude anthropology from the realm of ethics proper. Indeed, in the Metaphysics of Morals Kant refers to “moral anthropology” as a necessary counterpart to a “metaphysics of morals” within the larger field of practical philosophy (MM, 6: 217). Moral anthropology, so construed, makes available the rules of application by which pure moral laws may be applied to human nature. The passage does not shed further light, however, on the connection between the metaphysical and anthropological side of ethics, or on how a strictly moral anthropology relates to the “anthropology” he elsewhere calls “empirical” (cf. CPJ, 5: 479). Nor does the term ‘pragmatic’ offer decisive help, given the variety of its meanings across the critical corpus, including writings of the 1790s. These meanings range from usages that explicitly exclude moral ends or maxims5 to usages in which the term’s moral resonance is more positive.6 The term is often paired with cleverness (Klugheit), and with hypothetical imperatives that have their motive ground in the desire for happiness, as distinguished from morality.7 Even more morally positive uses of the term generally maintain a guarded ambiguity as to the precise relation of the pragmatic to the moral. On the one hand, pragmatic knowledge is a necessary and useful tool for the promotion of moral ends.8 On the other, pragmatic knowledge seems particularly susceptible to perversion toward non-moral ends, making it a source of immoral temptation in its own right,9 or otherwise darkening prospects for moral hope.10 One striking example of such morally ambiguous usage arises with regard to what Kant calls “(pragmatic) friendship” understood as a site of (potential) mediation between the mercenary and the genuinely moral; and it seems telling that it follows immediately upon a call on his part for

Kant am Pregelflusse  105 a “transition” (Übergang; transitio) that would “schematize, as it were,” the application of moral principles to the empirically varying “conditions” (Zustande) of human beings: These (duties of virtue) do not really call for a special chapter in the system of pure ethics; since they do not involve principles of obligation for human beings as such for one another. . . . They are only rules modified in accordance with differences of the subjects to whom principles of virtue (in terms of what is formal) is applied in cases that come up in experience (the material) . . . . Nevertheless, just as a passage from the metaphysics of nature to physics is needed—a transition having its own special rules—something similar is rightly required for the metaphysics of morals: a transition which, by applying the pure principles of duty to cases of experience, would schematize these principles, as it were, and present them as ready for moral use. (MM, 6: 468, emphasis added. All translations are my own) This (unmet) need for a transition from the a priori to the empirical, which Kant declares just prior to broaching the vexed question as to the possibility of true (moral) friendship, might suggest the “pragmatic” (as in “(pragmatic) friendship”) as the site of such a schematization. And yet any such easy conclusion is immediately blocked by Kant’s treatment, in the same paragraph, of “pragmatic learning” as one-sidedly mechanical and empirically delimiting (MM, 6: 469). A successful site of “transition” between moral principle and empirical action will have to operate, it seems, in both directions—from both the a priori laws of reason downward and the actual rules of human behavior upward. To be sure, there are certain works (Religion within the Boundaries of Bare Reason, Toward Eternal Peace and The Conflict of the Faculties come specially to mind) in which something like a moral anthropology, and with it an explicit “counterpart” (MM, 6: 217) to a moral metaphysics, is explicitly at issue. But these are also works that are manifestly occasioned by the peculiar circumstances in which an aging Kant found himself—circumstances that call upon his own worldly knowledge in a new and especially urgent way. Like the various senses of ‘anthropology,’ the shifting meanings of ‘pragmatic’ in Kant’s late work are themselves instructive, I would suggest, both for the light they shed on Kant’s later thought and as an indication of the lack of any final formal principle of demarcation between the moral and the empirical. This fact lends Kant’s final, published Anthropology from a Pragmatic Regard (1798/1800) particular systematic importance, at least potentially.

106  Susan M. Shell

2. The Preface to Anthropology from a Pragmatic Regard The Anthropology is not only Kant’s final book-length work; it is also the only version he saw fit to publish of his popular lecture course on anthropology, delivered in various forms over a period spanning three decades. As both Kant’s only published introduction to anthropology and a culminating farewell to his career as a published author, the Anthropology promises to yield important clues as to his final understanding of the relation between the a priori and the empirical as it applies to matters ‘practical.’ The spirit of Kant’s late anthropology is evident in Kant’s Anthropology (1798/1800), beginning with its Title and Preface. In identifying his regard (Hinsicht) here as specifically ‘pragmatic,’ Kant does not mean to limit his concern (as in some earlier uses on his part of the term ‘pragmatic’)11 to man’s natural end understood as the pursuit of happiness, but considers, rather, man as a freely acting being in the broadest sense. Pragmatic anthropology so understood does not aim to be a ‘valuefree’ science, potentially as useful to the Machiavellian as to the moral saint. It begins, rather, by taking man’s ability to make something of himself, and hence his responsibility, for granted. Knowledge of what man can do, ought to do, and does, addresses itself not only, and perhaps not at all, to techniques of human manipulation under ‘laws of nature’ but rather to the capacity of the species as a whole for individual and collective self-improvement under ‘laws of freedom.’ Pragmatic anthropology so understood would thus seem to be an intrinsically practical or moral science. But this makes all the more perplexing how, if at all, it is distinguished from the “moral anthropology” that constitutes the necessary yet elusive “counterpart” (MM, 6: 217) to Kant’s moral metaphysics. The Preface of the Anthropology has received little focused scholarly attention, despite its textual peculiarity: unlike the body of the text, neither the Preface nor the Contents are included in the handwritten manuscript of the Anthropology that has come down to us—one that contains many emendations and corrections of unknown authorship. Reinhart Brandt, in his detailed Commentary on the Anthropology, argues that the Preface is likely to have been composed just prior to the work’s publication (i.e., no earlier than late April of 1798) (Brandt 1999, 20, 22, 123). If this suggestion has merit, the Preface may well be not only the most reliably authentic portion of the published work, but also both Kant’s final statement on the systematic status of pragmatic anthropology and his last extended treatment, intended for immediate publication, of a major philosophical theme. The Preface occupies a little over three pages in the Academy edition, and includes two autobiographical footnotes—each unusual for an author who generally avoided personal reference. The first note draws

Kant am Pregelflusse  107 attention to Kant’s own place of residence as a site peculiarly suited to acquiring knowledge of the world (Weltkenntniß); the second to his own occupation as a scholar, at first “freely undertaken,” (Anth, 7: 122) and then in accordance with his duties as a university instructor. We will have reason to return to both footnotes later in this essay; for now, it will prove helpful to consider Kant’s argument as it unfolds in the body of the Preface, which consists of ten paragraphs of text, separated by a hiatus between paragraph five and paragraph six. 2.1.  Pragmatic Anthropology and Knowledge of the World The Preface opens with a precis of “world knowledge” in the specific “human” sense that is here at issue: All progress in culture12 through which the human being educates himself [seine Schule macht] has the goal of applying [anzuwenden] these acquired cognitions [Kenntnisse]13 and skills [Geschicklichkeiten] toward use for the world [zum Gebrauch für die Welt]; but the most important object in the world, on which he can employ [verwenden] each, is the human being; because he is his own final [letzter] end.—Therefore to cognize [erkennen] him according to his species as an earthly being endowed with reason especially deserves to be called knowledge of the world [Weltkenntniß], even though he only constitutes a part of earthly creatures. (Anth, 7: 119) Knowledge of the world (Weltkenntiß) of the sort intended here is especially concerned not only with knowledge of human beings but also with the application of such knowledge and skill for worldly use. By this Kant means not only use to the individual but also, by implication, to the world at large, and insofar as a person’s final end marks him as a citizen of the world (cf. Anth, 7: 120). Pragmatic anthropology aims to be useful, then, to the individual (whatever her ends may be) while also aiming toward her final (moral) end.14 The beginning of the Preface thus points directly to the gemeinnützigen Wissenschaft (i.e., altruistic science or, alternatively, science of general use) mentioned in the final paragraph (Anth, 7: 122). Pragmatic anthropology, one might already say, promises to span the “cleft” between the domains of nature and of freedom to which the third Critique (1790) had earlier systematically addressed itself (CPJ, 5: 175–6): A doctrine of knowledge of the human being [Kenntnis des Menschen], systematically composed (Anthropology), can exist either in a physiological or a pragmatic regard.—Physiological knowledge of the human being concerns the investigation of what nature makes

108  Susan M. Shell of the human being; the pragmatic concerns what he as a freely acting [freihandelndes] being makes of himself or can and ought make of himself.—He who ruminates over natural causes, for example, on what the capacity for memory might be based, can ratiocinate [vernünfteln] back and forth (like Descartes) over the traces left behind in the brain of impressions left by suffered sensations: but it must thereby be admitted that he is merely a spectator in the play of his representations, and that nature must be allowed, insofar as he neither knows the nerves and fibers of the brain, nor understands having them at hand [die Handhabung derselben . . . versteht] to his own purpose, as a consequence of which all theoretical ratiocination is a pure loss (reiner Verlust].—But if he makes use of perceptions concerning that which has been found to hinder memory or promote it in order to enlarge memory or make it agile, and if he needs knowledge of the human being for this, then this would constitute a part of anthropology in a pragmatic respect [Absicht], and it is with just this with which we here are occupied. (Anth, 7: 119) Pragmatic anthropology “systematically composed” makes up one part of “knowledge of the world” (Weltkenntiß); and is specifically distinguished both from the “theoretical” knowledge of non-human things (and that forms a part of “knowledge of the world” in the most comprehensive sense) and from physiological anthropology, or knowledge of what “nature makes of us,” a field of study whose relation both to anthropology profitably pursued and “knowledge of the world” more generally, is here left open. On the one hand, to ruminate over the natural causes of mental processes such as human memory, especially where no practical use is in the offing, is a sheer waste, as with the ratiocinations of Descartes, which are here implicitly contrasted, to the former’s disadvantage, with Kant’s own philosophic “dietetic.”15 On the other hand, “extensive knowledge” of plants, minerals, and animals in the various lands and climates, and even “the races of man” as produced “by the play of nature,” makes up what Kant calls “physical geography”—an academic subject that he introduced to Germany and taught for many years as a companion to his course on anthropology, as he informs us in a later footnote (Anth, 7: 122n). Knowledge of what has been found through the observation and study of human beings to hinder or promote memory would presumably constitute a part of that pragmatic anthropology with which Kant is here occupied—a “hands-on” understanding, as it were, by virtue of which we do not only “know” the world but also “have” it (Anth, 7: 120). The ratiocinating physical anthropologist, on the other hand, is the wasteful “spectator” of a “play of sensations” that, precisely as such, can never constitute the objective knowledge that he seeks. In what, then, does a

Kant am Pregelflusse  109 physiological anthropology that does not merely “rationalize back and forth” (as with Descartes) properly consist? Is it that part of physical geography that includes empirical knowledge of the human races (and hence falls within what is most deserving of the title “knowledge of the world”)? Or is it (also) something else? In short: might physiological anthropology be critically reconceived so as to advance our hands-on understanding of the physical conditions that either promote or retard what Kant elsewhere calls “the thinking self in life” (CPJ, 5: 461)? Such an anthropology would furnish a working, “hands-on” bridge between what “nature does for us” and what we “ought” and “can” do for ourselves, and would thus expand the definition of pragmatic anthropology beyond one of simple opposition to physical geography (as per the titles of Kant’s own paired courses). And it anticipates, as we shall see, his claim, in the note that follows, that his own geographic setting was particularly apt for promoting anthropological insight. Rather than immediately attempting to delimit the boundaries of an anthropology that is specifically pragmatic, Kant turns to “knowledge of the world” (Weltkenntiß) itself, a term that conventionally signified both a general knowledge of things and “knowing one’s way about.” Such an anthropology, considered as knowledge of the world, which must follow knowledge of the schools [auf die Schule], is not yet properly called pragmatic, if it contains extensive cognition of things of the world, for example, animals, plants, and minerals in the various lands and climates, but rather when it contains cognition of the human being as citizen of the world.—Accordingly cognition of the human races as products belonging to the play of nature is not yet counted as pragmatic knowledge of the world but only as theoretical knowledge of the world. (Anth, 7: 120) Pragmatic anthropology excludes extensive knowledge of things, including knowledge of the human races as “products belonging to the play of nature,” which is merely “theoretical knowledge of the world.” Weltkenntnis, then, also has a theoretical component that likewise (like its pragmatic counterpart) follows “knowledge of the schools.” Kant here sidesteps the conventional distinction (which he had himself once echoed) between Weltkenntnis and Schulkenntnis, or “knowledge of the world” and “knowledge of the schools” understood as useless pedantry. He also ignores, for present purposes, the older and arguably more fundamental distinction between secular and sacred wisdom. Progress in culture is itself, as he earlier notes, the great “school” of the human race (Anth, 7:119) to which Weltkenntnis not only contributes but also draws on. Knowledge of the schools thus has a double sense: both the collective scientific output of the universities, to which Kant will later allude, and

110  Susan M. Shell the collective product of the human race qua “culture” generally, or what Kant elsewhere calls nature’s “ultimate end” insofar as nature is judged “reflectively” (CPJ, 5: 431). 2.2.  Circumstantial Obstacles The pragmatic outlook acknowledges at the outset the inseparability, where knowledge of the world is concerned, of viewing the spectacle and playing a part in it.16 But just as Kant seems about to assign pragmatic anthropology the exalted function of thus uniting theory and practice, he breaks off into a digression on the political and social limitations that impede its fulfilling such a role. Knowledge of the world in the “most deserv(ing)” sense must be distinguished from the superficial “worldliness”—especially concerning the so-called great world—with which it is conventionally associated. The expressions “to know/be acquainted with the world” and “to have the world” [die Welt kennen und Welt haben] are rather far apart in their significance, in that one only understands the play that one has seen while the other has played along/participated [mitgespielt].—But in judging the so-called great world, the estate of the nobility, anthropology finds itself in a very unfavorable standpoint, because the nobles are too near one another, but too far from others. (Anth, 7: 120) “To have the world” conventionally meant to assume the manners that mark participation in the higher social ranks. As Kant put it in an early note outlining a prospective pragmatic anthropology: Knowledge of the world [Weltkentnis] is knowledge of human beings [Menschenkentnis]. “He has world” [Er hat Welt] means: he (knows fits) is dressed and turned out in the manner of the court. Vapid head contra the pedants. In order to know the world, one does not even need to travel. One must observe and home and make a study of this. Local knowledge and general knowledge. The former is limited and without rational use. Observation of general/common life [gemeines Lebens]. History and biographies. Threefold teaching: 1. skill [Die Geschikt]; 2. cleverness [die klug]; 3. to make wise [die weise machen]: scholastic, pragmatic, moral knowledge. (some knowledge cultivates; other civilizes, another moralizes. Pragmatic anthropology. Cleverness arises from the community in which we stand with human beings. All other pragmatic sciences are derived from this [borgen daher]. Pragmatic history. (Refl 1482, 15: 659; mid 1770s)

Kant am Pregelflusse  111 This extraordinary early reflection, composed five years prior to the publication of the Critique of Pure Reason, anticipates much of Kant’s later anthropological doctrine, including both the sequence (cultivation, civilization, moralization) stressed in the Idea for a Universal History (1784) (8: 26) and certain central features of pragmatic anthropology as indicated in the current preface, including an emphasis on the relation between “general” and “local” knowledge that will later prove crucial. Written at a time in which Kant could still describe his planned critique of reason as a “anthropologia transcendentalis” (Refl, 15: 395),17 this reflection, along with the extensive remarks surrounding it, can be said to represent a high point in his expectations as to what a philosophic anthropology might yield. At the same time, there are also signal differences between that early passage and what is presented in the Preface. For one: “having (the) world” is no longer, as in the early note, mainly pejorative; instead, Kant now highlights its active, participatory character at the apparent expense of “knowing the world” in the sense of passively observing it. Still, unlike the “play of representations” that the Cartesian spectator has in view, the “play” now at issue is one of human actions, suggesting the possibility of a unity of “knowing” and “having,” observation and action, that is not possible for what Kant here calls the “ratiocinating” theorist.18 Thus Kant’s early note is remarkable in another respect as well; namely, in as much as it was part of an extended reflection on “pragmatic anthropology” written when he was also evidently experimenting with a “transcendental anthropology” to replace the “cycloptic,” “ratiocinating” science of his predecessors (Refl, 15: 395). Was the Critique of Pure Reason that finally emerged the transcendental anthropology thus intended? Or did Kant’s eventual critique succeed, and thereby replace the need for such an attempt, in accordance with the more conventional view? Or, finally, is the summarizing transcendental question offered in the Jaesche Logic19 the realization of that earlier ambition, as Louden tentatively suggests? Whatever Kant’s considered understanding of what it means to “have (the) world” as distinguished from “knowing (it),” knowledge of the world, insofar as it requires acquaintance with the nobility, poses a difficulty for anthropology, which finds itself in an “unfavorable standpoint” for judging that estate, whose members are at once too close to one another and too distant from the rest (Anth, 7: 120). Kant will later provide a solution to this difficulty. For now he offers an alternative means of expanding the circle of one’s intercourse (Umgang)—namely, travel— while also suggesting the sort of “general”—as opposed to “local”— knowledge that must precede it: Traveling belongs to the means of expanding anthropology through social intercourse [Umgang] with others, even if it is only through

112  Susan M. Shell the reading of travel descriptions [Reisebeschreibungen]. But if one wishes to know [wissen] what to seek abroad in order to extend its greater scope one must still first have acquired knowledge of human beings at home through dealings with one’s fellow townsmen or countrymen [Stadt oder Landesgenossen]. Without such a plan [Plan] (that already presupposes knowledge of human beings] the citizen of the world always remains very limited [eingeschränkt] in regard to his anthropology. General knowledge [Generalkenntniß] always precedes local knowledge here, if the latter is to be ordered and directed through philosophy: without which all knowledge can yield nothing more than fragmentary groping around and no science [keine Wissenschaft]. (Anth, 7: 120) Unlike Machiavelli, who professed (with similar irony) to know the great precisely because he viewed them from beneath,20 Kant claims that social distinction is itself an obstacle to characterizing man “as a species.” Travel and travelogues (of which Kant was himself a famously avid reader) are at best partial substitutes for the arrival of the “world citizenship” that would show human beings in their true colors.21 Kant here indicates not only the formidable political and ethical obstacles that are faced by any attempt at a genuinely scientific anthropology but also its situated character—its inability to succeed without, as it were, bringing one’s Umgang with it. 2.3.  Generic Obstacles There are, however, other obstacles, perhaps even more intractable, to the advance of pragmatic anthropology as a scientific subject: namely, difficulties “dependent on human nature” itself (Anth, 7: 120). That such obstacles may differ in both severity and kind from those previously considered is suggested by the hiatus that separates the five earlier paragraphs from those that follow. But all attempts to arrive at such a science that has thoroughness [Gründlichkeit] encounter considerable difficulties dependent on human nature itself. 1. The human being that notices that someone is observing and seeking to study [erforschen] him will either appear embarrassed [verlegen] (made bashful), and then he cannot show himself as he is; or he dissimulates, and then he does not want to be known as he is. 2. Even if one wishes only to study oneself, one will come to a critical position/situation [eine kritische Lage],22 especially as

Kant am Pregelflusse  113 concerns one’s affectual condition, which normally does not allow for dissimulation: namely that when incentives are in action one does not observe oneself, and when one observes oneself, incentives are quiet. 3. Circumstances of place and time (Ort und Zeitumstände], when they are maintained [anhaltend sind], give rise to [bewirken] habits, that are, as is said, another nature and make it difficult for the human being to judge how to maintain himself [wofür er sich halten], but even more, to judge concerning how to form a concept of those with whom he is in exchange [Verkehr]; for the alteration of position [Lage] in which the human being is set down [gesetzt] by fate, or, if he is an adventurer, sets himself [sich . . . selbst setzt], makes it very difficult to raise anthropology to the rank of a formal [förmlichen] science. (Anth, 7: 121)23 All three of these difficulties, which respectively involve self-observation, observation of others, and the transformation of nature by culture, revolve around a common theme. Even if we wish sincerely to observe and to allow ourselves to be observed, our “situation” (Lage) as human beings seems to make a gründliche or ‘thorough and well-grounded’ science of man impossible. Every science, if it is not to be a mere “fumbling about,” must be guided by a plan and hence an idea. It seems, then, that in order to begin to study human beings systematically, we must already have a rational concept of what the human being is—to begin to study the human being, it seems, we must already know what she is capable of. But our ability to know the human being in this sense is precisely what is at issue here. The notion of human being as merely one of the earth’s creatures and that of it as an earthly being endowed with reason cannot obviously or easily be brought together into a single conceptual whole. The science of anthropology thus seems to involve us in a vicious circle. The anthropological “situation” is “critical” not only for particular kinds of psychological self-observation (whose peculiar hazards Kant will later note) but also generally for all of humanity’s self-knowledge. Each generic obstacle also raises difficulties peculiar to itself. The embarrassed self-consciousness of one who knows that he is being studied raises the issue of naivete more generally, one that the Critique of Judgment had touched on (CPJ, 5: 335) and that Kant will discuss at greater length in the body of the present work (Anth, 7: 132). The phenomenon of naïveté reveals with particular poignancy the dishonesty that blights even human being’s most conscientious efforts at sincerity. Equally pressing is the problem of self-observation, especially with respect to “affects,” which are by their nature hard to simulate. And yet what is advantageous for purposes of observing others is disadvantageous in one’s own case; since “when incentives are in action, one does

114  Susan M. Shell observe them, and when one observes them, they are not active.” In the body of the Anthropology, Kant will detail one way in which this “critical situation” can be overcome: namely, where the affect in question is an enthusiasm of “of good intention,” in which “reason still holds the reins” (Anth, 7: 254). An affect of this sort calls to mind the spectatorial enthusiasm (Enthusiasmus) that signals, according to the Conflict of the Faculties, mankind’s constant progress toward the better (CF, 7: 85); and it reconciles, perhaps uniquely, rational composure (animus sui compos) with the artless simplicity ordinarily found in the naïve (Anth, 7: 132). It also suggests how history progressively conceived might, precisely by uniting the respective roles of spectator and participant, provide a prospective answer to the question: ‘what is the human being?’24 Finally, circumstances of time and place impede human judgment with respect to both self and others: in the former case, as to how (or where) to maintain oneself; in the latter, of how to objectively conceive those with whom we deal. Alterations of location especially, whether it be the location in which fate has set us or one chosen adventitiously, makes it “very difficult” for anthropology to be a science at all. That time and location (Ort) (and hence the peculiarly situated character of anthropology) should raise special obstacles to the elevation of anthropology to the status of a “formal science” should not surprise us. For their vagaries, whether fated or adventurously chosen, bring home the sheer contingency of human life, a contingency that seems to stand in the starkest contrast with the a priori, systematic unity demanded of a formal science.25 When one adds these three generic obstacles to the class-related conditions, peculiar to his own place and time, that Kant had earlier described, the following conclusion suggests itself. Pragmatic anthropology cannot take a “concept” of the human being for granted in advance but must proceed from worldly intercourse (Umgang). To know what to look for abroad, one must already have acquired knowledge of people at home through Umgang with one’s neighbors and countrymen. But whence, then, the “general knowledge,” preceding local knowledge, that is necessary if anthropology is to become a systematic science?26 Whence the primary “knowledge of man” and accompanying “plan,” absent which “acquired knowledge,” however extensive, “cannot give rise to science” (Anth, 7:120)?

3. Kant am Pregelflusse Kant includes, in his otherwise terse prefatory statement, two extended autobiographical footnotes, that are both remarkable for their spatial and temporal specificity, and entirely without precedent in the many earlier versions of the Anthropology. And their inclusion here, from an author who generally avoided self-reference, suggests that they might

Kant am Pregelflusse  115 play a systematic role in accounting for how, given the formidable obstacles just cited, a formal science of anthropology might proceed. Local knowledge presupposes general knowledge; and both, if they are to give rise to a formal science, must be guided by a plan—one that bears a notable relation, as we shall see, to Kant’s own worldly Umgang, both geographic and professional. He describes the site of his own worldly “Umgang” as follows: A great city, at the center [Mittelpunkt] of a kingdom [Reichs], in which is found the government’s provincial councils, and that has a university (for culture and science) and thereby also a location [Lage] for sea trade—that, by way of rivers [Flüsse], favors exchange [Verkehr] both with the interior of the country [Landes] and with neighboring [angränzenden] outlying countries with different languages and customs—such a city as Königsberg on the Pregel River [Königsberg am Pregelflusse], can well be taken for a fitting place [schicklich Platz] for the expansion of knowledge of the human being as well as knowledge of the world, where this can be acquired [wo diese . . . erworben werden kann] even without traveling. (9: 120–1n; emphasis added) Kant here takes pains to draw attention to his own especially favorable location “for the expansion of knowledge of the human being as well as knowledge of the world,” or, in other words, both for the pursuit of “anthropology” as a “formal science” and for “worldly knowledge” generally. If such a pursuit is to succeed, it must combine knowledge of the “local” with a prior knowledge of the “general”: an intersection, as it were, of position (Ort), situation (Lage) and place (Platz), with their various connotations, both colloquial and systematic. Thanks to his fortunate place of birth and life-long domicile, favoring exchange both material and scientific with countries near and far, Kant is positioned—even without traveling—to overcome the more obvious circumstantial obstacles that ordinarily make elevation of anthropology to a formal science “very difficult.” In sum: Kant’s own geographic location was, as he here asserts, unusually “favorable” for acquiring the “general” knowledge of the human being without which “local” knowledge is scientifically useless. At the same time Kant’s claim as to Königsberg’s geographic and civil centrality to “a kingdom” poses a certain puzzle, as Brandt notes (Brandt 1999, 120), given that the political capital of the Kingdom of Prussia at the time was Berlin rather than Königsberg, which was then the political seat of the Province of East Prussia, recently expanded following the partition of Poland. The city had, to be sure, been the official capital of Prussia from 1525 to 1701, and also remained an important administrative and commercial center into the nineteenth century. Originally named for

116  Susan M. Shell the castle (or “king’s mountain”) built by the Teutonic Knights in 1255, the three surrounding settlements (Altstadt, Kneiphof, and Löbenicht) were municipally united only in 1724, the year of Kant’s birth, making Kant and his city twins, so to speak. Also known for the seven bridges suspended across the Pregel and its tributaries (giving rise to the mathematical puzzle that Euler solved in 1736) and across which Kant took his famous daily walks, Königsberg punched above its weight geographically, culturally, and internationally, owing to the historic influx of Lithuanians, as well as Scottish and Hanseatic merchants, followed in the more immediate past by recently incorporated Poles, Lithuanians, and Jews. Still, it seems a stretch to describe Konigsberg’s position as the at the center of a kingdom,27 unless Kant was using “Reich” in a more general sense (as in the “kingdom of ends,” the “animal realm” (Announcement of a course on physical geography 2: 9), etc.). And yet this too seems farfetched, unless Königsberg can also be conceived as the midpoint of an expanding “realm” of science and culture. That possibility—namely, that the “kingdom” in question is as much cultural and spiritual as it is administrative or juridical—is supported by Kant’s second footnote, which marks his own crucial role as teacher and presenter, while also drawing attention to an infirmity that situates him not only spatially but also temporally, and at the threshold of senescence and an accompanying illegibility to others. In my occupation [Geschäfte] with pure philosophy, at first freely undertaken, later as part of my teaching duties, I have for some thirty years given lectures aimed at knowledge of the world twice a year, namely (in winter) Anthropology and (in summer) Physical Geography, which, as popular presentations also found themselves recommended to other estates; the present handbook is from the former; as for the latter, it is scarcely possible at my age to produce a manuscript from my text which is hardly legible to anyone but myself. (Anth, 7: 122) Thanks to the “popular(ity)” of his instruction, Kant is acquainted with the upper classes (who were known to have attended his lectures and presentations in significant numbers), overcoming yet another serious obstacle that ordinarily makes the “standpoint” of anthropology “very unfavorable.” (Anth, 7: 120) For as members of his audience, intermingled with his other students, persons of noble rank are neither too near one another nor too distant from the rest to be studied and observed, making it possible to both “know the world” and also “have it.” But Kant’s second note is important in another respect as well: namely, in drawing attention to his own purely theoretical concern, an occupation “freely” assumed before being undertaken as an academic duty. Kant’s

Kant am Pregelflusse  117 occupational interest is not only, or primarily, in anthropology but also in “pure philosophy,” and hence in theoretical knowledge of the world in the broadest sense. That sense is not limited, then, to the topic of physical geography (the subject with which Kant’s consecutive courses on anthropology were paired academically), but includes the subject matter with which he was philosophically occupied prior even to his appointment as an extraordinary university lecturer in 1755; and it would embrace such early cosmological works as the Universal Natural History and Theory of the Heavens, and the New Elucidation, not to mention briefer cosmological and geological writings of the early to mid-1750s. In this respect as well—namely, with respect to “knowledge of the word” in a comprehensive sense—Kant’s “great city” seems to have proved an especially “fitting place.”28

4.  Knowledge of the World am Pregelflusse I now come to the most difficult—and frankly speculative—topic of my essay: namely, how, if at all, Kant’s situation might have fostered his evolving theoretical cosmology, and, ultimately, his final and most comprehensive understanding of ‘how to orient oneself in thinking.’ The issue is not the one of merely antiquarian biographic interest (along the lines of Newton’s fabled encounter with the apple); for if Kant had good reason to regard his own ‘fated’ placement in the world as having been especially conducive to acquiring knowledge of the world in a comprehensive sense, it would provide a more satisfying systematic answer to the question earlier left hanging: namely, how “general knowledge” of the world, and an accompanying theoretical “plan” (or map), for the expansion of pragmatic anthropology, and hence the latter’s elevation to the status of a “formal science,” might be acquired—and acquired, indeed, “without traveling” beyond one’s immediate birthplace. Such an inquiry is aided by a remarkable fact: namely, the appearance of the river Pregel not only in the above-quoted autobiographical footnote but also (whether explicitly or by implication) in certain key examples by which Kant chose to illustrate his evolving theories of space and motion, culminating in his famous juxtaposition, in the Second Analogy section of Critique of Pure Reason, of the “subjective sequence of apprehension” and the “objective sequence of appearances.” Suppose, he here writes: I see a ship driven downstream. My perception of its position [Stelle] downstream [unterhalb] follows the perception of its position upstream [oberhalb dem Laufe des Flusses], and it is impossible that in the apprehension of this appearance the ship should first be perceived downstream and afterwards upstream. The order in the sequences of the perceptions in apprehension is therefore here

118  Susan M. Shell determined, and the apprehension bound to it. In the previous example of a house my perceptions could have begun at its rooftop and ended at the ground but could also have begun below and ended above; likewise I could have apprehended the manifold of empirical intuition from the right or from the left. (CPR, A192/B237–8) Objective experience depends on one being able to distinguish, in accordance with the law of causality, between consecutive representations whose order is subjective, or up to us (as with consecutive apprehensions of the door and roof of a house, or its left window and its right) and sequences that are fixed in accordance with a rule independent of human choice (as with consecutive apprehensions of the positions of a ship relative to the fixed bank of a river (like the Pregel) that flows in only one direction [in this case from east to west, or right to left on a standard map]).29 The importance of the Second Analogy to Kant’s overall account of objective experience is generally conceded; less often noted is the peculiar (and sole) examples that Kant chose by way of illustration: a moving ship driven downstream on a river versus a stationary house. But this was not the first time that Kant had used the image of a moving ship against a fixed, directionally stable background to illustrate a fundamental theoretical principle. The passage from the Second Analogy, with its implicit autobiographical resonance, is anticipated by Kant’s only explicit published reference to the Pregel River, other than the aforementioned footnote to the Anthropology: in the brief 1758 essay on a “new doctrine of motion and rest” that accompanied his announcement of courses for the upcoming semester. (The full title reads: New doctrine of motion and rest and the conclusions associated with it in the fundamental principles of natural science while at the same time his lectures for this half-year are announced) Despite its brevity, Kant’s essay on motion and rest is theoretically pertinent for present purposes in three ways. First, it represents Kant’s earliest extended treatment of issues of motion and rest, relative and absolute space, that would continue to preoccupy him through his Metaphysical Foundations of Natural Science (1786) and beyond. Second, it presents a critique of Newton’s theory of absolute space that does not rely on the Leibnizian argument that motion and rest are derivative phenomena reducible to conceptual relations among monads. Instead, Kant emphasizes the relativity of all motion, given the infinite expandability of the spatial framework against which motion, understood as a “change of position [Ort]” (NDMR, 2: 16), can be measured. Third, he argues for what has been called “strong relationalism” (or motion’s irreducibly dyadic character, a view he shares with Huygens), as distinguished from the “weak relationalism” of his rationalist predecessors, for whom motion is “relative” in the sense of reducing to relations among metaphysical monads.30 Motion in a body

Kant am Pregelflusse  119 according to Kant can only be understood in relation to that of another body, making the motion of a single body a meaningless thought.31 That this is so can be shown directly, he thinks, through an analysis of what occurs when two bodies collide, rather than resting the argument (as with Leibniz and his followers) on independent metaphysical claims about the ultimate nature of reality; for example, such as the reducibility of motion to conceptual relations among non-extended, metaphysically independent monads. The latter two features of Kant’s argument are highlighted in the following passage, which for present purposes is worth quoting at length. Having “set (himself)” in the “framework” (Stellung) of “mere healthy reason” (blosse gesunde Vernunft), Kant continues: I soon grasp that the position [Ort] of a thing is known [erkennt] through its location [Lage], through its position [Stellung], or through its external relation to other things around it. Now I  can consider a body in relation to its nearest neighboring objects, and, if this relation does not change, I will say that it is at rest. But as soon as I regard it as in relation to a sphere of greater size, it is possible that that body will alter its position with regard to the objects close to it, and from this point of view [Gesichtspunkte] I will impart [mittheilen] motion to it. (NDMR, 2: 16) Continuing in this vein, Kant proceeds to urge that one is “free” to expand one’s “circle of vision” as “far as one wishes” without ever coming to an absolute resting point. In thus considering my body in relation to ever more distant circumferences, I “grasp that my judgment of the motion or rest of this body is never constant but can always be altered through new perspectives” (2: 16). Kant illustrates such perspectival alternation between motion and rest with the following example: Assume [setzet] . . . that I find myself on a ship lying at wharf on the Pregel. I have a ball lying on the table in front of me. I consider it in relation to the table, the walls, and the other parts of the ship and say it is at rest. Soon afterward, I look from the ship to the (river) bank [Ufer] and notice that the rope [Tau] by which it was secured has been untied, and that the ship is slowly being driven downstream by the current [Strom]; I then say: the sphere is moving, and indeed from east to west in accordance with the direction of the river. But someone says to me that the earth in its daily rotation turns with much greater velocity from west to east; I then change my mind and attribute [lege . . . ein] to the sphere an entirely opposed motion at a velocity that can be determined by the science of the stars [Sternwissenschaft]. But I  am reminded that the entire ball of the earth

120  Susan M. Shell is in even more rapid motion from west to east with regard to the structure of the planets. I am required to attribute this motion to my ball, and alter the velocity that I  previously ascribed to it. Finally, Bradley32 teaches me that the entire structure of the planets together with the sun is probably moving in relation to the fixed stars. I ask according to which side [nach welcher Seite] and which velocity? No one answers. And now I  become dizzy [schwindlicht]; I  no longer know whether my ball is at rest or in motion, toward where [wohin] and at what velocity. Now I begin to have insight that something is lacking in the expression “motion and rest.” I  should never use it in absolute terms, but always relatively [respective]. I should never say: the body is at rest, without also setting down in regard to which thing it is at rest; and never say it moves, without naming the object in regard to which it alters its relation [Beziehung]. Even if I should imagine a mathematical space empty of all creatures as a container for the bodies, it would not help me. For how might I distinguish its different parts and places [Plätze] that are not occupied by anything corporeal? (NDMR, 2: 16–17) Only given a pair of bodies is it possible to speak meaningfully about the motion or rest of either, and this, too, only respectively, rather than absolutely, given the inadequacy of absolute space mathematically conceived to provide the necessary frame of reference (NDMR, 2: 17). Kant’s students would have been familiar with the main University building (called the Albertina), which was situated on an island fronted on all sides by river canals, and would have been immediately familiar with his example and with the intuitive necessity, when judging motion or rest, of at least two objects of reference: for example, a river and its banks. In such a setting, the respective character of motion and rest is easily brought home by the illusions that thereby readily arise: for example, the illusion, in cases such as the aforementioned, that the bank is moving rather than the ship on which one is a passenger. Kant’s alternating train of thought culminates in a dizzying loss of all spatial orientation—reinforced, perhaps, by Kant’s admitted susceptibility to vertigo33—strikingly contrasts with a similar passage in the Metaphysical Foundations of Natural Science, composed 30 years later, and after the issue of absolute versus relative space had been critically settled. In that later passage, in which the Pregel is again implicitly in play,34 the accompanying mental experiment ends not in Schwindel, but in the steadying conclusion that “absolute space is nothing for all possible experience” though it remains necessary as a “regulative idea” (MF, 4: 487–8; cf. CPR, A689/B717). Kant’s earlier conclusion as to the inability of an imagined empty space to furnish measurements of motion and rest with an absolute frame of

Kant am Pregelflusse  121 reference would be amended ten years later, in the important essay On the ultimate ground of the differentiation of regions [Gegenden] in space (1768). Although his official position in that essay was that space was absolute (as Newton claimed and the Leibnizians denied), his argument was novel, laying the ground for his famous doctrine of the ideality of space (and time) that would first appear publicly in the Inaugural Dissertation of 1770. The novel argument of the 1768 essay rested on a new insight as to fundamental directionality of space, including, in the first instance, that occupied by our own bodies, which we “intuitive(ly)” recognize as having non-interchangeable, left and right sides. As Kant now insists: [I]n the most abstract [abgezogensten] sense of the term, region/direction [Gegend] consists not in the relation of one thing in space to another thing, which is properly the concept of position [Lage], but rather in the relations of the system of this position to absolute world space. UG (2: 377) Or, as he now wishes to prove one basis of what he calls “intuitive judgments about extension,” that “Absolute space, independent of the existence of all matter and as itself the first ground of the possibility of matter’s compounding (Zusammensetzung] has its own reality” (UG, 2: 378). That Kant’s proof for the independent reality of absolute space is itself based on inner (human) feeling—indeed, one as intimate as the “perceptible beating of one’s own heart” (UG, 2: 381)—points in a new philosophical direction that one is tempted to call anthropological. Such a direction would not be surprising, given the “revolution” in thinking, inspired largely by Kant’s reading of Rousseau (Notes, 20: 44), that separates the two essays. This temptation yields an additional suspicion: namely, that an intervening description of the human condition, in a series of remarks that strongly reflect Rousseau’s influence, may be relevant to our present inquiry. In that passage Kant compares the human world that surrounds us to a flowing river, without fixed points of nature or other indications as to where to stop: Everything goes by us in a flux/river [Flüsse] and the changing taste and differing shapes of human beings makes the whole game uncertain and deceptive. Where do I find fixed points of nature that the human being can never disturb in order to give himself signs as to which bank [Ufer] he should stop at [er sich zu halten hat]. (Notes, 20: 46) The discussion of relative and absolute cosmological magnitude that immediately follows (Notes, 20: 46) gives the individual human body a

122  Susan M. Shell primacy lacking in his earlier account of motion and rest—an anthropological reorientation born out elsewhere in the Remarks, which locates science’s necessary role in “teaching (man) properly to fulfill (and) the position [Stelle] that has been assigned to him in creation” rather than in stargazing: If there is any science that the human being needs it is that which teaches him properly to fulfill (and) the position [Stelle] that has been assigned to him in creation and from which he can learn what one must be in order to be a human being. Suppose [Gesetzt] he had become familiar with deceptive temptations over or under him that had, unnoticed, brought him away from his proper position, then this instruction would lead him back again to the state [Stande] of a human being, and however small or lacking he may then find himself he will still be quite good for his assigned post [Posten] because he is exactly what he should be. (Notes, 20: 45–6)35 Hereafter, as Kant here announces, he will set his Archimedean lever in neither this world or the next but “in a state of freedom”: The question is, whether in order to move my affects or those of others, I  shall take my point of support [Stützungspunkt] outside the world or in it. I answer: I find it in the state of nature, i.e., of freedom. (Notes, 20: 56) Kant would soon return to theoretical philosophy with renewed vigor, albeit one chastened with gained insight into the limits of human reason. Those limits, already evident in the essay on Directions in Space, would take more definitive form in the transformation of absolute geometric space from something independently real (Kant’s position in the 1768 essay), to a merely ideal though necessary condition of human knowledge (his position in the Dissertation of 1770 and all subsequent works). It was following upon the latter transformation, and with the ideality of space as the “mere formal condition of all appearances of outer senses” (such that “only from the human standpoint . . . we can speak of space, extended beings, etc.”) (CPR, A26/B42), firmly in place, that the Pregel river and its banks would reappear, albeit anonymously: both in the Metaphysical Foundations of Natural Science (MF, 4: 487), and in the Second Analogy section of the first Critique, of our sequential representations of the parts of a house to our sequential representations of a ship upstream and downstream. In that famous illustration of “the principle of temporal succession in accordance with the law of causality” (A 189/B 232), one can easily imagine Kant himself, consecutively attending

Kant am Pregelflusse  123 to the houses flanking the Pregel and to the ships driven by its steady downward current. But unlike his earlier invocation of the flowing river in illustration of the relativity of motion (NDMR, 2: 16–7), the current thought experiment, as with the similar riparian reference that would follow five years later (MF, 4: 487), does not culminate in “dizziness” (Schwindel, vertigo), thanks to Kant’s critical understanding of space as an a priori form of intuition offering a stable frame of reference for the apprehension of all bodily motion, one’s own included.36 And it would be followed, five years later, by the return of “absolute space” as a necessary regulative idea (MF, 4: 486–7). In sum: Kant’s riparian location seems to have proved especially favorable, in ways he marks, to the emergence of a series of theoretical insights, including the respective relativity of motion, the directionality of space; and the irreversibility that distinguishes objective from subjective sequences of apprehension. Such references, whether explicit or implied, lend further textual support to his own description of his own geographic setting as particularly “apt” (schicklich) for the “expansion” of “worldly knowledge,” including that aspect that pertains to “pure philosophy” (Anth, 7: 121–122n).37 It also seems to have provided him with a ready metaphor by which to express the philosophic uncertainty, and subsequent anthropological turn, that caused him briefly to abandon speculative metaphysics, and ultimately led to his critical reconceptualization of the “proper ends” of human reason.

5.  Vertigo and Seasickness But these are not the only passages that mark (whether directly or implicitly) his situation am Pregelflusse as one especially favorable to philosophic insight. The stabilizing regulative idea of absolute space calls to mind a third and final personal note, appended to the section of the Anthropology devoted to imagination (Einbildungskraft), whose force can be deliberately aroused or arise unbidden; as he goes on to note: I [here] pass over [übergehe] what is not a means to a goal [Absicht], but the natural consequence of the situation [Lage] in which someone is placed [gesetzt], and through which his mere imagination decomposes him [ihn ausser Fassung bringt]. To this belongs dizziness [Schwindel] from looking down from the edge of a steep height (and also even from a narrow bridge without railings), and seasickness. . . . Sea sickness, with its attack of vomiting, (to which I  myself experienced in a journey I  made from Pillau to Königsberg—if one can even call it a sea voyage),38 which came, as I believe I observed, merely through the eyes. Because the swaying of the ship seen through the cabin made me see now the (Vistula) lagoon, now the summit of Balga,39 and the recurrent falling after rising aroused,

124  Susan M. Shell by means of the force of imagination, an antiperistaltic movement of the intestines by the stomach muscles. (Anth, 7: 169n.) As Kant repeats, in less personal, geographically specific, mode, later in the work: with seasickness, as with vertigo, imagination brings about the real danger that it causes us to fear (Anth, 7: 263–4; cf. 169n.). It is thus a close cousin to “shuddering” (Schauern), albeit without the latter’s accompanying pleasure (and accompanying affinity with the sublime) (Anth, 7: 263). Read in light of his earlier reversions to the first person, one might conclude that the geographical and personal specificity of those earlier footnotes (Anth, 7:120–122n)—far from counting as regrettable lapses into the very “egoism” Kant proceeds to condemn (at the beginning of Book One, Anth, 7: 128–30)—are themselves local manifestations of the “general,” not to say “universal,” knowledge necessary to launch anthropology as a formal science. That a “sea-journey” in protected waters that did not extend beyond Kant’s own immediate “kingdom” could arouse such physical revulsion and sense of danger (in one whose susceptibility to vertigo and similar infirmities verged, by Kant’s own admission elsewhere,40 on the pathological), might well have seemed a fated, if not providential, sign to remain in place. Such, at least, was the excuse that Kant offered for refusing a professorship at Halle, after rejecting an earlier offer from the University of Erlangen, each of which might have enhanced his professional career, at least in the short run. As he wrote to Marcus Herz,41 his personal physician and former student: [In Königsberg] I have attained [gehalten habe] everything I wished for, namely a peaceful situation that is exactly suited to my needs, in turn possessed [besetzen] with work, speculation, and society [Umgang], where my easily affected, but otherwise carefree, mind and my even more capricious body, which however is never sick, will be occupied without strain. All change makes me anxious [bange], even if it gives the greatest semblance of improving my condition. I believe I must pay attention to this instinct [Instinkt] of my nature, if I still want to draw out somewhat the length of the thread that the Parcae [die Parzen] spin very thinly and delicately for me. (Letter to Marcus Herz, early April, 1778, L, 10: 231) Unlike Kant’s earlier, dizzying imaginative ascent beyond the fixed stars, culminating in the insight that “something is lacking in the expressions ‘motion’ and ‘rest,’ ” (NDMR, 2: 17) and subsequently corrected by his discovery of the ideality of space as the a priori form of intuition, in the cases of both vertigo and seasickness the effect of the imagination is both

Kant am Pregelflusse  125 morbid and unbidden.42 Indeed, that the transition (as in “I  here pass over [übergehe] what is not a means to a goal”) between voluntary and involuntary effects of the imagination at (Anth, 7: 169n.) proves to be merely a postponement, presaging his later treatment of that very topic in impersonal terms. In thus linking the personal and the impersonal, Kant calls attention to the transition between the local and the general without making their relation itself thematically explicit. This distinction between voluntary and involuntary perceptions, moreover, calls to mind that between the apprehensions of the simultaneous features of a house and the consecutive locations of a ship—a contrast Kant used to exemplify the related distinction between the subjective and objective order of succession as a condition of the very possibility of objective experience. To be out of “Fassung” (as with vertigo and seasickness) is to find oneself outside the framework that normally maintains the boundary between inner and outer sense and with it a stable means of regulating our own worldly interactions. At the same time, given favorable conditions (e.g., drifting down the Pregel rather than being tossed at sea while returning to it), and when put to work deliberately, the susceptibility to dizziness and nausea of which Kant complained to Herz may foster transcendental insight. Moreover, that Kant proved able to make deliberate philosophical use of such unbidden thoughts and feelings, which themselves limn the boundary between empirical psychology and physiology, to good effect, offers an exemplary alternative to the “ratiocinating” anthropology that he had earlier criticized (Anth, 7: 119).43 By thus uniting “knowing the world” and “having it,” in other words, Kant subsumes “physiological anthropology” within worldly knowledge in the sense most “deserv(ing)” of the name (Anth, 7: 119). Kant thereby can be said to model the proper articulation of relation between the study of physical geography, on the one hand, and the study of pragmatic anthropology on the other—subjects without whose appropriate juncture the elevation of anthropology to a formal science cannot proceed.

6. Conclusion A revealing sense of the distance Kant had travelled intellectually (if not physically) from his earlier period of anthropological study to the present one can be gleaned from a quick comparison of the Preface with the more expansive Introduction to the Lectures of Physical Geography, evidently based on a text from the mid-1770s.44 Like the briefer Prolegomena to the roughly contemporaneous Anthropology Pillau (1777/8), that Introduction lays out a confident program of “world knowledge” of which physical geography and pragmatic anthropology are mutually dependent, jointly comprehensive parts. Just as there is both inner and outer sense, so, correspondingly, is there nature (the object of the former) and

126  Susan M. Shell the human being (the object of the latter) (Physical Geography, 9: 156). Together, these two sciences, according to that earlier work, constitute a “propaedeutic for knowledge of the world” that by “anticipating our future experience in the world” gives us “a pre-conception (Vorbegriff), as it were, of everything,” a “plan of travel in advance.” (Physical Geography, 9: 157) The Prolegomena to Anthropology Pillau echoes these remarks in terms of a distinction between the local and the general, the former “empirically obtained,” is that enjoyed by merchants, the latter, which “the man of the world (Weltmann)” possesses, is not “empirical but rather cosmological.” (V-Anth Pillau, 25: 734) One does not yet see, in these early lectures, the complication in relating the inner (or subjective) and the outer (or objective) of which the first Critique will offer definitive expression (e.g., in the section devoted to the Second Analogy). Nor does the accompanying table of contents yet reflect the maturer articulation of their relation, as it appears in the final, published version, which assigns Part One (Anthropological Didactic) the secondary title “of the way of knowing internal as well as the external of human beings” and Part Two (Anthropological Characteristic) “on the way of knowing the internal of human beings from the external.” The later Kant, by way of contrast, no longer takes for granted that the “general” can be set forth prior to knowledge of the “local.” Nor does he compare physical geography and pragmatic anthropology to outer and inner sense, respectively, or promise a “fore-concept” of world knowledge. Instead, along with delicate indications of his own waning powers, the Preface yields a sense of gratitude for his (and our) propitious beginnings. If anthropology as a formal science is to be possible at all, local and general knowledge must arise, at least initially, in tandem. In short: the “idea” of a pragmatic anthropology is no longer “cosmological” but “cosmopolitan”; or rather, the cosmopolitan now encompasses the cosmological, both subsumed, as he puts it in the nearly contemporaneous Jäsche Lectures on Logic, within the larger question “what is the human being?” ( JL, 9: 25). The peculiar setting to which the Preface draws attention not only proved an apt point of departure for Kant’s early philosophical program, as outlined in a letter to Herz; it also proved to be especially, and perhaps uniquely suited, as he finally saw it, for articulating such a plan—that is, for overcoming the obstacles, both generic and specific, to raising pragmatic anthropology to the status of a “formal science” whose content is to be filled in progressively (and largely popularly) (Anth, 7: 121) by present and future generations. Only by such mass productive means, as Kant puts it at the Preface’s closing, by which [T]he labors are divided automatically [von selbst] among amateurs [Liebhaber] of this study and through the unity of the plan gradually

Kant am Pregelflusse  127 united into a whole, by which the growth of the generally useful [gemeinnützigen] science is promoted and hastened. (Anth, 7: 122) Pragmatic anthropology must proceed, in other words, from a setting that initially unites conditions for the acquisition of local and general knowledge, otherwise widely dispersed across space and time, in a single, spatiotemporal location. But Kant’s situation am Pregelflusse not only exhibits, in a single determinate site and concrete moment, the effectual reality of the a priori categories of quantity (universal, particular, singular) from which any objective formal science must take its primary bearings; it does so for a science that has the additional peculiarity, of making, as it were, its own truth. Pragmatic anthropology, according to the Preface, is knowledge of what man “makes of himself, or can and ought make of himself” (Anth, 7: 119). What man is can thus only be known by what man does in the course of history, be it one of cosmopolitan progress or of the lack thereof. Pragmatic anthropology as a science is set apart not only by the unique conditions of its founding but also by its necessary inconclusiveness, as indicated in the work’s sardonic final paragraphs (Anth, 7: 333). Not explicitly noted in the Preface, though indicated elsewhere, Kant’s location on a city of islands, surrounded by streams of water flowing in one constant direction, seems to have fostered both his early discoveries as to the relativity of motion and rest and his later, critical articulation of the relation of inner and outer sense, as well as bringing the law of causality into concrete focus. That the latter, critical insights played a crucial role in the ultimate subsuming of cosmology within a larger “cosmopolitan” framework of world knowledge further testifies, if less directly, to the pertinence of the Preface’s striking footnotes as to the aptness of a situation first given with his birth (as the place where he was “set down” (gesezt) by fate), and later freely chosen (as the place in which he “set himself” [sich . . . selbst setzt]) (cf. Anth, 7: 121). Also worthy of consideration in this regard—beyond knowledge of the world in the most comprehensive (cosmopolitan) sense—is that pragmatic worldly knowledge whose disjunctive parts consist of physical geography and pragmatic anthropology, respectively, or what Günter Zöller has felicitously called Kant’s “geo-anthropology.”45 The relation between what we are given and what we can, do, and ought to make of what we are given—a theme that runs like a red thread through all Kant’s work—was addressed, at more popular level, by the tandem courses he offered for over 30  years to beginning students and general auditors. Although presented under the two-fold heading of “physical geography” and “pragmatic anthropology,” they together constitute a unity, as he insisted in the Preface to the published Anthropology. Why, then, did he never make the principle underlying their unity explicit—let alone teach

128  Susan M. Shell a course under a combined heading with a single, integral plan? Perhaps, I  would suggest in closing, because the unity in question between the local and the general, or between what is given and what we can and ought to make of it (or between the realms of nature and of freedom), cannot be stated formulaically, but only modeled or exemplified. 6.1.  A Final Note What Kant leaves unspoken in the summary question: “what is the human being?,” and is today rarely noted (though it surely echoed in the minds of Kant’s initial readers) is the Biblical quotation from which Kant may have lifted it: namely, “What is man that Thou art mindful of him, the child of man that Thou carest for him?”46 If one understands Kant’s question in this light, his Preface for “lovers” of a unified science of “general utility”(or, alternatively, serving “the public good”) can also be read as a long delayed answer to the challenge once posed by his late friend and neighbor, Johann Georg Hamann. Hamann, newly lapsed at the time from their shared adherence to the broad principles of the enlightenment, had cautioned Kant (under the ironic mantle of “Socrates”) against giving way to “polypragmatic” sloth and vanity.47 If there is merit in this thought, it might indicate yet another way that Kant’s peculiar location proved apt, namely, by harboring within its neighborly ambit an ongoing and friendly spiritual goad.48

Notes 1. With special thanks to Professors Vadim Chaly and Nina Dmitrieva of the Immanuel Kant Baltic Federal University, and to the sponsors of the Twelfth International Kant Conference, Kaliningrad, Russia. 2. Foucault (2008). 3. A partial list of important recent studies includes works by Günter Zöller, John H. Zammito, G. Felicitas Munzel, Paul Guyer, Robert B. Louden, Holly Wilson, Richard Velkley, Peter Fenves, Alix Cohen, Patrick R. Frierson, Allen Wood, Scott M. Roulier, and Jennifer Mensch, among others. 4. Louden (2009, 60). See also JL, 9: 25. 5. See, for example, MM, 6: 281, 354, 363, 391, 429, 433, 468, 483, 485; Ped, 9: 945; Immanent Arrival of a Peace Treaty in Philosophy, 8: 421. 6. See, for example, CPJ, 5: 479. 7. Kant’s usage in the 1780’s exclusively follows this pattern; see, for example, CPR, A824/B852), and G, 4: 417. 8. See, for example, MM, 6: 444, 455, Ped, 9: 157. 9. See especially Kant’s treatment of “(pragmatic) friendship” at MM, 6: 472. 10. See TPP, 8: 380: the prospective execution of “rightful ideas” on a “pragmatic” basis culminates in the “despairing” thought that such a being as man should never have been created. 11. For example, CPR, A 800/B 828; A 806/B 834; A 824/B 852. See also A849/ B877 on “complete anthropology” as a temporary “pendant” to the empirical doctrine of nature, a doctrine which is eventually “to occupy its own domicile.” 12. Cf. CPJ, 5: 283, 319, 458.

Kant am Pregelflusse  129 13. Kenntnis and Erkenntis (along with related terms) will be translated as either cognition or knowledge, depending on the context and to facilitate fluency in English. 14. Compare Critique of Judgment, section  83 (CPJ, 5: 443–46); that nature has a final (letzte) end—namely human culture—is a reflective judgment ultimately supported by the moral end (Endzweck) the human being gives herself. 15. See Conflict of the Faculties, Part Three (The Conflict between the Philosophic and the Medical Faculties, or “how to overcome morbid feelings through sheer will-power.”) 16. Cf. CF, 7: 87, 82. 17. In regard to what he there calls the “critical physician (Medicus Critik)” Kant writes: “Not strength, but one-eyedness makes (the egoistic scientist) cycloptic. It is also not enough to know many other sciences, but rather the self-knowledge [Selbsterkenntnis] of understanding and reason. Anthropologia transcendentalis.” (Refl 903, 15: 395 (1776–78)); see also Brandt (1999, 117). 18. Kant later defines “rationalizing” [Vernünftelei] as “a use of reason that passes by its final end, partly due to impotence, partly due to a lapse in viewpoint [Verfehlung des Gesichtspunkts].” (Anth, 7: 200) 19. “The field of philosophy in (the) cosmopolitan sense can be brought down to the following questions: 1. What can I know? 2. What ought I do? 3. What may I hope? 4. What is man? Metaphysics answers the first question, morals the second, religion the third, and anthropology the fourth. Fundamentally, however, we could reckon all this as anthropology, because the first three questions relate to the last one.” ( JL, 9: 25) 0. Niccolo Machiavelli, The Prince, Dedicatory Letter. 2 21. Cf. Rousseau, Emile, 4: 428–32 (“On Travel”). 22. Kant often uses Ort, Lage, Stelle, and Stellung, along with the Latin terms situs, locus, and posito, more or less interchangeably to mean “site,” “location,” “position,” “situation,” or “place.” In the Metaphysical Foundations of Natural Science, his usual term for “position” (as in characterization of motion as a change of position (MF, 4: 482)) is “Ort,” but he also sometimes uses “Stelle” and “Lage” with similar meaning (e.g., at MF, 4: 519–20, 552). Kant’s clearest, albeit very early, discussion of the difference between “Ort” and “Lage” would appear to be in On the True Estimation of Living Forces, where he distinguishes “Ort” as the site of “the action of a substance on another” (1: 21) from “Lage” as a relative location on a shared spatial frame of reference (1: 22). At the same time, as Kant here goes on to insist, “place” cannot arise without “external connections, positions (Lagen), or relations.” When Kant elsewhere speaks of the soul having a “place” (Ort) in the body (Dreams, 2: 323), he may have something similar in relative to other bodies. A soul that has a “place” in this technically understood sense, “occupies” a space without “filling” it (i.e., necessarily resisting penetration by others). 23. Förmlich, according to Grimm and Grimm, means “legitimate” or “true.” By “förmlichen Wissenschaft,” Kant would seem to mean not a science that is merely “formal” (formal), such as general logic, which has no content other than the rules of thinking itself, but rather a body of knowledge capable of

130  Susan M. Shell yielding rules valid for worldly action generally as opposed to rules valid only contingently or in a given place and time. See also V-Anth/Pillau, 25: 734. 24. Cf. Idea for a Universal History, 8: 18, and CF, 7: 140. 25. On the contingency of our existence on the earth (e.g. via conception) and the difficulties this poses from the standpoint of man’s rational vocation, see A779–80/B807–8; Metaphysics L1, 28: 295. 26. On “local” versus “general” knowledge, cf. Reflection 1482 (Refl, 15: 659; early-mid 1770’s), and V-Anth Pillau (25: 734, Winter 1777/8): “(1) A local knowledge of the world [Local Weltkenntiß] that merchants have, which is also called empirical. (2) A  general knowledge of the world [general Weltkenntniß] that a man of the world [Weltmann] has, and that is not empirical but cosmological. The local is bound to place [Ort] and time, and also does not yield rules ready to hand for acting [handeln] in common life.” 27. Cf. MM, 6: 337. 28. In the Universal Natural History, Kant ascribes the principle causes of man’s general failure to achieve the “end” for which he was created to our “position” (Lage, Stellung) (UNH, 1: 362) in the universe (i.e., mid-way between the sun and the outermost planets). See UNH, 1: 352–63; and Shell (1996, 65–73). 29. Unlike tidal rivers, the Pregel flows into a body of water (the Baltic sea) that has no appreciable tide; hence its flow is unidirectional throughout (unlike that of a tidal river such as the St. Lawrence or the Amazon). See Physical Geography, 9: 219. On Kant’s own long-standing interest in tides and their effects, see for example his early essay on the Rotation of the Earth (1754) and later essay Something concerning the Influence of the Moon on the Weather (1794). On the geographic importance of rivers generally, see Physical Geography, 9: 216–21. A favored example on Kant’s part of “real opposition”—that of a ship exposed to the opposing forces of wind and current—would have been a relatively common occurrence on the Pregel, given the steady east-west current and the predominance of westerly winds (from May to October). (https://weatherspark.com/y/86511/Average-Weather-inKaliningrad-Russia-Year-Round, last visited 22 Oct. 2021.) 30. See Stan (2009). On Leibniz’s own argument, see Lyssy (2016). 31. To be precise, Kant's position is correct only as far as linear motion is concerned. On the difficult question of the possibility of non-linear motion, see Stan (2009). Given that physical monads, as he then conceived them, are not, strictly speaking, extended, Kant may not have regarded this as a possibility he needed to consider. 32. James Bradley (1693–1762), an English astronomer whose important discoveries included the aberration of light (through his study of the phenomenon of parallax) and the Earth’s notation (or variable rotation due to gravity). 33. Anth, 7: 166, 169n., 178, 264. On Kant’s vertigo, see Shell (1996: Chapter Nine). Marcus Herz, Kant’s friend and personal physician, composed a medical treatise on vertigo (der Schwindel). In the Anthropology Kant suggests that his reluctance to travel is at least partly due to his susceptibility to vertigo, prompted by situations (such as aboard a seagoing vessel) in which the horizon is imagined to be oscillating vertically. 34. In explaining the relativity of motion, Kant appeals once again to the example of a ball held by someone on a ship moving downstream whose motion is judged relative to the river’s banks (MF, 4: 487–488). 35. For discussions along similar lines, see 20: 22, 38, 41, 47–48, 146–47; on rivers and riverbanks see also 20: 105. 36. Cf. CPR, A613/B 641 and CF, 7: 109.

Kant am Pregelflusse  131 37. See also V-Met/ Dohna 28: 681: “We can determine our existence in time by nothing unless we base it on something perduring. Something must perdure with respect to which everything flows by”; cf. Refl 5871 (18: 373), and 5998 (18: 420). 38. Such a trip, beginning from the outer tip of the Vistula Spit (now bordering Poland), would have occurred within the relatively protected waters of the Vistula Lagoon. Königsberg at the time lays four miles upstream from the mouth of the Pregel. 39. An important historical site dating back to the era of the Teutonic Knights, and belonging at the time to the Province of East Prussia. 40. See, for example, CF, 7: 109. 41. In his intense correspondence with Herz during the 1770s, Kant was alternatively concerned with his emerging critique of pure reason, and his personal suffering from hypochondria and vertigo, for which he solicited Herz’s advice. Herz, himself an avid student and disseminator of Kant’s thought among the Berlin intelligentsia, was the author of several philosophical and medical works, including one on vertigo (Das Schwindel). For a fuller discussion of their correspondence, see Shell (1996, 275–81). 42. In the Metaphysical Foundations of Natural Science, absolute space will assume the character of a regulative idea. As James Messina puts it: “Kant takes us to be in the epistemic predicament of needing to seek ever more encompassing ‘relative spaces’ in order to determine the true relative motions of all bodies. The sort of sequence Kant has in mind proceeds, for example, from a ball moving with respect to a ship’s cabin, to the ship moving relative to the shore, to the earth moving with respect to the sun, and so on ad infinitum.” (MFNS, 4: 487–8) Nevertheless, as Messina adds: “Kant consistently explains what it is for a body to have a place in a manner that is incompatible with those places being absolute places. For Kant, a body’s place—or position—depends on and is determined by its spatial (and dynamical) relations to other bodies. Thus, the place of a body is always a relative affair.” (Messina 2018, 705, 710) 43. On the dependence of philosophic health on gaining mastery over the involuntary effects of the imagination, see also Philosopher’s Medicine of the Body, 15: 939–53. 44. See https://users.manchester.edu/FacStaff/SSNaragon/Kant/notes/notesGeography.htm#an-Holstein-Beck. Last visited on 22 Oct. 2021. 45. Zöller (2011) and Zöller (2013). 46. Psalm 8: 4, “Was ist der Mensch, dass du seiner gedenkst, des Menschen Kind, dass du dich seiner annimmst?” (Luther’s Bible). 47. Letter to Kant, Late December  1759 (L, 10: 27). A  recent convert to an idiosyncratic Christianity, Hamann (1730–88) was responding to Kant’s somewhat disingenuous invitation to collaborate on a book for children on Newtonian cosmology. Kant, with the support of their mutual friend Berens, hoped to thereby wean Hamann from his newly acquired religious enthusiasm. Hamann is the likely source of Kant’s initial exposure to the thought of Hume and (with somewhat less evidence) Rousseau; and the two

132  Susan M. Shell remained on warm terms, despite their serious philosophical disagreements. (Hamann chose to leave his own Metacritique of Kant’s critical philosophy unpublished. Herder’s Metacritique of the Critique of Pure Reason appeared in 1800.) That Hamann remains on Kant’s mind, if only indirectly, is suggested by his veiled reference in the Opus Postumum to Herder’s deceptive plagiarism of their joint friend (OP, 21: 225). See also CF, 7: 55; on Kant and Hamann generally see the Introduction by Klaus Reich to the Philosophische Bibliothek edition of the Streit der Facultäten (XII-XIII), cited in the Mary J. Gregor’s English edition of the Conflict of the Faculties (New York: Abaris Books, 1979), 218n. 48. See in this regard Anth, 7: 235.

References Brandt, Reinhard (1999). Kommmentar zu Kants Anthropologie. Hamburg: Meiner. Foucault, Michel (2008). Introduction à l’“Anthropologie” (Introduction to the “Anthropology”). Paris: Vrin. Louden, Robert B. (2009). “The Second Part of Kant’s Ethics.” In Brian Jacobs and Patrick Kain (eds.), Essays on Kant’s Anthropology. Cambridge: Cambridge University Press, 60–84. Lyssy, Ansgar (2016). Kausalität und Teleologie bei G.W. Leibniz. Stuttgart: Franz-Steiner Verlag. Messina, James (2018). “Kant’s Stance on the Relationalist-Substantivalist Debate and Its Justification.” Journal of the History of Philosophy 56.4, 697–726. Shell, Susan M. (1996). The Embodiment of Reason: Kant on Spirit, Generation, and Community. Chicago: University of Chicago Press. Stan, Marius (2009). “Kant’s Early Theory of Motion: Metaphysical Dynamics and Relativity.” The Leibniz Review 19, 29–61. Zöller, Günter (2011). “Kant’s Political Anthropology.” Kant Yearbook 3.1, 131–61. Zöller, Günter (2013). “Genesis und Klima: Geo-Anthropologie bei Herder und Kant.” In Stefano Bacin, Alfredo Ferrarin, Claudio La Rocca and Margit Ruffing (eds.), Kant und die Philosophie in weltbürgerlicher Absicht, Akten des XI. Kant-Kongresses 2011. Berlin, Boston: De Gruyter, 551–63.

Appendix: Königberg

Figure 6.1  1763 Map of Königsberg

134  Susan M. Shell

Figure 6.2  Detail from 1808 Map of Königsberg

7 Kant’s Philosophy of Religion—A Provocation to the Historical Religions1 Bernd Dörflinger

The avoidance of all misunderstanding is obviously quite important to Kant; for he makes clear, right in the first sentence of the preface to his treatise on religion (Religion within the boundaries of mere reason), that, however much his idea of rational religion may further develop, it will change nothing in morality from the point of view of its autonomous foundation in pure practical reason alone, and of the purely intrinsic motivation of its application. So far as morality is based on the conception of the human being as one who is free but who also, just because of that, binds himself through his reason to unconditional laws, it is in need neither of the idea of another being above him in order that he recognize his duty, nor, that he observe it, of an incentive other than the law itself. (Rel, 6: 3) A few lines later (ibid.) we find the corresponding passage: “Hence on its own behalf morality in no way needs religion (whether objectively, as regards willing, or subjectively, as regards capability) but is rather selfsufficient by virtue of pure practical reason.” Nevertheless, a few pages later in the same preface is found the famous conclusion, “Morality leads inevitably to religion.” (Rel, 6: 8n) In between these—going beyond the foundational aspect and the application aspect—is to be found a consequentialist reflection with respect to morality. Kant takes such a consideration to be absolute, “for it cannot possibly be a matter of indifference to reason how to answer this question, What is then the result of this right conduct of ours?” (Rel, 6: 5, the emphasis is Kant’s) His answer to this question, what should rationally be the consequence, is given as early as the first Critique: that “the system of morality is therefore inseparably . . . combined with the system of happiness” (CPR, A809/B837).2 Put another way, what is required is “the system of self-rewarding morality.” (Ibid.) Against this requirement, however, there stands the awkward insight we have into the irrational feature of life which a human being cannot overcome through his DOI: 10.4324/9781003044376-8

136  Bernd Dörflinger own power alone. This feature consists of the reason-adverse circumstance that the interconnection between the moral quality of a person and his sensible existence is not a necessary one, but is merely happenstance. This means: in life, a human cannot progress further through morality than to the worthiness to be happy. Neither is a moral being necessarily happy, nor is an immoral being necessarily unhappy. And it is precisely thus: [S]ince human capacity does not suffice to bring about happiness in the world proportionate to the worthiness to be happy, an omnipotent moral being must be assumed as ruler of the world, under whose care this result would come about. (Rel, 6: 8n) and this is why, Kant continues, “Morality leads inevitably to religion.” (Ibid.) Obviously this deals with religion considered as developed purely in thought, thus with natural religion, which claims no revelation, no communication from a god thought of from the beginning as external, in contrast to whom man would need to understand himself as merely passive and receptive. Inversely, it is through Kant’s rational idea of religion that God can be understood from the human perspective as the externalization in thought of human practical reason, which in the Conflict of the Faculties Kant characteristically refers to as “God within us.” (CF, 7: 48) The God conceived in accordance with the self-created idea of religion is only to be thought of as the hypostatization of practical reason supplemented by the ability to bring worthiness of being happy, and happiness itself, into a rational relationship; so the number of human duties is not increased by the thought of him. So the religion of reason recognizes no outer—that is, irrational— duties, which Kant calls statutory, but only the same duties that humans have with respect to other humans; these can also be derived from practical reason alone, entirely without appeal to religion. Accordingly, the God of rational religion requires, briefly stated, nothing more than the conduct of the good life. What it does not require are specifically religious practices which go beyond moral practices; this matter will be considered later. What has been said earlier corresponds to Kant’s standard explication of the religion of reason, in which our duties—that is, the duties that are already known to us thanks to practical reason—are considered simultaneously or subsequently as divine commands. The religion of reason which we have sketched out so far is what Kant calls the only true religion. From this it follows that everything going beyond this which claims the title of religion can only be false religion, only a pretense to religion. Put bluntly, the historical religions of revelation belong to this class, namely, those that are not developed from reason, but instead appeal to a communication to us, situated historically

Kant’s Philosophy of Religion  137 in space and time, from God Himself. On the basis of the spatiotemporal situatedness of this communication, Kant also refers to the belief in such communication as an “empirical belief which, it appears, chance [i.e., mere accident] has dealt to us” (Rel, 6: 110).3 Thus he also denies consistently to the species of such belief the name of “religion,” and calls them all species of mere church belief. In what follows they can nevertheless continue to be referred to as religions of revelation; for, first, Kant, as we know, could not carry through with his intention of language purification, and second, the species of church belief are religions according to their own self-understanding and pretension. When the matter under consideration here is in question, why do the historical religions not fulfill the conditions of the Kantian concept of religion, a preliminary inquiry into the concept of revelation suggests itself. In these empirical religions, as we have said, religion is not, as with Kant, derived from concepts of practical reason a priori; rather, they lay claim to theoretical experiences of revelation. Such supernatural experiences cannot, however, be legitimated according to Kant’s critical theory of experience. Experiences such as those of the supernatural fail to provide the human knower what he requires according to this doctrine, that is, the sensorium; neither outer nor inner sense can provide the necessary receptivity. “For,” according to Kant, [I]f God should really speak to a human being, the latter could still never know that it was God speaking. It is quite impossible for a human being to apprehend the infinite by his senses, distinguish it from sensible beings, and be acquainted with it as such. (CF, 7: 63) Yet not only are the limits of sensibility not respected when they are involved in a contradiction (when they are supposed to sense something non-sensible); in addition, Kant says, this experience is such that “(as supernatural) it cannot be traced back to any rule in the nature of our understanding and established by it.” (CF, 7: 57) The restrictions in particular to the causal rule of our understanding, according to which all appearing effects can only be related to causes which are in turn themselves appearances, are set aside where experiences of revelation are claimed. Kant’s formulation of his critique of such supposed experiences is especially sharp when it concerns such experiences as are viewed as occurring by means of inner sense, that is, when it is claimed that there is a “feeling of the immediate presence of the highest being.” (Rel, 6: 175) To “want to perceive in oneself” these kinds of “heavenly influences” is what Kant calls “a kind of madness,” (Rel, 6: 174) to which he gives the name “illuminatism of inner revelations.” In the case of such a revelation, “every one [of the illuminated subjects would have] his private,

138  Bernd Dörflinger inner revelations” with exclusive status, and there would no longer be found “any public touchstone of truth.” (CF, 7: 46) Over and above the deficit of an effusive cognition claim, Kant attributes to any virtuosos of inner revelation a moral deficit as well. The reproach addressed to them is, put generally, that of moral passivity. In view of the religious phenomena of Kant’s own time, that is, in view of the sect division between the “followers of Spener and Franck (Pietists), and the Moravian Brethren of Zinzendorf (Moravians),” (CF, 7: 55) both of which he counts as belonging to the genus of the “mystical theories of feeling,” (CF, 7: 56) Kant’s diagnosis is that each of these assumes that the human being “cannot hope to improve by his own powers.” (CF, 7: 55) Instead, they demand that anyone aspiring to revelation should make himself receptive to it, apply for this favor, to produce in himself certain feeling states: in the case of the one sect, feelings of the heartmelting kind; in the case of the other, the feeling of self-contempt. Kant’s commentaries to these are: whoever tends toward oversensitivity, toward “tender emotions,” whose “heart is soft,” is insensitive to the “rigorous precept of duty” (CPJ, 5: 273); and he who despises himself, takes a servile, groveling, favor-seeking attitude has abandoned “any confidence in his own power to resist evil.” Contrary to self-contempt, the human being should not disavow “the moral self-esteem of such a being” (MM, 6: 435), according to Kant in the Doctrine of Virtue.4 Since Kant cites self-esteem explicitly here as a duty of virtue and opposes it to the vice of servility, it can be recognized that he considers these aforementioned two varieties of sensible religion not merely to be false religious doctrines, which are in addition morally indifferent; rather he classifies them on the side of moral reprehensibility. A famous passage in the preface to the second edition of the religion treatise appears to contradict the opposition observed up until now between rational religion and revelation-religion; interestingly, this passage can then be invoked in the pursuit of harmony. In this passage Kant says in fact that “revelation can . . . comprise also the pure religion of reason,” and that it is possible “to start from some alleged revelation or other” and to “hold fragments of this revelation  .  .  . up to moral concepts,” and in this way to come back to the “rational system of religion.” (Rel, 6: 12) He expresses the potential relationship of the two using the image of two concentric circles of different circumferences. The smaller circle within the circumference of the larger thus represents rational religion which, as we have seen, is nothing more than the sum of the moral laws of practical reason, insofar as these are additionally conceived of as divine laws. Kant thereby expresses the possibility that the smaller circle coincides with the inner part of the large circle, which in turn stands for what is taken as a revelation; the possibility is that the lessons of such alleged revelations can also contain moral laws. One might think, for instance, of the prohibition against lying in the biblical

Kant’s Philosophy of Religion  139 Ten Commandments. The statutory religious lessons can be located in the part of the larger circle which lies outside of the smaller one, that is, those that are not to be developed from reason; and which consequently, without the revelation of that which God has arbitrarily ordained, are not to be professed. Another law from the Ten Commandments can serve as an example here, that is, the prescription requiring the ‘periodic veneration’ of God on every seventh day of the week. In view of the relationships sketched here, we can thoroughly agree with Kant’s statement that something taken as a revelation could contain something from the rational system of religion, since obviously the Ten Commandments contain the prohibition of pure practical reason against lying. Nevertheless, the agreement involved here is only a very restricted one between the presupposed revelation and the aspect of pure practical reason supposed to be included in it. For the laws in question coincide only with respect to their content, in that they forbid lying; with respect to their form, that which is of decisive importance, they do not coincide at all, for, in the first case, the ground of their validity is external, in the second case, it is an inner one. Apart from this, the differentiation which Kant continues to emphasize throughout, through the coincidence thesis on the one hand, the metaphor of the circles on the other, is not to be denied: the differentiation between heteronomy and autonomy. If the prohibition against lying were considered to have been originally revealed, and, in this regard, first made known to uncertain humans afterwards, it would be a heteronomous, statutory law, thus one not resting on free self-obligation, but instead one traceable back to the arbitrary choice of another being, that is, to God. Understood in this way, the prohibition against lying cannot have the status of a categorical imperative; it would have to stand as a hypothetical imperative merely under the condition of the constancy of the arbitrary choice of the other referred to above and of the belief in the validity of this revelation. Such consequences and their presupposition of a previous revelation are, however, in no way suggested by Kant’s determination of the relationship. Since he recommends holding up everything alleged to be a revelation to moral concepts, in order that eventually we may come back to the rational system of religion if the revelation-religion in question provides a way, what on the other hand is presupposed is the priority of pure practical reason and the rational idea of religion derived from it, and in this way to avoid prioritizing something considered to be a revelation. Kant’s model of the concentric circles is not suited overall to cover up all related differences or to blur a potential critique. This holds not only for his critique of the variety of laws he calls “statutory laws” in general and of certain specific statutory laws which he sees differently represented—variously specified—in all religions of revelation. To the last sort belongs the previously mentioned prescription for veneration of God or requirement for service to God. The historical religions, to the extent

140  Bernd Dörflinger that they ask for actions specially addressed to God, hence a specifically religious practice in the service of God, stand in just this point in sharp contrast to rational religion. The thought of coming through this moral religion to morality is only this: that through the fulfillment of duties grounded in pure practical reason, “divine commands” are also obeyed.5 But upon this thought depends neither the validity nor the practice of morality so that to someone to whom it does not appear convincing, no moral deficit can be attributed. Agnostics, who refrain from judgment, and atheists, who explicitly reject it, must then be considered as morally unchallenged thereby. The thought that through moral practice one also stands “in the service of God” (Rel, 6: 103) requires no outer manifestation corresponding especially to Him, that is, to God, thus requires no specifically God-serving religious practice which goes beyond moral practice as such, that is, the conduct of the good life. According to Kant, it is for human beings “absolutely impossible to serve him [sc. God] more intimately in some other way” than in the manner of fulfilling their duties to other human beings, “for they can act and exercise their influence on no other than earthly beings, not on God.” (Rel, 6: 103) On the other hand, the service of God (as it is commonly so called6) consists of those actions addressed directly to God, which is even characterized by Kant as “madness,” if consciousness is combined with it; it (and not the conduct of the good life alone) would be the true service of God. In particular, Kant is seriously concerned about the error that consists in construing something as a service to God that is not a service to God, through injecting a prayer as a plea for mercy. Convinced that a human being could stand in such a relationship to God that he, as if in a relationship with a fellow human, could reach for a means of influencing God, to achieve a desired reverse effect, here mercy—this Kant counts along with the belief in miracles and in secrets as among the mad beliefs.7 The vehemence with which Kant opposed any specifically God-serving religious practice can largely be explained by the fact that he saw in it a power that worked opposite to that which according to him constitutes the appropriate central part of human self-understanding, that is, pure practical reason, through which a human being commits himself (and so autonomously) to assuming a moral duty toward human beings. For the actors within the kind of God-serving practice Kant has rejected, on the other hand, the fulfillment of duty consists in the “transaction of an affair of God, not of humans.” (Rel, 6: 103) They cannot indeed conceive their obligation except as directed to some service or other which they must perform for God—wherein what matters is not the intrinsic worth of their actions as much as, rather, that they are performed for God to please him through

Kant’s Philosophy of Religion  141 passive obedience, however morally indifferent the actions might be in themselves. (Rel, 6: 103) When no inner moral motivation can be assumed, the second kind of inducement must be supposed, that is, that of self-love. Where duty is not pursued as an inner concern of the human being, but instead is conceived to be heteronomously established through God, the request that is made thereby is basically incomprehensible; to pursue this kind of duty despite its being externally commanded can only mean to apply in the spirit of passive obedience, as has been seen, to obtain favor and reward, and avoid disfavor and punishment. As noted earlier, the fulfillment of duty, when pursued in the uneasy God-serving manner and in the spirit of passive obedience, is done for the sake of moral duty such as the one previously mentioned: not to lie; but especially when it concerns alleged duties completely unrelated to morality. Among such duties Kant counts “festivities,” “public games,” “doing penitence,” “self-chastisement,” “pilgrimages,” etc. Among these, “self-torments” are especially suspect for Kant; of these he says: “just because they have absolutely no use in the world, and yet cost effort, they seem to be aimed solely at attesting devotion to God.” (Rel, 6: 169) These previously named appearances invite us to repeat that rational religion bears witness to no religious phenomena at all. To follow Kant’s explanation, it is due to a “weakness of human nature” (Rel, 6: 103) that human beings are not easily persuaded to restrict themselves to the purely mental sphere of the idea of the moral religion of reason. In his words, it is the weakness of their “impotence in the cognition of supersensible things” (Rel, 6: 103) that leads them to “demand for even the highest concepts and grounds of reason something that the senses can hold on to, some confirmation from experience or the like” (Rel, 6: 109). To represent duty–obligation to oneself in such a way that it is heteronomously proclaimed from God as a command to the passive subject, which later then complies with it as a way of serving God, that is, the duty is fulfilled for the sake of this external God—this Kant portrays as the transmission of a superordinate or subordinate relationship as it appears in the sensible world of humanity. Kant sees that [E]very great lord of this world has a special need of being honored by his subjects, and of being praised through signs of submissiveness; nor can he expect, without this, as much compliance with his orders from his subjects as he needs to rule over them effectively. (Rel, 6: 103) To represent God anthropomorphically in this way would mean to attribute properties to him taken from the morally less advantageous

142  Bernd Dörflinger sensible nature of the human being; put generally, from self-love. God would then need to be represented, following the model of the great lords of the world, as vain; as driven by the need to seek pleasure through demonstrations of obsequiousness; additionally as driven by the fear of losing his lordship. And he would welcome signs of obsequiousness as indications of his subordinates’ increasing adoption of the spirit of passive obedience, that is, of the mentality of one who receives commands. By the standards of pure practical reason, which should be decisive for this question, a figure burdened with these properties who treasures the vice of servility in his subordinates and fears the virtue of self-esteem cannot be God. Pure practical reason possesses, according to Kant, the highest normative authority in one further respect: that is, when we are concerned to interpret texts of the historical religions which venerate these texts as holy. The general principle of this hermeneutica sacra (sacred interpretation) is that these texts must be interpreted “in a practical way, according to rational concepts” (CF, 7: 46) and that “when conflict arises about the sense of a scriptural text, philosophy, [understood as the champion of practical reason] . . . claims the prerogative of deciding its meaning.” (CF, 7: 38) This privilege of reason reaches so far that some scriptural passages can even be rejected, when, for example statutory doctrines “conflict with duties which reason prescribes unconditionally. . .,” then “the former must yield precedence to the latter.” (Rel, 7: 154) An example of such a conflict is for Kant the biblical passage from which we can elicit that God, when it pleases Him, may demand a human sacrifice. This is the passage in which there is demanded of Abraham “on God’s command” the sacrifice of his son. Here Kant proposes nothing less than a revision of the Holy Scripture, by saying what Abraham “should have replied,” namely, “That I ought not to kill my good son is quite certain. But that you, this apparition, are God—of that I am not certain.” (CF, 7: 63n) For cases of moral ambiguity, that is, for not entirely hopeless, but still dubitable cases, Kant’s proposition is to place a moral sense underneath the text, even when that sometimes seems forced, and may really be forced, even if the interpreter imposes this sense on the text contrary to the probable intention of the author.8 A biblical prayer can count as an example, one in which the praying person is pleading for revenge on his enemies. Kant mentions here a symbolic interpretation, in which the enemies could stand for the evil inner inclinations of the one praying, or if not, to express the renunciation of the praying person’s right of revenge, if the revenge is left up to heavenly justice.9 Kant is aware that placing senses underneath a text or forcing a moral interpretation on one “seems to violate outright the highest rules of exegesis.” (CF, 7: 40) Thus the question arises of how this creative textual interpretation approaching the limits of manipulation can be justified.

Kant’s Philosophy of Religion  143 According to the proposal, it is to be justified as an attempt to improve the text which arrives with a claim of holiness, namely, on the basis of that which, according to Kant, originally has earned the entitlement of holiness, that is, the moral consciousness of what determines its will autonomically through the Moral Law. The attempt at an improvement of the text by the interpreter is thus to be considered as an attempt to carry out the improvement of a text taken seriously in its claim to holiness. Kant contrasts the normative philosophical interpreter with another kind of expositor, which he, with reference to Christianity, calls the biblical theologian. Such a one is completely oriented to the facticity of his revelatory scripture, and prefers the literal kind of interpretation to the symbolical, following the principle “Thus it is written.” (Rel, 6: 107; cf. Rel, 6: 163) In addition, he, as a theological academic specially educated in textual interpretation, will be granted an eminent position among the followers of his type of church belief on the basis of his favored access to God’s word, thus enjoying a special rank in the ecclesiastical hierarchy. According to Kant, this “degrading distinction between laity and clergy” (Rel, 6: 122), which is not to be found anywhere in Kant’s hierarchy-free conception of the religion of reason, is also no part of his hermeneutica sacra, since all those gifted with pure practical reason are equally authorized to contribute to its interpretation. Kant’s preference for this kind of rational-religion interpreter over the biblical-theologian kind implies that any claim of exclusive interpretive competence by the revelation-religion specialists, and any attempt by these to immunize themselves against lay interpreters in general, and philosophers in particular, are to be rejected. The principle “Thus it is written” has especially far-reaching consequences in the case of religious laws that are statutory in every respect. These laws are statutory (like the revealed prohibition against lying) because their validity has a heteronomic foundation according to their form (even though according to their content they are moral, and thus coincide with the proscription against lying). They are statutory also because, according to their content, they are in no way comprehensible for reason. This means that they must be completely traced back to the arbitrary will of God as shown in a revelation that provides this support. Since this religious statutory law is dependent on a self-revelation historically situated in this way, it is to be qualified as empirically accidental. As a law outside of morality, and so not coinciding with pure practical reason, it can be either amoral or indifferent to morality: it is amoral in a case which is not atypical even today in the history of species of church doctrine, when the alleged divine instruction states: non-believers, believers in a different religion, or heretics within the churches or species of church doctrine, are to be combatted; it is morally indifferent when it involves something like what is still today a widespread type of rule concerned with dietary regulations tracing back to God. These last rules, according to Kant’s standards (to be sure not in accordance with

144  Bernd Dörflinger the standards of the church beliefs of whatever kind) produce no credit in the case in which they are observed, and no guilt in the case in which they are violated. Especially under the conditions of various presumptive revelations and species of church belief, Kant nevertheless considers even morally indifferent statutory religious legislation to be not just harmless, but in fact to be dangerous. In his judgment, “conflict over historical dogmas can never be avoided” (Rel, 6: 115), and according to him “the so-called religious struggles which have so often shaken the world and spattered it with blood have never been anything but squabbles over ecclesiastical beliefs.” (Rel, 6: 108) One ground for this is that, according to Kant, each of the historical religions “refuse[s] to recognize as a law anything that differs from theirs” (CF, 7: 50) so that they insist on the exclusivity of their own alleged statutory laws as having originated from God. At the same time, they contest the right of the other species of belief legitimately to appeal on their own behalf to revelation. How could even God issue competing or even contradictory statutes? Over and above this, there lies in the consequence of the claim to exclusivity the assumption that the life of the other, according to their own statutory life rules, must not merely appear as a simply foreign one, but as a false, godless life. If the statutory laws, which, to repeat, are not to be derived and intuited from pure reason, nor considered as established by human beings on the basis of pragmatic empirical purpose-directed considerations, but rather as proclaimed by God—if, accordingly, the obedience to them is practiced as a concern of God’s—then they receive, in comparison to human-established rules, an excess of signification driven to the absolute. This is what explains the especially heated nature of religious wars. With respect to the class of statutory rules created by human beings rather than by God, to which belongs the entire sphere of the positive judicial laws, it is possible for there to be a continuous reconsideration furthering the carrying out of changes. It would then be possible here, in the human arena, for a search to be made for consensus and compromise in the carrying out of these changes—something that would seem to be hopeless when each of the participants in a conflict considers himself to speak for God, and that consideration is contested by the opposition. In view of the inevitability of conflict and the lack of any expectation of arriving at an understanding, Kant’s final demand (which is to be understood as normative and as a project for perpetual religious peace) is that “religion will gradually be freed of all empirical grounds of determination, of all statutes that rest on history [ . . . so that] finally the pure belief of religion will rule over all.” (Rel, 6: 121) The gradual character of the process can be explained by the fact that the means of reaching such a final situation can only be a progressive enlightenment, which continually proposes the reforming of thinking and through which it is firmly excluded that rational religion would, for the sake of peace, do

Kant’s Philosophy of Religion  145 something like appearing itself as a physical party of religion in a war being waged by historical religious parties. Though the path that should lead to this final state of continual reform is a moderate one, the idea of such a final state itself is thoroughly radical. If religion should really be freed from any empirical grounds of determination, and from any statutory laws, such a situation would mean that the historical religions of revelation would not exist at all, and that there would be no more specific appearances of religion—since the manifestations of the religion of reason differ in no respect from those of pure morality. In such a final state as we have under consideration, without any empirical elements, the appeal to revelation would lapse—for such an appeal, which would be, in accordance with its claim, historically situated in space and time, and so must be something empirical—so that there would be no statutory religious legislation to contribute to conflict. Right here at the end I would like to refer explicitly for the first time to the intended results of my contribution, by considering its title. In addition to a list of partial attributions that should be recognizable in the course of the arc which I have tried to follow, there is the fundamental provocation which Kant’s philosophy of religion presents to the historical religions: this philosophy, by treating these religions under a decidedly normative viewpoint through which such a provocation expresses itself as a demand of pure practical reason, is one which confronts these religions with their end.

Notes 1. This article is translated from the German original by Hoke Robinson, in collaboration with the author. An important deviation from the CUP translation is in the translation of “der Glaube.” The glossaries of both the CUP translations of the Kritik der reinen Vernunft (Critique of pure reason) and of Die Religion innerhalb der Grenzen der bloßen Vernunft (Religion within the boundaries of mere reason) give two possible English translations of “der Glaube”: “belief” and “faith.” In English, “faith” suggests acceptance of some doctrine of a historical religion such as Christianity, as it were from “within”: for instance, the existence of God, or of his creation of the world in six days; a believer who has faith has belief as well. But it would seem odd to say that Galileo had “faith” that the earth moves around the sun rather than vice versa, and that Newton had “faith” that for every action there is an equal and opposite reaction. Since Kant is here discussing God’s existence and creative acts as it were from “without” or “above,” the doctrines are being regarded as more on a par with empirical science, at least to begin with. Accordingly, “der Glaube” is rendered regularly here as “belief.” Example: “Kant [nennt] den Glauben in solche Mitteilungen auch ‘empirischen Glauben, den uns dem ansehen nach ein Ungefähr . . .’ (Rel, 6:110)” is rendered: “Kant also refers to the belief [not “faith”] in such communication as an ‘empirical belief which, to all appearances, chance . . . has dealt to us.’ (Rel, 6:110).” 2. “Morality” here translates “Sittlichkeit,” elsewhere it translates “Moral.” 3. On the translation here of “Glaube” as “belief,” see Note 1.

146  Bernd Dörflinger 4. The Doctrine of Virtue (Tugendlehre) containing the phrase cited in note 22 aforementioned is Part II (of two) of the Metaphysics of Morals (Metaphysik der Sitten). 5. Cf. Rel, 6: 103. 6. Cf. Rel, 6: 192. 7. Cf. Rel, 6: 194. 8. Cf. CF, 7: 46. 9. Cf. Rel, 6: 110.

8 Hume and Kant on Utility, Freedom, and Justice Paul Guyer

1. Introduction Hume’s account of our concept of and belief in causation was one of the chief targets of Kant’s theoretical philosophy. Hume’s particular version of moral-sense meta- and normative ethics was not an explicit target of Kant’s moral philosophy—Kant often named Shaftesbury and Hutcheson as his targets, although the Critique of Practical Reason does remind us about his argument with Hume about causation. Hume’s political philosophy is not an obvious target for Kant’s theory of right (Rechtslehre)—here Hobbes and Locke are obvious targets, along with his domestic sources as well as targets, Christian Wolff and Gottfried Achenwall.1 Nevertheless, a comparison of Hume’s and Kant’s political philosophies can be informative, although the differences between them may not always lie where we would expect. Of course there will be obvious meta-ethical differences between any empiricist methodology and Kant’s approach to practical philosophy founded on a priori principles of pure reason. One claim I will argue, however, is that the normative differences between Hume’s presumably proto-utilitarian approach to justice and Kant’s approach based on the innate right to freedom of every human being is not as great as might be expected. Yet there are important differences between their views at the level at which their general principles of justice are applied, in particular in their treatments of the relation between subjects and rulers, specifically on the issues of paternalism and of the right to rebellion. But their treatments of these issues do not always line up with their general principles in ways that might be expected. A review of Hume’s 1751 Enquiry concerning the Principles of Morals appeared in the Göttingische gelehrte Anzeigen in 1754, and a German translation appeared the same year. That was included in the fourvolume edition of Hume’s works published in 1754–56, which was in Kant’s library at the time of his death.2 So it is likely that Kant knew Hume’s treatment of justice from this work rather than from Book III of the earlier Treatise of Human Nature, no copy of which was reported in Kant’s library and to which he never referred. Be that as it may, the Enquiry DOI: 10.4324/9781003044376-9

148  Paul Guyer certainly includes statements on justice that sound as far as from Kant as may be imagined: That justice is useful to society, and consequently that part of its merit, at least, must arise from that consideration, it would be a superfluous undertaking to prove. That public utility is the sole origin of justice, and that reflections on the beneficial consequences of this virtue are the sole foundation of its merit; this proposition, being more curious and important, will better deserve our examination and enquiry. (EPM, 3.1; Hume 1998, 13)3 A few pages later, Hume restates that the virtue of justice “derives its existence entirely from its necessary use to the intercourse and social state of mankind,” and then, making clear that we can take utility, in standard fashion, to mean productive of happiness, adds that “The USE and TENDENCY of that virtue is to procure happiness and security by preserving order in society,” and argues further that justice is a virtue only in circumstances, namely, those of only relative scarcity of resources and limited but still some benevolence, where it can in fact preserve happiness and security (EPM, 3.1; Hume 1998, 15): Thus, the rules of equity or justice depend entirely on the particular state and condition in which men are placed, and owe their origin and existence to that UTILITY, which results to the public from their strict and regular observance. Reverse, in any considerable circumstance, the condition of men: Produce extreme abundance or extreme necessity: Implant in the human breast perfect moderation and humanity or perfect rapaciousness and malice: By rendering justice totally useless, you thereby totally destroy its essence, and suspend its obligation upon mankind. (EPM, 3.1; Hume 1998, 16) This certainly sounds different from Kant. He begins his 1797 Metaphysics of Morals, which contains his systematic statement of his political and legal philosophy, with the general remarks that moral laws “hold as laws only insofar as they can be seen to have an a priori basis and to be necessary” and that “They command for everyone, without taking account of his inclinations, merely because and insofar as he is free and has practical reason,” thus “a practical philosophy . . . has not nature but freedom of choice for its object” (MM, 6: 215–6). Kant states that although in the derivation of the particular duties of human beings [W]e shall often have to take as our object the particular nature of human beings, which is cognized only by experience . . . this will in

Hume and Kant  149 no way detract from the purity of these principles or cast doubt on their a priori source. (MM, 6: 217) He then continues that moral laws are laws of freedom, and in the case of the laws of justice, they are laws for freedom in “external actions,” that is, laws for the coordination of the free actions of persons physically interacting with each other: “As directed merely to external actions and their conformity to law [the laws of freedom] are called juridical laws. . . . The freedom to which the former laws refer can be only freedom in the external use of choice” (MM, 6: 214). Kant insists that “The concept of freedom is a pure rational concept,” and that “observance or transgression” of duty grounded on this concept are “indeed connected with a pleasure or displeasure of a distinctive kind (moral feeling)” but that “in practical law we take no account of these feelings (since they have nothing to do with the practical basis of practical laws. . .).” (MM, 6: 221) Thus Kant clearly rejects any empirical foundations for the fundamental principles of practical law, including juridical law or the laws of justice; and since happiness seems to lie in the satisfaction of inclination and he insists that the normative force of freedom, as the foundation of all moral law and therefore of juridical law, has nothing to do with inclination, he would seem to imply that justice has nothing to do with happiness, or mere utility—the exact opposite of Hume’s position. And yet . . . Although Kant obviously rejects Hume’s empiricist methodology, at least when it comes to the fundamental principles of morality and justice, the difference between their normative principles may not be so great as initially appears. This is because Hume in fact recognizes the freedom to conduct our own affairs as we see fit insofar as that is compatible with a general desire for security as a major component of human happiness, while for Kant freedom in external action is the freedom to set one’s own ends, and happiness is at least the ideal outcome of being free to pursue one’s ends, with the maximal happiness throughout the world ideally resulting from maximal freedom throughout the world, that is, the maximal freedom of each compatible with equal freedom for all. So although their normative and meta-ethical foundations are different, Hume’s utility- or happiness-based rules of justice do make room for freedom of action, and Kant’s freedom-based laws of justice do make room for happiness or utility. In the following section, I  will provide evidence for these claims.

2.  From Happiness to Freedom and Back Again First I will argue that Hume’s foundation of justice in utility or happiness is compatible with recognition that it functions by securing freedom of action, and that Hume himself recognizes this. Then I will show

150  Paul Guyer that Kant’s denial of any direct connection between justice and happiness does not preclude an indirect connection between the two, and that Kant makes this clear in his overall approach to morality, of which justice is a part, even if he does not make this explicit in his treatment of justice itself. As we saw, in the second Enquiry Hume explains the “origin and existence” of the rules of justice in terms of their “utility” in the actual circumstances of human existence, namely, relative scarcity of resources and limited benevolence. This parallels his view that the virtue of benevolence—the disposition to do good to or for others even when, like other “social virtues,” it is not required by justice—is also to be explained by utility: “the UTILITY, resulting from the social virtues, forms, at least, a part of their merit, and is one source of that approbation and regard so universally paid to them” (EPM, 2.2; Hume 1998, 10). Furthermore, it is part of his general theory that all virtues are dispositions of character or behavior that are either useful or agreeable to oneself or (perhaps better “and/or”) others: “every quality of the mind, which is useful or agreeable to the person himself or to others, communicates a pleasure to the spectator, engages his esteem, and is admitted under the honorable denomination of virtue or merit” (EPM, 9.1; Hume 1998, 78)—the “spectator” here being understood as what Adam Smith would subsequently designate the “impartial spectator,” that is, any normal human observer, including oneself, of any human action, including one’s own, as long as her response is not distorted by a personal interest in the action. It should be clear, however, that utility is utility for something, and the latter can be nothing but agreeableness: virtues that are useful are those for which “the end, which they have a tendency to promote, must be some way agreeable to us, and take hold of some natural affection” (EPM, 5.1; Hume 1998, 34). Thus the Humean virtues can be regarded more simply as those dispositions to action that are directly or indirectly agreeable, and if being agreeable in turn means producing happiness of some kind or other, then Humean virtues are simply those dispositions to action that produce happiness more or less directly. Hume’s fourfold classifications of virtues is thus a happiness-based version of utilitarianism. In the case of the virtue of justice, where the utility of general observance of the rules requires satisfaction of duties even in particular cases where that seems contrary to utility, such as a poor man having to pay back a debt to a rich man who hardly needs the money, this would seem to be rule rather than act utilitarianism—except since their tendency to weaken or strengthen general conformity to a rule is certainly part of the consequences of particular actions, rule utilitarianism is simply enlightened act utilitarianism, or, more bluntly put, this distinction is meaningless. The need for a separate theory of rule utilitarianism arises only from not considering all the consequences of particular acts (see EPM, Appendix 3; Hume 1998, 96–7; see also T, 3.2.2.22; Hume 2007, 319–20).4

Hume and Kant  151 But Hume makes it clear that the rules of justice can also be understood as establishing the conditions of “peace and security” in which individuals can pursue their own ends as they see best. “RULES of justice or property would best promote public interest, and establish peace and security among mankind” (EPM, 3.2; Hume 1998, 19)—rules of justice “abridge our native freedom” (EPM, 4; Hume 1998, 28), but only for the sake of greater “peace and security” for our remaining freedom of action. This point is clear from several passages in the Treatise, even though Hume does not state it as explicitly as I have just done. One such passage is Hume’s discussion of the three basic kinds of human goods that are candidates for protection by enforcement of the rules of justice: There are three different species of goods, which we are posses’d of; the internal satisfaction of the mind, the external advantages of our body, and the enjoyment of such possessions as we have acquir’d by our industry and good fortune. We are perfectly secure in enjoyment of the first. The second may be ravish’d from us, but can be of no advantage to him who deprives us of them. The last only are both expos’d to the violence of others, and may be transferr’d without suffering any loss or alteration. . . . As the improvement, therefore, of these goods is the chief advantage of society, so the instability of their possession, along with their scarcity, is the chief impediment. (T, 3.2.2.7; Hume 2007, 313) Hume adopts the ancient view that one cannot be robbed of one’s peace of mind by the actions of another so that insofar as one can separate one’s peace of mind from external circumstances, that does not need to be protected by rules of justice. But others can rob one of the otherwise possible use of one’s body, though to no direct advantage of themselves, or of the use of external objects that one possesses, in this case perhaps very much to their own advantage; thus these categories of goods need to be protected by the rules of justice. Hume’s distinction between the latter two kinds of goods is too sharp, and strikingly overlooks the cases of slavery, or wage slavery, that is one person’s use of the body and bodily capacities of another very much for his own advantage. Nevertheless, what is implicit here is that these goods, but particularly the latter two kinds, are resources that one can use as one sees fit for the end of one’s own happiness or for the happiness of others, and that anyone who would rob one of these goods is limiting or destroying one’s own freedom to use them as oneself sees fit. Justice does not simply protect one’s goods, but one’s freedom to use one’s goods as one sees fit—subject, of course, to such abridgement of one’s own freedom to use those goods to the disadvantage of others, or unjustly. Justice permits me the use of my sword and buckler in self-defense, but not to rob others (EPM, 3.1; Hume 1998, 16). The same is implicit a page later when Hume says that justice, as

152  Paul Guyer “a convention enter’d into by all the members of the society to bestow stability on the possession of those external goods,” leaves “every one in the peaceable enjoyment of what he may acquire by his fortune and industry” (T, 3.2.2.9; Hume 2007, 314): the point of the convention or rules of justice is neither simply to secure the stability of our possessions nor directly to augment our happiness, but to augment our happiness by securing our freedom to use and therefore enjoy our possessions as we ourselves see fit. In this way justice works by securing our freedom of action with our own bodies and our possessions.5 Hume’s concern for freedom is even more explicit in two later passages in the Treatise, one in his discussion “Of the source of allegiance” and the other in his discussion “Of the objects of allegiance.” The first of these sections contains the argument that our obligation of allegiance to government does not depend on an antecedent obligation to keep promises that Hume would also present in his famous essay “Of the Original Contract,” first published in 1748 in Three Essays, Moral and Political.6 Hume’s position is rather that we see the value of observing the conventions of justice in the same way as we see the value of keeping personal promises, thus that the source of allegiance to both of these conventions is the same, but that the former does not depend on the latter. Just as with regard to promises “We blame all treachery and breach of faith; because we consider, that the freedom and extent of human commerce depend entirely on a fidelity with regard to promises” (T, 3.2.8.7; Hume 2007, 349), so in the case of justice it is “the freedom and extent of human commerce” that we aim to secure by our allegiance. But surely “commerce” here is not to be understood narrowly as buying and selling goods and services, or the import–export business (important as that was to eighteenth-century Glasgow or London, for example), but as all forms of human interaction that can affect the bodily and external goods of the actors, to use Hume’s previous classification. In other words, the point of the rules of justice is to secure the greatest possible freedom and extent of human interaction, or just action. This will ultimately serve the cause of human happiness because our happiness is best served by greater rather than lesser freedom to use our resources as we please. Finally, under the rubric “Of the objects of our allegiance,” Hume discusses our obligation to existing governments. He continues his argument against the idea that the legitimacy of government is grounded on any promise or contract, and that in fact the origins of most governments will not bear close examination (a point on which Kant will agree), but holds that nevertheless government generally deserves our allegiance— No maxim is more conformable, both to prudence and morals, than to submit quietly to the government, which we find establish’d in the country where we happen to live, without enquiring too curiously

Hume and Kant  153 into its origin and first establishment. Few governments will bear being examin’d so rigorously. (T, 3.2.10.7; Hume 2007, 357) I will return to this topic later, but for the moment I just want to cite Hume’s further statement that “nothing is more essential to the public interest, than the preservation of public liberty” (T, 3.2.10.16; Hume 2007, 361). My point here is that Hume does not describe the role of government as the preservation of security in our possession of goods, including the good of our own bodies; he describes it as the preservation of liberty, which will of course include the liberty to use our bodies and our possessions as we see best. Justice is to be useful, and what is useful is what is conducive to what is agreeable, that is, to happiness; but in particular justice conduces to happiness by securing liberty, for the liberty to act and use our resources as we see fit is our path to happiness. So the utility of justice lies in its preservation of freedom of action, which is a precondition of an agreeable life. Or as Neil McArthur puts it, Hume is certainly a liberal in one sense of the term: he believes that government must allow people to form their own private conceptions of the good and give them the freedom to pursue these conceptions in the context of an open society and a free market. McArthur (2016, 500)7 Let us now turn to Kant. There is no problem in showing that for Kant the rules of justice are designed to secure the maximum freedom of action for each compatible with an equal degree of freedom of action for all; that is explicit in Kant’s definition of the condition of justice or right (Recht) and in his statement of the “Universal Principle of Right.” The former says that “Right is . . . the sum of the conditions under which the choice of one can be united with the choice of another in accordance with a universal law of freedom” (MM, 6: 230), and the latter is that “Any action is right if it can coexist with everyone’s freedom in accordance with a universal law, or if on its maxim the freedom of choice of each can coexist with everyone’s freedom in accordance with a universal law” (MM, 6: 230). There is also no difficulty in establishing that for Kant the concept and principle of justice have no direct concern with anyone’s happiness; in the paragraph preceding his definition of the condition of right, Kant makes this clear by stating that right concerns only the form of action, that is, the interpersonal consistency of action, and not the matter of action, that is, the particular ends that individuals use their freedom to pursue, in the hope of their own happiness. Kant makes this point at length: The concept of right, insofar as it is related to an obligation corresponding to it (i.e., the moral concept of right), has to do, first,

154  Paul Guyer only with the external and indeed practical relation of one person to another, insofar as their actions, as deeds [Facta], can have (direct or indirect) influence on each other. But, second, it does not signify the relation of one’s choice to the mere wish (hence also to the mere need) of the other, as in actions of beneficence or callousness, but only a relation to the other’s choice. Third, in this reciprocal relation of choice no account at all is taken of the matter of choice, that is, of the end, each has in mind with the object he wants; it is not asked, for example, whether someone who buys goods from me for his own commercial use will gain by the transaction or not. All that is in question is the form in the relation of choice on the part of both, insofar as choice is regarded merely as free, and whether the action of one can be united with the freedom of the other in accordance with a universal law. (MM, 6: 230). Kant’s first point, that justice concerns only the external and indeed practical relation of the actions of one person to those of others, can be taken as a statement of his point, already made in the Introduction to the Metaphysics of Morals as a whole, that justice does not concern itself with people’s ultimate motivations, only with the effects of their actions (see MM, 6: 214, and 6: 218–9), although his reference to maxims in the statement of the Universal Principle of Right itself can be taken to mean that justice may properly concern itself with the intended actions of persons and their intended effects on the freedom of action of others, rather than just with actual outcomes (as is indeed the case in much criminal law, for example, in the crimes of attempted murder or conspiracy under U.S. law). The second two points are relevant to my present concern; however, they make clear that justice concerns only the preservation of the freedom of everyone to set their own ends, not their actual needs, as may be the case with a duty of virtue such as beneficence, nor with the wisdom of their choice of ends—thus justice does not directly concern the happiness of others. Yet Kant certainly assumes that happiness is what people are aiming at in their own choice of ends, or perhaps better that happiness is simply the state in which one’s ends are realized, or even more precisely the state in which some maximally compossible set of one’s ends is realized. Kant defines happiness in terms of some sort of global satisfaction of an individual’s desires—in the Groundwork for the Metaphysics of Morals, as “an absolute whole, a maximum of well-being [Wohlbefindens] in my present condition and in every future condition” (G, 4: 418), in the Critique of Practical Reason, adding the requirement of awareness of such a condition, as “a rational being’s consciousness of the agreeableness of life uninterruptedly accompanying his whole existence” (CPracR, 5: 22) and as “the general name for subjective determining grounds,” but such

Hume and Kant  155 that “in what each has to put his happiness comes down to the particular feeling of pleasure and displeasure in each . . . which can and must be very different in each” (CPracR, 5: 25). If justice allows all to pursue their own ends, constrained only by the requirement that everyone be allowed to do this to the maximal yet equal extent possible, and the ends are suggested by the desires of each, then what justice allows and indeed requires is simply that each be allowed to pursue her own happiness to the greatest extent compatible with all doing the same—although anyone actively promoting the happiness of anyone, themself or anyone else, is no part of the direct part of the requirement of justice. Justice, as Kant says, concerns the form of action, specifically its interpersonal compatibility—but of course all action has its matter as well as its form, the specific end at which it aims, the realization of which would in turn be part of the happiness of the agent. And this is why Kant could already say in the Critique of Pure Reason that in an ideal world, in which everyone is moral and actions have their intended outcomes, [A] system of happiness proportionately combined with morality can also be thought as necessary, since freedom, partly moved and partly restricted by moral laws, would itself be the cause of the general happiness, and rational beings, under the guidance of such principles, would themselves be the authors of their own enduring welfare [Wohlfahrt] and at the same time that of others. (CPR, A809/B837) —morality, of which justice is a part, is the framework—the form—for the individual pursuit of happiness—the matter. On Kant’s account, justice’s formal constraints on individual action are threefold: first, there is the innate right to freedom, which derives from the obligation of all persons to treat the humanity in each as an end and never merely a means, and which is comprised by the three requirements of equality before the law, being beyond reproach unless one has performed any act wronging or affecting the rights of others, and “being authorized to do to others anything that does not in itself diminish what is theirs, so long as they do not want to accept it,” that is, as long as one leaves them as free as one is oneself (MM, 6: 238); second there are the rights which each has the innate right to acquire, namely, the rights to acquire property, to enter into contracts, and to enter into long-term relations of marriage, parenthood, and employment consistent with that innate right to freedom of all involved or affected; finally, “public right” or law then concerns the governmental structures needed to make these innate and acquired “private rights” determinate and secure, and the rights and duties of both rulers and subjects within governments dedicated to this end and this end only.8 It can go largely without saying that the effect of this scheme is to leave individuals as free to pursue their own

156  Paul Guyer ends and therefore their own happiness as they can be consistent with the requirement that others be left equally free. But sometimes Kant does make it at least partially explicit that people will use the freedom that justice allows them to use objects and even each other for their own purposes. Thus in his explication of the “postulate of practical reason with regard to rights,” which is that “It is possible for me to have an external object of my choice as mine” as long as I satisfy the formal constraint of justice, Kant writes: For an object of my choice is something that I  have the physical power to use. If it were nevertheless absolutely not within my rightful power to make use of it, that is, if the use of it could not coexist with the freedom of everyone in accordance with a universal law (would be wrong), then freedom would be depriving itself of the use of its choice with regard to an object of choice, by putting usable objects beyond any possibility of being used . . . even though in the use of things choice was formally consistent with everyone’s freedom in accordance with universal laws. (MM, 6: 250) That is, to deprive ourselves of the use of objects for any reason except to preserve the consistency of our own use of choice with that of others would be irrational. Why would it be irrational? Because it would deprive us of the use of objects when we have no reason to do so. And why would it be irrational for us to deprive ourselves of the use of objects in such a case? Because we can use objects to realize our ends—and realizing our ends is what brings us happiness, the aim of our ordinary and empirical use of practical reason. In other words, the empirical use of practical reason is to realize our ends and thus realize happiness; the pure use of practical reason constrains that empirical use by the requirement of interpersonal consistency, but it does not do away with our ordinary pursuit of happiness at all. It just subjects the matter to its necessary form.9 So Hume holds that justice ultimately aims at human happiness, but recognizes that this goal can be achieved only through securing freedom of action; Kant holds that justice aims directly at equal freedom of action, but recognizes that freedom of action is ultimately aimed at happiness. In this regard, even though their terminology and the details of their accounts are different—Hume focuses on different methods of rightfully acquiring property given the peculiarities of human psychology, while Kant focuses more on the general classes of acquired right and then on some details of the public enforcement of right—the normative implications of their theories of justice are not as different as they may initially seem. Nevertheless, there are some significant differences between their views, especially when it comes to the rights of individuals against the

Hume and Kant  157 mechanism for the enforcement of their own rights, that is, government. But here the results are not exactly what one might expect, specifically the theory—Kant’s—which places more direct emphasis on freedom as the aim of justice does not always concede more liberty to the subjects of government than the theory that does not so directly emphasize freedom—Hume’s. I will illustrate what I mean by this in the following section.

3.  Paternalism and Resistance Kant’s direct emphasis on freedom as the aim of justice leads him to a visceral rejection of paternalistic government that we do not find in Hume. Yet Kant stops short of endorsing a right to forcible resistance against unjust government, while Hume emphatically defends such a right. Thus Kant is the greater friend of justice as freedom in one context, but Hume in another. Paternalism: Kant’s opposition to “paternalistic government” (väterliche Regierung, imperium paternale) is emphatic. His sensitivity to this issue is not surprising: as a citizen and indeed Beamte of Prussia, he would have been intimately familiar with paternalism in practice, especially with the efforts of Friedrich Wilhelm I and Friedrich II to develop the industrial basis of Prussia (which had such side benefits as their liberal immigration policies, for example, with regard to the displaced French Huguenots, although not with regard to Jews); and as a critical student of the philosophy of Christian Wolff, he would also have been familiar with Wolff’s conception of the monarchical state as an instrument of perfectionism.10 Kant writes: A government established on the principle of benevolence toward the people like that of a father toward his children—that is, a paternalistic government (imperium paternale), in which the subjects, like minor children who cannot distinguish between what is truly useful or harmful to them, are constrained to behave only passively, so as to wait only upon the judgment of the head of state as to how they should be happy and, as for his also willing their happiness, only upon his kindness—is the greatest despotism thinkable (a constitution that abrogates all the freedom of the subjects, who in that case have no rights at all). Not a paternalistic but a patriotic government . . . is the only one that can be thought for human beings, who are capable of rights, and also with reference to the benevolence of the ruler . . . only so as to consider himself authorized to protect its rights by laws of the common will but not to subject the use of it to his unconditional discretion. (TP, 8: 290–1)

158  Paul Guyer By paternalistic government, Kant means one where the ruler acts not merely to enforce the rights—the innate right to freedom and the rights to property, contract, and service acquired under the rubric of private right—of his subjects, but also acts to improve their well-being, although on his own conception of the latter. For Kant this is a violation of both the principle of right and the principle of the virtue of beneficence, which is the proper locus for concern for the happiness of others. First, Kant’s understanding of the principle of right, that each person must be allowed as much freedom in external action as is compatible with the equal freedom of others, is that right is limited to the enforcement of our perfect duties to others, indeed to a subset of our perfect duties to them, namely, our obligation not to destroy their existence and their freedom in the form of their innate and acquired rights. Right does not extend to the enforcement of our duties of respect to others—the duties to avoid arrogance, defamation, and ridicule (MM, 6: 464–8)—even though these are perfect, that is, determinate duties of omission (as with all perfect duties, the rule with regard to the duties of respect is completely determinate: just don’t do it); and it certainly does not extend to the positive duties of commission, namely, beneficence, gratitude, and sympathy (MM, 6: 452–8), which are indeterminate in the dual sense that no one owes them to specific other persons and no one owes an antecedently specifiable, precise amount of these to anyone else. Thus the positive duties—which Kant calls in the Doctrine of Virtue the duties of love—are unsuitable for enforcement by the coercive instrument of the state, especially in the hands of a ruler of the state who is not really likely to distinguish between the interests of his subjects and his own interests. Second, the ruler’s inevitable self-interest makes it particularly likely that he will violate Kant’s cardinal rule for the observance of the duty of beneficence even in the cases in which it is appropriate to act upon this duty, namely, that when anyone is to attempt to augment the happiness of others, “It is for them,” that is, for the recipients of beneficence, “to decide what they count as belonging to their happiness,” although it is also open to the would-be benefactor “to refuse them many things that they think will make them happy” (MM, 6: 388). That is, no one has an obligation to do whatever someone else thinks will make her happy, but insofar as one is to be beneficent one can be so only by doing something that the recipient does count as part of her happiness. This is really analytic: anyone’s happiness consists in the satisfaction of their own desires, so the only way anyone else can contribute to someone’s happiness is by helping the latter satisfy one of her desires, not by satisfying one of the former’s own desires. But powerful rulers will be tempted by their own power to do precisely that. Kant thus sees the confinement of government to the enforcement of the rights of the subjects and the proscription of paternalism as an essential expression of the promotion of freedom by justice. Hume does not worry about paternalism at all; perhaps this was a natural response to the

Hume and Kant  159 much more limited powers of the crown in the eighteenth-century United Kingdom of Great Britain and Ireland than in contemporaneous Prussia. On the contrary, Hume is more attentive to the way in which government can promote collective well-being than Kant is. At least as far as what he explicitly discusses goes, Kant limits government efforts to promote well-being to the use of taxes “to support organizations providing for the poor, foundling homes and. . . pious institutions generally (such as widows’ homes, hospitals, and the like),” based on the argument from the premise that “the general will of the people has united itself into a society which is to maintain itself perpetually” (MM, 6: 326) to the conclusion that each member of the society must be preserved as long as possible. Hume takes a much more generous view of how the government can provide for the welfare and happiness of the people, namely, by undertaking collective projects to provide common goods that would not be provided by individual enterprise. Hume says nothing about governmental support for the poor and the helpless, but does describe how government can unite people to produce collective goods.11 Indeed, he describes this in unabashedly paternalistic terms: [I]n the execution and decision of justice, men acquire a security against each other’s weakness and passion, as well as against their own, and under the shelter of their governors, begin to taste at ease the sweets of society and mutual assistance. But government extends farther its beneficial influence; and not contented to protect men in those conventions they make for their mutual interest, it often obliges them to make such conventions, and forces them to seek their own advantage. For example, while “two neighbors may agree to drain a meadow, which they possess in common,” thereby improving the condition of each, only the collective power of government led by magistrates can bring it about that “bridges are built; harbours open’d; ramparts rais’d; canals form’d; fleets equip’d; and armies disciplin’d” (T, 3.2.7.8; Hume 2007, 345). Kant has nothing to say about how such public goods will be provided in his regime of freedom. But he would surely balk at Hume’s confident statement that “Magistrates find an immediate interest in the interest of any considerable part of their subjects. They need consult no body but themselves to form any scheme for the promoting of that interest.” Kant is afraid precisely that if magistrates need consult nobody but themselves, they will do what is in the interest of themselves and their dynasties, which may or may not be in the interest of their subjects. Here is a point, then, at which the difference between Kant’s foundation of justice and the power of government on individual freedom and Hume’s foundation of these on a concern for happiness makes a difference between them, as well as a point at which their empirical assumptions differ,

160  Paul Guyer Kant assuming that governors with unbridled power will put their own interests ahead of those of their subjects while Hume assumes that magistrates will take that “immediate interest in the interest of any considerable part of their subjects.” The last clause—any considerable part of their subjects—might suggest that Hume supposes that magistrates will consult with their subjects and follow some democratic procedure—seeking majority support—for any large project, which will of course require taxation, taking property by eminent domain, and so on. But he gives no hint of that—this seems to be a point on which his concern for happiness trumps any concern for the liberty of the subjects, or more precisely any concern for their consent to expenditures on their behalf, even if such projects would ultimately increase their liberty to engage in commerce of whatever kind. The Right to Resistance: On this issue, however, positions are reversed, and Hume comes out looking like the greater advocate for freedom. Kant notoriously rejects any right to forcible resistance to an existing government. In the essay on theory and practice, Kant argues that the social contract is an idea of reason, not an historical fact: this means, on the one hand, that the idea “has its unbounded practical reality,” as “a touchstone of any public law’s conformity with right,” namely, “to bind every legislator to give his laws in such a way that they could have arisen from the united will of a whole people” (TP, 8: 297), but, on the other hand, that dissatisfied subjects cannot simply withdraw their obedience to their government as they might withdraw from a contract, because they have no actual contract either with each other or with their government. Kant then goes on to say that subjects cannot resist an “actual legislation,” but presumably also their government in general, because of their judgment that “it is in all probability detrimental to its happiness,” because the purpose of government is not the happiness of the people in the first place: For what is under discussion here is not the happiness that a subject may expect from the institution or administration of a commonwealth but above all merely the right that is to be secured for each by means of it. (TP, 8: 298) This seems like a red herring, for it would not be an objection to resistance or rebellion against a government on the ground that it is unjust, for example, arbitrary in its administration of law or completely lawless, rather than ineffective in the promotion of happiness. However, Kant has a second argument, which is that a right to rebellion would be selfcontradictory, because government exists to arbitrate disputes, but if the people take action against their head of state, then it would be “judge[d] in its own suit,” and “Hence there would have to be another head above

Hume and Kant  161 the head of state, that would decide between him and the people; and this is self-contradictory” (TP, 8: 300). In other words, recognizing a right to rebellion would be a formula for anarchy: if “the previously existing constitution has been torn up by the people, while their organization into a new commonwealth has not yet take place,” then “the condition of anarchy arises with all the horrors that are at least possible by means of it” (TP, 8: 302n). In the Metaphysics of Morals, Kant develops this line of argument. Here he affirms the “postulate of public right,” that people have a moral obligation to enter into and remain part of a state if they cannot simply avoid interaction with other people, which of course they usually cannot (MM, 6: 307). He then argues that there must be a monopoly of coercion in a state, thus that although the executive who is to exercise that monopoly is merely an agent of the legislature, which is the true expression of the people’s ultimate sovereignty, “it would be self-contradictory for him to be subject to coercion” from the legislature (MM, 6: 317); and all the more so would it be self-contradictory for the people acting directly, as a rabble, rather than through their legislature, [T]o be authorized to resist, that is, the highest legislation would have to contain a provision that it is not the highest and that makes the people, as subject, by one and the same judgment sovereign over him to whom it is subject. (MM, 6: 320) The freedom of each citizen that a state is to secure does not extend to the freedom to overturn the state and replace it with another, even if in the view of the citizens a replacement might be more just than their current state.12 Of course, Kant insists upon freedom of speech and freedom of the pen, and the right of citizens, at least as scholars using public reason, to bring injustices to the attention of their government and to petition for redress (see again TP, 8:304, and of course What is Enlightenment?). But this leaves it up to the ruler to respond, and even if the ruler may himself be regarded as having a moral obligation to reform his state in the direction of greater justice, the people have no right to coercively enforce this obligation. The ruler retains a monopoly on coercion. On this point, Hume takes the more radical line that we also associate with John Locke. Hume’s position is duplex. First, appealing to his derivation of the rules of justice and their enforcement by government as “a mere human invention for mutual advantage and security,” he infers that government “no longer imposes any obligation, either natural or moral, when once it ceases to have that tendency,” thus “that in the case of enormous tyranny and oppression, ‘tis lawful to take arms even against supreme power.” If the government is not doing its job of providing for mutual advantage and security, if all the worse it is promoting the contrary of that, then it has no lawful claim on the allegiance

162  Paul Guyer of its soon-to-be-former subjects, and they therefore have every right to rebel against it. Then, in a move reminiscent of Hobbes, he states that even though no constitution, even “the laws in limited monarchies,” will explicitly grant subjects the right to resistance, nevertheless “the people still retain this right of resistance; since ‘tis impossible, even in the most despotic governments, to deprive them of it” (T, 3.2.10.16; Hume 2007, 360; see also T, 3.2.10.1; Hume 2007, 354).13 That is, whatever the law may say and whatever a government may try to do, people simply will not suffer the deprivation of justice, at least past a certain point, indeed a bar set rather high by the natural conservatism of the human imagination, without attempting resistance. “The same motive of self-preservation, and the same motive of public good, give them the same liberty in the one case as in the other”; that is, the natural laws of self-preservation and interest in the welfare of others (“limited benevolence”) remain in force regardless of the civil laws of a government, and are just as or more powerful than the latter. The laws of nature are not suspended by civil law. I say this will be true at least past a certain level of injustice, though, because it is part of Hume’s empirical, psychological argument that familiarity will breed acceptance rather than contempt, that is, that people get used to existing circumstances and it will take a high degree of discomfort to get them to change. Thus, as I quoted Hume earlier, “No maxim is more conformable, both to prudence and to morals, than to submit quietly to the government, which we find establish’d in the country where we happen to live” (T, 3.2.10.7; Hume 2007, 257). Nevertheless, in virtue of both the natural drive for self-preservation and the logic of the function of government, the people retain the right of resistance against a sufficiently unjust government, and “Those, therefore, who wou’d seem to respect our free government, and yet deny the right of resistance, have renounc’d all pretensions to common sense, and do not merit a serious answer” (T, 3.2.10.16; Hume 2007, 361). Thus Kant founds his conception of justice on the demand for the greatest possible freedom for all, but places such a value on the coherence and continuity of government that he rejects a right to rebellion in favor of a right to criticize and petition, while Hume grounds our allegiance to government on the utility of security but concedes a much more liberal right to resistance. This could show just that things do not always come out as you might expect. Or it could show that in spite of the different theories, in practice freedom and happiness are so closely linked that a liberal conception of freedom can be justified just as or even more easily from the starting point of happiness than from that of freedom itself. Or it could show that even the most rigorous philosophers are influenced as much by their contexts as by their theories, thus that while it was unthinkable for Kant to recognize or at least espouse a right to resistance in absolutist Prussia, it was no big deal for Hume to trumpet such a right in a Great Britain that had already had its rebellion 50 years earlier and

Hume and Kant  163 was in no danger of having another any time soon, say within the next 280 years. Or all of the above.

Notes 1. Thus Kant’s response to Hume’s moral and political philosophy is not discussed in Kuehn (2005). For my own approach to Kant’s response to Hobbes, see Guyer (2012). On Achenwall as both source and target for Kant’s philosophy of right, see Guyer (2020a), and my “Introduction” in Gottfried Achenwall, Natural Law, Guyer (2020b). 2. The contents of Kant’s library were actually reconstructed by Arthur Warda in Warda (1922) from the sale catalog of his legatee Johann Friedrich Gensichen, so is somewhat speculative, and is in any case hardly a complete guide to Kant’s lifetime reading, since we know that Kant read many books at the bookseller Kanter’s without buying them, that he was also the university under-librarian for many years, in which capacity he could order and read books as he pleased, and he may have given away books that he owned before he died. There are many books that we are certain he did read which are not to be found on Warda’s list. 3. References to Hume’s Enquiry concerning the Principles of Morals (EPM) will be given by Section and Part number, where the Section is divided into Parts, and the page number from the edition by Tom L. Beauchamp, Oxford: Clarendon Press (Hume 1998). References to A Treatise of Human Nature (T) will be by Book, Part, Section, and paragraph number as well as the page number from Volume I of the edition by David Fate Norton and Mary J. Norton, Oxford: Clarendon Press (Hume 2007). References to Hume’s Essays Moral, Political, and Literary will be by title and page number in the edition by Eugene F. Miller, Indianapolis: Liberty Fund, revised edition (Hume 1987). 4. In his discussion of whether Hume is an act- or a rule utilitarian, Roger Crisp understands rule utilitarianism as requiring that action which would produce the best results if everyone followed the same rule, even if in fact they do not, and denies that Hume is a rule utilitarian on that ground; see Crisp (2019, 219). I am rather taking either strengthening or weakening the general disposition to follow a rule to be part of the actual consequences of any particular action, under its real normal circumstances (i.e., not worrying about the ring of Gyges or actions performed in secret in some thought-experiment), and therefore denying that there is a real distinction between act utilitarianism and rule utilitarianism. 5. Thus I would take an even stronger position than Richard H. Dees. He writes that in addition to being useful for peace and security, justice and its enforcement by government “is essential for a very different kind of public good: liberty. . . . Suddenly, peace and security are not the only issue. Government should provide for liberty as well”; see Dees (2008, 397). My suggestion is rather that for Hume security is in good part the security of liberty, so liberty is not a separate good from security.   Knud Haakonssen emphasizes Hume’s tripartite analysis of the goods of the mind, of the body, and external goods, but argues that Hume does not erect a traditional, natural law theory of rights to these goods; see Haakonssen (1996, 117–19). It is correct that Hume does not use the language of rights, but this terminological point misses his substantive insistence that government must provide security of liberty. Haakonssen’s statement that “the interest in government” is “an interest in external and internal protection”

164  Paul Guyer (ibid., 112) similarly misses Hume’s emphasis that government is to protect liberty. 6. See Hume (1987, xiii). 7. Elsewhere, however, McArthur qualifies this statement, referring to Adam Smith as well as Hume: “Hume and Smith take a pragmatic, consequentialist approach to the role of the state, one very distant from the more dogmatic libertarianism adopted by many later free-market advocates. For these two thinkers, the presumption against state action is a defeasible one that must be tested against available evidence and reasonable predictions about what will promote the general happiness.” McArthur (2011, 324) 8. That Kant treats innate right so briefly, in only two pages of the Introduction to the Doctrine of Right, rather than assigning it its own “Part” as he does for both acquired and public right, should not mask its foundational role for the subsequent categories of right, nor should the sequence of private and public right lead to the misunderstanding that public right is meant to secure only private right; much in both criminal and civil law, including criminal and civil procedure, is meant to secure elements of innate right such as freedom of the person and freedom of speech. Thus Arthur Ripstein correctly states that “The innate right of humanity is the basis of any further rights a person must have” (Ripstein 2009, 56), and Sharon Byrd and Joachim Hruschka state that “The assumption that everyone has a right to external freedom is the logical starting point for Kant’s Doctrine of Right” (Byrd & Hruschka 2010, 77). 9. For a relatively early passage in which Kant describes the whole of morality, of which right will be a part, in terms of the subjection of empirical matter to a priori form, see Reflection 6820 (1776–78?): “Moral philosophy is the science of ends insofar as they are determined through pure reason. Or of the unity of all ends (that they do not conflict with themselves) of rational beings. The matter of the good is empirical, the form is given a priori.” (Refl, 19: 172) 0. See Behrens (1985, 178–9). 1 11. Dees stresses this element of Hume’s view of government, although without any qualm about potential paternalism; see Dees (2008, 397). See also McArthur (2016, 501). 12. Both Ripstein and Byrd and Hruschka attempt to downplay Kant’s resistance to a right to resistance by arguing that Kant’s stricture applies only to governments that “satisfy the postulate of public right” rather than merely enjoying a “near monopoly of force” (Ripstein 2009, 337), or are a genuine “state” rather than a mere “den of thieves” (Byrd & Hruschka 2010, 183). It seems fair enough to claim that a state that allows a condition of anarchy is not a genuine state at all, thus that in such a case rebellion cannot be prevented from sliding into anarchy because there is already nothing but anarchy; but Kant’s position seems to be rather that there can be no transition from a worse to a better state without passing through a condition of anarchy that is worse than the former, thus that any state is better than none at all. 13. The latter passage is also cited by McArthur (2011, 321).

Hume and Kant  165

References Behrens, Catherine B. A. (1985). Society, Government, and the Enlightenment: The Experiences of Eighteenth-Century France and Prussia. New York: Harper & Row. Byrd, Sharon and Joachim Hruschka (2010). Kant’s Doctrine of Right: A Commentary. Cambridge: Cambridge University Press. Crisp, Roger (2019). Sacrifice Regained: Morality and Self-interest in British Moral Philosophy from Hobbes to Bentham. Oxford: Clarendon Press. Dees, Richard H. (2008). “ ‘One of the Finest and Most Subtle Inventions’: Hume on Government.” In Elizabeth S. Radcliffe (ed.), A Companion to Hume. Oxford: Blackwell, 388–405. Guyer, Paul (2012). “ ‘Hobbes Is of the Opposite Opinion’: Kant and Hobbes on the Three Authorities in the State.” Hobbes Studies 25, 91–119. Guyer, Paul (2020a). “Achenwall, Kant, and the Division of Governmental Powers.” In Margit Ruffing, Annika Schlitte and Gianluca Sadun Bodoni (eds.), Kants Naturrecht Feyerabend. Berlin: Walter de Gruyter, 201–28. Guyer, Paul (2020b). “Introduction.” In Pauline Kleingeld (ed.), Gottfried Achenwall: Natural Law, translated by Corinna Vermeulen, with an Introduction by Paul Guyer. London: Bloomsbury Academic, xiii–xxxi. Haakonssen, Knud (1996). Natural Law and Moral Theory: From Grotius to the Scottish Enlightenment. Cambridge: Cambridge University Press. Hume, David (1987). Essays Moral, Political, and Literary. Ed. by Eugene F. Miller, revised ed. Indianapolis: Liberty Fund. Hume, David (1998). Enquiry Concerning the Principles of Morals (EPM). Ed. by Tom L. Beauchamp. Oxford: Clarendon Press. Hume, David (2007). A Treatise of Human Nature (T). Ed. by David Fate Norton and Mary J. Norton. Oxford: Clarendon Press. Kuehn, Manfred (2005). “The Reception of Hume in Germany.” In Peter Jones (ed.), The Reception of David Hume in Europe. London: Thoemmes Continuum, 98–138. McArthur, Neil (2011). “The Scottish Enlightenment.” In George Klosko (ed.), The Oxford Handbook of the History of Political Philosophy. Oxford: Oxford University Press, 319–30. McArthur, Neil (2016). “Hume’s Political Philosophy.” In Paul Russell (ed.), The Oxford Handbook of Hume. Oxford: Oxford University Press, 489–504. Ripstein, Arthur (2009). Force and Freedom: Kant’s Legal and Political Philosophy. Cambridge, MA: Harvard University Press.

9 Reading Fichte Today. The Prospect of a Transcendental Philosophy1 Marco Ivaldo

Addressing the theme of ‘Reading Fichte today. The prospect of a transcendental philosophy’ presupposes first of all a brief outline of the philosophical landscape in our age, within which transcendental philosophy has its own role to play and its ‘destination’ (Bestimmung) to seek. This is done in the first part of this chapter. Then in part 2, a fundamental account is given of what Günter Zöller has called the ‘Geist’ (mind/spirit) or ‘the inner form’ of transcendental thought in Kant and Fichte. Finally in part 3, by means of developing the consequences of the second part, it is my intention to focus in particular on Fichte’s science of knowing (Wissenschaftslehre), putting the emphasis on some of the aspects that still (could) invite or impel us to read Fichte afresh today.

1.  The Current Framework for Philosophizing In order to delineate the philosophical framework in which the question of the current meaning and definition of a transcendental philosophy is to be considered, I shall briefly take up a thesis of the Polish philosopher Marek J. Siemek. He is, among other things, the author of a pioneering study on Die Idee einer Transzendentalphilosophie bei Fichte und Kant in 1984 (Hamburg: Meiner). In his essay Das Denken der zweiten Hälfte des 20. Jahrhunderts—in the remarkable volume Vernunft und Intersubjektivität. Zur philosophisch-politischen Identität der europäischen Moderne of 2000 (Siemek 2000)—Siemek espouses the following main thesis. The substantial transformation that has occurred in the realm of philosophical thinking in recent decades justifies the assumption that today, before our very eyes, a new self-understanding of philosophizing itself is taking shape, which is ‘qualitatively different’ from the thinking in the first half of the twentieth century. Within the complex process of a gradual remodeling of the entire theoretical reality that characterizes the state of philosophy in our time, not only is a new ‘object’ and a new ‘method’ for practicing philosophy being developed but also a new level of its thinking, that is a new standpoint of its mental ‘seeing.’ In brief, the whole structure of the ‘previous’ field of philosophical theory DOI: 10.4324/9781003044376-10

Reading Fichte Today  167 is, according to Siemek, being radically transformed at this turning point in time and a new field of thought is being created. What is it, however, that distinguishes the ‘previous’ field or old paradigm of philosophy? Siemek says that its hallmark is the coexistence and connection between two polarities of thought, namely, the unity and the opposition at the same time of (roughly put) a ‘philosophy of knowing’ and a ‘philosophy of human living.’2 ‘Knowing’ and ‘living’ are the two opposite poles in the old field of thought, which refer to and ground one another reciprocally. Between them there exists, as it were, a kind of ‘disunited unity’ (concordia discors). It is striking that the theoretical breakthrough that is such a characteristic feature of philosophy in the second half of the last century was made manifestly at the expense of the two extreme but also most influential currents of thinking in the first half: neopositivism as successor to ‘philosophy of knowing’ and philosophy of existence as successor to ‘philosophy of human living.’ This is the post-Cartesian paradigm of the world constituted by an unbridgeable dualism of pure subjectivity and absolute objectivity, which in the twentieth century reached its acme in the duality of neopositivism and philosophy of existence (ontology of existence) and has been radically challenged by the philosophical breakthrough since the second half of the last century. The ‘disunited unity,’ in which the positivist ‘philosophies of knowing’ coexist with the existential-anthropological or ontological ‘philosophies of human living,’ is of an internal-theoretical nature. Both standpoints and modes of thinking to some extent generate one another as they both originate and function within the confines of that difference by which they are constituted. Now, where the convergent decline of these two apparently very divergent formations leads to is not only the collapse of both but also the disintegration of their common foundation, namely, the entire mental paradigm itself in the first half of the twentieth century with all its historical ontological background. In other words, the division of the constitution of the world into the independent spheres of a pure subjectivity of thought and an absolute subjectivity of being. What is it, then, that in Siemek’s view characterizes the new field of thought, the ‘new level of thinking?’ A somewhat metaphorical answer is that this new level of thinking came into being as a consequence of the opening up of an essential ‘dimension’ of philosophical questioning that reaches beyond the old paradigm. At this new level, questions are posed that concern the entire structure and the range of the purely ontic level of thought, which (still) in reality presupposes the division into the subjective and the objective. Siemek calls the new level ‘epistemological’ or also ‘transcendental’ and the level that still comes from knowing and being ‘epistemic.’ From the epistemological or transcendental perspective, thought is seen to be characterized by a ‘two-dimensional’ structure. Its starting point is

168  Marco Ivaldo always a form of the simple episteme, that is our knowledge of the world. The starting point for transcendental thought is always some concrete form or other of the theoretical–practical appropriation of reality by man in his understanding experience. Yet, from the epistemological point of view, that simple episteme (= knowing about the world) is transposed simultaneously to the pre- and the outer-epistemic horizons of its possible self-understanding. Thus, in the new paradigm philosophy is not thought that is merely found and moves in the forms of experience as is the case in everyday knowing or in individual spheres of knowledge. It is a thinking or self-thinking of these forms themselves, a reflection (Besinnung) on them as significant moments of a more encompassing whole, to which they on each occasion belong but which never itself appears directly in them. At the transcendental level of thinking, thought attempts to reconstruct reflexively the integral system of relations and differences, within which the forms of consciousness and knowledge can also emerge as relevant basic units of possible sense. Understood in this way, epistemological philosophy or transcendental philosophy would thus at the same time be the right ontology for the epistemic forms of human experience. For here thought tries to penetrate through everything that these forms ‘say’ directly to what they actually do in this saying, that is what they really are. Important for my present purpose is that, according to Siemek, it was precisely Kant and Fichte who were the main initiators of this turning point in thought, which is still active in contemporary thinking and that— if I  understand him correctly—should consequently develop a current form of philosophizing. It is with Kant’s critique of reason and Fichte’s science of knowing that the new paradigm of thought called epistemological or transcendental first comes clearly into view. It provides neither a (realistic) ontology of the ‘object’ nor an (idealistic) ontology of the ‘subject’ but, rather, encompasses and unfolds an ontology of the subject (thought) and of the object (being) as a whole. My aim in what follows is to bring to light the ‘inner form’ or the ‘Geist’—as Fichte would say—of this transcendental position of thought in Kant and Fichte.

2.  The Inner Form of Transcendentality Fully aware that only a very selective and limited exposition is possible, the sense of the transcendental will be explained on the basis of the “Thirty Theses on Transcendental Philosophy in Kant and Fichte,” which Günter Zöller has recently formulated anew and which, in my view, provides a compact theoretical elucidation of the meaning of transcendentality itself.3 These theses are divided into three groups on “Transcendental Philosophy in general” (Theses 1–6), “Transcendental Philosophy in Kant” (Theses 7–17), and “Transcendental Philosophy in Fichte” (Theses 18–30). They are expressly formulated by the author as a

Reading Fichte Today  169 “basis for discussion” and state the fundamental notions of the common transcendental philosophical project in Kant and Fichte, which is to be apprehended beyond, or rather, prior to any conceptual or methodological differences. I shall look at only some passages. The starting point is the thesis that transcendental philosophy in general is ‘emended First Philosophy or Metaphysics.’ There is no insurmountable opposition between transcendental philosophy and metaphysics. Only the ‘old,’ ‘dogmatic,’ ‘ontic-realistic’ metaphysics is rejected from a transcendental standpoint. Instead, for Kant transcendental philosophy is prote philosophia, first philosophy, which lies objectively at the center and systematically at the beginning of philosophical knowledge. Compared to the ‘old’ metaphysics, transcendental philosophy as first philosophy is in Kant modern in a very specific sense: not dogmatic but epistemological such that it is appropriate to call it the ‘metaphysics of metaphysics’ or metametaphysics. Transcendental philosophy is characterized as ‘emended’ metaphysics provided that it recasts the old question of metaphysics as to the forms and principles of being into the question concerning the conditions for the possibility of knowledge about the forms and principles of being. This recasting—from being to knowledge of being—contains a turning to the transcendental subject. Nonetheless, in transcendental philosophy, subjectivity is closely correlated with objectivity so that it can be argued its objects are the forms and principles of the grounding of objectivity through (transcendental) subjectivity. Transcendental philosophy is hence at one and the same time a transcendental theory of subjectivity and objectivity and at its origin a theory of subjective objectivity—as Fichte would say. With regard to Kant, it is Zöller’s fundamental conviction that transcendental philosophy is confined by Kant’s critique of reason to being a pure theoretical philosophy through exclusion of non-pure theoretical philosophy, practical philosophy, and the reflective faculty of judgment. Kant in fact has a ‘rigorous idea’ of the transcendental owing to which he limits transcendental philosophy to the grounding of the possibility of theoretical synthetic a priori judgments. In this respect, it is in my view worth discussing whether Kant’s moral philosophy also belongs in principle to transcendental philosophy although Kant never refers to his critical moral philosophy as transcendental (the same can also be said of the theory of the reflective faculty of judgment). However, as Zöller maintains, a crucial point in Kant’s critique of reason is that transcendental philosophy limits theoretical knowledge of subject and object to things as they appear under the conditions of sensibility (phenomena) as opposed to things as they are independently from the conditions of sensibility (things in themselves, noumena). The consequence according to Zöller is that with respect to the classical themes of metaphysics (God, soul, world/freedom) transcendental emended metaphysics in Kant proceeds

170  Marco Ivaldo first in a limitative and second in a postulatory manner. Limitative, to begin with, in that it excludes atheism from theology, materialism from psychology, and fanaticism from cosmology. Restricting transcendental metaphysics to a general metaphysics of appearances and to a limitative specific metaphysics (of noumena) also has, however, a positive function: the grounding of the possibility of moral philosophy as theory of freedom. Restriction to the theory of phenomena and extension of moral philosophy to the theory of individually conceived things in themselves (purposes in themselves) thus correspond to one another. In addition, transcendental emended metaphysics proceeds in a postulatory fashion as the combination of the restriction on the theoretical side and extension on the moral practical side leads in Kant to a theoretical–practical mixed metaphysics of postulates. Accordingly, the themes of metaphysics that are excluded from the epistemology of transcendental philosophy (God, soul, world/freedom) are reinstated in a theoretical–practical or postulatory manner—in Kant the postulate of practical reason constitutes precisely a theoretical–practical ‘mixed form’ of thought.4 Fichte takes up the Kantian combination of a theoretical limitation of knowledge to the sensible and a practical extension of knowledge to the supersensible. Zöller’s main thesis is that the mixed form of postulatory thought, which Kant restricts to positing the existence of God and the immortality of the soul, is turned by Fichte into the universal principle of method of transcendental philosophy as such. As a result, the disciplines of transcendental philosophy and moral philosophy that are still separately carried out in Kant are integrated by Fichte into a practical-theoretical basic science, which is called the ‘science of knowing’ (Wissenschaftslehre). This transcendental philosophy modified to a science of knowing becomes in Fichte a ‘system of freedom.’ With respect to the Kantian theory of freedom, in Fichte freedom is not only—as it is for Kant—the principle of practical philosophy5 but also the principle of theoretical philosophy. It could be said that for Fichte freedom plays the role of a general systematic principle.6 On the other hand, Zöller points out that in its function of setting principles the concept of freedom in the science of knowing is precisely Kant’s idea itself of freedom as autonomy, as self-legislation of reason, which is by Fichte fully thought through and extended further. As a system of freedom, the science of knowing thus becomes the theory of freedom’s self-legislation (and self-fulfillment)— both in the theoretical (knowledge) and in the practical (will). Fichte’s modified transcendental philosophy (i.e., science of knowing) shows itself to be the theory of the conditions for the possibility of knowledge, in terms of both its mode and its domain of objects. This theory is knowledge of knowing or knowing-knowledge, which presents itself as meta-knowledge of all other knowledge and forms of knowing. To take up a term used by Siemek, which we have already come across, it is neither ontology nor gnoseology in the traditional sense but

Reading Fichte Today  171 ‘epistemology,’ that is science of knowing in itself and for itself. As such, the science of knowing is a theory of the absolute nature of theoretical and practical knowledge or, as Zöller puts it, “critical theory of absolute knowledge,” knowledge in and for itself, which has to reach completion as knowledge’s speculative self-understanding of its conditions and limits—a completion that leads or has to lead from knowing to ‘wisdom.’ From this it emerges that the science of knowing has itself to pursue a specific integration of the ‘maximal idealism’ of absolute knowledge and the ‘minimal realism’ of the absolute. The absolute norm for this reciprocal integration of the absolute and absolute knowledge states that the absolute shall become knowledge, that is it shall appear; and knowledge shall be incorporated into the absolute, that is it shall become an increasingly transparent appearance of the absolute.

3.  On the Idea of a Science of Knowing Zöller’s 30 theses provide us with a full picture of the inner form of transcendentality in relation to Kant and Fichte. He is quite frank, which I  find praiseworthy and elucidatory, about the fact that he critically combines transcendentality and metaphysics in the form of an emended first philosophy. Kant’s transcendental philosophy is accordingly interpreted by him as “Prima philosophia moderna.” Now, in the last part of my chapter I would like to point out some aspects of this inner form of the transcendental, focusing in detail on the idea itself of the science of knowing. By idea I do not mean the system’s specific content or relevant main concepts but, rather, its generating form of thought. I hope that by doing so, the critical-constructive potential of a reading of Fichte will now become more apparent. With his science of knowing Fichte is not advocating merely a philosophy of subjectivity, which is opposed to a prior ‘ontic-metaphysical’ philosophy of objectivity. The science of knowing constitutes instead a theory—or a praxis of reflection (Besinnung)—that has as its object the constituting acts or performances of the object-subject-relation or—as seen earlier—of ‘knowing.’ Consequently, the science of knowing is modified transcendental philosophy,7 as is said in the first lecture of the Second presentation of the Science of knowing 1804: neither theory of being nor theory of thought as separate, but integrated and at the same time differentiated understanding of (pure and concrete) being-conscious—construed as a reference to being or as ‘relation of being.’ In the late phase of Fichte’s philosophy, this reference to being is—as is well known—characterized and interpreted, among other things, as ‘image,’ ‘being-image,’ and ‘image-forming’ (Bilden).8 According to Reinhard Lauth, the entire science of knowing should be conceived of more as a “theory of imageforming (Bildenslehre)” (Lauth 2004, 81), namely, as self-critical understanding of image-forming in its theoretical and practical activities.

172  Marco Ivaldo At the beginning of the Halle-Manuscript-copy of the Science of knowing 1813, such a theory is “also there to see knowing, the one and general knowing in its origination, to see it coming out from what is not knowing, from what is principle (Prinzip), not derived (Prinzipiat)” (GA IV/6: 273). The science of knowing is thus conceived of as genetic knowledge, as knowledge of the genesis (= ‘principle’) of knowing facts (or experience). ‘Seeing knowledge in its origination’: the science of knowing goes back to the genesis of the fact of consciousness, therefore the idea, in order to make its coming into being clear or to reconstruct it. As genetic knowledge, the science of knowing does not coincide with the phenomenological description of the facts of consciousness; the latter only acts, rather, as an introduction to the science of knowing. The science of knowing does not coincide either with what Fichte calls in this connection ‘theory of being,’ although the science of knowing in itself contains an ontology,9 which is however never to be separated from the Kantian moment of criticality (criticism). For the science of knowing, being is always understood as being in being-conscious. Accordingly, objective being makes up only one side of transcendental unity, the other side of which is formed by (subjective) thought. The a priori analyticsynthetic unity of being and thought is ‘pure knowledge’ or ‘absolute knowledge,’ namely, knowing understood in its essence and immediacy and in its identity in becoming. The philosophers who want to advocate a realistic-objective theory of being—Fichte refers in this respect more often to Spinoza,10 though also Schelling is meant11—in reality neglect to reflect on that image of being that they themselves are; they omit in actu to reflect on their thinking on being. This transcendental criticism of the realistic-ontological standpoint could, as I see it, also be extended to those post-Kantian fundamental ontologies that claim to proceed without any critical-reflective or ‘epistemological’ mediation of their propositions. For Fichte, ontology and critique of reason, emended metaphysics, and transcendentality must remain indivisible. On the other hand, a one-sided idealistic elucidation of reality by the science of knowing, in which the difference between being and thought is rendered indistinct to the advantage of thought, is explained as being equally inadequate. Fichte’s basic conception is that we always have access to reality (to being) only through (pure and particular) knowing and in knowing—therefore in image-forming—within the frame of a transcendental unity of being and thought, which has to be conceived of not as fact but as act (= transcendental apperception in actu). Seen transcendentally, ontology (theory of being) can always only be obtained from epistemology (= in the sense of knowing in its genesis). In light of this forming unity of being and thought—thanks to this opening up of the ‘transcendental field’12—neither a subject as creator of the object or of being, nor an object of which the subject or self is a mere reflection or product may be hypostatized. Rather, in light of this unity, the sense

Reading Fichte Today  173 of the subject–object relation and the practical, intuitive, and theoretical forms in which it appears (called phenomena) have to be examined. The science of knowing proves to be philosophy of the sense of human experience. In the introduction to the so-called Staatslehre (Theory of the State, see GA II/16: 15ff) Fichte explains that (transcendental) philosophy does not deal with ‘things’ but with ‘images,’ that is with ‘cognitions’ of things. On the basis of transcendental unity, these are in fact to be understood as the (intended) correlate of the images. Hence the philosopher has to distinguish between cognitions that present themselves by their mere presence and cognitions that are not directly known of and whose essence lies in the fact that from them it is possible to know the basis of the specificity of the first kind. In other words, cognition in its facticity (its ‘single being’) is to be separated from cognition ‘in its origin.’ The first is direct showing-itself or offering-itself of the image, the appearance of something, a phenomenon in its elementary immediate givenness. The second is cognition in so far as in its origin it is understood from its own conditions of being possible. Renewing his philosophical vocabulary, Fichte calls the first form of cognition “intuition” (Anschauung), the second “understanding” (see GA II/16: 22ff). Intuition means factual cognition of something; it generates the first showing-itself of something as image and in an image. Understanding, in contrast, is genetic cognition: it is the comprehension of facticity, that is of the (first) appearance, as appearance, starting from the law of its origination. Understanding has to do, then, with the law or basis of appearance. Accordingly, (transcendental) philosophy is—as mentioned earlier— the knowledge of all cognition or the cognition of the image-being in and through understanding, that is in and from principles. It is knowledge of the transcendental genesis of image, science of the origination of the phenomenon from what Fichte calls ‘essence’: the existential act of existing life. But if philosophy is such a theory, then it shows itself to be a liberating science, indeed as ‘science of freedom.’13 That is, this science frees us from the ‘chain of things in themselves,’ for example from the belief that the self, consciousness, is the mere product of natural processes following the laws of nature, as current ‘naturalism’ claims to demonstrate. However, philosophy also frees us from the fascination of the image itself since as self-critical thinking, it precludes the risk that appearance becomes absolute, that it hypostatizes and is equated with essence. Philosophical reflection explains to the contrary that image as the image of something, as phenomenon, only comes into being in accordance with a law of (practical and theoretical) thinking—a law that the image-forming (Bilden) does not at all make for itself but that governs it internally. Genetic knowledge does not, however, in any way mean to let oneself be guided by that law in conformity with which the respective given image originates, as is the case in natural cognition. Genetic is rather that cognition

174  Marco Ivaldo which remains at the same time conscious of the image and its law of origination. Philosophy is therefore free knowledge but, more than that, it is also a liberating act. Genetic philosophy comes into being through exercising the freedom of reflection and self-reflection. The late Berlin lectures bear clear witness to this definition of philosophy.14 Now, the theory of knowledge, the knowledge of freedom, shows itself in its concrete unfolding as not only the system of transcendental cognitions but also, and at the same time, a never finally determined and determinable result of the being active of thought, which—never satisfied with the results achieved—contends with new issues, opens up original perspectives, and probes new domains of thinking. The different expositions of the theory of knowledge within the manuscripts of Own meditations (Eigne Meditationen), which Fichte wrote throughout his lifetime and which the historical–critical edition has made accessible—for example, the Diaries of 1813 and 1814—provide clear evidence of this essential feature of the theory of knowledge—not only as doctrine (theory) but also as activity of thought, as praxis of reflection understood and carried out. As self-critical penetration of knowledge, as knowingknowledge, the theory of knowledge (as indicated) is to be developed through a definite praxis of reflection and self-reflection, which must be consciously exercised by philosophers in their research and teaching in order to achieve systematic content, that is, for the very depiction of transcendental knowledge. Accordingly, the theory of knowledge is system and praxis of reflection or ‘praxeology’ in one, whereby the latter, praxeology, presents itself as the dynamic premises, the accompanying precondition for the origination of the former, thus of the system itself. This ‘praxeological’ dimension in the construction of the science of knowing—for Fichte meaning philosophy itself—has to be explicitly taken into consideration, reflected upon critically and made use of methodically by transcendental philosophy. That the science of knowing is not only a theory of reason but also and at the same time a praxis of reason, and in fact a self-reflective praxis of reason, concerns not only the conveying of an already constituted content ready to be taught publicly. Its main function is the very construction, or constitution, of the transcendental system. It is, in other words, no lifeless structure of fixed concepts but a living organism of creative thinking, which rests on the intelligence’s intellectual intuition or on looking-at-oneself (or looking-into-oneself) and reproduces the immanent systematicity of the human mind itself. The ‘system of image-forming’—taking up again Reinhard Lauth’s definition—has to be consistently acquired and formulated in the living performance of thought, thus in image-forming itself.15 In the Second presentation of the Science of knowing 1804 one finds the following proposition, which seems to me to be of considerable importance as it brings to light a kind of ‘teleology of human reason’: “Only right knowing or wisdom is of value” (GA II/8: 379).16 Wisdom

Reading Fichte Today  175 or right knowing means here the unity of living insight and the correct conduct of living, of well-founded theory and true praxis. As theory of absolute knowledge, the science of knowing is to lead—in the act itself of self-fulfillment and self-conquest—to wisdom. Fulfillment takes place in practical self-surrender, in the act of right intention and action (“doing what is right”), which is called wisdom. Equally necessary for the whole idea of the science of knowing is the theoretical–practical continuation of the systematic fundamental standpoints in life, the transition from knowing to wisdom, as well as the thesis of an original, mutually referring affinity between life and philosophy.

4. Conclusion At the end of his outstanding portrayal of Fichte in Fichte lesen, referring to the three great systemizers in classical German philosophy (Fichte, Schelling, and Hegel), Günter Zöller notes that from a distance of two centuries their thought “really seems near and far at the same time.” He explains: [N]ear in its critical and self-critical focus on reason and freedom, but also far in the belief and confidence that it can provide a vision of the whole in which it shows and proves itself to be a malleable object of human co-determination. Zöller (2013, 104) Now, it is exactly this tension between near and far, holding firmly onto reason and freedom on one side and the untiring attempt to contemplate the whole in thought on the other, that in my view makes for the fascination and power of attraction which still today radiates from these great philosophies. This dialectic between free reason (or freedom based on reason) and the striving to grasp the principles of the whole of reality in thought takes on specific aspects, even alternative features in each of the three great systemizers. In this regard, Wolfgang Janke talks of three “forms of achievement” of idealist metaphysics,17 namely, Hegel’s ontologic, Schelling’s late philosophy, and Fichte’s theory of the appearance of the absolute, which may be neither reduced nor classified in genealogical succession to one another. They follow their own particular paths to truth; they are independent forms of a fundamental standpoint of reason. Janke puts forward the thesis that of the three forms of achievement in German idealism mentioned, only that of Fichte keeps to transcendental reflection. To put it another way: only Fichte’s science of knowing maintains from Kant the unity of transcendentalism and (emended) metaphysics. Given this state of affairs, I hold it to be appropriate and productive to read Fichte today, to grasp him in his transcendental core, and “to critically, comparatively, and creatively measure him” Zöller (2013, 104)

176  Marco Ivaldo against the other competing achievements in thought. The goal has to be as before the “understanding through reason of reality”—the title of a book by Reinhard Lauth18—and to fulfill the destination (Bestimmung) of philosophy creatively.

Notes 1. I am greatly indebted to Erich Fuchs for reading the original text in German and improving the language. 2. All translations are my own. 3. Cf. Zöller (2017, 175ff). See also Zöller (1998). 4. On the theory of postulates, see Guyer (1997). 5. See Ivaldo (2009) on the concept of freedom. 6. See Ivaldo (2012) on this. 7. See on the Fichtean idea of transcendental philosophy: Lauth (1965), Ivaldo (2014). 8. For an overview of the theory of images, see Bertinetto (2010). See also Drechsler (1955). 9. On the theme of Seinslehre by Fichte see Brachtendorf (1995) and Marcuzzi (2018). 10. On Fichte and Spinoza I refer the reader to my essay Ivaldo (1992). 11. See Ivaldo (1990). 12. I am grateful to Jean-Christophe Goddard for his suggestions as to how to interpret transcendentality from this perspective. See Goddard (1999). 13. See Tilliette (2003). 14. On the late Berlin phase see d’Alfonso, Ivaldo (2014). See also Furlani (2004). 15. On this see Ivaldo (2017). 16. On the concept of wisdom in Fichte see Rametta (2012) and Zöller (2013). 17. See Janke (1993, 2009). 18. Lauth (1992).

References Bertinetto, Alessandro (2010). La forza dell’immagine. Argomentazione trascendentale e ricorsività nella filosofia di J. G. Fichte. Milan: Mimesis. Brachtendorf, Johannes (1995). Fichtes Lehre vom Sein. Eine kritische Darstellung der Wissenschaftslehren von 1794, 1798/99 und 1812. Paderborn: Ferdinand Schöningh. D’Alfonso, Matteo and Ivaldo, Marco (2014). “Fichte 1810–1814. Theoretical Philosophy.” Rivista di storia della filosofia 4, 579–583. Drechsler, Julius (1955). Fichtes Lehre vom Bild. Stuttgart: Kohlhammer. Furlani, Simone (2004). L’ultimo Fichte. Il sistema della Dottrina della scienza negli anni 1810–1814. Milan: Guerini. Goddard, Jean-Christophe (1999). La philosophie fichtéenne de la vie. Le trans­ cendantal et le pathologique. Paris: Vrin. Guyer, Paul (1997). “In praktischer Absicht: Kants Begriff der Postulaten der reinen praktischen Vernunft.” Philosophisches Jahrbuch 104, 1–18. Ivaldo, Marco (1990). “La Dottrina della scienza nella controversia con Schel­ ling.” Paradigmi 23, 252–271.

Reading Fichte Today  177 Ivaldo, Marco (1992). “Transzendentalphilosophie und realistische Metaphysik. Das Fichtesche Spinoza-Verständnis.” In Manfred Walter (ed.), Spinoza und der Deutsche Idealismus. Würzburg: Königshausen und Neumann, 59–79. Ivaldo, Marco (2009). Libertà e moralità. A partire da Kant. Padua: Il prato. Ivaldo, Marco (2012). Ragione pratica. Kant, Reinhold, Fichte. Pisa: Ets. Ivaldo, Marco (2014). Fichte, Brescia: La Scuola. Ivaldo, Marco (2017). “Filosofia trascendentale e formazione al senso per la filosofia.” Arete 2, 47–57. Janke, Wolfgang (1993). Vom Bilde des Absoluten. Grundzüge der Phänomenologie Fichtes. Berlin, New York: De Gruyter. Janke, Wolfgang (2009). Die dreifache Vollendung des Deutschen Idealismus. Schelling, Hegel und Fichtes ungeschriebene Lehre. Fichte-Studien Supplementa. Amsterdam, New York: Rodopi. Lauth, Reinhard (1965). Zur Idee der Transzendentalphilosophie. Munich, Salzburg: Pustet. Lauth, Reinhard (1992). Vernünftige Durchdringung der Wirklichkeit. Fichte und sein Umkreis. Neuried: Ars una. Lauth, Reinhard (2004). Marco Ivaldo (ed.), Con Fichte, oltre Fichte. Turin: Trauben. Marcuzzi, Max (ed.) (2018). Fichte et l’ontologie. Aix-Marseille: Presses Universitaires de Provence. Rametta, Gaetano (2012). Fichte, Rome: Carocci. Siemek, Marek J. (1984). Die Idee einer Transzendentalphilosophie bei Fichte und Kant. Hamburg: Meiner 1984. Siemek, Marek J. (2000). Vernunft und Intersubjektivität. Zur philosophischpolitischen Identität der europäischen Moderne. Baden Baden: Nomos Verlagsgesellschaft. Tilliette, Xavier (2003). Fichte. La science de la liberté. Preface by Reinhard Lauth. Paris: Vrin. Zöller, Günter (1998). Fichte’s Transcendental Philosophy. The Original Duplicity of Intelligence and Will. Cambridge: Cambridge University Press. Zöller, Günter (2013). Fichte lesen. Stuttgart-Bad Cannstatt: Frommann-Holzboog. Zöller, Günter (2017). “Weder Methode noch Disziplin. Zum historisch-systematischen Ort transzendentalen Philosophierens.” In Marco Ivaldo, Hans Georg von Manz, Ives Radrizzani (eds.), Vergegenwärtigung der Transzendentalphilosophie. Das philosophische Vermächtnis Reinhard Lauths. Würzburg: Königshausen & Neumann 2017, 167–183.

10 Fichte’s Original Presentation of the Foundational Principles of the Wissenschaftslehre The Question of Method Marina F. Bykova For the entirety of his philosophical career, Fichte claimed to be true to the ‘spirit,’ if not the ‘letter,’ of Kant’s Critical philosophy. And he was indeed faithful to this commitment, even if in his own—often surprising but no less remarkable—way. Seeking to resolve the issues emerging in the wake of Kant’s Copernican revolution in philosophy, Fichte developed a philosophical system that would allegedly meet the demands of a rigorous science, an ‘all-encompassing’ philosophy called the Wissenschaftslehre. In the “First Introduction to the Wissenschaftslehre,” he describes the task of philosophy as demonstrating the basis “of the system of representations accompanied by the feeling of necessity” or displaying “the foundation of all experience” (IWL, 8; GA I/4: 186). This conception of philosophy’s task agrees with Kant’s definition of the purpose of philosophy, despite the two thinkers’ radically different understandings of what formed the content of the experience in question. It is also revealing that both philosophers turn to the reflective use of mind (of reason in Kant and of consciousness in Fichte) in order to organize laws of experience into a systematic whole. Echoing Kant, Fichte views his philosophical project (the project of the Wissenschaftslehre) as an ‘experimental inquiry,’ seeking to ground it on a secured course of a systematically arranged knowledge. Furthermore, Fichte employs a strategy whereby the way of laying down his philosophical system pursues the path of the Kantian appeal to the ‘condition of the possibility,’ thus emerging as distinctively transcendental. Yet the version of transcendental idealism Fichte delivers substantially diverges from the one elaborated by Kant. These many substantial overlaps between the key goals, strategies, and certain ways in which central ideas become established in Kant and Fichte inevitably raise the question about the methodology that both thinkers implement. My interest in this chapter is the style of reasoning and method—or perhaps the methods—of philosophizing employed by Fichte for the purpose of the explication of his foundational principles DOI: 10.4324/9781003044376-11

Fichte’s Original Presentation  179 in the 1794/95 Foundation of the Entire Wissenschaftslehre, Fichte’s first full-scale presentation of his philosophical system where clarification of methodological issues becomes an urgent matter. Revealing the philosophical methodology that Fichte uses to establish the first principles of the Wissenschaftslehre can shed light on the type of transcendental idealism he develops and its potential in addressing the challenges of Kant’s Critical philosophy. The question of method(s) employed in the early Jena Wissenschaftslehre is a valuable instrument in understanding Fichte’s philosophical project and the approach he chooses to achieve his goals as well as an important point of comparison between Kant and Fichte’s philosophical systems. It is also instructive in terms of the development of German idealism and the impact that Fichte’s methodology had on both his immediate and more distant devotees. In the Foundation and other early expositions of the Jena Wissenschaftslehre, Fichte openly associates the methodology of his systematic presentation of philosophical derivations with what he calls ‘the synthetic method.’ As it is well known, it was Kant who pointed to the synthetic method as the proper method of a genuinely transcendental philosophy, rigorously insisting that the Critique of Pure Reason “had to be composed according to the synthetic method” (Prol, 4: 263). For him, “to philosophize” means “to reflect discursively”: “there is, to be sure, a transcendental synthesis from concepts alone, with which . . . only the philosopher can succeed” (CPR, A718–9/B746–7). The questions that motivate my present study are whether Fichte’s methodology is fully coherent with Kant’s and to what extent their accounts of the synthetic method are similar. To be sure, there is no consensus among scholars on how to properly understand Fichte’s ‘synthetic method,’ especially when it comes to his 1794/95 presentation of the Wissenschaftslehre, which does not provide a single and well-elaborated account of the method. What divides commentators is that they interpret the method Fichte calls ‘synthetic’ in a variety of ways, assigning to it different features and characteristics. Fichte’s method is sometimes said to be ‘phenomenological,’ focusing on what can be discovered by means of ‘inner intuitions’ and reflection. Daniel Breazeale even sees “the obvious parallels between [Fichte’s] method of philosophical self-observation and description and Husserl’s method of phenomenological reduction and eidetic intuition” (Breazeale 2015, 82; see also Breazeale 1998). However, a more common understanding of Fichte’s ‘synthetic’ methodology is that it proceeds ‘genetically.’ On this view, Fichte’s method is conceived as a systematically organized process of explaining how the mind rises to consciousness, or more specifically, a derivation of the categories describing ordinary experience from the foundational principle of the I (see Kim 2020; Phillips 2017; Breazeale 2015, 78, 83–4, 87; Zöller 2002; Martin 1997, 134).1 Certain critics also point to the ‘dialectical characteristics’ of the ‘synthetic

180  Marina F. Bykova method’ employed in the 1794/95 Wissenschaftslehre. Yet they believe that these features remain rather undeveloped in Fichte’s system. A ‘dialectical’ trend in interpretation of Fichte’s method is not new. It goes back to at least the 1960s–1980s, when a group of—mostly German— scholars claimed to find the source of Hegel’s dialectic in Fichte. They suggested that the ‘synthetic method’ of Fichte’s Jena system might be viewed as the ‘distinctive dialectical method.’ However, this idea has not received proper attention and development in the literature (Stiehler 1962; Hartkopf 1967; Hammacher 1988; Lauth 1989a; see also Seebohm 1994). Most recently, Daniel Breazeale contended that in Fichte’s Jena writings “one encounters a distinctively ‘synthetic’ kind of thinking, which can only be described as ‘dialectical’ in character” (Breazeale 2015, 85). Yet, while commending Fichte’s “dialectic-synthetic thinking,” Breazeale assigns to it a limited significance within the early Wissenschaftslehre (and beyond), rather pointing to “the distinctive genetic-phenomenological method” (Breazeale 2015, 87) as the true method of Fichte’s transcendental philosophizing. My goal in this chapter is to closely examine the methodology Fichte uses in his first Jena presentation of the foundations of his Wissenschaftslehre and clarify the meaning of the ‘synthetic method’ that he employs there. I will be mainly focusing on his 1794/95 Foundation of the Entire Wissenschaftslehre, the first and the only detailed account of the principles of his philosophy published during his lifetime. Originally conceived as a “manuscript for the use of listeners” to be distributed in fascicles which were to accompany Fichte’s private lectures during his first semester at Jena, the text was eventually published in two installments that appeared in October 1794 and July 1795, respectively (Breazeale 2021a, 23). My primary focus in this chapter will be on the “First Installment” of the Foundation, especially on sections dealing with the presentation of the first principles and their foundational demonstration, which are important for understanding the style of reasoning and the methodical procedure employed there. While Fichte’s 1794/95 foundational account largely remains a work in progress, we do find there a distinct methodology. Fichte himself often describes it as a “synthetic way of proceeding,” which is often directly equated with the synthetic method followed by Kant. I  will argue that Fichte’s conception of ‘synthesis’ is not identical to Kant’s and that the way he enacts the ‘synthetic’ procedure in his first presentation of the Wissenschaftslehre points to his departure from Kant’s usage of the ‘synthetic method.’ I  will further suggest that the ‘way of proceeding’ Fichte uses to establish and demonstrate his foundational principles can be properly understood only as a combination of methods construed dialectically and that the ‘synthesis’ he has in mind is not a simple ‘unity’ of the opposing principles (affirmation and negation; ‘the I’ and ‘the Not-I’) but rather a dialectical resolution of the arisen contradictions.

Fichte’s Original Presentation  181 I will begin with a brief discussion of the theoretical challenges Fichte attempted to address with his project of a systematic philosophical science which grounds all human cognition and of the requirements necessary to achieve this goal. I will next consider the style of reasoning Fichte chooses for his foundational presentation of the system and only then turn to a closer examination of Fichte’s first principles as he formulates them in Part I of the Foundation. My main interest in this section is the procedure Fichte uses to justify his starting point and to present the ‘primordial unity’ of the original (‘synthetic’) act of the I’s self-positing with which he commences his system. I  will comment on possible ways to explain Fichte’s reliance on the rules of general logic for the ‘derivation’ of the foundational principles—one of the most puzzling gaps in Fichte’s argumentation. I will then discuss how Fichte’s approach changes once he realizes that the logic-based demonstration of the first principles is not able to resolve the opposition produced as the result of the first two foundational principles. This is what prompts him to reformulate his procedure as a dynamic triad progressing from thesis to antithesis and then to synthesis, interpreted as the ‘synthetic method.’ I will argue that this method is dialectic in its design and that ‘the synthetic way of proceeding’ that Fichte refers to in the “First Installment” of the Foundation could be properly understood only as a dialectical construction.

1. Fichte saw the goal of his philosophical project as the realization of Kant’s unfulfilled objective to raise transcendental philosophy to the level of a science. Early in his career—with the help of Karl Reinhold and Gottlob Ernst Schulze—he became aware of serious problems in the Kantian system and the challenges that those problems presented for philosophy. The most troubling was Kant’s inability to provide a sound transcendental (a priori) account of human knowledge, that is, his failure to explain human cognition that would go beyond appearances. Fichte’s Wissenschaftslehre is an attempt to reformulate and complete Kant’s transcendental philosophy by developing a unified understanding of experience, which could account for everything, including that which might be originally thought to be beyond our capacity to know. Arguments developed by Reinhold and especially Schulze demonstrated to Fichte the need to search for a proper foundation for philosophy if it was to become a science and survive skeptical doubt. Thus, rather than advancing a skepticism implied by Kant’s thing-in-itself, Fichte postulates the original unity of the human mind, or the ‘I,’ which is radically free and uncaused. The self-reflective productive activity of the I becomes then the starting point of his philosophical system. This signifies his choice for (transcendental) idealism over dogmatism, indicating the kind of philosophy he pursues and the goal he has in view. For him, what comprises “the material of

182  Marina F. Bykova all philosophy” is “itself the human mind or spirit, considered in all its affairs, activities, and modes of acting. Only after it has made an exhaustive inventory of all of these modes of acting is philosophy Wissenschaftslehre” (EPW, 200; GA II/2: 227). The task of a philosopher is then to “observe the way in which the human mind works” and that involves “holding still for examination that which is changeable and transient within the mind” (ibid.). In this context, the question of a proper philosophical method becomes crucial. Without a suitable method the sought philosophical science could not be developed and the pursued goals of transcendental idealism could not be achieved. The earlier quote is taken from Fichte’s public lectures on “Morality for Scholars” delivered during his first semester at Jena, around the same time he was working on the “Foundations” of his new system. However, convinced that philosophy must be advanced as systematically as a rigorous science following a distinct method, Fichte began pondering on the procedure of the development of such a system even before moving to Jena. In a letter to his colleague Johann Friedrich Flatt composed in November or December of 1793, he famously wrote: “I am convinced that philosophy can become a science only if it is generated from one single first principle (Grundsatz), but that it must then become just as selfevident as geometry” (EPW, 366; GA III/2: 1; no. 168a; trans. slightly modified). Here Fichte clearly points to two requirements he considers crucial for a scientific philosophy: it must be grounded in a foundational principle; and, it must have a geometric presentation. Fichte views both of these requirements as not just sufficient but necessary criteria for developing his Wissenschaftslehre. He advances these ideas further in Concerning the Concept of the Wissenschaftslehre, written in early Spring 1794, while still in Zürich. This “programmatic” work (Breazeale 2021a, 19) is like a “blueprint” of Fichte’s future system, though not a systematic part of the Wissenschaftslehre itself. It describes the task, method, and object of philosophical inquiry central to the Wissenschaftslehre. Here he goes on to “scientifically” explicate the concept of his philosophy as the systematic development of the “unified and whole” Wissenschaftslehre from a foundational principle (Grundsatz) through a chain of logical inferences in such a methodical way that the certainty of the initial principle is transferred to the claims and propositions derived from it (CCW, 158–9; GA I/2: 114– 5). One does not need to have any specific knowledge of a scientific methodology to recognize in the described way of proceeding an approach reminiscent of the geometric method that Spinoza uses in the Ethics, with the only difference that in Concerning the Concept of the Wissenschaftslehre, Fichte speaks of a single premise (Grundsatz) from which to begin the proofs. This apparent ‘inconsistency’ becomes resolved a few months later in Part I of the Foundation where Fichte explains the exact meaning

Fichte’s Original Presentation  183 of his “Grundsatz” and demonstrates how it could be known with the self-certainty required for the foundational principle. Yet, before turning to examination of Fichte’s first principles and how they are demonstrated in the Foundation, it is important to clarify the style of reasoning Fichte employs in his foundational presentation of the system.

2. In discussing Fichte’s original presentation of the Wissenschaftslehre, scholars typically pay most attention to its overall strategy and specific content, leaving the question of the style of this presentation beyond their consideration. However, the style of reasoning that the philosopher utilizes provides crucial insights into the ‘mechanism’ of his thought process. As an important device of philosophical practice, it allows one to identify central ideas and patterns of inferential relations that guide the inquiry. In this sense, explicating the style of reasoning that Fichte employs in his foundational presentation can offer a better understanding of his specific methodology, shedding light on the character of his demonstration and the procedure(s) he adopts for this purpose. What is under consideration is a distinct manner in which Fichte proceeds with his presentation, and I would argue that the style of reasoning we observe in his foundational (1794/95) presentation of the system can be described as geometric. Considering Fichte’s admiration for the rigorous rules of inference in geometry and his aim for the Wissenschaftslehre to “become as self-evident as geometry” (EPW, 366; GA III/2: 1; no. 168a), the preference for the geometric demonstration should not come as a surprise. Because of geometry’s reliance on the self-evident and universally valid first principles, its strict deductive form, and highly methodical and lucid process of reasoning, Fichte views geometry as a “model of certitude and scientific thinking” (Wood 2012, 5). There is much textual evidence from the Zürich and Jena periods—the time of his extensive work on his original presentation of the Wissenschaftslehre—where he speaks of a geometric demonstration as the only suitable strategy for a scientific philosophy and describes it as a procedure he plans on pursuing in the Wissenschaftslehre.2 Moreover, in an official announcement of the “First Installment” of the Foundation that appeared on October  1, 1794 of the Jenaische Allgemeine Literatur-Zeitung/Intelligenzblatt, Fichte’s system of philosophy was explicitly brought into connection with Euclid’s system of geometry: “even a hasty study of this work is sufficient to convince every competent reader that if any mortal is able to become for philosophy what Euclid has become for mathematics, then this is a task for Mr. Fichte.” (FG, 1: 209, note 1; cf. FG, 7: 19–20; cited in Breazeale 2021a, 26). As reported by Daniel Breazeale, the authorship of this announcement belonged to

184  Marina F. Bykova Fichte’s younger colleague, Friedrich Karl Forberg. Yet he composed this piece “undoubtedly in close consultation with the author,” that is, Fichte himself (Breazeale 2021a, 25). Given that by the time of the publication of the “First Installment,” the presentation of the foundational principles (Part I) and the “theoretical” part (Part II) were already complete, it seems that Fichte’s reliance on geometric methodology, at least in these sections of the work, is undeniable. Even a quick look at the structure of the Foundation allows one to see that the text has an unusual layout. Instead of a coherent flow of text, its sections and paragraphs are broken up into axioms, postulates, propositions, and derivations, even if not all of them have these special headings. This layout, uncommon for philosophical texts, is actually typical of a geometric style of presentation. Another apparent parallel between Fichte’s foundational demonstration and geometry is his insistence on the “unconditioned” character and self-evident certainty of the first principle (Grundsatz). In the opening lines of §1, Fichte states: We have to seek out the absolutely first, purely and simply [schlechthin] unconditioned foundational principle of all human knowledge. If this is the absolutely first foundational principle, then it cannot be proven or determined [by anything else]. (FEW, 200; GA I/2: 255) His employment of the term “foundational principle” (Grundsatz) in this context is also revealing. Since Christian Wolff, the term “Grundsatz” has been in use as a German word for “axiom” (Breazeale 2021b, 469) which, in mathematics, indicates a statement that everybody will admit as self-evidently true. Fichte demonstrably presents his foundational principles (and his own philosophical system in general) in a manner comparable to that of geometry, which he decidedly considers a gold standard for clear and rigorous reasoning. Accordingly, the Foundation begins with the three foundational principles—“axioms” of the system (Part I)—and then derives from them eight theorems, comprising Part II (§4. First Theorem) and Part III (§§5–11. Second Theorem—Eighth Theorem). Fichte was interested in the geometric style of presentation from a very early date, exploring the parallels between the method of the geometer and the transcendental philosopher. There is little doubt that Kant served as one of his immediate sources of inspiration regarding the practical utility of geometry and the methodical power that geometric reasoning can offer. As it is well known, Kant himself paid serious attention to geometry, and mathematics in general, underscoring the success of mathematics in producing knowledge, a success that he associated with the “happy and well-grounded” mathematical method, which he often compared to the method of logic. While Kant sharply differentiated between philosophical and mathematical lines of reasoning, he recognized the

Fichte’s Original Presentation  185 benefit of mathematics as a methodological tool, praising it for a specific synthetic character. The latter is not only characteristic of mathematical definitions but is also an essential feature of its procedures in general. The most important among them, and in which Kant saw the uniqueness of mathematical reasoning, is that mathematical cognition derives from the “construction” of its concepts: “to construct a concept means to exhibit a priori the intuition corresponding to it” (CPR, A713/B741). Because philosophical cognition does not rely on intuition, philosophy for Kant is not capable of constructing its concept. Several months before his arrival to Jena, during a lecture Fichte held in Zürich on “Wissenschaftslehre in Relation to Geometry” (GA IV/3: 30–5), he critically reflected on Kant’s view, praising his conception of “demonstration” for stressing its constructional component, but at the same time dismissing Kant’s denial of the ability of philosophy to construct its concepts. For Fichte, philosophy is in a position to produce “construction-based demonstrations” supported by intuitions, establishing an appropriate relationship of its concepts to experience. Only philosophy possesses intuitions that do not pertain to ordinary consciousness; they are not “sensible” ones but rather “intellectual” (or “inner”) intuitions that Fichte identifies as a central component of the transcendental philosopher’s “observational” or “descriptive” method of construction. In the case of the Wissenschaftslehre, the necessary acts of construction are actually the actions that the I as “the demonstrating subject” performs itself. Furthermore, this is the self-referential activity, which is directed not at something outside the I, but at the I itself. Fichte often discusses such activity of the I in terms of ‘self-observation’ and ‘self-reflection,’ viewing it as the only true approach to be used by the transcendental philosopher. The goal of the sought reflection is not merely the self-awareness of the I (self-consciousness) but rather the task of ‘deducing’ the world in thought, philosophically explaining human experience as it is captured in the conceptual content with which our mind operates. Yet when we speak about the ‘deduction’ of concepts and principles in the context of Fichte’s Wissenschaftslehre, it cannot be reduced to only the Kantian sense where to ‘deduce’ means to account for and to justify. What is often overlooked in scholarly literature is that the ‘deduction’ that Fichte has in mind is much closer to Spinoza’s appropriation of the same in his geometric method (more geometrico) employed in the Ethics. Kant’s philosophical insights into mathematics and his ideas about the relation of geometry to spatial intuition of course greatly influenced Fichte’s own philosophical position and the method of its presentation, prompting him to address several of Kant’s difficulties in his Wissenschaftslehre by reformulating the start and end points of philosophy itself. But when it comes to Fichte’s original (foundational) presentation of the system, it was perhaps not just Kant but also Spinoza who significantly helped shape its method in a number of important ways.3

186  Marina F. Bykova Spinoza’s distinct employment of the geometric style of reasoning (more geometrico) in the Ethics does not go unnoticed by Fichte, and he explicitly admits it in the Foundation by insisting that “there are only two fully consistent systems: the Critical system . . . and the system of Spinoza” (FEW, 207; GA I/2: 264). Spinoza’s efforts to formalize philosophical argument and use deduction for the purpose of the efficient philosophical demonstration of the individual facts of human experience inspire Fichte’s attention to his distant predecessor’s expository strategy.4 The close similarity between the structural organizations of Spinoza’s Ethics and Fichte’s Foundation is difficult to ignore. The existing parallels between the two writings are not of a merely formal nature. Both Spinoza and Fichte employ the geometric procedure, progressing from axioms and definitions to propositions, corollaries, and conclusions because both believe that only on this deductively constructed path does philosophical demonstration reach the height of consistency. Much like Spinoza, who refused to remain a humble learner in a world incapable of being completely grasped by the human mind and who aspired to lay down a system to describe human experience in the most didactic way, Fichte assumes the role of a system-maker, imposing upon himself the task of the philosophical demonstration of how representational content can be constructed through the activity of the mind itself. In his Ethics, Spinoza attempted to deduce the system and explain experience not from empirical data but from some primary and fundamental principle, which includes all that is to be deduced. For him, the principle that serves as a logical starting point of his system is the concept of God, which grounds everything else and from which all conceptual content can be derived by proceeding according to the demands of deductive thought and rigorous analysis of the construction of demonstrable proofs. Spinoza’s application of the method of geometry in philosophy thus amounts to first accepting the most fundamental principles as true and indisputable and then demonstrating—through a rigorous deductive procedure—how the phenomena of experience result from these very principles, progressing from primary, self-evident truths to general conclusions. Fichte’s foundational presentation unfolds in a similar demonstrable way of philosophical proof, relying upon a rigorous deduction as the true order of the methodical thought in the active process of its constructionbased exposition of factual reality. It is worth stressing that the geometric approach Fichte employs in the Foundation is not just the application of the purely geometric techniques and operations in use in Euclid’s geometry but also rather a methodology of geometrico demonstrata appropriated for philosophical purposes. This puts Fichte in direct connection to Spinoza and his rigorous and self-evident geometric style of reasoning employed in the Ethics. Like Spinoza, Fichte uses philosophical demonstration in his work not for the sake of the demonstration itself. In the case of the presentation of his

Fichte’s Original Presentation  187 foundational principles in §§1–3 (Part I), he seeks to convince his readers that the premises from which he starts, and which are supposed to ground the rest of the system, are true and self-evident. He does not undertake the demonstration to know it but to know it with certainty. Ensuring the cognitive certainty of the foundational principles actually becomes Fichte’s key aim in the Foundation. Yet he is not satisfied with just postulating this certainty as some kind of axiomatic proposition; it must be properly “constructed” by employing the correct procedure and deductively demonstrated to be “valid unconditionally and purely and simply” (FEW, 202; GA I/2: 258). Fichte insists that “philosophical proofs are deductions,” akin to those performed in geometry, and thus the “Wissenschaftslehre is and must be fully the equal of geometry with respect to the precision of its proofs” (GA IV/3: 36). His specific philosophical appropriation of the geometrical style of argument becomes a distinctive element of his foundational presentation. In his demonstration of the self-evident certainty of the first principles and all other propositions derived from them, he constantly appeals to geometric postulates and constructions using them as verification criteria and as an ultimate confirmation of the ‘scientific nature’ of his own experimental enterprise. This is not to say that geometry becomes for him a foundational science. The Wissenschaftslehre evidently holds the highest status in his hierarchy of the sciences, but geometric methodology is instrumental as a model of the rigorous standards necessary for philosophical reasoning if philosophy is to become a strict science. Fichte’s methodology shares distinct internal parallels with Euclidean geometry, whereas being more firmly rooted in Spinoza’s philosophical application of geometric method. Yet his geometric style of reasoning is not a simple reproduction of the principles of Euclidean geometry or the approach that Spinoza employed in his Ethics. Fichte’s reliance on a geometric style of reasoning in his first presentation of the Wissenschaftslehre and beyond5 is not merely an appropriation of the ‘presentational conventions’ of geometers or mathematicians in general, but rather concerns the internal ‘logical’ structure of the Wissen­ schafts­lehre and the manner in which its content is demonstrated. What distinguishes Fichte’s model of geometric reasoning is its constructivist character. Discussing Fichte’s model of geometry, David Wood calls it “a constructivist and synthetic model” (Wood 2012, 11).6 While I generally agree with this characterization, it needs some further clarification, especially in what concerns the meaning that Fichte associates with the term “synthetic.” I will return to this issue later and discuss it in relation to the “synthetic way of proceeding” he claims to choose in his foundational presentation of the Wissenschaftslehre. Yet we shall first see how Fichte’s geometric style of reasoning becomes instrumental in his demonstration of the first principles of the Wissenschaftslehre in the Foundation. The thinker’s later elaborations of his system are beyond the scope of this chapter. Here, my focus is on Fichte’s methodology in

188  Marina F. Bykova the first Jena presentation of the Wissenschaftslehre, more specifically its explication in the “First Installment” of the Foundation.

3. The question about the meaning of Fichte’s foundational principle remains a central topic of scholarly discussion. The thinker is traditionally viewed as a systematic philosopher seeking to ground ‘transcendental idealism’ on a single foundational principle of the I interpreted merely as an individual subject or finite subjectivity. On this reading, his transcendental idealism is pictured as the archetype of the philosophy of the subject, in which all that is external to the subject (‘the I’) has no independent status. External reality is then taken as being in some way conditioned or even ‘created’ by the subject that is presented not only as the ‘originator’ of all what is ‘Not-I’ but also as the arbiter of all meaning.7 This interpretation is exceedingly inaccurate, and it leads to a number of misconceptions about Fichte’s Wissenschaftslehre and the methodology employed there. As becomes apparent as early as in his 1794 “Review of Aenesidemus,” Fichte’s Grundsatz is not a single postulate, but rather a complex phenomenon that he introduces by using a technical term “Tathandlung” (a fact/act) expressing an original cognitive act, to counter Karl Reinhold’s proposal to begin with a basic “fact of consciousness” (Tatsache).8 Fichte first elaborates on the concept of the Tathandlung in Part I  of the Foundation making clear his intention to demonstrate the entire Wissenschaftslehre from this “absolutely first foundation.”9 The novelty of Fichte’s approach is that he introduces the Tathandlung as the original unity of the three foundational principles: one stating that the I posits itself purely and simply (or absolutely, schlechthin); another holding that in opposition to itself, the I also posits a Not-I; and a third asserting that the I posits within itself both a divisible I and a divisible Not-I. These three principles are not merely connected, but are inseparably linked to each other, forming a single totality. There is no time gap between the I’s positing itself and its own opposition (or more correct: negation), as well as divisibility of both. All three ‘modes’ of positing occur at once, in a single act of the I, expressing the original act of the unconditioned and unconditional self-positing.10 Thus, Fichte’s Tathandlung as the original ‘synthetic act’ expresses the concept of the absolute and pure infinite I as a spontaneous free activity. Establishing the three foundational principles as his starting point, Fichte then takes it upon himself to precisely elaborate every single claim made within the Wissenschaftslehre, showing how each of them is properly derived from and systematically grounded in the preceding ones. In Part II of the Foundation, he explicitly refers to the sought systematic exposition as to the “synthetic way of proceeding” (FEW, 225; GA I/2: 283), which many commentators equate with Fichte’s method of philosophizing in general.

Fichte’s Original Presentation  189 Indeed, there is much textual evidence which shows Fichte himself describing his own method of presentation as “synthetic” (EPW, 394; GA III/2 338; no. 292; FEW, 217, 225–6; GA I/2: 275, 283–284). The problem is that Fichte’s “synthetic way of proceeding” comes to be frequently viewed as equal to the synthetic method employed by Kant in his transcendental deduction in the Critique of Pure Reason. On this reading, Fichte became acquainted with the synthetic method through his encounters with Kant’s Critical philosophy. Convinced that this is the only workable method of transcendental philosophizing, he adopted this approach in his early presentations of the Wissenschaftslehre, making it a permanent feature of his philosophical system. My aim is to challenge this view of Fichte’s methodology in relation to this presentation. In order to clarify the nature and the characteristic features of Fichte’s methodology, I  shall examine the method Fichte uses to establish the foundational principles in Part I of the work. Given the importance of Kant’s heritage for Fichte and the frequent association of the methodological approaches used by both thinkers, it helps first to reflect on the method of philosophizing employed by Kant.

4. Generally, the method associated with Kant’s Critical philosophy is widely known as ‘transcendental.’ For Kant, transcendental cognition “is occupied not so much with objects but rather with our mode of cognition of objects insofar as this is to be possible a priori” (CPR, A11/ B25). A  transcendental method then allows us to examine the conditions of the possibility of this a priori cognition. Yet insisting that this can “at least some day be exhibited both analytically and synthetically” (CPR, A12/B26), Kant points to both analytic and synthetic methods as legitimate and productive procedures to be employed in transcendental philosophy. Despite him famously declaring that his Critique of Pure Reason uses the synthetic method, while the Prolegomena proceeds analytically (Prol, 4: 263, 274–5), there are in fact quite a few places in the First Critique that proceed following an analytic method (Gava 2015, 730, 742). However, Kant’s own description of the two methods is somehow vague and lacks any systematic rigor, which is one reason why his distinction between a synthetic and an analytic method has been a continuous source of difficulty for commentators (Guyer 1987; Ameriks 2003).11 Still, from numerous remarks and often ambiguous claims that Kant makes about the two methods throughout his writings, it should be possible to gather several ideas regarding their positive characterizations. He describes these methods as the two argumentative procedures moving in two different—essentially opposite—directions. The analytic method takes the necessity and universality of some propositions for granted and identifies which conditions are necessary for accounting

190  Marina F. Bykova for those propositions (Guyer 1987, 6–7). It thus regresses from the accepted body of knowledge (some general principles and laws) to its necessary conditions. In contrast, the synthetic method begins with the highest principle and then deduces the necessary element required for experience, thus progressing “from a condition to its conditioned” (Kim 2020, 338). In other words, in Kant’s Critical philosophy, the distinction between a synthetic and an analytic method seems largely to refer to two different expository strategies. Kant also makes clear that each of these two methods has a specific functionality. He recognizes the significance of the analytic method as explanatory tool and methodology, while an important sense in which he understands the synthetic method is related to his ongoing striving for systematicity. In his 1780s Lectures on Logic, he stresses: Analytic method is also a means of discovery and of exposition, in that I speak popularly. The true method of exposition is synthetic, however, for even if I have thought the thing analytically, the synthetic method is what first makes it a system. (HL, 492–3) For Kant, only the employment of the synthetic method yields the systematic presentation required for a scientific philosophy. His Critique of Pure Reason is then a demonstration of the suitability of the synthetic method for the purposes of transcendental philosophizing. It would perhaps be wrong to think that Fichte was unaware of Kant’s conclusion, and the idea of linking synthesis to the systematic presentation of a science was certainly there for him to grab. However, I want to challenge the view that the method Fichte adopts in his first presentation of the system is a Kantian synthetic method. The strategy Fichte uses to introduce his foundational principles in Part I of the 1794/95 Foundation has numerous features that make it incompatible with Kant’s definition of synthetic method. To see the difference, it helps to closely examine the way Fichte proceeds with his presentation of the first principles.

5. One of the difficulties in understanding Part I of the Foundation is associated with its beginning. Paradoxically it commences not directly with the original act of Tathandlung (the foundational principles themselves), but rather with the discussion of the logical principle of identity expressed by the abstract proposition “A is A” (FEW, 201; GA I/2: 256–7). This seems to be a problematic starting point for a number of reasons. Most troublesome of these is that while attempting to develop philosophy as the ‘foundational science,’ the one in which all sciences, including general

Fichte’s Original Presentation  191 logic, can be systematically grounded, Fichte turns to ‘rules’ or ‘laws’ of general logic for justification of the first principles of his system. It is worth noticing that relying on the rules of logic does not pertain to Fichte’s presentation of only the first foundational principle. In addition to the principle of identity, in Part I he also employs two other rules of general logic: the “principle of opposition” (a version of the principle of noncontradiction; FEW, 208–9; GA I/2: 266–7, §2) and the principle of sufficient reason (or “grounding principle”) (FEW, 214–5; GA I/2: 272–3). In all three cases, he insists that the “laws” of logic he uses as the starting point for his presentation must derive from the foundational principles,12 and later in the text, these rules are indeed demonstrated to follow from the first (§1), second (§2), and third (§3) foundational principles of the system, respectively. However, Fichte insists that the foundational principles “can be correctly established only on the condition that these same [logical] rules are correct” (FEW, 200; GA I/2: 92; my italics). Is this procedure, the circularity of which Fichte himself affirms, logically legitimate, and if so, what strategy might be associated with it? Can it be validated from the perspective of a rigorous geometric reasoning that Fichte seems to endorse? Those are important questions, and they need to be somehow addressed if we want to make sense of the way in which Fichte proceeds with his foundational presentation. Perhaps one could argue that Fichte’s reliance on the rules of general logic in demonstrating the first principles of his system is a result of his indebtedness to Kant, who viewed formal (general) logic as the true paradigm for the methodology of science. For him, logic owes this role to its formal and synthetic character associated with its apodictic (‘proven’) certainty. This is a reason for him to ground his transcendental logic in the most general principles of formal logic claiming that the latter provides the former with ‘clues’ for establishing its fundamental concepts.13 Several commentators even point to Kant’s effort to reconstruct mathematics (most notably geometry) as a logical theory, trying to find some parallels between Kant’s and Fichte’s applications of logic. However insightful this explanation might be, it remains unconvincing, especially given that earlier on, Fichte clearly indicates his disagreement to relate the principles of a foundational transcendental philosophy to those of general logic, casting serious doubts on the systematically motivated use of the laws of logic in the exhibition of his first principles.14 I would rather suggest that Fichte’s appeal to general logic for the purpose of explication of his foundational principles is related to his ambitious goal of making his presentation philosophically rigorous yet ‘crystal clear’ and indisputable for everybody attempting to apprehend it. This certainly requires the employment of a specific strategy that might involve a combination of methods. In Part I  of the Foundation, he sets forth to demonstrate the foundational nature of the first principles and their self-evident certainty

192  Marina F. Bykova proceeding from the general propositions that “everyone will concede without objection” (FEW, 200; GA I/2: 256). He calls such a general proposition, known as the rule of general logic, a “fact of empirical consciousness” (FEW, 207; GA I/2: 264). Daniel Breazeal explains: “Fichte’s strategy . . . is to begin with a ‘fact of empirical consciousness’ and then seek out the original power and resulting act of the mind that underlines and make possible the ‘fact’ in question” (see Breazeale 2021b, 477n). Indeed, the approach that Fichte employs throughout Part I of the work, which is not merely a “blueprint” for the future Wissenschaftslehre— as was the case with the Concerning the Concept of the Wissenschaftslehre—but rather an essential part of it, is not prescriptive. The sought conclusion should be derived but not by the means of any conventional procedures. It involves discursive reasoning that does not merely concern one’s own thinking but requires being completely “immersed” in a thought. In focus here is what Fichte calls the cognitive process of reflection and abstraction, which occurs by consciousness reverting its attention inwards, back onto itself, back to its own pure inner activity. In an attempt to explain the strategy of his presentation, in his 27 June, 1795 letter to Schiller, Fichte writes, “I had to begin by ascending to the basis . . ., for I did not wish to proceed by descending from [the] foundation” (EPW, 394; GA III/2: 358). I take this to suggest that the method Fichte employs in his foundational presentation is not the synthetic method Kant followed in the transcendental deduction of the Critique of Pure Reason, progressively descending from a condition of experience and knowledge to what is conditioned. Kant himself would rather classify the procedure described earlier as an analytic method that begins with an accepted body of knowledge and proceeds to its necessary conditions. If we take a closer look at how Fichte moves forward with his presentation of the foundational principles, we will see that he begins with propositions of general logic recognized “to be completely certain and agreed-upon” (FEW, 201; GA I/2: 256) and then “ascends” to the principles themselves. He demonstrates that these propositions are absolutely certain, or as he puts it, “certain for no further reason” (FEW, 201; GA I/2: 256), only because they are grounded in the active character of the human mind, in “the pure activity of the I” itself (FEW, 203; GA I/2: 259). The activity in question is the purely self-referential and self-productive activity, which expresses the Tathandlung, “what is active and what is brought about by means of this activity” (ibid.), action that is simultaneously deed. Fichte actively pursues this procedure in abstracting and affirming the three highest principles of his system in their self-evident certainty in §§1, 2, and sections A though D in §3. But once he establishes the foundational principles, he apparently changes his course and—somewhere closer to the end of §3 (FEW, 216–8; GA I/2: 274–6)—launches a transcendental derivation of the “system of representation” from the highest principles.

Fichte’s Original Presentation  193 This rarely noticed shift in Fichte’s strategy that apparently relates to the transition from the use of the analytic method to the employment of the synthetic method is of great importance. In addition to highlighting the experimental and probing character of Fichte’s first presentation that requires numerous adjustments, further clarification of goals, etc., it also reveals Fichte’s concerns about, first, the accessibility of his foundational presentation of the Wissenschaftslehre and, second, the suitability of the employed methods for the purpose of making philosophy into a rigorous science. In his already mentioned letter to Schiller, Fichte distinguishes between popular and philosophical presentations pointing to differences in their procedures and explaining his varying approaches to each of these expositions. He writes: The essence of a popular presentation seems to me to lie in proceeding synthetically. . . . The popular aspect of my system lies chiefly in my procedure. . . . Once the strictly philosophical part of the exposition has been completed, I  then proceed to its popular treatment according to quite different principles. I start with a very common experience and develop the argument through what appears to be an association of ideas (though in fact the system keeps invisible watch over these associations). I specify nothing with more precision than is immediately required, until at the end a sharply defined specification emerges on its own. (EPW, 394–5) Fichte indicates that in his first (foundational) presentation of the system he employs two procedures—‘strictly philosophical’ and ‘popular’—which follow different principles corresponding to two distinct methods. In other words, in his presentation of the Wissenschaftslehre, Fichte employs not one single method, but at least two, using them in a unique combination that becomes apparent in the early paragraphs of the Foundation. Furthermore, his application of both methods does not appear to correspond directly to any of the methods used by Kant. Even if we consider the analytic and synthetic methods utilized by Kant, the ways of proceeding Fichte describes here are not identical to those captured by Kant’s conception of the same. Fichte begins with the established rules of general logic first ‘ascending’ to the activity of human mind (the pure I as the Tathandlung) that underlies and grounds them and only after demonstrating the pure I as the unconditioned and utterly independent entity, he derives all experience and all knowing (the content of the Wissenschaftslehre) from this original unity. The first procedure might be described in terms of the analytic method yet used in a very specific—reflective—meaning; the second one is synthetic in its approach.

194  Marina F. Bykova In order to establish the foundational principles, Fichte invites us “to attend to oneself” and one’s own mind by the way of philosophical reflection. One must “think the I,” undertaking a careful analysis of one’s own cognition and observing what transpires. The employment of reflection as the means of this “analytic self-observation” leads some scholars to consider this method “phenomenological” (Breazeale 2015, 82). But what is involved here is more than a pure self-observation associated with imposed awareness. The procedure in question is a transcendental reflection that calls for the I’s self-referential activity of abstracting from anything that does not belong to it. This abstraction, based on the analysis, is not merely a contemplation; it is a dynamic act of making or producing. As Fichte puts it in his second Jena presentation of the Wissenschaftslehre: “I have to construct the concept of the I” (FTP, 119; GA II/3: 340; my italics). The result of this “production” is the (concept of the) I, which is able to explain and render intelligible itself along with its own characteristics and nature. The self-conscious I is thus self-grounded in its own activity of self-referential analysis and reflection; it is a condition of its own being and self-awareness. This is the meaning of Fichte’s principle of the self-positing I: the I is simply what it posits itself to be. The I  in question is the original or pure I, the Tathandlung discussed earlier, that Fichte makes the absolute foundation and the starting point of the Wissenschaftslehre. As we saw, this pure I is not exhausted by the single act of self-positing; it is constantly determining itself through positing Not-I as its own limitation and then overcoming the emerged opposition by achieving the ‘synthetic’ unity of the I and the Not-I in which the subject and object are not two but one and the same. Drawing all of this together, Fichte concludes that this utterly unconditioned, independent, and self-sufficient pure I  can impart certainty to the whole of our conscious experience both of ourselves as empirical individuals and of a world of spatio-temporal material objects. This conclusion culminates a largely ‘analytic’ presentation of the foundational principles (FEW, 200–16; GA I/2: 255–73) and marks a transition to a genuinely ‘synthetic’ part of the Foundation, where Fichte, pursuing his original goal of giving an account of our experience, proceeds from Tathandlung to a derivation of the whole world of experience, which is nothing else but a derivation of the content of the entire system. Interestingly, Fichte’s methodology undergoes a significant change already in the process of establishing his three foundational principles, transitioning from a mostly ‘analytic’ way of proceeding to the ‘synthetic’ one. Fichte himself indicates this move in §4 at the beginning of Part II by pointing out that the second and third foundational principles are “mutually grounded in the first” one. Summarizing the results of Part I and stressing the importance of the undertaken demonstration of the first principles, he writes: “It is the latter two foundational principles

Fichte’s Original Presentation  195 that establish and ground the form of the synthetic way of proceeding” (FEW, 225; GA I/2: 283). The manner of Fichte’s presentation might indeed be synthetic, but the syntheticism involved differs fundamentally from the one we observe in Kant’s transcendental deduction in the Critique of Pure Reason. In order to accomplish the task he sets himself of establishing how rational cognition comes about, Kant begins with our most basic experience and advances to its a priori and necessary conditions. Fichte’s presentation, on the other hand, moves in the opposite direction: from the foundational I that is absolutely unconditioned and spontaneously free to the manifold of the experience. Then, our experience of the world is shown to come about from the actions of the I “as its transcendental conditions” in a specific procedure. Fichte chooses a way of proceeding that involves the employment of both the analytic and synthetic methods taken in a unique combination where analysis and synthesis are intimately connected to each other, and one is not possible without another (FEW, 217; GA I/2: 274). They substantially sustain and complement each other. The stressed complementarity of the two methods might at first appear odd, especially given Fichte’s introduction of both methods in §3 by sharply distinguishing them as representing essentially the opposite ways of proceeding. By comparing two things, the analytic method (which Fichte prefers to call ‘antithetic’ to emphasize the specific of this approach) focuses on “the characteristic feature through which they are posited as opposites,” while the synthetic method aims to discover the quality “in which they are the same” (FEW, 216; GA I/2: 273, 274). Note that Fichte turns to this discussion as a means of his presentation of the third foundational principle. He already established his first two principles: (1) the I posits itself (FEW, 203; GA I/2: 254); and (2) “a NotI . . . posited purely and simply in opposition to the I” (FEW, 209; GAI/2: 266) as the I’s own negation. Since both the I and the Not-I are posited by the I itself, now the task at hand is to achieve the ‘unity’ of the I and the Not-I posited in opposition to each other. Otherwise, the identity of consciousness, which is “the sole, absolute foundation of our knowledge” (FEW, 212; GA I/2: 269) cannot be demonstrated. This is when Fichte describes both methods—antithetic and synthetic—as necessarily connected by mutually conditioning each other. He describes the sought ‘unity’ as the synthesis of the limited I and the limited Not-I antithetically posited in opposition. Rarely noticed, Fichte’s insistence on the complimentary correlation between the two methods brings to attention the distinctive character of his methodology. At the core of this methodology lies a ‘synthetic’ procedure which is dialectical in its character and attributes; a dialectical synthesis that ‘brings together’ the two concepts (or terms) posited in opposition. Next, we shall briefly consider this procedure attempting

196  Marina F. Bykova to explicate specific features of the way of proceeding that Fichte calls ‘synthetic.’

6. As we have seen, Fichte’s ‘synthesis’ involves the positing of opposed elements and then undertaking an action aiming at ‘connecting’ these antithetically posited elements. The ‘synthetic’ procedure in question is certainly not the traditional method of logical inferences that is based on the ‘laws’ of general logic. Similarly, it goes beyond a mere explication of the conclusions drawn from the original premises. In §3 (section D.7; FEW, 218; GA I/2: 276), Fichte himself explicates his ‘synthetic’ method by establishing the triadic formula, thesisantithesis-synthesis, the basic pattern to describe the dialectical process. In Fichte’s presentation, the triad refers to the acts of the I: the act of self-positing (thesis), the act of positing the Not-I as its own negation (antithesis), and the act of positing the relation of mutual limitation (a divisibility) between the I and the Not-I (synthesis). Fichte considers this tripartite account of the I’s positing as the fundamental synthetic act that has a higher status in the hierarchy of all other syntheses undertaken in the system. Such a disposition entails that all of the other syntheses must be contained in this original one. As such, this original synthetic act serves as a basis for introducing some important systematic distinctions, including distinguishing between the theoretical and practical parts of the Wissenschaftslehre to be covered in Parts I and II of the Foundation, respectively (see Klotz 2016, 72). What is important for our discussion is that at this point in the foundational presentation of the system, it becomes plain that Fichte’s understanding of ‘synthesis’ is very different from Kant’s conception of the same. Fichte radically transforms the Kantian account of synthesis in at least three key aspects by reinterpreting its structure, goal (or function), and the role of the subject in the synthetic process. I have already mentioned Fichte’s novelty with regard to the structure of the fundamental synthetic act, which is a triadic process and involves thesis and antithesis as its necessary elements. Without them no synthesis is possible at all: as an act that ‘furnishes this whole,’ a thesis grounds the process, whereas an antithesis is responsible for a differentiation that needs to be synthesized. This, in turn, also points to a very different function that synthesis has in Kant’s and Fichte’s systems. Recall that Kant describes synthesis as a spontaneous act “of putting different representations together, and grasping what is manifold in them in one cognition” (CPR, A77/B103; my italics); it is a process that “gathers the elements for cognition, and unites them to form a certain content” (CPR, A78/B103). For Fichte, the goal of the synthetic act lies not in the unification of the manifold of a given intuition, but rather in connecting

Fichte’s Original Presentation  197 the two elements posited in opposition to each other and unifying itself with its own differentiation. Fichte also goes beyond Kant in how he understands the role of the subject in the process of synthesis. According to Kant, a ‘manifold’ of presentations that the synthetic act connects or unifies are given independently and although both a faculty (the understanding) that yields synthesis and the categories by means of which manifolds of our representations are synthesized belong to the subject, for Kant this subject is distinct from its representations. In contrast, in Fichte’s system, the original synthetic act involves the power of productive imagination, by means of which the I unites the opposing elements (the I and the Not-I) in the third one, a new concept (of divisibility) also posited by the I itself. In other words, for Fichte the I does not merely play a crucial role in the synthesizing process, but synthesis has its source in the I’s acts of positing. This explains an internal unity of the triadic synthetic process and the very possibility of the synthesis itself. As Fichte puts it: “these terms that are the same would not be the same had they not first been posited as the same by means of a synthetic act” (FEW, 217; GA I/2: 274; my italics). It is not to say, however, that this original unity (‘the sameness’) should be understood as something that is found or given as a “fact” of consciousness. Instead, this signifies the important point regarding the nature and character of the pure (original) I  that can impart certainty to the whole totality of our experience by being present in all that derived from it. Yet the actual ‘sameness’ has rather to be achieved and demonstrated as the I’s own ‘synthetic’ act of connecting the conflicting (opposite) results of its own positing and reconciliation of the emerging conflicts between its own ‘real’ and ‘ideal’ activities. The synthetic act is a constructive process: the relation between the I and the Not-I is not theoretically assumed or merely presupposed but actually constructed through the activity of the I itself, the activity by the means of which the I constitutes and shapes itself and the world of its own experience. In this sense, Fichte’s synthesis has a crucial constructive function, which is correlative with the activity of the I. Discussing the procedure of the Foundation in its entirety, Daniel Breazeale summarizes this idea as follows: The Foundation begins with a posited contradiction between the I and the not-I and then proceeds ‘inward’, as it were, first redefining the conflict between the ‘directions’ of the I’s activities or between the ‘real’ and ‘ideal’ activities of the I, and finally locating it within the necessary internal structure of I-hood itself. Breazeale (2016, 122) Approached from the philosophical (transcendental) standpoint, the task of the philosopher is to bring the opposed elements into a unity by the means of a synthesis that will necessarily move the process to a higher

198  Marina F. Bykova level. Thus, Fichte tells us that “our entire way of proceeding from now on will be synthetic” (FEW, 217; see also 225; GA I/2: 275, 283; my italics). The discussion of the details of his proceeding is beyond the scope of this chapter. But I would like to briefly comment on the type of synthetic procedure that Fichte has in mind. It is dialectical both in its conception and execution. The dialectical qualification of Fichte’s synthetic method is not limited only to its triadic structure, which is indeed often identified as a dialectical one. Three other crucial features of Fichte’s account of synthesis that point to its dialectical character are its focusing on a contradictory relation between opposing elements, relying on negation as a means of positing in opposition, and presenting a progressive process leading from less sophisticated conceptions or views to more sophisticated ones. Let me briefly describe each of these features. Since the first two are directly related, I will consider them together. As we have seen, Fichte’s foundational synthesis (Grundsynthesis) is that of the I and Not-I which are not merely posited as opposed but also as being in a relationship of contradiction. Recall that in logic, a pair of propositions that cannot both be true or both be false constitutes a contradiction. On the basis of the discussion of the foundational principles Fichte offers in Part I of the Foundation, it is apparent that his conception of positing as opposed directly correlates to the idea of positing as contradictory. For him, the I and the Not-I present a contradiction that has to be overcome. Furthermore, in order to explicate what the “positing in opposition” means, Fichte employs negation. He states: The Not-I is posited in opposition to the I. Just as the I contains reality, the Not-I contains negation. If the absolute totality of reality is posited in the I, then the totality of negation is necessarily posited in the Not-I, and this negation must itself be posited as an absolute totality. (FEW, 229; GA I/2: 288) The treatment of negation has long been linked to the treatment of opposition between two terms, but in Fichte’s time, the primary type of this opposition was contradiction. The fact that he demonstrates the second foundational principle of his Wissenschaftslehre from the proposition “-A is not = A” (FEW, 207; GA I/2: 264) seems to support my suggestion about his treatment of opposition not only as negation of each other but also as contradiction. For him, each contradiction needs to be reconciled (in Fichte’s own terminology, “unified”) by means of synthesis, and because he believes that the mind necessarily generates contradictions, he thinks that new premises will indeed produce new contradictions, thus leading to further syntheses. This is directly related to the third crucial dialectical feature of Fichte’s account of synthesis indicated earlier.

Fichte’s Original Presentation  199 If we consider how Fichte describes the progression of the original (foundational) synthetic act, we will notice that the contradiction arising between the antithetically posited I and the Not-I becomes resolved only by positing a third element—the concept of divisibility—that unites the initial two (FEW, 214; GA I/2: 272). The newly introduced element brings into existence a new concept, which eventually gives rise to a new synthesis that, in turn, grounds a new one, and so on. Fichte argues that a synthetic concept that unifies the opposed elements in a dialectically generated contradiction does not completely cancel the contradictory sides, but only limits (or, as he says, “determines”) them. Yet for him, “[t]o limit something means to nullify its reality by means of negation, not entirely but only in part” (FEW, 213; GA I/2: 270). Thus, instead of leading to the dismissal of both sides of a contradiction, the synthetic concept retroactively validates the opposing elements by demonstrating their limit and by determining which portions of the reality pertain to them (FEW, 214; GA I/2: 271). In Fichte’s system, the dynamic that sustains the progression of philosophical knowledge is, in fact, the activity of the I  itself. This dynamic is posited from the very beginning, in the very act of selfpositing. It becomes most apparent, however, in the I’s positing of the Not-I as not simply mutually exclusive elements (by the means of purely logical negation) but rather as ‘mutually determining’ each other (the relation of the Wechselbestimmung). Such a ‘determination’ necessarily involves a limitation of each other’s activity, simultaneously demonstrating their intimate connection. The I  and the Not-I mutually ‘restrict’ each other’s activity, causing some ‘inertia of passivity’ but at the same time this mutual limitation makes them mutually ‘dependent’ upon each other. This ‘dependence’ is not absolute; it rather stresses the dynamic nature of the relation between the I and the Not-I. It is through the synthetic character of this dynamic involved in the mutual determination of the I and the Not-I that Fichte can introduce the distinction between the ‘theoretical’ and the ‘practical’ parts of the Wissenschaftslehre, to the exposition of which he devotes Part I  and Part II of the Foundation respectively. This is where he demonstrates the synthetic unity of the I  proceeding from two opposite directions: first he begins from the principle of the I  positing itself as determined by the Not-I (the derivation of the ‘theoretical’ part) and then from the principle of the I  positing itself as determining the Not-I (the derivation of the ‘practical’ part). Despite moving in opposite directions, both expositions employ Fichte’s distinct synthetic method, manifesting its constructive power and dialectical dynamics. Considered systematically from the perspective of the development of the Wissenschaftslehre, the result of this very complex and often obscure demonstration is the synthetic derivation of all human experience and all knowledge from the

200  Marina F. Bykova original foundational principle of the pure I, common to both theory and practice. From the point of view of the transcendental philosopher, the dialectical synthetic approach to the I thus entails a procedure whose purpose is to show how the observed I grows in its own self-awareness and realizes itself in concrete experience as a free individual. Thus, we have to begin with the original subjective spontaneity and freedom of the I and then proceed to a transcendental derivation (transcendental deduction) of its objective necessity and (self-)limitation as a condition necessary for the possibility of the I itself. For Fichte, the synthetic method is much more than simply a formal tool of transcendental reasoning. Radically reformulating the notion of ‘synthesis’ in terms of dialectical dynamics, Fichte makes it instrumental for the explication of specific issues relevant to our inquiry into the activity of the human mind. As such, it reveals a dynamic structure of representational consciousness itself (Klotz 2016, 74–6) and demonstrates the intimate connection between our subjective consciousness and its universal and necessary features in all our cognitive activities (see Kim 2020, 341). But what is more, the distinct synthetic way of proceeding that Fichte employs in his system underscores the dynamic character of the foundational presentation of the Wissenschaftslehre, which largely unfolds dialectically.

7. Conclusion As we have seen, while both Kant and Fichte claim to proceed ‘synthetically’ and for their systematic presentation to employ the ‘synthetic method,’ the meaning they assign to ‘synthesis’ varies dramatically. In his deduction of categories, Kant begins with our most elementary experience and basic representations, attempting to synthesize manifolds of these representations in one cognition. He thus proceeds in an ascending manner, progressing stepwise. And although cognition for him is synthetic in its character—structured by the unifying work of transcendental imagination and understanding—his synthetic exposition is a linear narrative. Fichte’s procedure is manifestly different. For him, synthesis is present from the start as the original synthetic act of the I, yet it is construed in its most radical form of posited opposites. Although the antithetic relation between the I  and the Not-I is contradictory in its character, its distinct feature lies in the intimate interconnection and the direct joining of the two opposites. Such a disposition gives rise to the foundational synthesis unfolding as a dialectical triadic procedure of thesis, antithesis, and synthesis. The philosophical significance of Fichte’s dialectical synthetic procedure of reasoning lies in his ability to move beyond Kant who in his Transcendental Dialectics claimed that reason’s contradictions (the antinomies of reason) lead to nothingness and thus cannot count as knowledge. In contrast, Fichte argues that the move to

Fichte’s Original Presentation  201 resolve contradictions with synthetic concepts is not only possible, but it is also necessary and grounded in the original act of the I itself. Similarly, in his appropriation of geometric methodology, Fichte does not just blindly adopt Spinoza’s more geometrico as a formal method of philosophizing. For Fichte, geometry provides a needed plane of necessity that also entails certainty of knowledge, and this allows him to advance his Wissenschaftslehre as a new type of philosophy. Following Spinoza, he does capitalize on the compelling power of the geometric demonstration, but for him the geometric method is more than simply a form of presentation. It rather serves as a specific framework within which true knowledge becomes possible. He considers geometric reasoning as the high road to certainty in human knowledge while raising caution about unreflected sense experience and the partial interpretation of mere empirical facts. Fichte’s endorsement of more geometrico also provides an important clue for his preference in methods of presentation. During the time of Spinoza, the geometric demonstration was described as involving two aspects, namely, resolution (resolutio) and composition (compositio), also known as the analytic and synthetic sides of the geometric method (Cassirer 1974, 1: 136–44).15 Likewise, Fichte’s approach is not limited to solely using the synthetic method; it also involves a specific appropriation of the analytic method. Starting with the existing (original) synthesis of the act/fact (the Tathandlung or the agency of the I), Fichte has first to analyze that synthesis, discovering and recognizing its opposing elements, and then widening the opposition into a contradiction that becomes an antithesis of his synthetic triad. This intricate unfolding of the synthesis warrants Fichte’s reliance on the analytic method and its employment for establishing the first two foundational principles of his system. However, the ‘analysis’ in question is not a typical analytic method, that is, a genetic process that involves a logical analysis of concepts. Fichte’s analytic method requires reflection and abstraction, which are interpreted as the self-referential activity of the I. The result of this activity is the self-awareness of the I  and the recognition of its own self-constituting agency manifested through positing itself and its own negation in the antithetic relation. Understood in this way, analysis becomes a crucial part of synthesis, the component that actually ‘exhibits’ its dialectical core, that is, the contradiction of the antithetically posited elements that must be overcome, or, more precisely, ‘reconciled,’ through synthesis. Fichte radically reformulates both the analytic and the synthetic methods. The methodology he employs in his foundational presentation of the system is a distinct combination of the two methods where analysis becomes integrated in synthesis, and synthesis is framed as a dialectical way of proceeding. Fichte’s ‘synthetic’ approach is thus inherently dynamic, and this dynamic correlates with the constructive activity of our human mind.

202  Marina F. Bykova

Notes 1. See also two recent works of G. Anthony Bruno (2018, 2021). Both discuss Fichte’s method of philosophizing in his later writings. 2. See Fichte’s Letter to Heinrich Stephani, written in mid-December  1793: “I believe that in a few years we shall have a philosophy which will be just as self-evident as geometry” (EPW, 371; GA III/2: 28, no. 171). Another example is Fichte’s Letter to Karl August Böttiger from February 4, 1794, where he reveals the main goal of his project of the Wissenschaftslehre: “this project concerns nothing less than a scientific philosophy that could be measured against mathematics itself” (GA III/2, 55). 3. Interestingly, the geometric method was the most characteristic feature of Christian Wolff’s philosophical system. While Wolffians preferred to call this method “the mathematical method,” it followed the same procedure as Spinoza’s more geometrico (see Frketich 2019, 335ff; Gava 2018). To be sure, the mathematical method was very popular in the eighteenth century. Many intellectuals in a variety of disciplines considered it very attractive to be able to arrive at positive proofs by the means of logically verified reasoning. Often mathematicians and scientists would first grow interested in the method and then apply it to other disciplines, which became a norm in the environment where there existed no sharp divide between different disciplines. One such example was that of Martin Knutzen, a professor in Königsberg and Kant’s teacher, who is known for his work in mathematics and astronomy, as well as in philosophy. An adherent of Newtonian physics in science and mathematics, in philosophy he was influenced by Wolff, utilizing Wolffian methodology in his own philosophical and religious writings (see Frängsmyr 1975, 662–3). In this regard, one may plausibly wonder if Wolff and his application of the geometric method for the construction of his metaphysical system may have had a more immediate impact on Fichte than Spinoza’s more geometrico. I would argue that Wolff’s influence on Fichte, if it exists at all, is minimal. To my knowledge, there is neither mention of Wolff nor even indirect reference to him in Fichte. It is also worth mentioning that Fichte never had a broad education in philosophy and was not interested in the previous philosophical tradition. He mainly focused on current philosophical debates, trying to promote his Wissenschaftslehre and prove himself as a ‘true builder’ of a scientific philosophy. For an illuminating discussion of Fichte’s main philosophical interests at the turn of the century, see Di Giovanni (2021, 25–56). I would also like to thank George di Giovanni for his helpful advice concerning a possible connection between Fichte and Wolff. 4. To be sure, Fichte was also strongly influenced by Karl Reinhold who insisted in his Elementarphilosophie that in order to make philosophy scientific, it should be grounded in one single principle, from which all other propositions and concepts must be derived based on rules of logic. For a concise discussion about the indebtedness of Fichte to Reinhold see Bykova 2020, 26–8. 5. The influence of geometric reasoning is also apparent in Fichte’s later presentations of the system up to the final ones composed in 1813/1814. At the end of 1800, Fichte even undertook a so-called geometric exposition of his system, the unfinished manuscript of the new version of the Wissenschaftslehre known under the title Neue Bearbeitung der Wissenschaftslehre (see GA II/5: 331–401). 6. For an illuminating discussion of Fichte’s original geometry and how it compares to Kantian and Euclidean versions of the same, see Wood (2012, 79–120; ch. 2).

Fichte’s Original Presentation  203 7. While this simplistic reading of Fichte is most directly and crudely presented by Bertrand Russell in his famous History of Western Philosophy (Russell 1945), it continues to prevail in textbooks and other instructional materials on Fichte. This reading also appears in more recent post-structuralist literature as well as scholarly publications by representatives of speculative realism who heavily draw from Fichte. 8. For an excellent discussion of Fichte’s conception of Tathandlung and its role in the context of the Wissenschaftslehre see Wood (2019, 171ff). 9. In the second (1796–99) Jena presentation of the system (known as the Wissenschaftslehre Nova Methodo), Fichte employs another term to describe his foundational principle(s). He drops ‘Tathandlung’, and uses ‘subject/object’ instead. 10. For more details about Fichte’s conception of the self-positing I and the meaning of Tathandlung see Bykova (2009, 131–5) and Bykova (2019, 155–63). 11. For an illuminating account of recent discussion about Kant’s distinction and a description of a key position presented within this discussion, see Gava (2015). 12. Fichte famously makes the claim about the need to ground the ‘laws’ of logic in the foundational principles—the claim that Daniel Breazeale calls “provocative”—already in his Concerning the Concept of the Wissenschaftslehre, but only in the Foundation he presents it in a demonstrable way. See Breazeale (2021a, 20). 13. For Kant’s project of transcendental logic see Tolley (2012, 418ff). 14. Recall Fichte’s remarks about the relationship between the Wissenschafts­ lehre and logic in his Concerning the Concept of the Wissenschaftslehre. See especially §6 of this writing: CCW, 178–81; GA I/2: 137–40. In the most explicit form, Fichte expressed his deep doubts about the use of general logic for justification of the foundational principles and about the validity of this procedure in his Eigene Meditationen über Elementarphilosophie, composed between November 1793 and January 1794 as well as in his private lectures on the Critical philosophy he presented in Zürich in February 1794. For a general commentary on Fichte’s Eigene Meditationen see Lauth 1989b, 157ff. For a recent discussion about the role of general logic in Fichte’s early formulations of his foundational principles, see Sommer (2021). 15. The resolutive and the compositive methods came into use in the late sixteenth century. It was introduced by Jacopo Zabarella, the key figure of Renaissance Italian Aristotelianism, primarily known for his writings on logic and methodology. Zabarella considered both methods to be intellectual instruments with the purpose to produce knowledge of the unknown. He distinguished between these two methods of presentation based on the path they follow: the resolutive method proceeds from the desired end to its first principles, while the compositive method begins from general principles and proceeds to particular beings (Mikkeli 2018). The former is said to be helpful for discovery and invention of new truths, and the latter guarantees the certainty of the results as a consequence of a complete deduction of propositions from axioms and self-evident principles.

References Ameriks, Karl (2003). “Kant’s Transcendental Deduction as a Regressive Argument.” In Karl Ameriks (ed.), Interpreting Kant’s Critiques. Oxford: Clarendon Press, 51–66.

204  Marina F. Bykova Breazeale, Daniel (1998). “Fichte’s Nova Methodo Phenomenologica: On the Methodological Role of ‘Intellectual Intuition’ in the Later Jena Wissenschaftslehre.” Revue Internationale de Philosophie 4, 586–616. Breazeale, Daniel (2015). “The ‘Synthetic-Genetic Method’ of Transcendental Philosophy. Kantian Questions/Fichtean Answers.” In Sebastian Gardner and Matthew Grist (eds.), The Transcendental Turn. Oxford: Oxford University Press, 74–95. Breazeale, Daniel (2016). “The Wissenschaftslehre of 1796–99 (nova methodo).” In David James and Günter Zöller (eds.), The Cambridge Companion to Fichte. Cambridge: Cambridge University Press, 93–138. Breazeale, Daniel (2021a). “Editor’s Introduction.” In Daniel Breazeale (ed.), J. G. Fichte: Foundation of the Entire Wissenschaftslehre and Related Writings (1794–95). Oxford: Oxford University Press, 1–97. Breazeale, Daniel (2021b). “Endnotes.” In Daniel Breazeale (ed.), J. G. Fichte: Foundation of the Entire Wissenschaftslehre and Related Writings (1794–95). Oxford: Oxford University Press, 461–568. Bruno, G. Anthony (2018). “Genealogy and Jurisprudence in Fichte’s Genetic Deduction of the Categories.” History of Philosophy Quarterly 35.1, 77–96. Bruno, G. Anthony (2021). “Facticity and Genesis: Tracking Fichte’s Method in the Berlin Wissenschaftslehre.” Fichte-Studien 49, 177–97. Bykova, Marina F. (2009). “Fichte’s Doctrine of Self-Positing Subject and Concept of Subjectivity.” Fichte-Studien 32, 129–39. Bykova, Marina F. (2019). “Kant’s ‘I Think’ and Fichte’s Principle of Self-Positing.” Anuario Filosófico 52.1, 145–65. Bykova, Marina F. (2020). “Fichte’s Life and Rise to Philosophical Prominence.” In Marina F. Bykova (ed.), The Bloomsbury Handbook of Fichte. London, New York: Bloomsbury, 21–42. Cassirer, Ernst (1974). Das Erkenntnisproblem in der Philosophie und Wissenschaft der neueren Zeit. 4 vols. Darmstadt: Wissenschaftliche Buchgesellschaft. Di Giovanni, George (2021). Hegel and the Challenge of Spinoza: A  Study in German Idealism, 1801–1831. Cambridge: Cambridge University Press. Frängsmyr, Tore (1975). “Christian Wolff’s Mathematical Method and Its Impact on the Eighteenth Century.” Journal of the History of Ideas 36.4, 653–68. Frketich, Elise (2019). “Wolff and Kant on the Mathematical Method.” KantStudien 110.3, 333–56. Gava, Gabriele (2015). “Kant’s Synthetics and Analytic Method in the Critique of Pure Reason and the Distinction between Philosophical and Mathematical Syntheses.” European Journal of Philosophy 23.3, 728–49. Gava, Gabriele (2018). “Kant, Wolff, and the Method of Philosophy.” In Daniel Garber and Donald Rutherford (eds.), Oxford Studies in Early Modern Philosophy, Volume VIII. Oxford: Oxford University Press, 271–304. Guyer, Paul (1987). Kant and the Claims of Knowledge. Cambridge: Cambridge University Press. Hammacher, Klaus (1988). “Zur transzendentallogischen Begründung der Dialektik bei Fichte.” Kant-Studien 79, 467–75. Hartkopf, Werner (1967). “Die Dialektik Fichtes als Vorstufe zu Hegels Dialektik.” Zeitschrift für philosophische Forschung 21, 173–207. Kim, Halla (2020). “Transcendental Method.” In Marina F. Bykova (ed.), The Bloomsbury Handbook of Fichte. London, New York: Bloomsbury, 337–44.

Fichte’s Original Presentation  205 Klotz, Christian (2016). “Fichte’s Explanation of the Dynamic Structure of Consciousness in the 1794–95 Wissenschaftslehre.” In David James and Günter Zöller (eds.), The Cambridge Companion to Fichte. Cambridge: Cambridge University Press, 65–92. Lauth, Reinhard (1989a). “Der Ursprung der Dialektik in Fichtes Philosophie.” In R. Lauth (ed.), Transzendentale Entwicklungslinien von Descartes bis zu Marx und Dostojewski. Hamburg: Meiner, 209–26. Lauth, Reinhard (1989b). “Die Entstehung von Fichtes ‘Grundlage der gesammten Wissenschaftslehre’ nach den ‘Eignen Meditationen über Elementar Philosophie’.” In R. Lauth (ed.), Transzendentale Entwicklungslinien von Descartes bis zu Marx und Dostojewski. Hamburg: Meiner, 155–79. Martin, Wayne (1997). Idealism and Objectivity: Understanding Fichte’s Jena Project. Stanford: Stanford University Press. Mikkeli, Heikki (2018). “Giacomo Zabarella.” In Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy, Spring 2018 edition. https://plato.stanford.edu/archives/spr2018/entries/zabarella/ (Last accessed: 23. Oct. 2021). Phillips, Rory (2017). What Is Fichte’s Thesis of the Primacy of the Practical and How Does He Argue for This Thesis? Master’s Thesis, University College London. https://discovery.ucl.ac.uk/id/eprint/10046382/ (Last accessed: 23. Oct. 2021). Russell, Bertrand (1945). A History of Western Philosophy. New York: Simon & Schuster. Seebohm, Thomas W. (1994). “Fichte’s Discovery of the Dialectical Method.” In Daniel Breazeale and Tom Rockmore (eds.), Fichte: Historical Contexts/Contemporary Controversies. Atlantic Highlands, NJ: Humanities Press, 17–42. Sommer, David (2021). “General Logic and the Foundational Demonstration of the First Principle in Fichte’s Eigene Meditationen and Early Wissenschaftslehre.” In David Wood (ed.), The Enigma of Fichte’s First Principles. Leiden, Boston: Brill, 32–58. Stiehler, Gottfried (1962). “J. G. Fichtes synthetische Methode als Keimform der Dialektik.” Deutsche Zeitschrift für Philosophie 10, 252–70. Tolley, Clinton (2012). “The Generality of Kant’s Transcendental Logic.” Journal of the History of Philosophy 50, 417–46. Wood, David W. (2012). “Mathesis of the Mind.” A Study of Fichte’s Wissenschaftslehre and Geometry. Leiden, Boston: Brill. Wood, David W. (2019). “Fichte’s Absolute I  and the Forgotten Tradition of Tathandlung.” In Manja Kisner, Giovanni Pietro Basile, Ansgar Lyssy and Michael Bastian Weiß (eds.), Das Selbst und die Welt—Beiträge zu Kant und der nachkantischen Philosophie (Festschrift für Günter Zöller). Würzburg: Königshausen & Neumann, 167–92. Zöller, Günter (2002). Fichte’s Transcendental Philosophy: The Original Duplicity of Intelligence and Will. Cambridge: Cambridge University Press.

11 The Idea of Universal Monarchy in Fichte’s Practical Philosophy David James

Fichte discusses the idea of universal monarchy in his Contribution Toward the Correction of the Public’s Judgments concerning the French Revolution (Beitrag zur Berichtigung der Urteile des Publikums über die französische Revolution) from 1793. This idea also appears in later texts, including the Addresses to the German Nation (Reden an die deutsche Nation) from 1807 to 1808. In what follows, I  shall discuss Fichte’s explicit references to the idea of universal monarchy and then argue that, in The Closed Commercial State (Der geschloßne Handelsstaat) from 1800, he implicitly extends this idea by identifying the same expansionary tendency as a feature of states engaged in international commercial activity. Fichte himself suggests this in The Fundamental Characteristics of the Present Age (Die Grundzüge des gegenwärtigen Zeitalters), which consists of a series of lectures delivered in 1804–05. In the 14th lecture, Fichte describes a situation in which states within the area of Christian Europe exhibit the same tendency to expand their territory and dominion. Such expansion is facilitated by a common culture rooted in the Christian religion and in similar forms of political organization. This tendency is not an accidental one. Rather, there is “the necessary tendency [die nothwendige Tendenz] in every cultivated state to extend itself universally [allgemein], and to incorporate everything that exists within its civil unity [bürgerliche Einheit]” (CPA, 228; GA I/8: 355; translation modified). Although the Catholic Church once acted as a barrier to this tendency, in so far as its authority provided the states of Christian Europe with an incentive to behave peaceably toward one another, these states have become independent and powerful enough to assert themselves in opposition to it, to the point at which the “tendency toward a ChristianEuropean universal monarchy” has become “the real animating principle of our history,” for even if modern states might not consciously act in accordance with this principle, only it can explain their actions (CPA, 229; GA I/8: 356; translation modified). Fichte then identifies gaining control over international commerce and trade (Welthandel) as one way DOI: 10.4324/9781003044376-12

The Idea of Universal Monarchy in Fichte  207 in which a state may seek to extend its power and dominion by commanding the productive powers of the citizens of other states without needing to defeat these states militarily (CPA, 232–3; GA I/8: 358). Even if Fichte is here assuming that all such states are monarchies, there does not appear to be any reason to think that states with a different political form could not exhibit the same expansionary tendency. I  shall argue that Fichte himself implies in The Closed Commercial State that any state can indeed exhibit this tendency. The idea of universal monarchy features either in a ‘popular’ writing or in a text that represents the further application of principles identified elsewhere that can nevertheless, to some extent, be understood independently of Fichte’s transcendental philosophy. Ultimately, however, there is, in each case, a connection with the philosophical system that Fichte sought to construct. Although it predates the first version of his Wissenschaftslehre, even Fichte’s defense of the French Revolution has been said to anticipate central features of this philosophical science.1 Fichte himself declares in a letter that My system is the first system of freedom. Just as France has freed man from external shackles, so my system frees him from the fetters of things in themselves, which is to say, from those external influences with which all previous systems—including the Kantian—have more or less fettered man. . . . [I]t was while I was writing about the French Revolution that I was rewarded by the first hints and intimations of this system. (EPW, 385–6; GA III/2: 298) The German national education described in the later Addresses to the German Nation is described not only as the means of securing the German nation’s independence but also as “the only possible education for philosophy and the only means of making it universal” (AGN, 129; GA I/10: 228), while the economic proposals outlined in The Closed Commercial State represent the further application of principles identified in one of the philosophical sciences founded on the Wissenschaftslehre, namely, the theory of right set out in the Foundations of Natural Right according to the Principles of the Wissenschaftslehre (Grundlage des Naturrechts nach Prinzipien der Wissenschaftslehre) from 1796 to 1797. Yet is there any deeper connection between the idea of universal monarchy and Fichte’s philosophical system, in that elements of this system explain the expansionary tendency that is a defining feature of universal monarchy? I  shall argue that the theory of drives presented in Fichte’s System of Ethics according to the Principles of the Wissenschaftslehre (Das System der Sittenlehre nach den Prinzipien der Wissenschaftslehre) from 1798, which is the cornerstone of his practical philosophy, provides one possible explanation of this expansionary tendency.

208  David James

1. The Idea of Universal Monarchy in Fichte’s Defense of the French Revolution The moral, rather than strictly political, nature of Fichte’s defense of the French Revolution is indicated by such statements as the following one: The French Revolution appears to me to be important for the whole of humanity. I am not speaking of the political consequences which it has had both for that land and for neighboring states, and which it would probably not have had without the uninvited interference and rash self-confidence of these states. All this is in itself a lot, but it is much less compared to what is incomparably more important about it. (GA I/1: 203) The moral significance of the French Revolution concerns how this political event exemplifies the requirement to found a state on the moral autonomy of its members and the inalienable rights that derive from it. By voluntarily subjecting themselves to a set of fundamental laws, that is, to a constitution (Verfassung), individuals become the citizens of a state that they are nevertheless free to leave on the same voluntary basis. The theme of universal monarchy relates to developments to which Fichte alludes in the passage quoted earlier. These developments concern the increasingly violent turn taken by the French Revolution with the onset of the Terror. Fichte implies that this violent turn can be explained in terms of the decision of reactionary European states to attack revolutionary France (“the uninvited interference and rash self-confidence of these states”), rather than in terms of any inherently violent tendencies found within the Revolution itself. Since these European states were all monarchies and the French nation acted defensively, even if territorial expansion was a by-product of its attempt to defend itself against military aggression aimed at restoring the status quo ante, it is not surprising to encounter a discussion of the idea of universal monarchy in Fichte’s defense of the French Revolution. Fichte introduces this idea in connection with the claim made by the rulers of monarchical states that they are merely seeking to maintain the existing balance of power in Europe. Yet each ruler makes this claim only as a pretext for increasing his or her power over his or her own subjects and for expanding the territory of his or her own state. Thus each ruler’s real intention is to establish “the most unlimited and extensive dictatorship [Alleinherrschaft],” not only domestically but also internationally, thereby reducing the maintenance of the existing balance of power in Europe to a means of concealing or justifying this intention (GA I/1: 246). Each monarch is, in fact, disposed not to prevent other states from

The Idea of Universal Monarchy in Fichte  209 seeking to expand their territory, because this will provide him or her with an excuse to seek to expand the territory of his or her own state as a defensive measure undertaken with the alleged aim of restoring the balance of power in Europe in the face of the threat posed by those monarchs who would disrupt it. Moreover, Fichte claims that this tendency toward dictatorship at home and expansion abroad is not an accidental feature of monarchies, and it is in connection with the monarchical state’s inherent tendency to seek to expand its territory and dominion that Fichte speaks of universal monarchy: “The tendency of all monarchies is toward internal unlimited dictatorship [nach Innen uneingeschränkte Alleinherrschaft], and universal monarchy abroad [nach Außen Universalmonarchie]” (GA I/1: 247). What are Fichte’s grounds for this claim? Fichte provides an argument that can be stated as follows: Even if it is granted that maintaining the existing balance of power in Europe is the end toward which territorial expansion serves as the means, it cannot be assumed that this is an end that the peoples ruled by monarchs would themselves adopt. There are, in fact, grounds for claiming that they would not adopt this end as their own end, thereby committing themselves to bearing the high costs attached to the pursuit of this end. Monarchs may claim that the failure to maintain the existing balance of power in Europe will result in a war of all against all that consumes everyone and everything. This is to assume that such a war would indeed be a consequence of the destruction of the existing balance of power, and that the costs of failing to maintain it would therefore prove costlier to the ruled than seeking to maintain it. Fichte denies the validity of these assumptions. He asserts that a complete suspension of the balance of power in Europe “could never become as disadvantageous for the peoples [Völker] as the disastrous maintenance of it has been” (GA I/1: 248). From this we can see that Fichte’s argument consists in showing that those who advocate the maintenance of the existing balance of power in Europe, even assuming that they are being sincere, introduce premises that they fail to justify. Thus these monarchs’ claims concerning the necessity of maintaining the existing balance of power turn out to be only a matter of the right of the stronger: “I know that you support your conclusions by means of standing armies, by heavy artillery, by shackles and by imprisonment in a fortress; but they do not therefore appear to me to be the sounder ones” (GA I/1: 249). In the absence of any genuine rational grounds for claiming that those peoples currently living under monarchical rule would adopt the end of maintaining the existing balance of power in Europe as their own end, why not appeal to different grounds that demonstrate the very opposite? Fichte himself does this, by claiming that We, in contrast, conclude the following: this constant striving to expand internally and externally is a great misfortunate for nations

210  David James [Völker]. If it is true that they must bear this misfortune, in order to escape an immeasurably greater one, then let us nevertheless seek the source of this great misfortune, and divert it, if possible. We locate this source in the unlimited monarchical constitution; every unlimited monarchy (you say it yourselves) incessantly strives for universal monarchy. Let us block this source, and thus is the evil from which we suffer rooted out. If no one any longer wants to attack us, we will no longer need to be armed; terrible wars, and the even more terrible constant readiness for war that we suffer in order to prevent war, will then become no longer necessary. (GA I/1: 248) Fichte here implies that the real problem is, in fact, the nature of monarchy itself, at least insofar as there are no effective constitutional constraints on the monarch’s powers. The abolition of monarchy will therefore be sufficient to prevent the evils of war. For the purposes of an argument that aims to call into question the justification of the existing balance of power in Europe provided by monarchs whose ultimate aim is to consolidate and extend their power and dominion, Fichte needs only to show that the claim that the abolition of monarchies would remove the causes of war is no less valid than the claim that territorial expansion and increasing dominion can be justified in terms of the necessity of maintaining the existing balance of power in Europe. We can nevertheless identify two further lines of argument present in Fichte’s attack on the idea of universal monarchy, one of which exposes the true intention of monarchs and thus strengthens the argument that they are not, in fact, seeking merely to maintain the existing balance of power, while the other argument concerns the claim that a state must be founded on the moral autonomy and associated rights of its members. The first of these arguments rests on the claim encountered in the passage quoted earlier that “every unlimited monarchy (you say it yourselves) incessantly strives for universal monarchy.” The appeal to the necessity of maintaining the existing balance of power in Europe, and thus the claim that it is necessary for there to be several monarchies whose power is sufficiently great to ensure this outcome, both reveal something about the monarchs themselves, given that each of them thereby acknowledges that this necessity derives from the threat posed by any monarch who has become sufficiently powerful to subjugate all others (“you say it yourselves”). This betrays a mutual suspicion whose ultimate source is a common awareness of the inherent tendency of monarchical states to strive for universal monarchy. Thus, in attempting to justify his or her actions by appealing to the threat that other monarchs pose to the existing balance of power in Europe, each monarch condemns himself or herself. The second argument involves several claims, beginning with the claim that if one accepts the ‘ought’ implied by the idea of universally valid

The Idea of Universal Monarchy in Fichte  211 norms of action, then one presupposes the existence of a fundamental law or principle that enables one to judge whether or not an action or state of affairs is morally justified. The law in question is the moral law, which can be employed both to judge the legitimacy of actions or states of affairs and to determine the nature and extent of a human being’s rights and duties. Each individual’s awareness of subjection to this law is the highest instance of his or her consciousness of his or her own moral autonomy, in relation to which all civil freedoms and political freedoms are subordinate, since they would not be possible in the absence of a rational agent’s capacity to impose norms of action upon itself and to act in accordance with these norms. The object of legislation concerns that regarding which the moral law is silent, that is to say, that which this law neither commands nor forbids, and that is therefore morally permissible (GA I/1: 220). The act of making a contract, through which individuals voluntarily alienate certain powers or goods, falls within the domain of the morally permissible. Thus a contractual theory of the state can provide an affirmative answer to the question as to whether or not a nation has the right to alter its constitution, as required of a moral defense of the French Revolution.2 The moral law here performs the negative function of determining what may or may not form the object of a contract because it is neither morally commanded nor morally forbidden, while the concept of moral autonomy itself is employed to justify the claim that individuals can be legitimately subject only to those laws that they impose upon themselves. The importance of the last requirement is evident from the following passage: Thus, the question was: from where arises the binding nature [Verbindlichkeit] of civil laws? I answer: from the freely willed acceptance of the same by the individual; and the right not to recognize any law except the law that one has given oneself is the foundation of that souveraineté indivisible, inalienable of which Rousseau speaks. (GA I/1: 238–9) In this passage, Fichte appeals to Kant’s idea of moral autonomy as the will that is not merely subject to the law but subject to it in such a way that it must be viewed as also giving the law to itself and just because of this as first subject to the law (of which it can regard itself as the author). (G, 4: 431) Fichte understands this idea of moral autonomy in such a way that the law in question does not generate any kind of unconditional moral obligation. Rather, it concerns that which is permissible, and so any obligations

212  David James incurred remain conditional on whether individuals continue to subject themselves to them. Fichte recognizes this implication of his claim that individuals can be legitimately subject only to those laws that they have imposed upon themselves, when he asserts that individuals have an inalienable right to annul unilaterally any contract into which they have entered, which would include the contract that establishes any specific form of political union into which they have agreed to enter. This entails that individuals have an inalienable right to change the constitution of a state if they should wish to do so, given that this constitution is itself the result of a contract into which they have voluntarily entered. Even if an individual cannot get others to agree to such a change, he or she remains at liberty to renounce the legal and political obligations that he or she had previously imposed upon himself or herself by accepting this constitution, though he or she must then be willing to bear the costs attached to taking this step. This argument reinforces Fichte’s objection to the assumption that those peoples ruled by absolute monarchs would adopt the end of maintaining the existing balance of power in Europe as their own end. This time, however, there is a normative claim about what genuine autonomy requires, rather than an empirical claim about the ends that people can be expected to adopt. Even if it is granted that maintaining the existing balance of power in Europe is the final end of those monarchs who appeal to it with the aim of justifying their attempts to expand their territory and dominion by military means, these monarchs would then have no right to demand that others adopt the same end or act as if they had adopted it by playing an active role in a monarchical state’s attempt to expand its territory and dominion, thereby exposing themselves to the evils of war. Rather, whenever the moral law is silent, as Fichte assumes is the case here, it remains up to each individual to adopt those ends that he or she judges to be right for him or her. Moreover, the idea of moral autonomy demands that each individual be the judge of what is likely to prove most advantageous or costly to himself or herself. It is not, therefore, up to monarchs to judge whether the consequences of failing to maintain the existing balance of power in Europe will prove costlier to the peoples whom they currently rule over than the consequences of seeking to maintain it. Forcing individuals to pursue ends that they could not reasonably make their own ends is to reduce them to “instruments in an alien hand,” which is precisely how Fichte describes the situation in which the ruled find themselves in relation to the European monarchs of his own time (GA I/1: 250). As well as forming the basis of an argument against the claims and pretensions of absolute monarchs, the idea of moral autonomy represents a solution to the problems associated with the idea of universal monarchy, in that the exercise of it may, and in fact ought to, result in the rejection of the ultimate cause of war, which is the institution of absolute

The Idea of Universal Monarchy in Fichte  213 monarchy itself. Indeed, once Fichte’s justification of the French Revolution is viewed in connection with his discussion of universal monarchy, it can be seen to assume the truth of the following two claims: (1) a political union founded on the moral autonomy of its members would, unlike an absolute monarchy, not exhibit the expansionary tendency associated with the idea of universal monarchy, and (2) a political union of this kind would therefore be able to coexist peacefully with other states founded on the same principle. Yet one may wonder whether Fichte’s defense of the French Revolution is truly successful in this regard. One practical problem concerns the individualistic nature of Fichte’s presentation of the agreement that forms the basis of a legitimate constitution. If every individual, by virtue of his or her moral autonomy, has the right to leave the political union whose terms are stated in this constitution, and thereby becomes no longer subject to these terms, then this political union would be very unstable indeed, to the point that it would lack the kind of collective or general will that is required to resist the expansionary tendency exhibited by surrounding monarchical states.3 Although Fichte might respond to this objection by claiming that no citizen would in fact want to leave a political union founded on the principle of moral autonomy, there remains no moral obstacle to his or her leaving it, since the obligations generated are not unconditional ones, whereas the desire not to risk one’s life in defense of this political union may provide individuals with a sufficiently strong incentive for renouncing their allegiance to it. It is therefore not surprising that reference to the idea of universal monarchy is later found in the Addresses to the German Nation, whose aim is to make the German nation conscious of both the bonds that already unite its members, despite the divisions that have emerged within this nation in the course of its history, and the measures required to strengthen these bonds in such a way as to make this nation more unified than it currently is. In this connection, Fichte speaks of a German national education whose aim is “to form the Germans into a totality that in all its individual parts is driven and animated by the same single interest” (AGN, 19; GA I/10: 114). Fichte later mentions a “new” universal monarchy, which he associates with the same nation whose revolution he had earlier sought to defend, but which now poses a threat to Germany and the rest of Europe because of the expansionary tendencies that it itself has come to exhibit under Napoleon, to whom Fichte alludes when he speaks of someone who “has determined that all the seeds of what is human in humanity be ground down in order that the yielding dough may be pressed into some form or other” (AGN, 172; GA I/10: 273). From what has so far been said, the fundamental problem that Fichte identifies with the idea of universal monarchy is a tendency toward territorial expansion that requires increasing the monarch’s power both internally and externally. In the next section, I shall argue that Fichte

214  David James attributes the kind of expansionary tendency associated with this idea to the modern state as such in so far as it engages in international commercial activity. Expansion here takes the form of an attempt to dominate international markets, rather than the form of actual territorial expansion, though, as with colonization, this type of expansion may accompany the first type of expansion. Fichte claims that only the commercial closure of the state can put a stop to this expansionary tendency. This act of closure must be performed by the state itself, whose authority does not depend on the free choice of its members. Rather, the state’s authority derives from how it meets certain normative requirements that can be specified independently of that to which individuals happen to consent. This extension of the idea of universal monarchy becomes evident once we compare certain claims found in The Closed Commercial State from 1800 with Kant’s remarks on universal monarchy in Toward Perpetual Peace (Zum ewigen Frieden) from 1795. Fichte reviewed this essay, and in his review, he tellingly speaks of “the commercial oppression of foreign nations and portions of the world” (RPP, 320; GA I/3: 227).

2. The Extension of the Idea of Universal Monarchy in The Closed Commercial State Fichte’s The Closed Commercial State and Kant’s Toward Perpetual Peace have been viewed as contributions to a broader debate about how to secure peace in Europe by establishing a law-governed international order (Nakhimovsky 2011). Fichte’s contribution to this debate is said to be the idea that Europe can transform itself into a peaceful federation of constitutional republics only if the economic life of states is disentangled from the competitive commercial relations between them. This is to be achieved by each state introducing a planned and largely self-sufficient national economy. This emphasis on international peace is misleading, however, because Fichte makes clear that the economic arrangements proposed in The Closed Commercial State, including the commercial closure of the state, follow in a quasi-deductive manner from the pure principles of right set out in the Foundations of Natural Right. One key principle is the concept of property,4 and it is this concept especially that provides the normative foundations of Fichte’s theory of the relevant type of state: All these assertions are founded on my theory of property. If this theory of property is correct, then, without doubt, there will also be good reason for these assertions. If the theory of property is false, then that which claims to be nothing more than a consequence of the theory of property will no doubt fall together with it. (CCS, 129; GA I/7: 84)

The Idea of Universal Monarchy in Fichte  215 Thus, although Fichte does indeed claim that the commercial closure of the state promises to bring about global peace, this peace is a by-product of the full, consistent application of the concept of right, rather than the direct aim of the commercial closure of the state. In this respect, Fichte’s extension of the concept of universal monarchy forms part of a more general turn to the economic sphere demanded by the application of pure principles of right to the empirical domain of the existing relations between states, which are products of history. I  shall now show how Fichte’s and Kant’s arguments concerning universal monarchy and the threat that it poses to international peace are motivated by certain common concerns, despite how their accounts of the removal of this threat substantially differ. In Toward Perpetual Peace, Kant locates the problem of universal monarchy within an account of the following three types of global order: (1) A  global legal and political order made up of independent states that have voluntarily agreed to enter into a federal union with one another and are thereby able to coexist peacefully. (2) A global order that consists of independent states that have not voluntarily entered into a political union that prevents war breaking out between them. Instead, these states confront one another in a hostile manner, in that each of them fears the other states that in turn fear it. (3) A global order that can ensure peace in the short term but only in a way that depends on the overwhelming power of one state that is able to subjugate all other states, as opposed to by means of a global legal and political order established by the voluntary agreement of each member state to establish this order and honor its terms. While (1) represents the ideal international political order, Kant prefers (2) to (3) despite how the former is a state of war and the latter is, like (1), a condition of peace, for the following reason: [T]his is nevertheless better, in accordance with the idea of reason, than the fusion of them by one power overgrowing the rest and passing into a universal monarchy, since as the range of government expands laws progressively lose their vigor, and a soulless despotism, after it has destroyed the seed of good, finally deteriorates into anarchy. Yet the craving of every state (or of its head) is to attain a lasting condition of peace in this way, by ruling the whole world where possible. (TPP, 8: 367) From this passage, Kant’s preference for (2) over (3) appears to rest on certain negative moral and political consequences of (3) that will eventually destroy the state of peace achieved by means of the dominant

216  David James position of one overwhelmingly powerful state. Since, according to both (2) and (3), states may exhibit the tendency to expand their dominion because of the desire to secure their independence and security in the face of the actual or potential aggression of other sufficiently powerful states, one may wonder how Kant’s first preference for a state of global peace achieved by means of a federal union of otherwise independent states could ever be satisfied. Kant explains the possibility of this federal union of independent states in terms of both the negative effects of the expansionary tendencies exhibited by states and certain countervailing forces. Kant identifies two ways in which the expansionary tendencies exhibited by states are countered by the continued separation of peoples. The first way concerns differences in language and religion. Although these differences are sources of conflict between these peoples, this conflict, which is accompanied by the advance of culture in the long run, leads human beings to discover the means of establishing global peace and motivates them to agree to the adoption of these means (TPP, 8: 367). It is here assumed that it is possible to achieve an appropriate balance between the dynamism of competition between human beings as speakers of particular natural languages, as adherents of particular religions, and as citizens of particular states, on the one hand, and the need for a global legal and political order that prevents wars between states, on the other. The second way concerns the self-interest of states, and how it is transformed into a mutual interest once they enter into commercial relations with one another. The pursuit of each state’s commercial interests is here assumed to be possible only if trading relations with other states are made sufficiently reliable and stable, for otherwise no state could ever be sure that the terms of any commercial agreement into which it has entered with another state will in fact be honored. Thus states have a mutual interest in establishing and maintaining the kind of international legal and political order that is a condition of global peace: It is the spirit of commerce, which cannot coexist with war and which sooner or later takes hold of every nation. In other words, since the power of money may well be the most reliable of all the powers (means) subordinate to that of a state, states find themselves compelled (admittedly not through incentives of morality) to promote honorable peace and, whenever war threatens to break out anywhere in the world, to prevent it by mediation, just as if they were in permanent league for this purpose. (TPP, 8: 368) As this statement shows, it is not the case that states directly intend to establish an international legal and political order of the relevant kind. Rather, initially it is a matter of necessity for them, leading Kant to speak of a purposiveness that lets “concord arise by means of the discord

The Idea of Universal Monarchy in Fichte  217 between human beings even against their will” (TPP, 8: 360). Although this necessity in the first instance concerns the need to avoid a state of war and its consequences, Kant implies that commerce and the conditions of it also serve as the means of establishing global peace, despite the expansionary tendencies otherwise exhibited by states. Commercial relations are not, however, viewed as being in and of themselves directly peace-promoting ones. Although Fichte’s position differs from Kant’s in that severing commercial relations between states is held to be a necessary condition of global peace,5 commercial relations between states have for him as well an indirect peace-promoting role. This is because the conflict between states caused by their commercial relations with one another demonstrates the necessity of the commercial closure of each state. This conflict may, moreover, eventually motivate politicians guided by the philosopher to introduce the relevant measures. Given that the source of conflict between it and other states is thereby removed, the economically self-sufficient state that has severed all commercial ties with other states provides neighboring states with “the guarantee that from now on it will not expand itself in any way,” whereas this is not the case with states that continue to operate within an international system of commerce and trade: A state that pursues the customary system of trade and aims at superiority at world trade will retain a continuing interest in expanding itself even beyond its natural borders, so that it may thereby increase its trade and by this means its wealth, applying this new wealth in turn to new conquests, which will then be put to the same use as before. A new evil is forever chasing the feet of the old, and the greed of such a state knows no boundaries. (CCS, 171; GA I/7: 119) From this passage, it is clear that, for Fichte, states manifest the tendency to expand their commercial influence in the pursuit of greater wealth, and that this expansionary tendency is typically accompanied by some form of territorial expansion. This tendency to expand may involve only an attempt to open up new markets or to dominate existing ones, and a state may achieve these ends on the basis of its greater economic power in relation to other states. The economic domination of other states is said to characterize the commercial relations that exist between the European states of Fichte’s own time and between these states and other states: Europe has a great advantage in trade over the remaining parts of the world, whose forces and products it takes for its own use without giving anywhere near a sufficient return payment. Every single European state, however unfavorable its own trade balance stands in

218  David James relation to the others, still draws some advantage from this common exploitation of the rest of the world, nor will it ever abandon the hope of improving the trade balance in its favor and thus drawing an even greater advantage. (CCS, 85; GA I/7: 44)6 There is no reason to think that this tendency to expand in the form of an attempt to achieve increasing economic power and to dominate global trade would not manifest itself equally as much in republics as in monarchies. All that matters is that a state engages in international commercial activity. In this respect, one can speak of an extension of the idea of universal monarchy, in that the tendency to seek to expand a state’s dominion is not reducible to the desire for territorial expansion exhibited by monarchical states. The only major difference concerns the grounds used to justify the state’s attempt to expand its dominion, for while monarchical states appeal to the need to maintain the existing balance of power in Europe, states that seek to dominate one another commercially appeal to such grounds as the benefits of free trade. If there is an expansionary logic that is true of all states, irrespective of whether they are monarchies or republics, then the following question arises: does Fichte provide any explanation of this general tendency that might cover all such expansionary tendencies exhibited by states? This explanation would have to be sought at a more fundamental level than the application of the concept of property to an existing state of affairs that is a product of history or a popular presentation of philosophical concepts and theories. I  shall now argue that an explanation can be found in the particular philosophical science called the doctrine (or theory) of ethics (Sittenlehre), which Fichte equates with the sphere of ‘practical’ philosophy that forms an essential part of the philosophical system which he sought to develop. This explanation concerns Fichte’s theory of drives.

3. The (Extended) Idea of Universal Monarchy and Fichte’s Theory of Drives The central task of the doctrine of ethics is “to provide an exhaustive presentation of that system of necessary thinking according to which a being corresponds to and follows from our representations” (SE, 8; GA I/5: 22). In other words, the task is to identify and explain the consciousness of ourselves as agents capable of producing events and states of affairs in accordance with the “concept of an end” (SE, 14; GA I/5: 27–8). This consciousness concerns how we think of ourselves and how the world in which we act ought to be. Given that practical philosophy here operates at the transcendental level, the fulfillment of the task in question demands the identification and explanation of the necessary

The Idea of Universal Monarchy in Fichte  219 conditions of this consciousness of ourselves. These conditions will be shown to be conditions of the identity of the I presupposed by any conscious act, but of which we ourselves can never become directly conscious because consciousness necessarily involves a separation of the subject of consciousness from the object of consciousness, whereas in the case of self-consciousness that which thinks and that which is thought are one and the same (SE, 11; GA I/5: 24; SE, 45; GA I/5: 56). The I’s identity must nevertheless be presupposed if we are to explain how the subject can be conscious of something as other than itself and recognize the act of distinguishing itself from this object as its own act. Fichte explains the relevant type of practical self-consciousness in terms of the concept or principle of morality, in which the subjective moment consists in the consciousness of oneself as free that is itself made possible by the consciousness of the “norm” or “law” of self-legislation, which is objective in that it is how a free agent must think its freedom. This consciousness of subjection to the norm or law in question is possible only if an agent thinks of itself as free, given how the norm or law commands that the agent engage in an act of self-legislation, and thus presupposes the freedom of its addressee. Thus, that which is subjective and that which is objective in the I are “thought of as purely and simply one,” and “[f]reedom does not follow from the law, no more than the law follows from freedom. These are not two thoughts, one of which can be thought to depend upon the other; rather this is one and the same thought” (SE, 55; GA I/5: 64–5). Since the I is not determined by anything external to itself when viewed purely as a free and rational agent, Fichte speaks of “absolute self-sufficiency [absolute Selbstständigkeit], absolute undeterminability by anything other than the I” (SE, 58; GA I/5: 67). This absolute self-sufficiency is not simply a state of the I, for it expresses itself in a norm or law to which the I is subject, and to which it ought to conform. This absolute self-sufficiency is therefore also a condition which the I strives to achieve, leading Fichte to speak of the I’s “tendency toward self-sufficiency” (Tendenz zur Selbstständigkeit) (SE, 61; GA I/5: 70). Fichte’s theory of drives is concerned with this conative aspect of human freedom. The concept of a drive is introduced in order to explain how the I can be causally efficacious and thereby “find” itself as free. From the transcendental standpoint there is one “original drive” (Urtrieb) (SE, 124; GA I/5: 125). This original drive nevertheless appears as two separate drives: the natural drive and the higher drive. The natural drive is one that the I has by virtue of being an embodied agent confronted with an external nature of which it itself forms a part. This is nature understood as “what is fixed and determined independently of freedom” (SE, 105; GA I/5: 108). The givenness of the natural drive means that it, too, fits this description of nature. As it stands, the natural drive is incompatible with the concept of freedom understood as absolute

220  David James self-sufficiency. This incompatibility manifests itself in how enjoyment becomes the goal of all activity, making a human being “dependent on something given: namely, the presence of the objects of his drive; and he is therefore not sufficient unto himself, inasmuch as the achievement of his end depends upon nature as well” (SE, 124; GA I/5: 125). Yet, as part of the ‘entire’ I, the natural drive cannot be eliminated. Rather, it must be made compatible with the freedom of the I. This compatibility is explained in terms of how the I becomes reflectively aware of the natural drive in such a way that it not only recognizes this drive as its drive but is also conscious of its independence of the natural drive in the form of the power to determine its response to this drive that “belongs to the I, but. . . is not the I” (SE, 134; GA I/5: 133). Thus, “the reflecting subject stands higher than what it reflects upon,” and for this reason Fichte terms the drive of the reflecting subject the “higher” drive (SE, 125; GA I/5: 126). Moreover, in connection with the I’s independence of, and elevation above, nature, Fichte speaks of the “pure” drive, which consists in “a drive to activity for activity’s sake, a drive that arises when the I internally intuits its own absolute power” (SE, 137; GA I/5: 136). The I’s intuiting of its self-sufficiency is explained in terms of a second act of reflection, whose object is the I’s act of reflecting upon its natural drive, thereby transforming this drive into the object of its consciousness. In reflecting on this act, the I abstracts from the natural drive altogether and thereby becomes conscious of “nothing but the pure, absolute activity that occurred in the first act of reflection” (SE, 133; GA I/5: 133). The higher drive and the natural drive must be unified, however, in such a way that the latter remains subordinate to the former. This unification can be achieved only if the purity of the higher drive’s self-activity is surrendered at the same time as the lower drive no longer seeks mere enjoyment as its end. There would then be the drive for absolute self-sufficiency, on the one hand, and the limitation of this same drive accompanied by an attempt to remove this limitation, on the other, thereby giving rise to “an objective activity, the final end of which is absolute freedom, absolute independence from all nature—an infinite end, which can never be achieved” (SE, 125; GA I/5: 126). Fichte’s theory of drives, as so far presented, bears no obvious relation to the idea of universal monarchy. There is a tendency that is explained in terms of the goal of achieving absolute self-sufficiency, and this tendency might be thought to imply an attempt on the part of an agent to expand its independence by increasing its power over nature. Yet there is no suggestion that this tendency must manifest itself in relation to anything other than nature understood in the very broad sense of that which “is fixed and determined independently of freedom”. I  shall now indicate some ways in which Fichte’s theory of drives and the absolute selfsufficiency that the I strives to attain can nevertheless be related to the idea of universal monarchy and the expansionary tendency associated

The Idea of Universal Monarchy in Fichte  221 with it, whether this tendency be viewed in territorial, political, or economic terms. The first way concerns how the I’s independence of, and power over, nature is a condition of the transition to a genuinely moral way of thinking and acting. This implies an intermediate stage at which the I is conscious of its independence of, and power over, nature without, however, the concept that it adopts as its immediate end being a moral one. The tendency toward absolute self-sufficiency may then assume a distorted, immoral form. Fichte himself identifies “the oppression of others” as one rule (or maxim) that an agent may adopt (SE, 57; GA I/5: 66), and the idea that the self-sufficiency of the I  explains immoral conduct of this kind is supported by other claims concerning the drive for absolute self-sufficiency. Fichte identifies three different levels that an individual may occupy, two of which directly concern the drive for absolute self-sufficiency. At the first level a human being is conscious of the natural drive and subjects this drive to the maxim of one’s own happiness. Here, “the human being becomes an intelligent animal” (SE, 171; GA I/5: 167). A human being can then reflect on the free activity that is already present at this level, thereby becoming conscious of this activity as such and directly aware of the drive for absolute self-sufficiency. In this way, a human being raises himself or herself “to a completely different kind of freedom” (SE, 175; GA I/5: 170). The consciousness of freedom thus achieved assumes, however, only the form of the consciousness of “a blind drive” that is not governed by any law (SE, 176; GA I/5: 171–2). Yet Fichte also speaks of the maxim of unrestricted and lawless dominion [Oberherrschaft] over everything outside us—a maxim which, to be sure, is not clearly thought, but which offers to an observer who occupies a higher standpoint the only ground for explaining [such a way of acting]. (SE, 177; GA I/5: 172) This claim is crucial because it implies that even when the drive for absolute self-sufficiency is governed by a norm or law, it can manifest itself in the form of a largely unconscious drive to dominate everything external to the I, which includes not only nature in general and specific instances of it (e.g., pieces of land and natural resources) but also other human beings. Although the agent may not be directly conscious of the norm or law in accordance with which he or she acts, the maxim identified earlier is, for Fichte, the only one that can sufficiently explain his or her desires and actions, and so it can be viewed as a maxim that this agent implicitly adopts. At the third and final level, in contrast, the drive for absolute self-sufficiency is present in such a way that it is subject to consciously adopted moral maxims that limit it. Fichte distinguishes between this moral drive for absolute self-sufficiency and the lawless one

222  David James by describing the former as wanting to achieve “freedom and independence . . . only gradually and in accordance with certain rules,” so that “it does not want to possess unconditional and lawless causality,” whereas the latter “demands unconditional and unlimited causality” (SE, 177; GA I/5: 172–3). The lawless drive for absolute self-sufficiency might explain not only the expansionary aims of absolute monarchs and someone such as Napoleon,7 but also the willingness of individuals to act in opposition to the maxim of one’s own happiness by exposing themselves to the dangers and enduring the hardships associated with the pursuit of such aims. Indeed, Fichte himself claims that “One could call it heroic, and it is also the usual way of thinking for the heroes of our history” (SE, 181; GA I/5: 175). The drive for absolute self-sufficiency in its lawless form can then be thought to explain the actions of “the extraordinary human being” who possesses a robust character without being the least bit better with respect to morality. He is neither indolent nor cowardly nor false, but in his exuberance he tramples upon everything around him and becomes the lord [Herr] and oppressor of those who prefer to be slaves. (SE, 193; GA I/5: 186) The superiority of this drive for absolute self-sufficiency, as compared to slavish obedience to the maxim of one’s own happiness and the dependence that it entails, consists in how, with respect to form, “there is no difference between this way of thinking and the genuinely moral way of thinking” (SE, 178; GA I/5: 173). Both ways of thinking are manifestations of the drive for absolute self-sufficiency, though in the second case this drive is materially limited by moral norms to which an agent subjects himself or herself. Although Fichte’s theory of drives appears to be the place in which to seek his transcendental explanation of the expansionary tendency that he associates with the idea of universal monarchy, it might be asked whether this theory promises to provide a good explanation of the phenomenon in question. One difficulty concerns the way in which Fichte extends the idea of universal monarchy by attributing this expansionary tendency to the state as such, irrespective of its political form, because of how it seeks to increase its economic power, resulting in conflict between it and other states. In Fichte’s ethical theory, the drive for absolute self-sufficiency is attributed to individual agents capable of becoming directly conscious of this drive and limiting it in accordance with moral norms. It is less clear, however, how such a drive, let alone a direct consciousness of it and the act of subjecting it to moral norms, can be attributed to a political entity

The Idea of Universal Monarchy in Fichte  223 such as a state. There are nevertheless ways in which the drive for absolute self-sufficiency can be related to Fichte’s account of the behavior of states. In The Closed Commercial State, Fichte identifies a two-stage process in which the state’s drive for absolute self-sufficiency manifests itself in different, but complementary, ways, with the second stage marking the transition from a lawless drive for absolute self-sufficiency to a law-governed one. The first stage concerns a state’s attempt to attain its “natural borders” prior to its commercial closure. These borders are understood in terms of a geographical area in which sufficient natural resources are available to the state for it to achieve “productive independence and self-sufficiency [productive Selbstständigkeit, und Selbstgenugsamkeit]” (CCS, 169; GA I/7: 117). Thus the ‘natural’ tendency to expand exhibited by states can be regarded as a legitimate one because it relates to material conditions that must be fulfilled if the state is to become a closed commercial state. Although this tendency can be explained in terms of how the natural drive is the cause of territorial expansion aimed at securing the material conditions of economic self-sufficiency,8 such an aim presupposes the concept of self-sufficiency, and so the tendency in question can be understood as a manifestation of the drive for absolute self-sufficiency. At the second stage, the drive for absolute self-sufficiency is limited by the law of economic self-sufficiency. This law concerns the measures required to achieve the commercial closure of the state and the conscious adoption of them. As we have seen, according to Fichte, if this law is applied, then a condition of global peace can be achieved, because each state will no longer have any incentive to extend itself beyond its natural borders. There would then be a lawful drive for self-sufficiency, as opposed to a lawless one, and this lawful drive may explain how the expansionary tendency associated with the idea of universal monarchy can be opposed. Yet the law in question is one of which state actors may fail to become directly conscious, or may not be willing to adopt even if they are conscious of it, resulting in a situation in which states continue to seek to extend their borders even though there is no longer any material need for them to do so. Rather, expansion has become an end in itself. Fichte himself speaks of how states may not only aim “obscurely at the acquisition of their natural borders,” but also “be driven by the thirst for conquest [Eroberungssucht]—be it blind and indefinite, or indeed clear in vision and very definite” (CCS, 170; GA I/7: 117–8). This is not sufficient to show that such expansionary tendencies can indeed be explained using the subject-centered approach adopted by Fichte in his ethical theory. One way of making this explanation more convincing would be to attribute the desire to increase their economic power exhibited by states in their commercial relations with one another to the lawless, and thus unlimited, absolute drive for self-sufficiency of a single ruler, or to a collective drive of the same kind that arises from the

224  David James aggregation of the individual drives of the most powerful members of the state. A  further difficulty then arises, however. This difficulty concerns how alternative explanations of the same behavior of states in relation to one another then become available. For example, it could be that one state actor or group of state actors believes that another state is exhibiting expansionary tendencies or is likely to do so in the future in such a way as to pose an existential or serious economic threat to one’s own state. When translated into the language of Fichte’s theory of drives, there might then be said to exist a mutual fear of another state’s lawless drive for absolute self-sufficiency, even though there is a willingness on the part of one or more states to limit this drive by adopting the law of economic self-sufficiency, the universal application of which is recognized to be a condition of lasting global peace. This fear becomes mutual because the state actors of one state can neither be directly conscious of the drives of other state actors nor know whether there is any genuine intention on their part to limit the drive for absolute self-sufficiency in accordance with the law of economic self-sufficiency.9 Each state may then come to exhibit expansionary tendencies that result from its desire to protect itself, despite its willingness to apply the law of economic self-sufficiency and thus initiate the commercial closure of the state once its natural borders have been attained. This is precisely the kind of reason that absolute monarchs might use to justify their attempts to maintain the existing balance of power in Europe by attacking other states and seeking to occupy their territory with the alleged intention of stopping them becoming powerful enough to disrupt this balance of power. In this case, Fichte suggests that fear of other states merely serves as a pretext for seeking to expand one’s dominion as far as possible in accordance with the lawless drive for absolute self-sufficiency. There appears, however, to be no compelling reason for thinking that such a pretext would not also be employed by any state that is seeking to gain the upper hand in its commercial relations with other states even when there is no longer any absolute material need for them to increase their economic power and dominance of markets.10

Notes 1. See, for example, Beiser (2016, 48); Buhr (1991, 62–70). 2. An important element of Fichte’s defense of the French Revolution concerns how an unalterable constitution would prevent the advance of culture and thus contradict the final end of humanity (see GA I/1: 252–3, 257). This relates to Fichte’s attack on universal monarchy, in that the full exercise of moral autonomy requires that certain civil and political conditions be in place, among which he includes freedom of thought and, we may assume, freedom of expression as well. Indeed, Fichte suggests that this freedom would be sufficient to prevent the tendency toward absolute monarchy from continuing to manifest itself: “Next to unlimited freedom of thought unlimited monarchy

The Idea of Universal Monarchy in Fichte  225 cannot exist” (GA I/1: 251). Thus, insofar as monarchical power is incompatible with freedom of thought, it is also incompatible with the advance of culture. I think, however, that this must be viewed as a subsidiary argument against universal monarchy, in that the right to moral autonomy is more fundamental and culture concerns the exercise of this right rather than the existence of it. Moreover, even the capacity to exercise this right must be presupposed in so far as it is a condition of changing a constitution in such a way as to enable culture to advance. 3. Although Fichte appeals to Rousseau’s notion of an indivisible and inalienable sovereignty whose source is the exercise of each individual’s right to give himself or herself the law to which he or she is subject, this claim has been treated with skepticism because of how Fichte fails to explain the idea of a general will, which is integral to Rousseau’s notion of popular sovereignty. The way in which Fichte appears to associate the general will with the autonomy of individual conscience rather than with the collective autonomy of the sovereign people demands an account of how the single will of each individual can be thought to accord with the wills of all other individuals so as to form a truly general will. By failing to address this issue, Fichte fails to justify the French Revolution, understood as the act of a sovereign people, which implies the idea of a unified whole. See Philonenko (1968, 204–5). 4. For more on this theory of property, see James (2011) and Merle (2016). 5. There are other important differences, including ones that relate to Kant’s identification of a republican constitution as a condition of global peace. By this he means a constitution founded on the following three principles that can be thought to follow from “the idea of an original contract”: the freedom of the members of a society; the dependence of all of them as subjects upon a single common legislation; and the law of their equality as citizens of the same state (TPP, 8: 349–50). A republican constitution also requires the separation of the executive power from the legislative power, so as to avoid the despotism that Kant associates with a “democratic” constitution (TPP, 8: 352). The pacific nature allegedly inherent to republics combined with their commercial interests can, in fact, be regarded as sufficient in Kant’s view to guarantee global peace in a world consisting of commercial republics, making any kind of external enforcement mechanism of international legal norms unnecessary. See Maliks (2014, 156ff, 164–5). Although Fichte also seeks to found the state on the three principles mentioned above and explicitly appeals to the idea of an original contract in his Foundations of Natural Right, he rejects the claim that the existence of commercial relations between republics would be favorable to global peace. 6. The way in which states here appear to form a predatory coalition may be thought to require that they limit their attempts to expand in relation to one another, thereby promising to establish a condition of peace at least among themselves. A coalition of this kind would, however, be arguably too unstable to guarantee a condition of lasting peace between them. 7. Fichte himself claims that it is only by invoking this drive and the way of thinking characteristic of it that we can explain certain key historical events and phenomena (SE, 180; GA I/5: 175). 8. In connection with a state’s need to attain its natural borders, Fichte uses language similar in kind to that which he uses in relation to the natural drive. The natural drive is associated with a longing, which Fichte identifies with the feeling of a need: “We feel that something—we know not what—is missing” (SE, 119; GA I/5: 120). Similarly, the rulers of a state that has yet to attain its

226  David James natural borders “will dimly feel that they are missing something” (CCS, 170; GA I/7: 117). 9. This would broadly correspond to Hobbes’s notion of the “diffidence of one another,” which he identifies as one of the principal sources of conflict in the state of nature. Fear of the harm that others may inflict upon one means that it is reasonable to seek to master others by force or fraud with the aim of achieving security. Although instituting a sovereign may remove this fear, it persists at the level of international relations, where individual sovereigns “are in continuall jealousies, and in the state and posture of Gladiators; having their weapons pointing, and their eyes fixed on one another” (Hobbes 2012, chap. XIII). In The Fundamental Characteristics of the Present Age, Fichte paints a similar picture of international relations when he describes the position of a previously weaker state that has now become powerful, thereby upsetting the existing balance of power, in the following way: [T]he newcomer excites the jealousy and distrust of other states more strongly than those powers with which they are already familiar. It must henceforth be always on its guard, keep the available forces of the state constantly ready and be prepared [to attack], and leave no means unemployed to add at least to its internal strength when no favourable opportunity presents itself for outward expansion (CPA, 233–4; GA I/8: 359; translation modified). 10. I would like to thank Jeffrey Church for his helpful comments on the penultimate version of this essay.

References Beiser, Frederick (2016). “Fichte and the French Revolution.” In David James and Günter Zöller (eds.), The Cambridge Companion to Fichte. Cambridge: Cambridge University Press, 38–64. Buhr, Manfred (1991). “Die Philosophie Johann Gottlieb Fichtes und die Französische Revolution.” In Manfred Buhr and Domenico Losurdo (eds.), Fichte—die Französische Revolution und das Ideal vom ewigen Frieden. Berlin: Akademie Verlag, 104–17. Hobbes, Thomas (2012). Leviathan, Volume 2: The English and Latin Texts (i). Ed. by Noel Malcolm. Oxford: Clarendon Press. James, David (2011). Fichte’s Social and Political Philosophy: Property and Virtue. Cambridge: Cambridge University Press. Maliks, Reidar (2014). Kant’s Politics in Context. Oxford: Oxford University Press. Merle, Jean-Christophe (2016). “Fichte’s Political Economy and his Theory of Property.” In David James and Günter Zöller (eds.), The Cambridge Companion to Fichte. Cambridge: Cambridge University Press, 199–221. Nakhimovsky, Isaac (2011). The Closed Commercial State: Perpetual Peace and Commercial Society from Rousseau to Fichte. Princeton, NJ: Princeton University Press. Philonenko, Alexis (1968). Théorie et praxis dans la pensée morale et politique de Kant et de Fichte en 1793. Paris: Vrin.

Index

Achenwall, G. 147, 163n1 acting 2, 74, 92, 106, 130n26, 161, 182, 221; agency 34, 201 activity 15–16, 100n2, 185–8, 192–201, 220–1; commercial activity 206, 214, 218; thinking as activity 35, 39, 174, 181 actuality 11, 12, 14, 16, 19–25, 27n13, 45, 54–9, 154, 160, 216 Analogies of Experience 56 analytic 13, 15, 43, 158, 189–95, 201 anthropology 5, 61, 86–99, 100n3, 100n4, 100n6, 103–18, 123–8, 128n11, 128n19 Anticipations of Perception 56 a posteriori 50, 54, 88, 91 apperception 13–14, 27n10, 33, 36–7, 41–8, 60–5, 80, 172 application 2, 4, 9–14, 17–25, 26n5, 27n17, 28n22, 28n23, 28n24, 51, 56–9, 66–7, 104–7, 135, 207, 215 a priori 14, 33, 50, 54–64, 68, 79, 87–8, 90–2, 100n2, 104–6, 123–4, 127, 137, 147–9, 164n9, 169, 172, 185, 189, 195 Aristotle 10, 15, 27n17, 96 attitude 40, 138 autonomy 22, 67, 72–4, 78, 135, 139, 143, 170; moral autonomy 208, 210–13, 224n2, 225n3 awareness 17, 19, 154, 210–11; self 185, 194, 200–1 Axioms of Intuition 56 Baumgarten, G. 52, 59, 68n5 body 41, 46, 57, 118–20, 129n22, 131n42; human body 121, 151 boundary 31–2, 39, 125 Bruno, G. 96

capacity 10–23, 26n7, 27n14, 28n19, 28n22, 74, 136, 181, 211, 224n2 categorical imperative 72–81, 139; see also moral law category 42, 51, 65–7; category of causality 64; modal categories 55, 57–8, 69n20 causality 21–2, 72, 222; category of 64; law of 118, 122, 127; noumenal 74, 78; principle of 55–8 certainty 56, 66, 187, 201; Cartesian 44, 46; self 183–4, 191–2 chance 38, 79 choice 11, 118, 139, 148–9, 153–6, 181, 214; power of 20, 24 church 137, 143–4, 207 cleverness 104, 110 cognition 10, 14, 17, 27n13, 32, 35–6, 38, 48n6, 58, 64–7, 99, 107, 129n13, 138, 173–4, 181, 185–200; a priori 54–5, 87; cognizer 45–8; practical 20–5, 72–9 community 21–5, 29n29, 110 concept see notion condition 4, 27n12, 39, 68n11, 148, 164n12, 189–94; formal 55–64, 122, 189; of the highest good 10, 62; human 98, 121; material and sensible 11, 15, 18–23, 169–71; of peace 151–3, 215–17, 223–4, 225n5, 225n6; of the possibility of experience 50–5, 64, 125; practical 24; unconditioned 53, 64–8, 76, 135, 184, 187–8, 195, 211–13, 222 conduct 21–5, 29n29, 135–6, 140, 175, 221 consciousness 9–13, 16–17, 23, 25, 27n11, 27n14, 32–6, 42, 81, 140, 168, 172–3, 179, 185, 200, 218–22;

228 Index of the categorical imperative 76; combination 46–8; diversity in sensible conditions 18–21; fact of 188, 192, 197; identity of 195; moral 143, 211; thinking 42 contingent 2, 10–11, 15, 36, 43, 52, 55–62, 67–8, 90, 114 contradiction 7, 25n2, 57, 61–2, 73, 137, 180, 198–201; principle of 51, 191 critical philosophy 2, 9, 53, 86–99, 100n4, 104, 178–9, 189–90 Critique of Pure Reason 13, 50, 52, 55, 79, 89–92, 111, 117, 155, 179, 189–95 Crusius, C. A. 68 deduction 13, 26n7, 33, 35, 64–5, 189, 192, 195, 200, 203n15 Descartes, R. 38, 108 desire 7–8, 26n7, 27n17, 28n24, 29n32, 74, 104, 154–5, 158; for security 213, 216, 218, 221, 224; sensible 20, 23–4 determination 10–12, 15, 20, 22, 24, 35, 66–7, 75–6, 139, 158, 174, 194, 199, 219; of experience 54, 145; moral 143; physical 72; power of judgment 78 dialectical ideas 64 disposition 1, 3, 24, 27n17, 94, 196, 200; of benevolence 150; moral 77 divisions of philosophy 9 duty 74–8, 81, 105, 116, 135, 138, 140–1, 149, 154, 158 embodiment 2, 81, 219 empirical 15, 25, 26n7, 50–6, 63, 66–7, 83n19, 86–8, 100n2, 104–6, 125–6, 130n26, 137, 144–5, 145n1, 149, 164n9, 186, 192, 212; anthropology 5, 86, 90–1, 104; judgment 43; religion 137; self 35; thinking 59; use of practical reason 156–62 empiricism 51 essential 52, 71, 80, 82, 92, 158, 167, 174, 185, 218 ethics see practical philosophy experience 20–3, 29n29, 29n30, 36, 59, 75–8, 88, 168, 173, 178, 194–5, 197; conditions of 55, 60–8, 68n11, 190, 192; human 185–6, 199; judgments of 17; objective

118, 125; objects of 2, 4, 51, 56–7; possibility of 50–5, 59–63; sense of 7; supernatural 137 faculty 14, 26n7, 27n12, 27n17, 45, 47, 83n7, 197; of judgment 72–3, 78, 169 feelings 5, 71–81, 125, 129n15, 149 foundation 71, 98, 136, 164n8, 167, 179–201, 203n9, 203n12, 203n14, 214; of justice 159; normative 6, 149 freedom 1–7, 24, 74, 92–3, 100n6, 122, 148–63, 164n8, 169–75, 176n5, 200, 219–22, 224n2; laws of 10, 92, 104, 106, 149; political 211; practical 72; science of 173; state of 122 French Revolution 2, 7, 206–13, 224n2, 225n3 Gassendi, P. 96 geometry 186–7, 191 God 29n29, 74, 97, 100n5, 136–45, 145n1, 186; proof of existence 52, 170 government 6, 152–63 Haller, A. 53 happiness 6, 20–5, 78, 136, 148–63, 164n7; desire for 104, 106 highest good 9, 20, 22–5, 62 history 5–6, 110–11, 114, 127, 206, 213, 215, 218 human being 86, 94, 98–9, 100n2, 104, 112, 141 humanity 89, 97–9, 113, 141, 155, 164n8, 208, 213, 224n2 human reason 1, 10, 26n4, 28n25, 97–8, 122–3, 174 Hume, D. 6, 17–18, 28n19, 28n20, 94, 98–9, 131n47, 146–63, 163n1, 163n3, 163n4, 163n5, 163n7 Husserl, E. 75–6, 179 Hutcheson, J. M. 94, 145 hypothetical 53, 55, 61, 104; imperatives 104, 139; of possible worlds 96, 97; of pre-established harmony 36, 38 identity 65–6, 190–1, 195, 219; of apperception 42–3; of reason 3, 10–13, 25, 28n28, 29n29; of subject and object 21–2

Index  229 individual 25, 63, 77, 107, 151–6, 159, 188, 194, 221–4, 225n3, 226n9; civil 208, 211–14; individuation 39; thinker 40 individualist 92, 213 inner sense 28n24, 35, 47, 66; and outer 126, 137 intelligible 23–5, 53–5, 58 intuition 20, 28n23, 33–8, 43–8, 54–7, 60–8, 68n11, 173–4, 185; a priori 123–4; moral 74–6, 79 judgment 13–23, 27n11, 27n12, 27n14, 28n28, 43–8, 97, 114, 121, 129n14, 169; power of 78–80 juridical laws 149 justice 6, 50, 142, 147–63, 163n5 knowledge 6–7, 129n13, 129n17, 168–75, 192, 199–201, 203n15; a priori 54, 61–2; argument 76–9; general and local 111–16, 126–7, 130n26; historical 88, 97; human 181–4; material and formal 10–25, 26n4, 26n5, 26n7, 27n17, 29n29, 29n30; moral 79; objective 108; pragmatic 104, 109; spheres of 168 knowledge of the world 6, 103, 107–17, 130n26 Lambert, J. H. 96 language 52, 137, 216, 224, 225n8; linguistic 34, 39, 97 law 104–6, 135, 148–9, 155–7, 164n8, 173–4; character of 17; civil 160–2, 211, 214; criminal 154; of freedom 10, 92, 104, 153; of justice 6; of logic 191–6; moral 5, 26n7, 28n25, 62, 72–8, 81, 82n12, 82n13, 138, 211–12, 219–24, 225n3, 225n5; of nature 5, 10, 36–8, 104; objective 75; religious 6; statutory religious 138–9, 143–5 legislative 7, 10, 160–1, 211, 225n5; form 17, 21, 29n28; religious 144–5; self 22–3, 170, 219 Leibniz, G. W. 36, 118–19, 130n30 Lichtenberg, G. C. 2, 31–48 life 26n7, 28n24, 29n32, 46, 92, 109–10, 114, 130n26, 144, 173–5; economic 214; good 135–6, 140; inner 77, 81

limitation of knowledge 23–4, 26n7, 68n11, 169–71, 194–6, 199–201, 222–4; of benevolence 148–50; of government 159–62; of human reason 122; of moral 73; of sensibility 137; social and political 110–12; unlimited 208–10, 222 Locke, J. 59, 96, 147, 161 logic 10–13, 18, 22, 94, 181, 184, 202n4; general 129n23, 191–6, 203n14; transcendental 51, 62, 203n13; universal 13 manifold 56, 60, 66, 195–7, 200; of intuitions 33, 37–8, 42, 47, 67, 118 mathematics 73, 98–9, 120, 183–8, 202n2; arithmetic 56; geometrical 56, 59, 182–7, 201, 202n3, 202n5, 202n6; principles 56–9 Meditations on First Philosophy 31 metaphysics 1–3, 59, 87–91, 95, 105, 123, 169–75 metaphysics of morals 1, 5, 77–8, 104–5, 146n4, 148, 154, 161 method 7, 147, 149, 166, 178–201, 202n1, 202n3, 203n15; of a priori construction 100n2, 100n3 mind 18, 33–7, 55, 62, 81, 166, 179, 198–201; peace of 151; philosophy of 71; states of 79; unity of human 181–2, 185–6 modal/modality 4, 14, 27n13, 50–68, 69n20 monad 36, 118–19, 130n31 Monadology 36 monarchy 2, 7, 206–24, 224n2 morality 22, 72, 81, 97, 104, 135–6, 140–5, 149, 155, 164n9, 219 moral law 5, 62, 72–8, 81, 83n12, 83n13, 104, 138–9, 143, 148–9, 211–12 necessity 4, 14, 17, 23–5, 27n12, 52–8, 67, 69n20, 189, 200, 216–17; absolute 52–3; practical 31 Nicholas of Cusa 96 normative 6, 71, 87, 90, 93, 96, 99, 142–5, 156, 214, 218–22; of action 211–14; ethics 148–9; rational 40 notion 11–12, 15–16, 19, 26n8, 39, 45, 52, 66, 113–14, 149, 199–200; of combination 42, 44; construction 185, 194; of existence 66–7; firstperson 39–42, 197; modal 50–1,

230 Index 57–8, 61–2, 65; of right 153, 214–15; unity of 13–14; world 98 noumena 21, 25, 82n3, 169–70; good will 73–4, 77–8 objective 68n10, 75–7, 83n12, 108, 114, 167–72, 219–20; experience 118, 125–6; objective agreement 21–2; validity 43, 55 objects of experience 2, 4, 56 obligation 5, 105, 139–41, 152–5, 158, 161, 211–13; and duty 72, 75, 81 ontology 7, 26n7, 167–8, 170, 172 paralogisms 65, 67 paternalism 147, 157–9, 164n11 persons 21–5, 116, 149, 154–8 phenomena 6, 21, 23, 62, 77–8, 118, 169–73, 186; religious 138, 141 philosophy in the cosmopolitan sense 98 pleasure and displeasure 20, 26n7, 28n24, 78, 124, 149, 155 political 7, 110–12, 221–2, 224n2; communities 25; freedom 211; organization 206–8, 212–16; philosophy 72, 147–8, 163n1 possibility 23–5, 45–6, 69n20, 71–2, 189, 216; of experience 50–9, 62–8, 68n10, 125; of friendship 105; of highest good 9; of human thought 37–8; of knowledge 6, 111, 169; of matter’s compounding 121, 197; of practical knowledge 12, 18, 170 postulates 170, 176n4, 184, 187; of pure practical reason 24–5, 29n29 Postulates of Empirical Thought 51–67 power 12, 22, 24, 28n19, 140, 158–61, 220, 226n9; of choice 24; of commerce 216–18, 222–4; in Europe 207–12; of mind 18, 192, 197; monarch 213–15, 225n2, 225n5; over nature 220–1; of sense perception 20 power of judgment 13–14, 78–9 practical philosophy 5–7, 9–10, 17, 23, 26n7, 62, 71–2, 103–4, 147–8, 169–70, 206–7, 218; applied ethics 72–4

practical reason 1–6, 9–12, 17–18, 22–5, 29n31, 29n32, 55, 72–82, 135–45, 148, 156, 170 pre-established harmony 36–9 principles 6–7, 56–63, 67–8, 87–8, 105, 118, 147–9, 207, 225n5; of being 169; categorial 43, 46; foundational 178–201; of knowledge 10, 26n7; material 16, 21; of right 214–15; reason 17; transcendental 55 property 151, 155–60, 214, 218, 225n4 psychology 26n7, 35, 79, 113, 124, 156, 162 pure reason 17, 23–5, 27n17, 36, 78, 88, 144, 147 qualia 77–8, 83n9 reality 20, 22, 54, 65, 69n20, 76, 121, 167–8, 198–9; external 188; factual 186; idealistic 172–6; institutional 88, 91; nature of 97, 119 reason 26n7, 27n12, 29n30, 29n31, 29n32, 39–40, 43, 52–3, 92–3, 129n17, 156; antinomies of 200; dialectical 62; fact of 76; praxis of 174; principle of sufficient 191; religion of 136–8, 141, 143, 145; theoretical 13–17, 27n14; unity of 3–4, 9, 26n3 receptivity 15, 20, 63, 136–8 reconciliation 9, 197 regulative principles 56, 120, 123, 131n42 Reinhold, K. 181, 188, 202n4 relation 11–12, 57, 63–4, 68, 106, 115, 119–21, 154; apperception 46–7; between I and Not-I 197–201; of inner and outer sense 127; mutually supplementary 86, 91–3; to objects 14, 19; subject object 171–3 religion 6, 129n19, 135–45, 206, 216 representation 18, 24, 26n7, 27n11, 27n14, 31–8, 41–8, 63, 73, 79–80, 118, 122, 196, 200, 218; aggregate of 15; receptive 20; simple 66; system of 178, 192 respect 75–8, 158 right 6, 135, 144, 147, 153–8, 160–3, 164n8, 164n12, 174–5, 224n2;

Index  231 inalienable 212; natural 207, 214; public 161–3, 213–15 Rousseau, J. 94, 121, 131n47, 211, 225n3 Schelling, F. W. J. 172, 175 schema 29n29, 29n30, 66–7, 103–5 science 7, 12, 16, 103–4, 129n17, 129n23, 187, 202n3; anthropology 87, 90, 94–9, 106–7, 112–15, 126–8; formal 114–15, 117; of knowing 168–76; natural 118–22; philosophical 182, 207, 218; scientific employment 10, 18, 186, 193–5 self-consciousness 15–18, 20–2, 31–5, 41–3, 45, 47–8, 79–82, 113, 185, 194, 219 self-knowledge 2, 11, 21–2, 113, 129n17 sensible 15, 18–25, 28n24, 74, 82n3, 136–7, 141–2, 170, 185; affection 23, 63; desire 24; perception 20–1; senses 13, 32, 44, 122, 137, 141–2 Schulze, Gottlob Ernst 181 site 52, 103–7, 115, 127, 129n22; situated 112, 114, 137 skeptical 2, 18, 33–5, 50, 181, 225n3 soul 1, 32, 38, 129n22, 169–70 spatial 33, 44, 54–6, 60–2, 92, 114, 117–24, 129n22, 131n42, 185 Spinoza, B. 172, 182, 185–7, 201, 202n3 spontaneous 2, 18, 35, 72, 188, 195–6, 200 state 8, 24, 173; of body 58; closed commercial 222–4; government 157–61, 164n7, 164n12, 206–18, 225n5, 225n8, 226n9; of human being 99, 122; mental 77 subject 3–5, 11, 17, 21–2, 26n8, 33–5, 38–9, 41–7, 141, 168, 172, 188, 211–12; to coercion 161; conscious 219–20; moral 72; role in synthetic process 196–7; sensible 19; transcendental 56, 64, 67, 68n10, 169 substance 47, 65, 129n22 supersensible 62, 64, 141, 170 supreme principle(s) (of reason) 60, 62–3, 65

synthetic 7, 18, 172; a priori judgements 169; method 179–81, 187–201; unity of apperception 13–15, 27n10, 47, 60; unity of the manifold 42–4 system 1–3, 90–1, 103–34, 174–5, 178–201; of commerce 217–18; Fichte 207; of freedom 7, 170, 207; political 148; of religion 139; sensory 47; unity of 16–17 teleology 29n32, 87, 90, 174 theoretical reason 9, 13–17, 24–5, 26n5, 27n14 thinking 1–3, 12, 15, 34–48, 79–82, 83n9, 83n21, 117, 129n23, 166–8, 172–4; dialectical 180; ego 65–71; empirical 59; moral 221–4; scientific 183 time 114; and space 37, 54–6, 60–3, 68n10, 69n20, 121, 127, 137, 145 transcendent 4, 7, 13, 25, 35, 41, 46, 50–68, 68n10, 69n20, 91–2, 166–76, 178–201, 207; anthropology 87, 111; argument 39; discourse 50; idealism 4, 32, 178–82; practical freedom 72; psychology 8 transcendental deduction see deduction transcendental philosophy 1–2, 4–5, 6, 8, 13, 26n7, 51, 67, 87, 91, 166–76, 176n7, 179–81, 189–91, 207 understanding 7, 11–19, 27n10, 28n22, 37–42, 51–5, 60–2, 66–8, 74, 78–9, 117, 129n17, 137, 173, 179–81 unity 18, 24–5, 43–4, 46, 114, 127–8, 159, 213; of apperception 27n10, 37, 42–6, 60; of concept 13–14; of consciousness 33, 27n11; disunited 167; of I and Not-I 194–5; of judgment 15; of knowledge 15, 22, 27n17; of the manifold 42; primordial 181, 188, 193; of reason 4, 9, 11, 17, 26n3; of system 16; of thought 15, 18, 20, 46–8; transcendental 172–3; unity of the I think 64–5

232 Index universality 16–17, 21, 73, 76, 90, 189 utility 6, 128, 147–53, 162, 184 virtue 24–5, 105; of benevolence 150, 158; duty of 138, 154, 158; of justice 148–50 volition 33, 73, 125, 129n15, 208, 215; good 73–5

Weltkenntniß see knowledge of the world Wolff, C. 32, 48n5, 147, 157, 184, 202n3 Zöller, G. 4, 6, 31–6, 39–42, 46–7, 50, 86, 90–3, 99, 127, 166–71, 175