The Highest Good in Kant’s Philosophy 9783110369007, 9783110374285

Table of contents :
Contents
Acknowledgments
Introduction
Abbreviations and Methods of Reference
Notes on Contributors
I. The Concept of the Highest Good and its Place in Kant’s Moral Theory
The Highest Good and the Notion of the Good as Object of Pure Practical Reason
Kant on ‘Good’, the Good, and the Duty to Promote the Highest Good
Kant on Happiness and the Duty to Promote the Highest Good
“Mixtum Compositum”: On the Persistence of Kant’s Dualism in the Doctrine of the Highest Good
The Determination of the Concept of the Highest Good
II. Kant’s Moral Arguments and the Postulates of Pure Practical Reason
God, the Highest Good, and the Rationality of Faith: Reflections on Kant’s Moral Proof of the Existence of God
Kant on “Moral Arguments”: What Does the Objectivity of a Postulate of Pure Practical Reason Consist In?
Kant, Mendelssohn, and Immortality
Life without Death: Why Kantian Agents Are Committed to the Belief in Their Own Immortality
III. Epistemology, Science, and Metaphysics
Kant on Opinion, Belief, and Knowledge
Must We Believe in the Realizability of Our Ends? On a Premise of Kant’s Argument for the Postulates of Pure Practical Reason
Applying the Concept of the Good: The Final End and the Highest Good in Kant’s Third Critique
“The supersensible … in us, above us and after us”: The Critical Conception of the Highest Good in Kant’s Practico-Dogmatic Metaphysics
Index of Names
Subject Index

Citation preview

The Highest Good in Kant’s Philosophy

The Highest Good in Kant’s Philosophy

Edited by Thomas Höwing

Printed with financial support from the Deutsche Forschungsgemeinschaft (DFG).

ISBN 978-3-11-037428-5 e-ISBN (PDF) 978-3-11-036900-7 e-ISBN (EPUB) 978-3-11-039274-6 Library of Congress Cataloging-in-Publication Data A CIP catalog record for this book has been applied for at the Library of Congress. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.dnb.de. © 2016 Walter de Gruyter GmbH, Berlin/Boston Cover illustration: Ivary/Thinkstock Printing and binding: CPI books GmbH, Leck ♾ Printed on acid-free paper Printed in Germany www.degruyter.com

Acknowledgments This book is the product of a conference that was organized as part of a research project funded by the Deutsche Forschungsgemeinschaft (German Research Foundation, DFG). By providing financial support for the conference, held in September 2013, the DFG helped to bring together an international group of Kant experts to reassess Kant’s account of the highest good from a philosophical perspective. Additional funding was provided by the Freunde und Förderer of Goethe University, Frankfurt am Main. The conference was planned and organized by the three members of the project, Florian Marwede, Marcus Willaschek, and myself. I should like to thank Marcus Willaschek for his constant encouragement, generous support and helpful advice concerning this volume. Thanks are also due to colleagues and staff at the Philosophy Department at Goethe University, Frankfurt am Main, where the conference took place. My special thanks go to Carolyn Benson for correcting the English of the non-native speakers’ contributions, to Luca Essig for helping me to edit the papers, and to Marion Seiche for her help in organizing the conference.

Contents Acknowledgments Introduction

V

1

Abbreviations and Methods of Reference Notes on Contributors

9

11

I The Concept of the Highest Good and its Place in Kant’s Moral Theory Federica Basaglia The Highest Good and the Notion of the Good as Object of Pure Practical 17 Reason Pauline Kleingeld Kant on ‘Good’, the Good, and the Duty to Promote the Highest Good Florian Marwede Kant on Happiness and the Duty to Promote the Highest Good

33

51

Birgit Recki “Mixtum Compositum”: On the Persistence of Kant’s Dualism in the Doctrine of the Highest Good 71 Stephen Engstrom The Determination of the Concept of the Highest Good

89

II Kant’s Moral Arguments and the Postulates of Pure Practical Reason Gabriele Tomasi God, the Highest Good, and the Rationality of Faith: Reflections on Kant’s Moral Proof of the Existence of God 111

VIII

Contents

Stephan Zimmermann Kant on “Moral Arguments”: What Does the Objectivity of a Postulate of Pure 131 Practical Reason Consist In? Paul Guyer Kant, Mendelssohn, and Immortality

157

Jochen Bojanowski Life without Death: Why Kantian Agents Are Committed to the Belief in Their Own Immortality 181

III Epistemology, Science, and Metaphysics Thomas Höwing Kant on Opinion, Belief, and Knowledge

201

Marcus Willaschek Must We Believe in the Realizability of Our Ends? On a Premise of Kant’s Argument for the Postulates of Pure Practical Reason 223 Andrea Marlen Esser Applying the Concept of the Good: The Final End and the Highest Good in 245 Kant’s Third Critique Günter Zöller “The supersensible … in us, above us and after us”: The Critical Conception of the Highest Good in Kant’s Practico-Dogmatic Metaphysics 263 Index of Names Subject Index

281 283

Introduction Kant’s doctrine of the highest good runs counter to a common picture of his moral philosophy. On this picture, Kant is seen as the natural opponent of consequentialist thinkers, such as Bentham and Mill. According to Kant, moral action is not about maximizing happiness but rather about acting in accordance with a formal principle of reason. Moreover, Kant is seen as giving strong philosophical expression to the criticism of religion that was envisaged by secular Enlightenment thought. According to Kant, moral consciousness is independent of religious belief; genuinely moral conduct is determined by rational requirements and not by divine commands. Kant’s doctrine of the highest good apparently relativizes this picture in important ways. For one thing, at the centre of the doctrine lies the view that the final end of all moral conduct, or the highest good, consists not only of moral virtue but also of happiness. This suggests that for Kant, happiness does have a moral standing, even though its pursuit is restricted by a formal law of reason. For another, Kant argues that in pursuing the highest good moral agents commit themselves to the belief that God exists and that there is life after death. In other words, Kant’s doctrine of the highest good entails the view that there is a deep affinity between moral consciousness and a genuinely religious world view. In light of these considerations, it is not surprising that Kant’s doctrine of the highest good has faced a great deal of criticism, especially from those who consider themselves sympathetic to Kant’s moral philosophy. The doctrine, it is widely held, corrupts major insights of Kant’s moral philosophy, not only by reintroducing happiness as an unnecessary reward for moral conduct, but also by burdening moral consciousness with heavy theological baggage. On the other hand, however, the doctrine of the highest good invites us to rethink the somewhat schematic picture that we have of Kant’s moral philosophy. What is more, taking a fresh look at the doctrine may reveal that it touches on a variety of issues of philosophical interest. To cite an example from moral philosophy, at the centre of Kant’s account of the highest good lies the view that an analysis of moral consciousness should not restrict itself to a local perspective on moral agency; it should not merely ask what accounts for the moral goodness of a particular action. Rather, very much like the great ancient philosophers, Kant insists that moral consciousness comes with a global perspective on moral agency. According to Kant, moral agents necessarily form the concept of an overarching end for the sake of which all moral action is undertaken. What is more, they conceive of moral agency as an essentially co-operative enterprise, aiming at the realization of a shared vision of the world. What the doctrine of the

2

Introduction

highest good adds to Kant’s account of moral agency is thus the thought that there is a community of moral agents that transcends the contingent borders of culture and society. Although the doctrine of the highest good touches on important issues in moral philosophy and Kant’s philosophy of religion, it is also of interest to those working on Kant’s metaphysics, his philosophy of history, and his epistemology. Again, to cite just one example, it has already been pointed out that Kant’s so-called ‘moral arguments’, along with his account of the postulates of pure practical reason, come with a much more general account of justification. On this account, we are justified in accepting a proposition on the basis of genuinely practical considerations, such as, for example, considerations about the realizability of the highest good. In other words, Kant’s moral arguments anticipate a more contemporary view in epistemology, namely, the view that justification is not solely a matter of sufficient evidence. The aim of this volume is to take a nuanced perspective on Kant’s doctrine of the highest good – a perspective that not only provides us with a more authentic picture of Kant’s moral philosophy but also highlights the various philosophical problems with which the doctrine is concerned, bringing to light some of the key philosophical insights contained within it. The volume is organized as follows. Chapters 1–5 deal with the place of the highest good in Kant’s moral theory. The contribution by Federica Basaglia (‘The highest good and the notion of good as object of pure practical reason’) explores how Kant’s account of the highest good relates to his more general account of good and evil. In the second Critique, Kant defines good and evil as the only two objects of pure practical reason, and the highest good as the unconditioned totality of the object of pure practical reason. It has often (and rightly) been suggested that there are many similarities between Kant’s notions of the good and the highest good. However, in this chapter Basaglia argues that there are also crucial differences between the two notions concerning both their content and their theoretical role in Kant’s moral philosophy. Kant’s inclusion of happiness in the highest good, in combination with his claim that it is a duty to promote the highest good, is widely seen as inconsistent. In the second chapter (‘Kant on ‘good’, the good, and the duty to promote the highest good’), Pauline Kleingeld argues that there is a valid argument, based on premises Kant clearly endorses, in defence of his thesis that it is a duty to promote the highest good. She first examines why Kant includes happiness in the highest good at all. On the basis of a discussion of Kant’s distinction between ‘good’ and ‘pleasant’, and in light of his methodological comments in the second chapter of the Critique of Practical Reason, Kleingeld explains how Kant’s conception of the good informs his conception of the highest good. She then argues

Introduction

3

that Kant’s inclusion of happiness in the highest good should be understood in light of his claim that it is a duty to promote the happiness of others. In the final section of her essay, Kleingeld reconstructs Kant’s argument for the claim that it is a duty to promote the highest good and explains in what sense this duty goes beyond observance of the categorical imperative. Kant’s claim that it is a duty to promote the highest good is also discussed by Florian Marwede (‘Kant on happiness and the duty to promote the highest good’). As Marwede points out, one problem with this claim is that it seems to entail the view that we have a direct duty to promote our own happiness, which is explicitly denied by Kant. However, he argues that Kant’s claim merely entails the view that we should promote our own happiness as part of universal happiness. In other words, we should transform the pursuit of our own personal happiness into a pursuit of universal happiness. As Marwede suggests, this is compatible with Kant’s claim that there cannot be a direct duty to promote happiness, since the latter would concern the agent’s own happiness in abstraction from the happiness of others. Birgit Recki considers Kant’s inclusion of happiness in the highest good from a different perspective (‘‘Mixtum compositum’: On the persistence of Kant’s dualism in the doctrine of the highest good’). As Recki argues, Kant’s account of the highest good contains two dualisms. On the one hand, Kant rejects Epicurean and Stoic accounts of the highest good on the grounds that both conceive of the relation between virtue and happiness in terms of an analytic connection or a relation of conceptual entailment. Instead, according to Kant, when it comes to the highest good we should think of the relation between virtue and happiness as a synthetic connection of cause and effect. As Recki suggests, this dualism in Kant’s account of the highest good takes seriously the legitimacy of the human desire for happiness. On the other hand, however, Kant’s account also comes along with the claim that the synthetic connection between virtue and happiness cannot be realized in this world and must instead be realized in another, noumenal world. As Recki argues, this latter dualism is more problematic since it relegates the moral agent’s legitimate claim to happiness to an afterlife in an intelligible world. The problem of dualism also lies at the centre of Stephen Engstrom’s ‘The determination of the concept of the highest good’. Kant criticizes ancient philosophers for heteronomously basing their moral investigations on the problem of the highest good, yet his own account of the highest good as a distribution of happiness in proportion to virtue can bolster the impression that his alternative approach, which requires that morality’s principle be investigated prior to the determination of the concept of the good, results in a dualistic separation of morality from the good. However, as Engstrom argues, closer scrutiny of Kant’s engagement with the problem of the highest good and with the ancients’ doctrines

4

Introduction

can free his alternative from the appearance of dualism and reveal a deep kinship it shares with the best of the ancient accounts. According to Engstrom, attention to the famous argument concerning the good will with which Kant begins his ethics can provide a context of interpretation within which his investigation of morality’s principle can be recognized as an analysis of the act of knowing the good. When his investigation is so understood, his doctrine of the highest good can be seen to be, not an attempt to secure an eschatological closure of a cleft opened up by a dualistic analysis, but rather a hylomorphic account according to which the highest good’s two elements, virtue and happiness, are united as form and matter of good conduct. Chapters 6–9 deal with Kant’s moral arguments for the existence of God and the immortality of the soul. Kant’s so-called ‘moral proof’ of the existence of God is often discussed in connection with his account of the postulates in the Critique of Practical Reason. What is less well known is that Kant offers an intricate analysis of his proof at the end of the Critique of Judgment. There, it is related to the so-called physico-theological proof, or to what nowadays is called the ‘argument from design’. As Gabriele Tomasi argues (‘God, the highest good, and the rationality of faith: Reflections on Kant’s moral proof of the existence of God’), both arguments ultimately aim at a vision of the world as purposively structured by a highest intelligence. However, Tomasi also points to the crucial differences that Kant sees between the two arguments. For one thing, although the physico-theological proof leads to the notion of an intelligent world cause, it cannot raise this notion up to the concept of a personal God. For another, while the physico-teleological proof fails to provide knowledge of God, the moral argument does not even aim at such knowledge. Rather, it aims at what Kant calls ‘Belief’ or ‘Faith’ (Glauben), which, according to Tomasi, is not so much a doxastic attitude as an essential ingredient of a moral agent’s commitment to the moral law. At the centre of Stephan Zimmermann’s contribution (‘Kant on ‘moral arguments’: What does the objectivity of a postulate of pure practical reason consist in?’) lies Kant’s claim that, although the moral arguments do not establish knowledge of God and immortality, they still provide an “increment” to pure theoretical reason. As Zimmerman argues, this increment consists in the gaining of objectivity on the part of the corresponding ideas of God and the human soul. However, Zimmermann also suggests that the objectivity in question must be distinguished from the objective reality that Kant attributes to empirical concepts and to schematized categories. Rather, according to Zimmermann, to postulate these objects is to represent their existence indirectly, i.e. by means of an analogy with empirical objects. Paul Guyer’s contribution addresses the historical background and the development of Kant’s argument for the immortality of the soul (‘Kant, Mendelssohn,

Introduction

5

and immortality’). In his Phaedon, Moses Mendelssohn argues that human beings must be granted immortality because they would need that to fully develop their capacities for both morality and happiness, and God would not have granted them such capacities only to deny them the possibility of fully developing them. According to Guyer, Kant’s expositions of the postulate of immortality in the Critique of Pure Reason and the Critique of Practical Reason can be interpreted as versions of Mendelssohn’s new argument, the first Critique grounding the postulate of immortality on the conditions for perfecting our happiness and the second on the conditions for perfecting our virtue. However, Guyer argues that Kant’s argument in Religion within the Boundaries of Mere Reason undercuts at least the second of these arguments, with its thesis that moral conversion and thus the perfection of virtue is always open to human beings because of their noumenal freedom, and that any idea of our own immortality as necessary for the perfection of virtue is only an aesthetic image, not a postulate of pure practical reason. According to Guyer, Kant accordingly downplays the postulate of immortality in much of his writings in the final decade of his career, indeed even beginning with the third Critique. Kant’s account of the immortality of the soul stands opposed to a view that is widely shared by contemporary moral philosophers. On this view, if our lives were to continue endlessly, they would become tedious and boring. The good life is a life that comes to an end. Kant, however, approaches the topic from a somewhat different angle. His question is not whether immortality would be conducive to our happiness. Rather, we are committed to believing in immortality because we have an obligation to become morally perfect. In his contribution, Jochen Bojanowski reassesses and defends Kant’s argument (‘Life without death: Why Kantian agents are committed to the belief in their own immortality’). Bojanowski first shows why moral perfection, even in the case of human beings, requires not just virtue as a “moral disposition in battle” but the idea of holiness. In addition, he explains the connection between the moral requirement to perfect our moral disposition and our belief in the immortality of the soul. According to Bojanowski, Kant’s idea is not that we have direct intuitive knowledge of our immortality; it is rather that the belief in our immortality makes possible an action (perfecting ourselves) which we practically cognize as obligatory. Finally, Bojanowski considers why Kant thinks that belief in our own immortality, although it does not amount to knowledge, still counts for Kant as a cognition, at least in some sense. Chapters 10–13 focus on the epistemological aspects of Kant’s doctrine of the highest good, as well as its relation to science and metaphysics. It has often been observed that Kant’s moral arguments are based on a more general account of doxastic attitudes and their roles in rational agency. Unfortunately,

6

Introduction

however, this account is hardly ever made explicit by Kant. In his ‘Kant on opinion, belief, and knowledge’, Thomas Höwing explores a distinction that lies at the centre of this account, namely, the distinction between opining (Meinen), Believing (Glauben), and knowing (Wissen). In presenting the moral arguments, Kant often points out that, unlike opinion and knowledge, Belief has a unique feature – it requires non-epistemic or ‘practical’ justification. Yet Kant’s official formulation of the tripartite distinction seems to run counter to this claim. It describes Belief in terms of a set of two features, each of which also pertains to either opining or knowing. Höwing’s aim is to provide a novel interpretation of Kant’s distinction which dissolves this exegetical puzzle. Marcus Willaschek’s paper (‘Must we believe in the realizability of our ends? On a premise of Kant’s argument for the postulates of pure practical reason’) focuses on another assumption that serves as a premise of Kant’s moral arguments, namely, what Willaschek calls the realizability principle. According to the realizability principle, it is a condition of rational agency that one believes that the end one pursues can be realized by one’s own actions. Willaschek discusses the role of the realizability principle in Kant’s arguments for the postulates and asks whether there is a version of this principle that is both sufficient for Kant’s argument and independently plausible. In addition, Willaschek offers a reconstruction of Kant’s argument for the postulates in the Critique of Practical Reason, which allows us to see in more detail how that argument depends on some version of the realizability principle. He then discusses three objections to the realizability principle, one of which requires a modification and weakening of it. However, Willaschek argues that there is a version of this principle that remains unaffected by these objections and is still sufficient as a basis for Kant’s argument for the postulates. The final two chapters concern the way in which the doctrine of the highest good is related to science and metaphysics. As the final end of morality, the highest good gains new systematic significance in the Critique of Judgment, where it is integrated into a more general account of purposiveness and reflective judgment. The aim of Andrea Marlen Esser’s contribution (‘Applying the concept of the good: The final end and the highest good in Kant’s third Critique’) is to explore this systematic integration in more detail. She clarifies Kant’s critical engagement with Spinoza, and she discusses three distinctive features of the doctrine of the highest good, as it is presented in the Critique of Judgment: (i) the connection that Kant establishes between the doctrine of the highest good and the purposiveness of nature, which is important for the intended completion of the critical system of philosophy; (ii) the logical relation established between the doctrine of the highest good and the theory of reflective judgment; and (iii)

Introduction

7

the critical implications for religion which emerge from Kant’s engagement with Spinoza and from his profound reservations regarding physico-theology. Finally, the contribution by Günter Zöller (‘‘The supersensible … in us, above us and after us’: The critical conception of the highest good in Kant’s practicodogmatic metaphysics’) presents and analyses Kant’s conception of the highest good in his late fragmentary work, Prize Essay on the Advances of Metaphysics (1793). Zöller’s contribution focuses on the teleological nature of the unity of theoretical and practical reason and on the practically based validation of theoretical concepts of reason in Kant. Zöller begins by detailing the change in teleological thinking between Aristotle and Kant. In addition, he explores the extent of the analogy drawn by Kant between theoretical and practical reason, focusing on the complementary relation between theoretical reason as the faculty of determining the object and practical reason as the faculty of determining the will. Finally, he discusses Kant’s teleological conception of the unity of theoretical and practical reason as well as the integration of the highest good into a critically warranted practico-dogmatic metaphysics.

Abbreviations and Methods of Reference Except for references to the first Critique, all references to Kant’s writings are to the German original of Kant’s gesammelte Schriften, Deutsche Akademie der Wissenschaften (formerly: Preußische Akademie der Wissenschaften), Berlin: Walter de Gruyter 1900 ff. References to the Academy Edition list the relevant abbreviation in italics, followed by volume, page, and, if necessary, line number(s) (e. g. KU 5:222.21). References to the Critique of Pure Reason refer to the A- and B-pages of the first and second editions (e. g. KrV A33/B49). Unless indicated otherwise, translations are taken from The Cambridge Edition of the Works of Immanuel Kant (ed. Paul Guyer and Allen W. Wood).

Abbreviations Anth

Anthropologie in pragmatischer Hinsicht. – Anthropology from a Pragmatic Point of View. Bemerkungen Bemerkungen zu den Beobachtungen über das Gefühl des Schönen und Erhabenen. – Kant’s Remarks on the Observations on the Feeling of the Beautiful and Sublime. Beobachtungen Beobachtungen über das Gefühl des Schönen und Erhabenen. – Observations on the Feeling of the Beautiful and Sublime. EEKU Erste Einleitung zur Kritik der Urteilskraft. – First Introduction to the Critique of Judgment. Ende Das Ende aller Dinge. – The End of All Things. Fortschritte Welches sind die wirklichen Fortschritte, die die Metaphysik seit Leibnizens und Wolffs Zeiten in Deutschland gemacht hat? – What Real Progress Has Metaphysics Made in Germany since the Time of Leibniz and Wolff? Frieden Zum ewigen Frieden. – Toward Perpetual Peace. Gemeinspruch Über den Gemeinspruch: Das mag in der Theorie richtig sein, taugt aber nicht für die Praxis. – On the Common Saying: That May Be True in Theory, But Is of No Use in Practice. GMS Grundlegung zur Metaphysik der Sitten. – Groundwork of the Metaphysics of Morals. Idee Idee zu einer allgemeinen Geschichte in weltbürgerlicher Absicht. – Idea for a Universal History with a Cosmopolitan Aim. KpV Kritik der praktischen Vernunft. – Critique of Practical Reason. KrV Kritik der reinen Vernunft. – Critique of Pure Reason.

10

Abbreviations and Methods of Reference

KU Log-Busolt Log-DW Logik MAN Met-L1 Mo-Collins Mo-Kaehler Mo-Mrong Mo-Mrong II MS Op. Post. Prol

Refl RGV Theodizee Ton Träume Verkündigung

Kritik der Urteilskraft. – Critique of the Power of Judgment. Logik Busolt. – Logic Lectures Busolt. Logik Dohna-Wundlacken. – Logic Lectures Dohna-Wundlacken. Jäsche Logik. – Lectures on Logic, edited by G. B. Jäsche. Metaphysische Anfangsgründe der Naturwissenschaften. – Metaphysical Foundations of Natural Science. Metaphysik L1 (Pölitz). – Metaphysics Lectures Pölitz. Moralphilosophie Collins. – Moral Philosophy Lectures Collins. Immanuel Kant: Vorlesung zur Moralphilosophie, ed. by Werner Stark, Berlin/New York: de Gruyter 2004. Moral Mrongovius. – Moral Philosophy Lectures Mrongovius. Moral Mrongovius II. – Moral Philosophy Mrongovius, Second Set of Notes. Metaphysik der Sitten. – Metaphysics of Morals. Opus Postumum. Prolegomena zu einer jeden künftigen Metaphysik, die als Wissenschaft wird auftreten können. – Prolegomena to Any Future Metaphysics that Shall Come Forth as Scientific. Reflexionen. – Kant’s notes in volumes 14–20, 23 of the Academy Edition. Die Religion innerhalb der Grenzen der bloßen Vernunft. – Religion within the Boundaries of Mere Reason. Über das Misslingen aller philosophischen Versuche in der Theodizee. – On the Miscarriage of All Philosophical Trials in Theodicy. Von einem neuerdings erhobenen vornehmen Ton in der Philosophie. – On a New Superior Tone in Philosophy. Träume eines Geistersehers. – Dreams of a Spirit-Seer. Verkündigung des nahen Abschlusses eines Traktats zum ewigen Frieden in der Philosophie. – Proclamation of the Imminent Conclusion of a Treaty of Perpetual Peace in Philosophy.

Notes on Contributors Federica Basaglia is a Lecturer at the University of Konstanz. She is the author of Libertá e Male Morale nella Critica della Ragion Pratica di Immanuel Kant (2009) and of numerous articles on Kant’s practical philosophy. Jochen Bojanowski is a Senior Research Lecturer at the University of Illinois at Urbana-Champaign. His main research interests lie in Kant, ethics, political philosophy and free will. He is the author of Kants Theorie der Freiheit (2006) and of numerous articles on Kant’s practical philosophy and his theory of freedom. Stephen Engstrom is Professor of Philosophy at the University of Pittsburgh. His main research interests lie in ethics, metaphysics, and modern philosophy (especially Kant). He is co-editor of Aristotle, Kant, and the Stoics: Rethinking Happiness and Duty (1996) and the author of The Form of Practical Knowledge: A Study of the Categorical Imperative (2009). Andrea Marlen Esser is Professor of Practical Philosophy at the University of Jena. She is co-editor of Deutsche Zeitschrift für Philosophie and author of Kunst als Symbol. Die Struktur ästhetischer Reflexion in Kants Theorie des Schönen (1997) and Eine Ethik für Endliche. Kants Tugendlehre in der Gegenwart (2004). In addition, she has written numerous articles on Kant’s ethics and aesthetics. Paul Guyer is Jonathan Nelson Professor of Humanities and Philosophy at Brown University. He is the author, editor, and translator of numerous works on Kant, including Kant and the Claims of Taste (1979), Kant and the Claims of Knowledge (1987), Kant and the Experience of Freedom (1993), Kant on Freedom, Law, and Morality (2000), Kant’s System of Nature and Beauty (2005), Values of Beauty (2005), and Kant (second edition, 2014). His most recent work is the three-volume A History of Modern Aesthetics (2014). Thomas Höwing is a Post-Doctoral Fellow at the University of Frankfurt. He is the author of Praktische Lust. Kant über das Verhältnis von Fühlen, Begehren und praktischer Vernunft (2013) and of articles on Kant’s practical philosophy and his philosophy of mind. Pauline Kleingeld is Professor of Philosophy at the University of Groningen. Her publications focus on Kant and Kantianism, as well as on themes in ethics and political philosophy. She is the author of Kant and Cosmopolitanism: The Philo-

12

Notes on Contributors

sophical Ideal of World Citizenship (2012) and Fortschritt und Vernunft: Zur Geschichtsphilosophie Kants (1995). She is also the editor of Immanuel Kant, ‘Toward Perpetual Peace’ and Other Writings on Politics, Peace, and History (2006). Florian Marwede is a graduate student at the University of Frankfurt. He is currently working on a dissertation on the role of the highest good in Kant’s deontological ethics. Birgit Recki is Professor of Practical Philosophy at the University of Hamburg. Her publications focus on ethics, aesthetics and the philosophy of culture. Her book publications include Ästhetik der Sitten. Die Affinität von ästhetischem Gefühl und praktischer Vernunft bei Kant (2001), Die Vernunft, ihre Natur, ihr Gefühl und der Fortschritt. Aufsätze zu Immanuel Kant (2006), and Cassirer (2013). Gabriele Tomasi is Associate Professor at the University of Padua. His research interests lie in aesthetics, Kant, German Idealism and the philosophy of Ludwig Wittgenstein. His books include Ineffabilità. Logica, etica, senso del mondo nel Tractatus di Wittgenstein (2006) and Un bicchiere con Hume e Kant. Divertissement estetico-metafisico (2010). He is the author of numerous articles in aesthetics and co-editor (with Massimiliano Carrara and Antonio Nunziante) of Individuals, Minds and Bodies: Themes from Leibniz (2004), (with Elisa Caldarola and Davide Quattrocchi) of Wittgenstein, l’estetica e le arti (2013), and (with Alberto L. Siani) of Schiller lettore di Kant (2013). Marcus Willaschek is Professor of Philosophy at the University of Frankfurt. He is co-editor (with Stefano Bacin, Georg Mohr and Jürgen Stolzenberg) of the new three-volume Kant-Lexikon (2015) and author of Praktische Vernunft. Handlungstheorie und Moralbegründung bei Kant (1992) and Der mentale Zugang zur Welt: Realismus, Skeptizismus und Intentionalität (2003). In addition, he has authored numerous articles on Kant’s theoretical and practical philosophy. Stephan Zimmermann is Assistant Professor of Philosophy at the University of Bonn. His research interests lie in German Idealism, twentieth-century German philosophy (Wittgenstein, Luhmann, Habermas), hermeneutics (Heidegger, Gadamer, Dilthey), practical philosophy, and social and political philosophy. He is author of Kants ‘Kategorien der Freiheit’ (2011) and of articles on Kant’s practical philosophy and aesthetics. Günter Zöller is Professor of Philosophy at the University of Munich. His publications, comprising 35 authored, edited and co-edited books and over 300 auth-

Notes on Contributors

13

ored articles in journals, essay collections and reference works, have appeared in sixteen languages worldwide. His recent book publications include Fichte lesen (2013) and Res publica. Plato’s ‘Republic’ in Classical German Philosophy (2015).

I The Concept of the Highest Good and its Place in Kant’s Moral Theory

Federica Basaglia

The Highest Good and the Notion of the Good as Object of Pure Practical Reason In the second chapter of the Analytic in the Critique of Practical Reason, Kant introduces the notion of the “good” as an “object of pure practical reason” (KpV 5:57 f.). In the Dialectic, he defines the “highest good” as the “unconditioned totality of the object of pure practical reason” (KpV 5:108.11 f.) and as “the whole object of pure practical reason” (KpV 5:109.21 f.). These two elements of Kant’s moral theory have traditionally, and rightly, been connected. In this paper, I intend to point out some fundamental differences between these two elements of Kant’s moral philosophy – differences which, in my opinion, can help us to better understand them. In order to do so, it is first of all necessary to understand what an “object of pure practical reason” is. The first and second sections of the paper will serve this purpose: by taking a closer look at Kant’s response to Hermann Andreas Pistorius’s criticism of his conception of a “good will”, which occurs in the Preface to the Critique of Practical Reason, I will try to clarify the topic of the second chapter of the Analytic and the definition of “objects of pure practical reason”. In the third part of the paper, and on the basis of the results of the first and second sections, I will compare the notion of an “object of pure practical reason” with that of the “highest good” in order to underline their essential differences. I will argue that, despite the similarities and systematic connections between these two notions – such as, for example, the fact that they both refer to objects of pure practical reason – there are also crucial differences between them. The notion of the good presented in the Analytic refers exclusively to what can be judged to be good in itself (and hence belongs exclusively to morality), whereas the highest good refers to the realization not only of morality, but also of (empirical) happiness, which, according to the premises of Kant’s moral philosophy, is not part of morality.

18

Federica Basaglia

1 Kant’s Reply to Pistorius in the Preface to the Critique of Practical Reason ¹ In his Preface to the Critique of Practical Reason, Kant refers to an objection from a “truth-loving and acute” critic of his moral philosophy, who complains about the fact that the notion of the “good” is not the foundation of the practical law in Kantian theory (KpV 5:8.25–10.2; cf. Sala 2004: 63; Beck 1995: 27; Bittner/Cramer 1975: 16). The critic Kant is referring to is the German parish priest, theologian and philosopher, Hermann Andreas Pistorius.² In his review of Kant’s Groundwork of the Metaphysics of Morals, Pistorius opposes the Kantian conception, according to which nothing can possibly be called “good” in an absolute and unconditional sense except the “good will”. As a matter of fact, what makes a will good is, according to Kant, not what it might bring about, or its “fitness to attain some proposed end”, but solely the way it wills: the good will is good in itself (GMS 4:393.5–394.31). By contrast, Pistorius holds the view that a coherent and meaningful moral philosophy should instead start with the notion of what has to be considered good: only on the basis of this definition can we judge whether a will is good or not. Pistorius doesn’t mean that “good” is what is commonly accepted to be good: the moral philosopher, in his opinion, has to investigate the reasons on the basis of which something is considered good and must decide whether they are in fact good reasons. What Pistorius finds problematic in Kant’s view is the definition of “good will” as a will that is good independently of its objects: “I wish that the author had chosen to discuss above all the general concept of that which is good, and to determine more closely what he means by this, for we obviously must come to an agreement on this before we make something of the absolute worth of a good will. […] I do not see here how one can accept something as being utterly and in an absolute sense good, or call something good, when in fact it is good for nothing, just as I do not see how one can accept the idea of an absolutely good will, considered merely in itself. The will should be considered as good only with respect to some sort of object, not with respect to its principle or to a law, for the sake of which it acts” (Pistorius, Rezension der Grundlegung zur Metaphysik der Sitten, 27; third emphasis is mine, my translation).³

 This section and the one that follows draw upon material that was published in Basaglia . I would like to thank Thomas Höwing for his helpful comments on this paper.  On Pistorius’s life and works, see Gesang : VII–XLIII.  I thank Michael Walschots and Carolyn Benson for their great help in translating Pistorius’s quotations into English.

The Highest Good and the Notion of the Good as Object of Pure Practical Reason

19

According to Pistorius, if, following Kant, the will were good exclusively in virtue of the principle or law guiding it, the question whether this principle or law is good or bad would remain open: “[…] [T]o establish a will as the good, is it merely sufficient that it act according to some principle or another, or from respect for some law, be it good or evil? Impossible. It must therefore be a good principle, a good law, the following of which makes a will good. The question “what is good?” turns back around, and if we have pushed it back from the will to the law, then we must answer it in a sufficient way here; i. e. we must eventually come to some kind of object or to an ultimate end of the law, and we must avail ourselves of what is material, because for us neither a formal will nor a formal law will suffice” (Pistorius, Rezension der Grundlegung zur Metaphysik der Sitten, 27; my emphasis, my translation).

An analysis of morality, according to Pistorius, should start with an analysis of the concept of “good” (Pistorius, Rezension der Grundlegung zur Metaphysik der Sitten, 33). To do so, the moral philosopher must first consider the question whether anything exists that is universally, without exception, and in every circumstance good for feeling and thinking beings like humans. This something – and not the Kantian good will determined by a merely formal principle – should be called the “highest and absolute good” (Pistorius, Rezension der Grundlegung zur Metaphysik der Sitten, 33). For Pistorius, such an absolute highest good is grounded in the collective nature of human beings and in their universal interest as rational beings, because something can be good for a human being only if it accords with both. Through this notion of the highest good, it is finally possible to develop the highest moral principle and to define a will which acts in accordance with it as good: “If, through this examination, something is discovered that is universally, without exception and in every cirtumstance good for feeling and thinking beings, this somenthing must be called the highest and absolute good. If there is such a highest good, then a collective nature of all rational beings must also exist, in which is grounded a universal interest, for only in accordance with the former and in conformity with the latter can something be good for such a being at all” (Pistorius, Rezension der Grundlegung zur Metaphysik der Sitten, 33, my translation).

Hence, for Pistorius, in order to define “good” – i. e. the concept of “good” – the moral philosopher must first investigate the highest object (or final aim) of human action. Moral investigation should start with an investigation into the highest good for human beings, i. e. that which human beings should realize, or at least try to realize, through their actions. Only after having defined the highest good are we able to establish – in accordance with its definition – whether

20

Federica Basaglia

other things are good as well. In other words, according to Pistorius, something is “good” – a good principle of action or a good will – only in virtue of its relation to the highest good. In the Preface to the Critique of Practical Reason (KpV 5:8.25–10.2), Kant claims to answer this objection in the second chapter of the Analytic of Pure Practical Reason, to which he gives the title “On the Concept of an Object of Pure Practical Reason”. Accordingly, in the following section, I will present my analysis of this chapter, in which Kant defines “good” and “evil” as objects of pure practical reason.

2 The Objects of Pure Practical Reason Before I turn to the interpretation of the second chapter of the Analytic and attempt to understand what “good” and “evil” as objects of pure practical reason might be, it is necessary to clarify what an object of practical reason is for Kant.⁴ In fact, Kant does not deliver any precise definition of an “object of the will” or an “object of practical reason” (the various German expressions Kant uses here being the following: Gegenstand der Handlung, Gegenstand/Objekt des Wollens, Gegenstand/Objekt der praktischen Vernunft and Gegenstand des Willens). In fact, these expressions seem to be equivalent for Kant. They refer generally to a determination of the faculty of desire by empirical inclinations, or to the empirically determined will. In the Groundwork of the Metaphysics of Morals, he explicitly defines an “object of the will” as the motivation of the will (Bewegursache, GMS 4:458.16–18) or its matter (Materie, GMS 4:461.29 f.). Kant explains that we have an interest in the object of an action when the will is determined to action by empirical inclinations. In the case of an empirically determined action, what moves us to act is what we want to accomplish through our action, i. e. our purpose; our interest is a pathological interest in the object of our action stemming from a certain inclination or desire. In this case, practical reason only provides the rule in accordance with which we should act in order to satisfy our need. In the case of moral action, the motive (Beweggrund) of practical reason is not what we want to achieve through the action, but the moral law itself; the interest is a practical interest in the action: we are interested in the action itself, not in its purpose (GMS

 For a more detailed discussion of the literature on the “objects of pure practical reason”, see Basaglia : –.

The Highest Good and the Notion of the Good as Object of Pure Practical Reason

21

4:413.26–414.36). In both cases, I think, it can be held that the matter of the will, its object, is the purpose or end (Zweck) of the action.⁵ In the second Critique, an “object of the will” indicates the “matter” of a practical principle, which can become the ground of determination (Bestimmungsgrund) of the will. Where this occurs, the determination of the will is non-moral and heteronomous. In cases where the will is not determined by the object, the determination is a moral, autonomous one. “If a rational being is to think of his maxims as practical universal laws, he can think of them only as principles that contain the determining ground of the will not by their matter but only by their form. The matter of a practical principle is the object of the will. This is either the determining ground of the will or it is not. If it is the determining ground of the will, then the rule of the will is subject to an empirical condition (to the relation of the determining representation to the feeling of pleasure and displeasure), and so is not a practical law. Now, all that remains of the law if one separates from it everything material, that is, every object of the will (as its determining ground), is the mere form of giving universal law” (KpV 5:27.3–14; first and third emphases are mine)

Other passages in the second Critique help us to understand what Kant means when, in the above-quoted passage, he claims that the will is subject to empirical conditions when its object is its ground of determination. At the beginning of the first Theorem, for instance, Kant affirms that all practical principles that presuppose an object (i.e. a matter) of the faculty of desire as a ground of the determination of the will are empirical practical principles and cannot function as practical moral laws (KpV 5:21.14–17).⁶ Moreover, he clarifies: “By ‘the matter of the faculty of desire’ I understand an object whose reality is desired” (KpV 5:21.17 f.). Only a few pages prior to this, Kant defines a practical rule as “always a product of reason because it prescribes action as a means to an effect” (KpV 5:20.6–8). These passages show clearly that for Kant the object of the will is what one intends to produce through the action, i.e. its end or purpose. Independently of the morality of the action, human beings need a practical principle, i. e. a rule given by reason. This rule prescribes the action, which is needed in order to reach what the will aims at (its object). Thus, the object of the will – the object  In this article, I use the term “purpose” synonymously with “end” (both of which are a translation of the German “Zweck”).  This passage does not contain the term “moral” and instead refers to the “practical”. However, it is clear from the use Kant makes of the expression “practical laws” in §§– – in opposition to “practical rules” (KpV :.–) – that he refers to practical laws, which have a moral quality: “In a practical law reason determines the will immediately, not by means of an intervening feeling of pleasure or displeasure, not even in this law, and that it can as pure reason be practical is what alone makes it possible for it to be lawgiving” (KpV :.–).

22

Federica Basaglia

of the faculty of desire – is also the object of the practical principle, since the latter is a rule that reason prescribes to the will in order for it to reach what it aims at. This object can be what moves us to act, that is, the ground of determination of the action, as in the case of non-moral (heteronomic) action. In the case of moral action, which is autonomous, i.e. only determined by pure reason, the ground of determination is not material, i. e. not the object of the will, but the moral law itself. Kant, however, does not deny that moral actions also have objects, i.e. that they serve a material purpose. It is rather that this purpose does not set the action in motion and is thus irrelevant to its moral quality. In both cases, the will has an object, which is what we intend to achieve through the action. In the case of an empirically determined action, the object of the will is at the same time the ground of determination of the action. Of course, a moral action also has an object, but this object does not serve as the ground of determination of the action, for in order to be moral an action has to be determined a priori by the moral law exclusively. It is relatively easy to understand how an empirical determination of the will works. The purpose or the end of the action, i. e. a particular object that we intend to achieve through our action, determines the action: we desire something and we are prepared to act in a manner so as to achieve it. In this case, practical reason provides us with certain practical rules or principles; that is, it determines what we should do in order to achieve the end in question. As Pistorius’s review shows, understanding what the object of a moral action is and how a moral determination of an action works are more complex tasks. Kant himself admits that, even if moral actions are not determined by objects, each and every action – whether moral or non-moral – has an object, i. e. a purpose: “All the matter of practical rules rests always on subjective conditions, which afford it no universality for rational beings other than a merely conditional one (in case I desire this or that, what I would then have to do in order to make it real), and they all turn on the principle of one’s own happiness. Now it is indeed undeniable that every volition must also have an object and hence a matter; but the matter is not, just because of this, the determining ground and condition of the maxim” (KpV 5:34.7–13; second emphasis is mine).

Each matter (i. e. each object, each purpose) of a morally or an empirically determined action rests on subjective conditions. How should we picture the matter, i. e. the purpose, of a moral action, which is determined a priori, if the former is not the determining ground of the action but merely subjective and accidental? The second chapter of the Analytic of Pure Practical Reason discusses exactly this question. Kant’s line of reasoning in these pages has a clearly dualistic structure. His argumentative strategy rests on the second Critique’s fundamental distinction between general practical reason and pure practical reason, as the be-

The Highest Good and the Notion of the Good as Object of Pure Practical Reason

23

ginning of the Preface makes clear. Here Kant distinguishes between a general use of practical reason, which needs to be “criticized” in order to show that an a priori moral theory (and in the end morality itself) is possible, and a pure, a priori use of practical reason. The aim of this work is to “criticize” the first, in order simply to show that the latter exists, i. e. that the human will can be determined by pure practical reason alone and can give rise to genuine moral actions (KpV 5:3.2–13). That Kant is ultimately referring to this distinction in his chapter on the objects of pure practical reason is already clear in the very first lines: “By a concept of an object of practical reason I understand the representation of an object as an effect possible through freedom. To be an object of practical cognition so understood signifies, therefore, only the relation of the will to the action by which it or its opposite would be made real […]” (KpV 5:57.17–21; my emphasis).

The connection between the object of practical reason and the will is clearly understood in this passage as a causal one. The action is the means by which the willed object can be produced. Therefore, the “object of practical reason” indicates here what, in the first chapter of the Analytic, Kant called an “object of the will” (KpV 5:27.6) and a “matter of the faculty of desire” (KpV 5:21.17 f.). The reason for the different description is, in my opinion, that Kant concentrates here on the realization of the object, whereas in the previous chapter he intended to emphasize the determination of the will by the object. I do not think it can be called into question that what can be realized through an action has to be something the realization of which is desired before the action is set in motion. Hence, in my opinion, the object of practical reason is to be understood as the possible purpose of an action. With this in mind, I think, Kant affirms that something’s being an object of practical reason hinges on the relation of the will to the action by which the object or its opposite would be realized: the action is the means by which the purpose can be produced. Nothing ensures that the action will produce the intended effect, but the purpose of the action remains the same even if the effect is not realized through it. This is why Kant uses the adjective “possible”: the object of practical reason is the possible effect, and not the effect of the action as such, because the latter might not coincide with what was intended. Once he has clarified what an object of practical reason is in general, Kant proceeds with the definition of an object of pure practical reason: “[…] and to appraise whether or not something is an object of pure practical reason is only to distinguish the possibility or impossibility of willing the action by which, if we had the

24

Federica Basaglia

ability to do so (and experience must judge about this), a certain object would be made real” (KpV 5:57.21–5).

Among all the objects of practical reason in general, namely among all possible purposes that can be achieved through actions, there are some that can be realized through actions that can also be willed in accordance with the moral law, and hence a priori. According to my interpretation, the field of general practical reason is the set of all objects of the will that can possibly be realized through actions. The field of pure practical reason is the set of those objects of the will that can be realized through actions that, in addition to being adequate to their purposes, can also be willed in accordance with the moral law. The object of pure practical reason is the purpose of an action that conforms to the categorical imperative. The question whether a purpose is an object of pure practical reason as opposed to an object of practical reason in general is therefore determined not by its features but solely by the fact that the action thought to lead to its realization is determined exclusively by the moral law and conforms to this law. Kant concludes by defining the objects of pure practical reason as “good” and “evil”: “The only objects of a practical reason are therefore those of the good and the evil. For by the first is understood a necessary object of the faculty of desire, by the second, of the faculty of aversion, both, however, in accordance with a principle of reason” (KpV 5:58.6–9).

Summing up, according to my interpretation the objects of pure practical reason – good and evil – are for Kant purposes of actions that are determined solely by the moral law.⁷ That Kant defines not just good but also evil as the object of practical reason (KpV 5:58.7 f.) is not unproblematic. As Kant himself explains, the concepts of good and evil are “consequences of the a priori determination of the will” (KpV 5:65.5) and presuppose “a pure practical principle and hence a causality of pure reason” (KpV 5:65.6 f.). It is easy to understand that pure practical reason brings about, as the effect of its determination of the will, something which is good; this is, for Kant, what a moral determination of the will consists in (KpV 5:41.30–8). On the other hand, it is not at all clear how a causal determination of the will by pure reason might bring about an effect that has to be considered

 On the evolution of the “way of thinking” (Denkart) and the Copernican revolution in metaphysics, see Brandt : –. On the objects of pure practical reason, see Brandt : .

The Highest Good and the Notion of the Good as Object of Pure Practical Reason

25

evil. I will not discuss Kant’s problematic definition of evil as an object of pure practical reason here, which would require a much more detailed analysis of his conception of moral evil;⁸ instead, I will concentrate on the notion of good as a concept of pure practical reason. The second chapter of the Analytic is rightly considered one of the most confused and unclear parts of the Critique of Practical Reason (cf. Torralba 2009: 215): in trying to make sense of it, we cannot always count on a coherent use of terms or a straight line of argument. In his analysis of the “Concept of an Object of Pure Practical Reason”, in fact, Kant sometimes refers to “the concept of the object of pure reason” (KpV 5:57.17, 58.10, 60.37, 62.37–63.1) and sometimes to the “object itself” or to the “Good in itself” (KpV 5:58.6 f., 60.13 f., 62.8), without distinguishing clearly between the object and its concept. Nonetheless, it is in my opinion clear that, given his definition of the objects of pure practical reason, Kant clarifies his understanding of the concept “good” in this chapter.⁹ The beginning of the second chapter of the Analytic states, in fact, that the “concept of an object of practical reason” is the “idea of an object as a possible effect of freedom” (KpV 5:57.17 f.). Hence we can deduce from this definition that the concept of an object of pure practical reason is the idea of an object as a possible effect of an a priori determination of the will. In other words, according to Kant, the concept of the good quite generally refers to a determination of the will solely by the moral law: the good can only be something we aim at or realize when our will is determined by the moral law (cf. KpV 5:62.8–18, 62.36–63.4). After he defines the objects of pure practical reason, Kant explains why he believes, contra Pistorius, that the strategy of first defining the moral good and then deriving the moral law from the former fails to provide a solid foundation for a coherent theory of pure morals. If the moral law were to derive from what we think the good is, as Pistorius suggests, rather than the concept of the good’s deriving from the moral law, the good would simply be an object that determines the moral will. According to Kant, any determination of the will by purposes we want to achieve represents a material determination of the will (cf. KpV 5:21.14–31). A material determination of the will – i. e. a determination of the will by an object – can also take place where the object in question is not empirical, as occurs, for

 On the problematic question about the possibility of acting immorally out of freedom, see Basaglia : – and Brandt : –.  I thank Pauline Kleingeld for interesting remarks on the difference between “good” as the object of practical reason and the concept of “good”, which she provided in her comments on my paper.

26

Federica Basaglia

example, when what determines our action is the will to conform to the idea of moral perfection (cf. KpV 5:39.5–41.28). Furthermore, according to Kant, any object that determines the will is something that causes some kind of pleasure. Hence, in the case of the moral law’s deriving from the notion of good, this “good” would indicate the existence of something that promises pleasure and thus determines the causality of the subject to realize it. It follows from this that, since it is impossible to know a priori what causes us pleasure or pain, the ground of determination of the will would depend on experience and could thus not be a moral (a priori) ground. In this case, the judgement concerning what is morally good and evil would be a mere matter of experience (related to a faculty which Kant calls the “feeling of pleasure and displeasure”) rather than a matter of morality. The maxims we could formulate according to these notions would concern merely means-ends relations: the good would be a mere “good-for-something-else”, not an absolute moral good (KpV 5:58.10–62.35). In short, Kant thinks that if the moral law were derived from the notion of good, the foundation of morality would be empirical, which would completely contradict the Kantian project of a pure, a priori morality. This would be the case if the moral law were derived, as Pistorius suggests, from the idea of the highest good. This concept would refer, according to Kant, to an object – the final aim of human actions – and would represent a material ground of the determination (Bestimmungsgrund) of the will. All this is well summarized in the following passage: “In this appraisal of what is good and evil in itself, as distinguished from what can be called so only with reference to well-being or ill-being, it is a question of the following points. Either a rational principle is already thought as in itself the determining ground of the will without regard to possible objects of the faculty of desire (hence through the mere lawful form of the maxim), in which case that principle is a practical law a priori and pure reason is taken to be practical of itself. In that case the law determines the will immediately, the action in conformity with it is in itself good, and a will whose maxim always conforms with this law is good absolutely, good in every respect and the supreme condition of all good. Or else a determining ground of the faculty of desire precedes the maxim of the will, which presupposes an object of pleasure or displeasure and hence something that gratifies or pains, and the maxim of reason to pursue the former and avoid the latter determines actions only mediately (relatively to a further end, as means to it), and such maxims can in that case never be called laws but can still be called rational practical precepts” (KpV 5:62.8–26).

The Highest Good and the Notion of the Good as Object of Pure Practical Reason

27

3 “Good” as Object of Pure Practical Reason and the “Highest Good” As we have seen, Kant introduces the notion of the good in the second chapter of the Analytic as an object of pure practical reason. In the Dialectic of Pure Practical Reason, he defines the “highest good” as the “unconditioned totality of the object of pure practical reason” (KpV 5:108.11 f.) and as “the whole object of pure practical reason” (KpV 5:109.21 f.). Thus, it cannot be doubted that the analysis of these two notions – “good” as object of pure practical reason, and the “highest good” – gives us the opportunity to clarify what kind of role objects play in Kantian (traditionally considered a formal) ethics.¹⁰ Accordingly, these two elements of Kant’s moral theory have been connected in the secondary literature (cf. Watkins 2010: 158 f.; Sala 2004: 142 f., 236). The way Kant himself defines them seems to suggest this connection: indeed, both notions refer to an object of pure practical reason. Moreover, as we saw above, the notion of good introduced by Kant in the second chapter of the Analytic indicates the purpose of moral actions. In the context of the theory of the highest good, Kant explicitly defines the highest good as the supreme end of the will (KpV 5:115.9–11; 134.8–13). Instead of concentrating on these well-known connections, I intend to point out some central differences between these two elements, which, in my opinion, can help us to understand them better. As we have seen, in the Preface, Kant explains that he intends to respond to Pistorius’s criticism in the second chapter of the Analytic by giving an account of good and evil as the objects of pure practical reason. As mentioned above, Pistorius holds the view that a coherent moral theory should have as its starting point the concept of the highest and absolute good as the ultimate end of the moral law (Pistorius, Rezension der Grundlegung zur Metaphysik der Sitten, 33). Kant mentions his theory of the highest good just once in this chapter – which contains his reply to Pistorius – almost at the end of the chapter. Referring to the objects of pure practical reason, Kant makes use of notions such as “absolutely good” (KpV 5:62.16), “good in every respect” (KpV 5:62.17 f.) and “good in itself” (KpV 5:62.8), but it seems to me that he tries in this chapter to avoid establishing a parallel between “good” as an object of pure practical reason and his concept of “the highest good”.

 On the problematic relation between eudaimonism and teleology in Kant’s ethics, and on the discussion of this topic in the literature, see Oggionni .

28

Federica Basaglia

In the last few pages of his analysis of the objects of pure practical reason, Kant seems to turn directly to Pistorius and expresses sympathy for his criticism: he admits that his method of defining the concept of good in the Critique of Practical Reason might sound paradoxical: instead of first defining the concept of good and then deriving the moral law from it, he defines the good only after having formulated the moral law and only in the light of that law (KpV 5:62.36–64.5). As we have seen above, Kant proceeds in this way because he views any object that determines the will as something we desire because we expect to gain pleasure from it, be it intellectual or physical. In other words, he views it as something that, in some way or other, relates to our feelings rather than reason. In this way, if a moral determination of the will is possible, it must be explainable in terms of an a priori law of reason rather than an object that we desire. Pistorius’s criticism provides Kant with the opportunity to clarify a crucial point in his theory: he explains that this note about his paradoxical method reveals “the occasioning ground of all the errors of philosophers with respect to the supreme principle of morals” (KpV 5:64.7–9).¹¹ Moral philosophers have traditionally focused on finding an object of the will in order to make it the matter and foundation of the moral law. Independently of what they took this object to be – happiness, human perfection, an object of moral sense or of God’s will – it could only serve as the foundation for a heteronomous principle of the will (KpV 5:64.6–25). Hence, when Pistorius suggests that Kant should begin his examination of the highest moral principle with an investigation into the highest good, he commits the same kind of mistake as that made by previous moral philosophers. For Kant, this mistake is particularly evident in ancient moral philosophy. Ancient moral philosophers, Kant explains, concentrated on the definition of the highest good, making it the ground of the determination of the moral will. At this point, Kant mentions his concept of the highest good, which he will present in the Dialectic of Pure Practical Reason: “The ancients revealed this error openly by directing their moral investigation entirely to the determination of the concept of the highest good, and so of an object which they intended afterwards to make the determining ground of the will in the moral law, an object which can much later – when the moral law has first been established by itself and justified as the immediate determining ground of the will – be represented as object of the will, now determined a priori in its form; and this we will undertake in the Dialectic of pure practical reason” (KpV 5:64.25–34; second and third emphases are mine).

 “[D]en veranlassenden Grund aller Verirrungen der Philosophen in Ansehung des obersten Prinzips der Moral” (KpV :.–).

The Highest Good and the Notion of the Good as Object of Pure Practical Reason

29

In my opinion, these lines can also be read as an answer to Pistorius. Kant appears to explain to his critic that his account in the Dialectic articulates the only way the concept of the highest good, as the object of the moral will, can find a place in a moral theory. Once it is established that the pure moral law is the highest principle of morality and is justified as the direct ground of the determination of the moral will, the “highest good” can be straightforwardly defined as the object of the will determined a priori by the moral law. In the above passage, Kant clarifies the difference between the notion of “good” as object of pure practical reason (presented in the second chapter of the Analytic) and the “highest good” as it is introduced in the Dialectic (KpV 5:108.11 f.). Here Kant makes explicit that what Pistorius wished to be the foundation of each moral theory (and of morality), namely the highest good, can be treated only in the context of the Dialectic of Pure Practical Reason, not in the Analytic. This clarification is very important when it comes to understanding the difference between “good” as object of pure practical reason and the “highest good”. The two different parts of the Doctrine of Elements – the Analytic and the Dialectic – deal with different problems concerning the doctrine of morals. As Kant states very clearly in the Introduction to the Critique of Practical Reason: “We shall therefore have to have a Doctrine of Elements and a Doctrine of Method for it [i. e. for the Critique of Practical Reason, F.B.]; and within the former as the first part, an Analytic, as the rule of truth, and a Dialectic, as the exposition and resolution of illusion in the judgements of practical reason” (KpV 5:16.18–20; trans. modified).

The task of the Analytic is to unfold the principles and concepts of the pure theory of morals, whereas the Dialectic deals with the illusions created by pure practical reason when it seeks the unconditioned condition for its judgements and decisions.¹²

 Cf. Beck : , . The result of my analysis is roughly in line with the interpretations that Michael Albrecht and Ottfried Höffe have proposed. According to Albrecht, whereas the Analytic in the Critique of Practical Reason works out the concept of the moral “ought” and proves that the moral law is the “determining ground” of the moral will, the Dialectic searches for the “unconditioned totality of the object of pure practical reason” (KpV :.–), without making it a determination of the will (Albrecht :  f.; cf. Düsing : ; Zobrist : ). Otfried Höffe has recently stressed the same distinction, pointing out that Kant has already treated the topic of the Dialectic in the Canon of the Critique of Pure Reason (Höffe : ). In the Canon the topic of the Analytic of the Critique of Practical Reason is presented as the answer to the purely practical question “what ought I to do?”, whereas the topic of the “Dialectic” is presented as the answer to the question “what may I hope?”, which is both practical and theoretical (KrV A ff./B ff.; cf. Beck : –).

30

Federica Basaglia

In his review of the Groundwork, Pistorius inquired into the notion of the good (derived from the notion of the highest good) as the foundation of the highest moral principle and as the condition for every judgement concerning what is morally good. Kant’s answer in this regard is very clear and straightforward: there is no such thing as a highest good guiding our moral judgement (of moral principles, maxims, actions, etc.) or our moral deliberation. What we might call, in our moral judgements and deliberations, “the good in itself” (KpV 5:62.8) can only be the object of pure practical reason (KpV 5:58.6 f.) as a “consequence of the a priori determination of the will” (KpV 5:65.5 f.). Kant further clarifies the difference between the two notions in the Dialectic, where he defines the highest good as “the unconditioned totality of the object of pure practical reason” (KpV 5:108.11 f.) and as “the whole object of pure practical reason, that is, of a pure will” (KpV 5:109.21 f.). The highest good is the supreme object of the will (KpV 5:115.9–11) and its realization and promotion (KpV 5:109.24 f.) is the supreme end of the moral will. Here it is crucial to notice that the expression “realization and promotion” refers not merely to the realization of virtue, but also to the realization of the whole of the highest good, i. e. of both virtue and happiness (KpV 5:110.22–31). What Kant means by “happiness” is expressed clearly in the paragraph on the postulate of the existence of God: “Happiness is the state of a rational being in the world in the whole of whose existence everything goes according to his wish and will, and rests, therefore, on the harmony of nature with his whole end as well as with the essential determining ground of his will” (KpV 5:124.21–5).¹³

It seems clear to me that, at least in the Critique of Practical Reason, happiness is understood by Kant as empirical happiness (cf. Düsing 1972: 33 f.; Albrecht 1978: 51 f.; Förster 2002: 180), to which skillfulness, health and even wealth and prosperity belong (KpV 5:93.15–19; cf. KpV 5:25.12–20).¹⁴ In my opinion, the supreme condition of the highest good – i. e. virtue – can certainly be understood, with Watkins, as the coherent and systematic unity of the plurality of the objects of practical reason which belong to our moral actions,

 “Glückseligkeit ist der Zustand eines vernünftigen Wesens in der Welt, dem es im Ganzen seiner Existenz alles nach Wunsch und Willen geht, und beruht also auf der Übereinstimmung der Natur zu seinem ganzen Zwecke, imgleichen zum wesentlichen Bestimmungsgrunde seines Willens” (KpV :.–).  For a different interpretation of this issue see: Zobrist : – and Oggionni .

The Highest Good and the Notion of the Good as Object of Pure Practical Reason

31

such as “helping somebody in need” or “telling the truth”¹⁵ (cf. Watkins 2010, 158). Hence, the connection between the notion of “good” as an object of pure practical reason and the “highest good” is clearly strong. The highest good is in part (maybe for the most part) the totality, the unconditioned condition of all good objects of volition (cf. Watkins 2010: 160). The crucial point in Kant’s theory, which marks the essential difference between “the good in itself” as an object of pure practical reason and “the highest good”, is that, as a matter of fact, the possession of the summum bonum consists for Kant not only in being virtuous, but also, and at the same time, in being happy in exact proportion to our virtue. Hence, as an object of pure practical reason, the highest good clearly differs from the notion of good in the second chapter of the Analytic. The former is the single and supreme end of the moral will and consists in the realization of both virtue and (empirical) happiness. The latter refers to the various purposes of moral actions. The notion of the good presented in the Analytic refers exclusively to morality – to what, on the basis of the moral law, can be judged to be good in itself. The highest good refers to the realization not only of morality but also, although it is proportioned to morality, of empirically understood happiness.

References Albrecht, Michael 1978, Kants Antinomie der praktischen Vernunft, Hildesheim: Olms Basaglia, Federica 2009, Libertá e Male Morale nella Critica della Ragion Pratica di Immanuel Kant, Roma: Aracne Beck, Lewis White 1960, A Commentary on Kant’s Critique of Practical Reason, Chicago/London: The University of Chicago Press Bittner, Rüdiger/Cramer, Konrad (eds.) 1975, Materialien zu Kants “Kritik der praktischen Vernunft”, Frankfurt a. M.: Suhrkamp Brandt, Reinhard 2007, Die Bestimmung des Menschen bei Kant, Hamburg: Meiner Brandt, Reinhard 2010, Immanuel Kant – Was bleibt?, Hamburg: Meiner Düsing, Klaus 1971, “Das Problem des höchsten Gutes in Kants praktischer Philosophie”, Kant-Studien 62, 5–42 Förster, Eckart 2002, “Die Dialektik der reinen praktischen Vernunft (122–148)”, in: Otfried Höffe (ed.) 2002, Kritik der praktischen Vernunft, Berlin: Akademie-Verlag, 173–86 Gesang, Bernward 2007, “Einleitung”, in: Bernward Gesang (ed.) 2007, Kants vergessener Rezensent. Die Kritik der theoretischen und praktischen Philosophie Kants in fünf frühen Rezensionen von Hermann Andreas Pistorius, Hamburg: Meiner, VII–XLIV

 Under the condition that we help somebody and tell the truth out of duty, i. e. that our will is determined solely by the moral law.

32

Federica Basaglia

Höffe, Otfried 2012, Kants Kritik der praktischen Vernunft. Eine Philosophie der Freiheit, München: Beck Oggionni, Eva 2014, “Das höchste Gut: Kants Eudämonismus unter besonderer Berücksichtigung der Jahre 1781–85”, Philosophical Readings IV.1, 76–89 Pistorius, Hermann Andreas 1786, Rezension der Grundlegung zur Metaphysik der Sitten, Allgemeine Deutsche Bibliothek 66, 447–63; reprinted in: Bernward Gesang (ed.) 2007, Kants vergessener Rezensent. Die Kritik der theoretischen und praktischen Philosophie Kants in fünf frühen Rezensionen von Hermann Andreas Pistorius, Hamburg: Meiner, 26– 38 Pranteda, Maria Antonietta 2009, Il legno storto. I significati del male in Kant, Cittá di Castello (Perugia): Leo S. Olschki Torralba, José Maria 2009, Libertad, objecto práctico y acción. La facultad de juicio en la filosofía moral de Kant, Hildesheim/Zürich/New York: Georg Olms Watkins, Eric 2010, “The Antinomy of Practical Reason: Reason, the Unconditioned and the Highest Good”, in: Andrews Reath/Jens Timmermann (eds.) 2010, Kant’s Critique of Practical Reason. A Critical Guide, Cambridge: Cambridge University Press, 145–67 Zobrist, Marc 2009, “Kants Lehre vom höchsten Gut und die Frage moralischer Motivation”, Kant-Studien 99, 285–311

Pauline Kleingeld

Kant on ‘Good’, the Good, and the Duty to Promote the Highest Good Kant’s account of the highest good is often regarded as a failure. He argues that the highest good comprises both virtue and happiness, and he also claims that we have a duty to promote it.¹ These claims raise difficult questions. First, given that the account of the highest good in the Critique of Practical Reason is phrased, in part, in terms of individual happiness in proportion to virtue (that is, in terms of desert), it would seem that any duty to promote the highest good would require us to act as moral judges in a way that Kant himself adamantly rules out as impossible. Furthermore, the inclusion of happiness seems to introduce “non-moral goods” into the highest good (Beck 1960: 242–3), and this makes it hard to see how Kant could consistently call it a moral duty to promote it. Finally, Kant claims that this alleged duty goes “beyond” obedience to the moral law and that it cannot be analytically derived from the moral law (RGV 6:7n.; Gemeinspruch 8:280n.). This makes it even harder to see how this duty can be justified, and what exactly this duty amounts to. In this essay, I argue that there is a valid argument, on the basis of premises Kant clearly endorses, for the conclusion that it is a duty to promote the highest good, and I explain what this duty consists in. Because the main difficulties for such an argument are often considered to stem from the inclusion of happiness in the highest good, it is necessary to examine why he regards the highest good as consisting of both virtue and happiness. This, in turn, requires a clear account of what Kant means by ‘good’ in the first place, and the conditions under which, on his understanding of ‘good’, happiness is indeed good. Therefore, I start by examining Kant’s distinction between ‘good’ and ‘pleasant’, and his account of what is good (section 1). This provides the key to understanding why happiness, as one of the two constituent parts of the highest good, is good – indeed morally good – and not merely pleasant. The answer lies in the fact that it is a duty to promote the happiness of others (section 2). Building on Kant’s account of the good and the highest good, I then reconstruct a valid argument to the effect that it is a duty to promote the highest good. I illustrate the structure of the argument with an example that illuminates why this is a genuine duty even though it is ‘not contained in the moral laws’ and how it goes beyond obedience to the Categorical Imperative even though it does not add a separate duty (section 3). Kant’s improbable-sound KpV :, , , ; KU :, , n.; RGV :, n.

34

Pauline Kleingeld

ing combination of claims starts to make sense only once we approach his account of the highest good via his account of the good (section 3). For the purpose of this paper, I draw on a distinction between two different conceptions of the highest good that are found in Kant’s moral theory (Reath 1988; Kleingeld 1995). In many of his discussions of the highest good, Kant defines it as a moral world, by which he means a world populated by fully virtuous agents whose virtuous action causes happiness (KrV A808/B836; RGV 6:5; Gemeinspruch 8:280n.). At other times, most notably in the Dialectic of the second Critique, he discusses the highest good “in a person” (KpV 5:110, emphasis added), defining it as happiness in proportion to virtue. The first conception of the highest good can more easily be developed on the basis of the principles of Kant’s moral theory than the second. In this essay, I limit my discussion to the conception of the highest good as a moral world. Kant’s conception of the highest good “in a person” requires separate treatment.² Progress in our understanding of Kant’s conception of the highest good as a moral world may also shed light on the idea of the highest good ‘in a person’, but this issue lies beyond the scope of this essay. Furthermore, I do not address the issue of the possibility of the highest good, or associated issues regarding the nature, role, and justification of the postulates of God and immortality.³ Instead, I focus on the preliminary issues of the concept of the highest good and, especially, the duty to promote it.⁴

1 Is Happiness Good? And What is Good Anyway? In order to understand why happiness is part of the highest good, we need to understand why, and in what sense and under which conditions, happiness, as part of the highest good, is deemed good, and not merely agreeable. The distinction between ‘good’ and ‘agreeable’ is of fundamental importance to Kant’s moral theory, for reasons that I shall clarify, in this section, on the basis of Kant’s discussion with Hermann Andreas Pistorius (1730–1798). Kant starts the Groundwork with the famous claim that the only thing that can be held to be good without limitation is a good will (GMS 4:393). He also claims that his conception of a good will can already be found in every “natural

 See Engstrom  for a detailed discussion. In Kleingeld , I suggest an account of why Kant, in the Critique of Practical Reason, discusses the highest good “in a person” in terms of the proportionality between virtue and happiness (Kleingeld : ).  See Willaschek  for a discussion of the postulates.  The argument in this essay elaborates and partially modifies the account I sketched in outline in Kleingeld : –.

Kant on ‘Good’, the Good, and the Duty to Promote the Highest Good

35

sound understanding” and that it needs not so much to be taught as only to be clarified (GMS 4:397). As the title of the first section indicates, Kant starts his argument in the Groundwork from “common moral rational cognition”. In his review of the Groundwork Pistorius criticizes Kant for proceeding in this way. He complains that Kant should have first explained what he means by ‘good’, and that an appeal to general public agreement can never be philosophically decisive. Pistorius asks: “What is good anyway?” (Was ist überhaupt gut?). This question is ambiguous and could mean “what does ‘good’ mean?” or “which objects are good?” It seems that Pistorius has both in mind. He writes: “In this regard I wish the author had liked to discuss first of all the general concept of what is good, and to determine more precisely what he understands by it; because obviously, we would first have to agree on this before we can make out anything concerning the absolute value of a good will. Therefore, I am entitled to ask first: What is good anyway, and what is a good will in particular? Is it possible to conceive of a will that is good in itself and regarded without relation to any object? If one says: good is that which is generally approved and valued, then I am permitted to ask further why it is approved and valued, does that happen rightly [mit Recht] and with reason [mit Grunde] or not? General unanimous approval, if this would occur or be possible on anything, would never be able to count as the ultimate decisive reason for a philosophical researcher” (Pistorius, Rezension der Grundlegung zur Metaphysik der Sitten, 449).⁵

Pistorius criticizes Kant for following a bad method. He claims that we cannot determine whether a will is good unless we have first determined what is good, and that we cannot determine what is good unless we have first settled on a definition of ‘good’. Pistorius has the impression that Kant’s appeal to widespread intuitions about the unconditional goodness of a good will, in the first section of the Groundwork, implies that he develops his moral theory on this basis. This, Pistorius claims in the quoted passage, is bad philosophy: morality cannot be grounded in mere intuitions that just happen to be widespread. A philosophical grounding of morality should be able to provide an account of what is good and why. It should be able to provide the reason why something that is widely claimed to be good is indeed rightly regarded as good. Anything less is beneath the professional standards of philosophy. Furthermore, Pistorius believes that it is impossible, without the help of “material” presuppositions, to provide such an account of what is morally good. He asserts that we cannot decide whether a will is indeed good except by reference to the goodness of the object or purpose of this will. If lawfulness as such were suffi-

 The translations of the quoted passages from Pistorius’ review are my own.

36

Pauline Kleingeld

cient to make a will good, Pistorius adds, then a will could be called good even if it obeyed a bad law, and this would be absurd. Therefore, he writes, “it must be a good principle, a good law, obedience to which makes a will good. […] we must finally arrive at some object or at the final end of the law, and we must also make use of the material [aspect], because we cannot make do with the formal [aspect] of either the will or the law” (Pistorius, Rezension der Grundlegung zur Metaphysik der Sitten, 449).

Kant feels the sting of Pistorius’ criticisms and replies extensively in the second chapter of the Critique of Practical Reason, addressing the question “what is good?” in both senses.⁶ In response to the call for conceptual clarification, he refers to the German semantic distinction between “the good” (das Gute) and “well-being” (das Wohl) (KpV 5:59–60). Other languages, such as Latin, do not have separate words for these, which leads to confusion, Kant argues, because saying that an action is good (or evil) is fundamentally different from saying that it serves our well-being (or our ill-being). In the latter case, we say something about the expected effects of an action on our feeling of pleasure or displeasure. To say that some course of action is good, by contrast, is to say that there is sufficient reason to act in this way. This is true regardless of whether the action is judged good in light of some antecedent end or good in itself (KpV 5:58). In both cases, the question whether an action is good differs from the question whether I regard it as conducive to pleasure. Thus, I may decide to take a course of action towards which I feel aversion, because on the basis of practical reasoning I recognize that doing so is good. Kant gives the example of someone who needs to undergo surgery: he regards this as an ill (Übel) and as highly unpleasant; yet on the basis of his reasoning he nevertheless declares it to be good to undergo it (KpV 5:61). Given the conceptual distinction between the good and the pleasant, and given the fact that the judgment that an action is good – whether good in light of an antecedent end or good in itself (morally, unconditionally, absolutely) – requires reference to reasons for action, it follows that it is impossible to ground a moral theory in a conception of the good that somehow precedes rational standards. Unless we can give a reason in support, the assertion that a certain object is morally good would be merely arbitrary – philosophical footstamping without rational warrant. If the assertion is grounded in feeling, moreover, the alleged good would actually be pleasant, and it would not deserve to be called good – let alone absolutely and unconditionally good. In order for something to deserve being called morally good, therefore, we need to be able to indicate why it is good,  See Kant’s comment in the Preface, KpV :.

Kant on ‘Good’, the Good, and the Duty to Promote the Highest Good

37

and such an account necessarily has to rely on rational standards. Therefore, morality cannot be grounded in a ‘material’ conception of the good, independently from underlying rational criteria. Instead, as Kant claims to have shown in the first chapter of the Critique of Practical Reason, the principle of morality is a formal a priori principle of pure practical reason itself. Once the principle of morality has been established, however, Kant argues, this principle can be used to develop an account of what is morally good. This is what he calls the “paradox of method” in a critique of practical reason (KpV 5:62–3). This paradox consists in the fact that it is impossible for a conception of the good to ground moral principles, and that the conception of the good must instead follow the determination of moral principles. He adds that this methodological insight explains why he starts his Critique of Practical Reason by considering moral principles, in the first chapter – showing both that no material principle qualifies as a suitable basis for a moral law and that an a priori principle of pure practical reason does qualify as the moral law – and why he discusses the notion of the good only afterwards, in the subsequent chapter. What, then, is the good, on his conception? The good is what we have reason to do, whether conditionally (dependent on antecedent ends) or unconditionally (morally, absolutely). Strictly speaking, the object of practical reason is always an action or a way of acting, and never a thing (KpV 5:57, 60). Actions, in turn, may be directed to the realization of something, of course. In such cases, the object of practical reason is the action, and the object of the action is the realization of the thing. If the action is good, then the object of the action may also be called good by virtue of the goodness of the action. In that case, the goodness of the object of the action derives from the goodness of the action. The object of pure practical reason is that which is good “absolutely” and “in every respect and without any further condition”. Again, strictly speaking this is a “way of acting” (Handlungsart), not a thing (Sache) (KpV 5:60), because it is that way of acting which it is morally possible to will (KpV 5:58). Whether it is physically possible to act in such a manner or achieve the end at which the action is directed, is a different matter (KpV 5:57–8). In other words, Kant equates the question whether an action is morally good with the question whether it is an object of pure practical reason and with the question “whether we are allowed to will [wollen dürfen] an action which is directed to the existence of an object if the object were within our power” (KpV 5:58, translation altered). In other words, morally good action is action under the guidance of the moral law. The upshot of Kant’s account of the good, for the question regarding the moral status of happiness, is the following. Promoting happiness is morally good (that is, it is an object of pure practical reason) if and only if it is morally allowed to will it. Furthermore, the mere fact that we desire our own happiness

38

Pauline Kleingeld

does not justify calling happiness good – it does not even justify calling it a “nonmoral good”.

2 From the Good to the Highest Good In the Dialectic of the Critique of Practical Reason, Kant proceeds to a discussion of the “highest good”, which he defines as the “unconditioned totality of the object of pure practical reason” (KpV 5:108). In the Analytic, Kant determined the good in itself formally as the object of pure practical reason, that is, as action under the guidance of the moral law. He now determines the highest good as the unconditioned totality of the object of pure practical reason (KpV 5:108). On the basis of the conception of the object of pure practical reason as action under the guidance of the moral law, the highest good is the world in which all rational agents universally act under the guidance of the moral law. What would the world be like, if every moral subject always acted as the Categorical Imperative demands? According to Kant, this is a natural question for finite rational beings to ask. Reason always seeks the unconditioned, striving – also in its practical employment – to establish systematic unity and totality (see Watkins 2010; Kleingeld 1998a). In this case reason forms the idea of a moral world, by synthesizing the totality of moral demands into a single moral ideal (KrV A808/B836). Kant argues that the resulting notion of a moral world, as the highest good, has two elements. First of all, it includes virtue, as the “supreme” good, that is, as that which is unconditionally good. In order to be the “complete” good, however, Kant adds, the highest good must also include happiness. The two elements are to be connected in such a way that the first is the cause of the second (KpV 5:110). It is not surprising that Kant designates virtue as the “supreme” good. The highest good is conceived as a world in which all agents act morally, so virtue is necessarily part of it. Moreover, moral agents have the duty to strive for their own perfection, so doing so is clearly a morally good way of acting and virtue is clearly an object of pure practical reason. Hence, universal virtue is included in the “absolute totality” of the “object of pure practical reason”. As mentioned in the introduction, Kant’s account of the highest good has been the target of many objections, however, because of his inclusion of happiness into the notion of the highest good and because of his claim that it is a duty to promote the highest good. How can happiness be designated as good, rather than as pleasant? Moreover, is it consistent for Kant to claim that it is a duty to promote the highest good if the latter includes happiness? I shall address these questions in turn, starting with Kant’s inclusion of happiness in the highest good.

Kant on ‘Good’, the Good, and the Duty to Promote the Highest Good

39

Several strategies for defending Kant’s inclusion of happiness in the highest good are found in the literature. Some authors have argued that happiness, in the highest good, is limited to the satisfaction of one’s morally permissible ends (Reath 1988), which means that its inclusion does not run counter to moral demands. Others have argued that we regard virtue alone as less good than virtue plus happiness, because we desire happiness, and therefore – so the argument goes – the highest good must also include happiness (this line of reasoning is suggested, among others, by Beiser 2006: 595). On both of these interpretations, however, happiness is included on the basis of the expectation of its agreeableness, and not because of its moral goodness. Moreover, if happiness is included in the highest good on the basis of our desire for it – even if this desire is limited to what is morally permissible – this still makes it hard to see how we could have a duty to promote the highest good.⁷ A very different strategy for defending Kant’s inclusion of happiness in the highest good could be thought to be possible on the grounds that Kant argues that it is a duty – albeit an “indirect” one – to promote one’s own happiness insofar as it is necessary for the sake of morality (KpV 5:93; GMS 4:399). If promoting one’s own happiness (even if only to a degree) is a duty, then this would seem to warrant designating happiness as good. Nevertheless, this strategy for defending the inclusion of happiness in the highest good does not succeed. Kant defends the indirect duty to promote one’s own happiness on the basis of the fact that one may be tempted to violate the moral law if even one’s most basic human needs are not met. The concept of the highest good as a moral world, however, is derived exactly by abstracting from such temptations (KrV A808/B836); its first element is perfect virtue. Therefore, the indirect duty to promote one’s own happiness (to a certain degree) cannot explain the inclusion of happiness in the highest good. Furthermore, the indirect duty to promote one’s own happiness is limited to the degree that doing so is required for the sake of morality. By contrast, Kant’s definition of happiness includes the complete realization of all of an agent’s ends. In the Dialectic of the Critique of Practical Reason, he defines happiness as follows:

 An additional difficulty for this strategy is that if happiness is included in the highest good on the grounds that virtue-plus-happiness is preferable to virtue-without-happiness, this does not explain why Kant calls the relation between virtue and happiness in the moral world a necessary causal connection. For happiness might then in principle also be brought about by other means than by virtuous action. It could, for example, be regarded as a reward or supplement distributed by God (see Beiser : –).

40

Pauline Kleingeld

“Happiness is the state of a rational being in the world in the whole of whose existence everything goes according to his wish and will, and rests, therefore, on the harmony of nature with his whole end as well as with the essential determining ground of his will” (KpV 5:124).

This definition is important in two other respects as well. First, Kant defines happiness in terms of the harmony between an agent’s ends and the state of the world, not in terms of how the agent feels. Agents count as happy when things go the way they want them to go. Second, the reference to “essential determining grounds of the will” indicates that happiness should not be understood merely as a matter of the satisfaction of morally permissible contingent desires. When happiness is conceived as element of the highest good, that is, as the happiness of virtuous agents, it also includes the realization of their moral ends. This, in turn, suggests a different strategy for making sense of Kant’s inclusion of happiness in the highest good, a strategy which yields a rather straightforward explanation. If we conceive of that which is morally good as the object of pure practical reason, and of the highest good as the “unconditioned totality” of this object, then the highest good is, as Kant puts it in the Religion within the Boundaries of Mere Reason, the “world [which one] would create, were this in [one’s] power, under the guidance of practical reason” (RGV 6:5). The highest good, then, is the world that all moral agents, acting under the direction of the moral law, bring into existence when it is in their power to realize the object of their actions. In order to know what this world comprises, therefore, we should examine what morality demands that we do; that is, we should look to our list of duties. This list includes our own moral perfection first of all, of course, and, as we saw above, Kant includes virtue in the highest good. It is also the duty of every moral agent, however, to promote the happiness of others by adopting others’ (permissible) ends as their own. Kant argues on many occasions that promoting the happiness of others is a duty (KpV 5:34–5; GMS 4:423; MS 6:385–8, 453). And in the Metaphysics of Morals Kant designates “one’s own perfection” and the “happiness of others” as the two “ends that are at the same time duties” (MS 6:385–8). If it is a moral duty to promote the happiness of others, then the highest good, conceived as an ideal moral world populated by virtuous agents, does include the happiness of all. In a moral world, I promote the happiness of others, and others promote mine. The happiness of each is good in the eyes of the others, and vice versa. This means that the virtuous agents in this world collectively aim at the happiness of all. ⁸ In this way, the inclusion of happiness in the highest  This duty is restricted to promoting the morally permissible ends of others, and so it might seem to lead to the problem that the others’ conception of their happiness comprises more

Kant on ‘Good’, the Good, and the Duty to Promote the Highest Good

41

good can be defended on the grounds that the moral law demands that it be promoted. This in turn means that happiness, brought about by virtuous action, is indeed morally good and not merely pleasant.⁹ This interpretation of Kant’s grounds for including happiness into the highest good is confirmed by passages in which Kant says as much. In the Critique of Pure Reason, Kant describes the highest good in terms of a moral world in which rational beings, “under the guidance of such [moral] principles, would themselves be the authors both of their own enduring welfare and at the same time that of others” (KrV A809/B837). This world is a “system of self-rewarding morality”, and it is an idea “the realization of which rests on the condition that everyone do what he should” (KrV A809/B837). The rational inhabitants of this world are virtuous: they all act under the guidance of moral principles, and their virtuous activity causes general happiness. If happiness is included in the highest good on the grounds of duty, this also explains Kant’s use of causal language describing the relation between virtue and happiness. It is often overlooked that Kant calls happiness, as component of the highest good as a moral world, an “effect” or “result” of virtue (KpV 5:115, 119; RGV 6:7n.), as being “caused” by virtue (KpV 5:111, 114). These expressions suggest that happiness is indeed conceived as the end and the result of virtuous action.¹⁰ In sum, the highest good, when conceived as a moral world, is the world that moral agents would bring into existence if their agency faced no obstacles, that is, if all moral agents were fully virtuous and their actions would achieve their moral ends. The highest good includes happiness because morality demands that we make the happiness of others our end, while making it a duty on the part of others to promote ours (as part of their duty to promote the happiness of others). Thus conceived, the idea of the highest good as comprising both virtue

than moral agents are willing to endorse. Because Kant conceives the agents in the ideal moral world as fully virtuous, however, these perfect agents will not have morally impermissible ends, and so this problem does not emerge.  Importantly, this point also sheds light on Kant’s inclusion of happiness in the notion of the highest good “in a person”. Kant justifies this on the grounds that a virtuous person’s happiness is good in the eyes of another rational being who is impartial and perfect. Here, too, the virtuous person’s happiness is called morally good because the other imagined rational being regards the virtuous person as an end and wills the virtuous person’s happiness (KpV :).  This is not the only way in which Kant speaks of the connection of virtue and happiness: when discussing the highest good for an individual (the highest good “in a person”, KpV :), Kant mentions God as the cause of happiness (e. g., KrV A/B, A/B). But as mentioned in the introduction, in this essay I focus on the notion of the highest good as a moral world.

42

Pauline Kleingeld

and happiness is defined completely in terms of that which is morally good, that is, in terms of action under the guidance of the moral law. On this interpretation, developing the idea of the highest good requires a specific form of rational activity, namely, thinking through what morality requires and then synthesizing this into a single ideal. It requires determining systematically what way of acting the moral law requires, which leads to the insight that moral agency can be brought under the most general headings of pursuing our own moral perfection and the happiness of others. The idea of the highest good is developed by pure practical reason, when it extends itself “beyond observance of the formal law to production of an object (the highest good)”, that is, when it conceives of the world that universal moral agency would realize and then conceives that world as its final end (Gemeinspruch 8:280n.; cf. RGV 6:7n.). The idea of the highest good is formed by thinking through which duties the application of the moral law would yield, and then developing the idea of the world that would be brought into existence by universal virtue.¹¹ It is an idea that “rises out of morality and is not its foundation; it is an end which to make one’s own already presupposes ethical principles” (RGV 6:5).

3 The Duty to Promote the Highest Good Kant’s claim that it is a duty to promote the highest good is, on the face of it, very puzzling, even when it is related to a moral world instead of to a proportional relation between individual virtue and happiness. It raises the question whether this duty is just another way of saying that we have a duty to obey the Categorical Imperative, or whether it goes beyond this. On the one hand, it would seem that it cannot be an additional duty, over and above the duty to obey the Categorical Imperative. If it were, this would seem to imply that the latter is not fully sufficient as a guide in moral matters after all, because there would be at least one duty in addition to what it tells us to do. On the other hand, if the duty to promote the highest good does not go beyond the duty to obey the Categorical Imperative, then what is the point of mentioning this duty at all? Kant’s argument regarding these matters is very brief, and any interpretation of it will have to rely not merely on textual evidence but also on philosophical (re)construction.

 Relatedly, Kant argues not merely that every rational agent has a duty to promote the highest good, but also that this duty is a duty of the human race towards itself (RGV :).

Kant on ‘Good’, the Good, and the Duty to Promote the Highest Good

43

Kant claims that the Categorical Imperative commands that we promote the highest good,¹² but he does not actually spell out the details of the justification of this claim. At first sight it is confusing that he claims that this obligation is based on the Categorical Imperative while also insisting that it is not contained in it or in our particular duties. For example, Kant writes: “[the claim that humans ought to make the highest good their end] goes beyond the concept of the duties in this world, and adds a consequence (an effect) of these duties that is not contained in the moral laws and cannot, therefore, be developed out of these laws analytically” (RGV 6:7n., translation modified).

Here he claims that the duty to promote the highest good “goes beyond” the moral laws: it “adds a consequence” that is “not contained in the moral laws”. Similarly, in “On the Common Saying”, Kant writes: “The need for a final end assigned by pure reason and comprehending the whole of all ends under one principle (a world as the highest good and possible also through our cooperation) is a need of an unselfish will extending itself beyond observance of the formal law to production of an object (the highest good)” (Gemeinspruch 8:280n.).

Again Kant speaks of an “extension” that goes “beyond” observing the formal (moral) law. This way of putting it presents two difficulties. First, Kant’s statement that this duty adds something that is “not contained” in the “formal law” (i.e., the moral law) seems to suggest that the duty to promote the highest good is at least partially based on something other than the moral law. If the claim is indeed that there is another source of moral duties, this would run counter to the results of the Groundwork and the Analytic of the Critique of Practical Reason. Second, Kant’s appeal to a “need” to motivate the “extension” seems to be at least as problematic. It seems to indicate that the duty is indeed based on a consideration that is foreign to the rest of Kant’s moral theory, because mere appeal to needs does not suffice to ground moral duties. Given that Kant seems to view the duty to promote the highest good as justified by a need (albeit a “need of the will”) and that he writes that this duty cannot be analytically derived from the formal principle of morality, the duty seems to be on shaky ground. Some commentators have suggested that the duty to promote the highest good can be defended in terms of its moral usefulness. They suggest that the duty to promote the highest good might be justified by the fact that promoting a moral world would bring about the social conditions needed for realizing fun-

 KpV :, , , ; KU :, , n.; RGV :n.

44

Pauline Kleingeld

damental moral ends in general (Reath 1988: 617 n.30, 619) or that it is the best way to promote virtue within ourselves (Fugate 2014: 151). The problem with any justification of the duty to promote the highest good in terms of some other, more fundamental moral endeavor is conceptual: the highest good would not be the highest good if promoting it were morally required in the service of something else, even if, in this case, ‘something else’ is obedience to the moral law itself. Others have suggested that the duty to promote the highest good does not really amount to a special duty at all. Sometimes this is meant as a criticism (Beck 1960: 244–5), but other authors argue, in defense of Kant’s position, that the duty to promote the highest good “coincides” entirely with the duties Kant identifies as an individual’s duties (Engstrom 1992: 776). This interpretation has the hermeneutic disadvantage, however, that it does not explain in what way the duty to promote the highest good goes “beyond” the duties that flow from the Categorical Imperative, as Kant claims it does. The analysis, in the previous section, of how the idea of the highest good is developed on the basis of the concept of the good, however, provides the basis for a valid argument in support of the duty to promote the highest good, an argument, moreover, that also explains how this duty goes “beyond” the list of duties that follows from applying the Categorical Imperative. If the highest good is defined as the world that would be brought into existence by universal virtue (if our powers were sufficient to achieve our ends), this implies not only that moral action promotes the highest good, but also that nothing else promotes the highest good. For if virtue is the supreme condition of the highest good, it is impossible to promote the highest good in any way independently from the Categorical Imperative. In other words, given how the highest good is defined, moral agency does not promote the highest good accidentally, but necessarily. The highest good is brought about through virtuous action and only through virtuous action, as its supreme condition. Given that acting in accordance with the Categorical Imperative necessarily promotes the highest good, and that the highest good cannot be promoted any other way (that is, it cannot be promoted in a non-virtuous manner), we can formulate the following argument: 1. We ought to act in accordance with the Categorical Imperative; 2. Acting in accordance with the Categorical Imperative, and only acting in accordance with the Categorical Imperative, necessarily promotes the highest good; 3. Therefore, we ought to promote the highest good.

Kant on ‘Good’, the Good, and the Duty to Promote the Highest Good

45

The second premise requires a good deal of further clarification, however, especially regarding the nature of the connection between acting in accordance with the Categorical Imperative and promoting the highest good. As stated, this premise could seem to posit an analytic connection. If acting in accordance with the Categorical Imperative is simply synonymous with promoting the highest good, then the connection is analytic. In that case, the argument goes through, but the duty in the conclusion does not go “beyond” the duty expressed in the first premise. “Promoting the highest good” would be just another way of saying “acting in accordance with the Categorical Imperative”. As is perfectly clear from the passages quoted above, however, Kant explicitly claims that the duty to promote the highest good cannot be derived analytically from the duty to act in accordance with the Categorical Imperative. On the other hand, if the connection between acting in accordance with the Categorical Imperative and promoting the highest good is not analytic, it could seem contingent, and then the argument would not go through. If acting in accordance with the Categorical Imperative merely contingently leads to the highest good, then the deontic status of the first premise does not transfer to the third. An example helps to clarify this: If I have a duty to help others in need, and if, when I do help others in need they (contingently) happen to give me flowers with a thankyou note, it does not follow that I have a duty to make others give me flowers with a thank-you note. This is why the second premise includes “necessarily”. In other words, if the connection between obedience to the Categorical Imperative and promoting the highest good is analytic, the conclusion is valid but the latter does not go beyond the former; and if the connection is contingent, the argument is not valid and does not lead to a duty to promote the highest good. To avoid both results, the connection stated in the second premise needs to be synthetic and yet necessary (a priori). This is precisely Kant’s contention, however. He writes: “[T]hat every human being ought to make the highest possible good in the world his own final end is a synthetic practical proposition a priori, that is, an objective-practical proposition given through pure reason” (RGV 6:7n., translation altered).

On the basis of the analysis in the previous section, we are now in a position to make sense of this statement. Because the idea of the highest good is constructed on the basis of the Categorical Imperative itself, the highest good is necessarily the result of action in accordance with the Categorical Imperative. The connection is nevertheless synthetic because one cannot derive the notion of the highest good by analyzing the Categorical Imperative itself. This is because the idea is developed through a synthesizing operation that involves first conceiving

46

Pauline Kleingeld

of everything the Categorical Imperative demands – that is, the moral duties it imposes on us – and then taking the result together into a complete whole. In this sense it is an extension “beyond” the Categorical Imperative. An analogy may help to make this clearer. Suppose my boss commands that I do what she says. She first gives me one task, and then another, and so on. First she tells me to make a wheel, and then another wheel; then she tells me to weld one metal tube to another tube, at an odd angle; then she tells me to weld a third tube to the first two, and on and on. I may complete each of my tasks just like that, seeing myself as simply following the formal command to “do what she says”, and as carrying out the specific duties of making one wheel, then another, then welding some tubes, and so on. But it would be natural for me to start wondering what my activities are eventually going to amount to – where all of this is going. Knowing the answer would change my understanding of what I am doing. I cannot derive the answer to this question from the injunction to “do what she says”, however. The imperative is merely formal and does not allow the inference to any specific end my boss has in mind. Moreover, a partial list of assignments would leave open many possible outcomes. At the start I might still think that I need to make a large number of wheels, or a bicycle trailer, or a modern work of art. Only once I see the complete list of assignments and mentally put all the steps together does it become possible for me to see that I am, in this example, building a bicycle. It then becomes possible for me to conceive of each of the partial tasks as contributions to the larger project of building a bicycle by following the specific steps. This project goes beyond the formal imperative to “do what she says” and the individual assignments my boss gives me, because it takes the list of concrete tasks as a whole, synthesizing the individual tasks into the overall project of building a bicycle. This conception of the larger project is not analytically contained in the formal requirement to “do what she says”. It “broadens” and “extends” my perspective as an agent from merely doing what she says, completing the individual assignments one by one, to a focus on the final end – I should build a bicycle. Furthermore, this final end is not contingently related to the requirement to do what she says, because the bicycle is a necessary consequence of following all the steps she lists (after all, the very end of building a bicycle was established on the basis of the list of steps). Finally, recognizing the overall end of my following the list of particular assignments does not give me license to build just any bicycle in whatever way I see fit. Rather, my task is to build a bicycle following this particular list of steps, by doing what my boss tells me to do. The analogy is not perfect, of course, because my boss’s imperative is a heteronomous command, it is directed at me rather than at all rational beings, and the bicycle is a product that will have independent existence once it is built (whereas a moral world, because it requires virtue, requires an ongoing effort).

Kant on ‘Good’, the Good, and the Duty to Promote the Highest Good

47

Yet there is an illuminating structural similarity between this example and the duty to promote the highest good. The duty to promote the highest good cannot analytically be derived from the formal Categorical Imperative as such. Only once we conceive of the absolute totality of everything that is morally demanded by the Categorical Imperative do we come to conceive of the highest good as the final end of moral agency. This gives us a new sense of purpose and a new understanding of what we are doing that goes beyond the formal command to obey the Categorical Imperative and our list of specific duties taken one by one. Just as the list of partial assignments, taken together as a whole, yields the assignment to “build a bicycle (by doing what she says)”, the absolute totality of all moral duties of all agents, taken together as a whole, yields the duty to promote the highest good (via the argument specified above). Recognizing this requires the distinctive reflective operation of synthesizing the “absolute totality” of the object of pure practical reason. This procedure explains why the synthetic relation between the Categorical Imperative and the duty to promote the highest good is not merely contingent but necessary. The example also illuminates why the duty to promote the highest good does indeed go beyond the Categorical Imperative without circumventing it. When I am faced with a formal meta-command (to do what my boss says) and a long list of discrete tasks, finding out at some point during the process of assembly (but before the end) that I am to build a bicycle does add essential new information about what my assignment consists in, without adding an additional step to the list. “Build a bicycle (by doing what she says)” goes beyond the assignment to “do what she says”, without introducing an additional assignment that is generated independently from my boss’s general demand that I do what she says. Similarly, from my perspective as an agent who ought to act in accordance with the Categorical Imperative and who is faced with a long list of discrete moral duties, finding out that I am to promote the highest good does add new information, without adding another separate duty to my list. As Kant put it, “harmonizing with this final end [viz., the highest good] does not increase the number of morality’s duties but rather provides these with a special point of reference for the unification of all ends” (RGV 6:5, translation altered).

One final issue that still needs to be addressed is Kant’s claim that the conception of the highest good is formed on the basis of a ‘need’. As became clear above, Kant writes that the conception of the highest good springs from the will’s “need for a final end” (Gemeinspruch 8:280n.). Elsewhere, Kant claims it

48

Pauline Kleingeld

“is one of the inescapable limitations of human beings and of their practical faculty of reason (perhaps of that faculty in all other worldly beings as well) to be concerned in every action with its result, seeking something in it that might serve them as an end […]” (RGV 6:7n.).

The fact that Kant calls this tendency an “inescapable limitation” of human beings might make it sound like a regrettable feature, an anthropologically based shortcoming it is unfortunately impossible to get rid of. If this is how the comment should be read, this might still be seen as causing trouble for Kant’s argument that it is a duty to promote the highest good. However, as the parenthetical comment and Kant’s mention of practical reason make clear, the limitation Kant has in mind is tied to the nature of reason, and to human nature only insofar as humans are finite rational beings – not to contingent sensible needs. This relates Kant’s comment to his broader description of reason as having “needs” and “interests” that it cannot dismiss (cp. KrV Avii), and in particular to his description of reason as striving for the unconditioned. The “need” of the will – and it is important to remember that Kant equates the will with practical reason – is, then, best understood against this background of Kant’s conception of the nature of reason itself. In the Dialectic of the first Critique, Kant provides an account of the ideas of speculative reason as based on reason’s “need” for systematic unity and completeness (KrV Avii, A796/B824). In the Dialectic of the second Critique, the same rational tendency leads to the idea of the highest good. The conception of the highest good does not spring from a sensible need, but, rather, from a tendency that Kant regards as characteristic for reason as such, whether in its speculative or its practical employment.¹³ In sum, on the basis of Kant’s account of the good we can explain Kant’s inclusion of happiness in the highest good – at least when the highest good is conceived as a moral world, which is the conception I focus on in this essay. On the interpretation I have argued for, happiness is morally good insofar as (and only insofar as) moral agents have a duty to promote each other’s happiness. Happiness is included in the highest good because the fully virtuous agents in this “moral world” collectively bring about the happiness of all. On the basis of this interpretation, it is possible to reconstruct a logically valid argument, with premises Kant clearly endorses, in support of his claim that it is a duty to promote the highest good. Finally, this interpretation explains the sense in which

 On Kant’s terminology of ‘needs’ and ‘interests’ of theoretical and practical reason, see Kleingeld a and b.

Kant on ‘Good’, the Good, and the Duty to Promote the Highest Good

49

this duty goes beyond obedience to the Categorical Imperative without increasing the number of our duties.¹⁴

References Beck, Lewis White 1960, A Commentary on Kant’s Critique of Practical Reason, Chicago: University of Chicago Press Beiser, Frederick 2006, “Moral Faith and the Highest Good”, in: Paul Guyer (ed.) 2006, The Cambridge Companion to Kant and Modern Philosophy, Cambridge: Cambridge University Press, 588–629 Engstrom, Stephen 1992, “The Concept of the Highest Good in Kant’s Moral Theory”, Philosophy and Phenomenological Research 52, 747–80 Fugate, Courtney 2014, “The Highest Good and Kant’s Proof of God’s Existence”, History of Philosophy Quarterly 31, 137–58 Kleingeld, Pauline 1995, “What Do the Virtuous Hope For?: Re-reading Kant’s Doctrine of the Highest Good”, in: Hoke Robinson (ed.) 1995, Proceedings of the Eighth International Kant Congress, Memphis 1995, Milwaukee: Marquette University Press, vol. 1.1., 91–112 Kleingeld, Pauline 1998a, “The Conative Character of Reason in Kant’s Philosophy”, The Journal of the History of Philosophy 36, 77–97 Kleingeld, Pauline 1998b, “Kant on the Unity of Theoretical and Practical Reason”, The Review of Metaphysics 52, 500–28 Pistorius, Hermann Andreas 1786, “Rezension der Grundlegung zur Metaphysik der Sitten”, Allgemeine Deutsche Bibliothek 66, 447–63 Reath, Andrews 1988, “Two Conceptions of the Highest Good in Kant”, Journal of the History of Philosophy 26, 593–619 Watkins, Eric 2010, “The Antinomy of Practical Reason: Reason, the Unconditioned, and the Highest Good”, in: Jens Timmerman/Andrews Reath (eds.) 2010, A Critical Guide to Kant’s ‘Critique of Practical Reason’, Cambridge: Cambridge University Press, 145–67 Willaschek, Marcus 2010, “The Primacy of Practical Reason and the Idea of a Practical Postulate”, in: Jens Timmerman/Andrews Reath (eds.) 2010, A Critical Guide to Kant’s ‘Critique of Practical Reason’, Cambridge: Cambridge University Press, 168–96

 I benefited greatly from comments by Thomas Höwing and Jochen Bojanowski. Work on this essay was funded by the Netherlands Organization for Scientific Research (NWO).

Florian Marwede

Kant on Happiness and the Duty to Promote the Highest Good In the Critique of Practical Reason, Kant seems to be making two apparently conflicting claims. On the one hand, he holds that there can be no direct moral duty to promote one’s own happiness (cf. KpV 5:93; MS 6:386). On the other hand, he writes that we have a duty to promote the highest good (KpV 5:125). And since the highest good includes not only a moral agent’s virtue but also her happiness (KpV 5:110), this claim seems to imply that there is a direct duty to promote one’s own happiness after all. How are we to reconcile these two claims? While it seems hard to deny that Kant endorses the first claim, the second leaves some room for interpretation. For example, we might argue that, rather than entailing a duty to promote one’s own happiness, the duty to promote the highest good merely entails a duty to promote the happiness of others. It thus adds nothing new to what the moral law already commands. Or we might defend the view that for Kant the highest good involves not happiness in an ordinary sense, but rather something like moral self-contentment, which necessarily accompanies virtue. As I will suggest, however, these solutions encounter problems of their own and cannot do full justice to Kant’s text (section I). I therefore suggest another solution. I will argue that we can derive a duty from the categorical imperative that is directed at our own happiness considered as part of universal happiness. This is compatible with Kant’s denying that there is a direct duty to promote one’s own happiness in abstraction from the happiness of others, because this denial does not entail that we do not have a duty to promote some greater good that includes one’s own happiness. The pursuit of individual happiness is, according to Kant, a “natural necessity” and thus unavoidable; it must therefore be promoted in some way or another (GMS 4:415). Since we are morally permitted to promote our own happiness only in ways that respect, and where possible advance, the happiness of others, it follows that we are obligated to promote our own happiness as part of universal happiness (section II). This derivation of the duty to promote universal happiness reveals that the dutiful promotion of the highest good includes the promotion of one’s own happiness. Since every morally good end is directed at universal happiness, and since the highest good can be understood as the totality of good ends (cf. KpV 5:108), virtue and universal happiness (including one’s own) are necessarily connected. A

52

Florian Marwede

duty to promote the highest good therefore implies a duty to promote my own happiness (section III).

I Here, in brief, are two ways in which one might respond to the apparent contradiction indicated above. Both accounts would resolve it. However, as I will argue in more detail below, these attempts at reconciliation conflict with other important parts of Kant’s ethics, which brings their ultimate usefulness into question. First, we can accept a duty to promote the highest good without accepting a duty to promote our own happiness by assuming a division of labour. If everyone were to fulfil their duty to promote the happiness of others, everyone’s happiness would be promoted by others, and the highest good as a whole could be promoted without anyone’s promoting their own individual happiness. This amounts to saying that the duty to promote the highest good coincides with the duties of virtue “and that the promotion of these ends exhausts an individual’s contribution to the highest good”.¹ This account does not necessarily exclude virtuous agents who promote their own happiness by merely permissible actions.² Indeed, certain passages seem to support such a reading.³ However, Kant does not relate them directly to the highest good. Their usefulness as textual evidence is therefore limited. Furthermore, there are passages that are related to the

 Engstrom : ; cf. also Denis : .  Beck pointed out in  that it is difficult to understand what the duty to promote the highest good adds to what the moral law already commands. In his view, this is only the distribution of happiness in proportion to virtue. But that “is the task of a moral governor of the universe, not of a laborer in the vineyard” (Beck :  f.). However, the fact that we cannot realize the highest good without divine assistance does not imply that we have no duty to promote it (cf. Wood : ). All that is required is a reasonable way to be morally obligated to promote not only the first component of the highest good (virtue, moral perfection) but also the second (universal happiness). A different approach, which denies the relevance of the duty to promote the highest good, involves distinguishing between an immanent and a transcendent concept of the highest good and then claiming (i) that the former only contains permitted ends of happiness that are morally neutral, and (ii) that the latter consists of holiness and beatitude (Seligkeit). This reading likewise renders such a duty meaningless because a holy will necessarily conforms to the moral law (Mariña ). However, neither this distinction nor the idea of morally neutral happiness is covered in Kant’s text.  “This does not mean that I am thereby under obligation to love myself […]; it means instead that lawgiving reason […] permits you to be benevolent to yourself on the condition of your being benevolent to every other as well” (MS :).

Kant on Happiness and the Duty to Promote the Highest Good

53

highest good and that support the contrary position.⁴ Something similar can be said for considering the pursuit of one’s own happiness as “merely permitted” and for referring to duties that address communities rather than individuals. As I will point out in the next sections, although this reading makes valid points, it still leaves out important aspects and can thus be misleading. Second, one might assume that the happiness that is part of the highest good differs from the happiness we naturally pursue; for instance, it might only consist in the satisfaction that accompanies our own good deeds and the consciousness of keeping our own inclinations sufficiently under control (e. g. Wentscher 1900: 46 f.). Such a “moral happiness” would necessarily accompany virtue. Following this idea, the duty to promote the highest good could be carried out simply by being virtuous. This is reconcilable with Kant’s denial of a direct duty to promote one’s own happiness. However, although Kant may have sympathized with an idea of moral happiness in his pre-critical period, he later clearly rejects it (MS 6:377). What comes closest to a concept of “moral happiness” in Kant’s critical philosophy is his concept of self-contentment. According to Kant, self-contentment is a necessary condition for the happiness of virtuous agents, but it cannot replace happiness in the highest good. Both moral self-contentment and the satisfaction of our desires influence our happiness, but happiness cannot be reduced to either of these alone.⁵ Another concept that might translate to moral happiness is beatitude (Seligkeit). But again, Kant gives an exact definition in the second Critique, describing it as a concept that, although related to happiness, is not to be confused with it. For Kant, beatitude is a kind of well-being that is independent of any sensual desire and is therefore the ultimate self-sufficiency (KpV 5:25). A moral person must be able to reject some of her own desires when they would prevent her from doing the right thing. But there is no further duty to transform one’s striving for happiness into a striving for beatitude. For this reason, we cannot simply resolve this issue by replacing happiness with beatitude in the highest good.⁶

 Cf. e.g. KrV A ff./B ff., where the highest (derived) good is described as a moral world in which everyone is the author of their “own enduring welfare and at the same time that of others”.  “[…] and in fact an upright man cannot be happy if he is not first conscious of his uprightness […]” (KpV :; cf. KpV : f.).  Beatitude is to happiness as holiness is to virtue in the sense that a pure will that is not bothered with sensual desires would be one for which there is no difference between “ought” and “want”; as such, it would be both holy and blessed. In another respect, there is a significant difference between holiness and beatitude: While we ought to strive for the former in an endless progress now and in this life, the latter is something we can only hope for as part of an afterlife (KpV : f.).

54

Florian Marwede

Both accounts outlined here share common problems. If the duty to promote the highest good could be completely reduced either to the duties of virtue or to being virtuous, one would have to question whether this duty has any function of its own. Given the number of passages where Kant claims that there is such a duty, this would be highly unsatisfying.⁷ Furthermore, the postulates of pure practical reason are introduced because the highest good cannot be realized by human agency alone (KpV 5:114 f.). This is one of the essential lines of argument of the Dialectic of the second Critique, and it would be undermined if we were to follow either of the above accounts. In both cases, the duty to promote the highest good can be fulfilled by human agency. Kant does not need the postulates to explain how the duties of virtue can be carried out or how it is possible to be virtuous (which might not be exactly the same thing; cf. e. g. MS 6:394 f.). For this reason, it is important to develop a solution that (i) explains how the highest good can be promoted by human agency and (ii) leaves room for Kant’s argument for the postulates. This is the task of the next section.

II 1. Is there a duty to promote one’s own happiness as part of universal happiness? If so, this duty should be derivable in some way from the categorical imperative, which is the general principle from which all our moral duties are derived. As a starting point for answering this question, we will therefore take a brief look at how Kant derives other duties. In the Groundwork, Kant suggests the following formulation of the categorical imperative: “[A]ct only in accordance with that maxim through which you can at the same time will that it become a universal law” (GMS 4:421).⁸

There is much controversy concerning the precise meaning of this formulation and how Kant’s examples of its application are to be understood in detail. However, there seems to be consensus with regard to the general procedure Kant uses  For the duty to promote the highest good, cf. KpV :, , , , n.; Gemeinspruch :; Verkündigung :, n. Without using the word “duty”, and instead using similar words like “demand”, “command”, “obligation” etc., Kant mentions the concept e. g. in KpV :, , , ; KU :; Ton :.  In the scope of this essay, I restrict consideration of the categorical imperative to this formula. Cf. Kant’s claim that the different formulations express “the very same law” (GMS :). Cf. also Stephen Engstrom’s approach to defending this claim (Engstrom : –).

Kant on Happiness and the Duty to Promote the Highest Good

55

in most of his examples.⁹ He suggests a thought experiment to the effect that the maxim at issue holds as a universal law. (This thought experiment is often referred to as the universalizability test.) If this thought experiment involves a “practical contradiction” (such that acting on that maxim, including the intention implicit in it, cannot be willed consistently, or even conceived of, once the maxim is universalized), the maxim is morally wrong and must not be acted upon.¹⁰ For example, in a world where everyone adopted the maxim “I shall borrow money by means of promises I cannot keep”, nobody would trust a given promise (GMS 4:422). Consequently, one cannot consistently intend to act on this maxim and will its universalization at the same time. In this way, it can be shown that the maxim at issue does not qualify as a universal law and that acting on it is morally impermissible.¹¹ We can call maxims that pass this test “universalizable”. Even though Kant does not explicitly say so, it is widely assumed that it is morally permissible to act on a maxim if and only if it passes the universalizability test. Moreover, it is assumed that there are such maxims.¹² Moral duties arise from considering maxims that are not universalizable. Given a maxim M1, the adoption of which requires you to F, we can call the maxim M2 that requires you not to F its “practical opposite” (and vice versa). It is a moral duty to adopt all maxims the practical opposites of which are not universalizable. For example, the maxim “I shall borrow money by means of promises I cannot keep” does not qualify as a universal law. It is therefore a moral duty not to borrow money by means of promises one cannot keep (cf. GMS 4:422). Consequently, in order to establish a duty to promote one’s personal happiness as part of the highest good, we need to show that a maxim expressing its practical opposite does not qualify as a universal law. 2. In the second Critique, Kant adds to this account of the universalizability procedure the thought that a maxim qualifies as a universal law if and only if its form is sufficient to determine the will (KpV 5:27). The idea behind this criterion

 Cf. e. g. O’Neill :  ff.; Timmons : ; Horn/Mieth/Scarano :  ff.  The term “practical contradiction” is favoured by Korsgaard (:  ff.) to designate the specific type of contradiction that is used here. In the scope of this essay, I will leave out the distinction between “contradiction in conception” and “contradiction in willing” (cf. e.g. O’Neill : – ).  The same procedure is also used to prohibit the maxims “to increase my wealth by every safe means” (KpV : f.) and “to shorten my life when its longer duration threatens more troubles than it promises agreeableness” (GMS : f.).  Cf. references in footnote .

56

Florian Marwede

can be very briefly sketched as follows. All maxims have a form and a matter. The matter consists in the end that is to be achieved. The form is a structural feature that derives from the ultimate principle that guides the will in adopting its maxims. If that principle is the moral law, the form of the maxim is that of universal validity; if the principle is that of maximizing one’s own happiness, the maxim lacks the form of universal validity.¹³ In the former case, the adoption of the maxim entails that its form is sufficient to determine the will; one acts on the maxim only because of its universal validity, irrespective of whether the end serves one’s own well-being. In the latter case, the matter determines the will, because one adopts the maxim only if acting upon it promises the realization of an end one happens to have. The form is not sufficient to determine the will in this case because the adoption of the maxim depends on the matter. Since it is possible for these two grounds of determination to come into conflict, either the form or the matter alone determines the will.¹⁴ 3. Nevertheless, a maxim that qualifies as a universal law still has a matter. But how are we to conceive of willing an end without allowing it to determine our will – that is, without making the pursuit of that end a necessary condition for acting on our maxim? This is the subject of a passage in the second Critique, which argues that my own happiness can be morally willed as an end only if I “include in it” the happiness of others (KpV 5:34). Because it is central to solving the problem of how there can be a duty to promote one’s own happiness, we will analyse this passage in detail in what follows. Let me begin by citing the whole passage: (1) “Thus, the happiness of other beings can be the object of the will of a rational being.

 Kant focuses on the form that makes a maxim “fit for a giving of universal law” (KpV :) and refrains from explaining the form of other maxims. Andrews Reath suggests that we fill this gap with the principle of happiness (Reath ; cf. Reath :  f.). Nevertheless, the passages where Kant explains the form of maxims are ambiguous. On the one hand, a maxim is only a subjective principle of volition (GMS :, ; KpV :) and is therefore only valid for the subject. On the other hand, the form of all maxims “consists in universality” (GMS :). To take both characterizations adequately into account, we might say that a maxim as such is “universal” insofar as the subject judges that to act on it is good all relevant things considered and insofar as it would be plausible, from the subject’s point of view, for everyone in similar conditions to act on it. The universality of a law differs from this because it is independent of any conditions of the subject (for a similar approach, see Engstrom : ). Cf. also Allison :  f.  Kant develops this line of thought in Theorems I to III in the Analytic of the second Critique (cf. KpV :, , ). Kant claims earlier in the Groundwork that maxims consist of “form” and “matter”, but these terms do not lie at the heart of his theory of the determination of the will (cf. e. g. GMS :).

Kant on Happiness and the Duty to Promote the Highest Good

57

(2) But if it were the determining ground of the maxim, one would have to presuppose that we find not only a natural satisfaction in the well-being of others but also a need, such as a sympathetic sensibility brings with it in human beings. (3) But I cannot presuppose this need in every rational being (not at all in God). (4) Thus the matter of the maxim can indeed remain, but it must not be the condition of the maxim since the maxim would then not be fit for a law. (5) Hence the mere form of a law, which limits the matter, must at the same time be a ground for adding this matter to the will but not for presupposing it. (6) Let the matter be, for example, my own happiness. (7) This, if I attribute it to each (as, in the case of finite beings, I may in fact do) can become an objective practical law only if I include in it the happiness of others. (8) Thus the law to promote the happiness of others arises not from the presupposition that this is an object of everyone’s choice but merely from this: that the form of universality, which reason requires as the condition of giving to a maxim of self-love the objective validity of a law, becomes the determining ground of the will; (9) and so the object (the happiness of others) was not the determining ground of the pure will; (10) this was, instead, the mere lawful form alone, by which I limited my maxim based on inclination in order to afford it the universality of a law and in this way to make it suitable for pure practical reason; (11) only from this limitation, and not from the addition of an external incentive, could there arise the concept of obligation to extend the maxim of my self-love to the happiness of others as well” (KpV 5:34 f.; numbering mine).

The first part (1–3) of our passage gives an example in which the will is determined by the matter of a maxim, and to this extent the maxim does not qualify as a universal law. The second part (4–7) gives an example in which the will is determined by the form of a maxim; consequently, this maxim does qualify as a universal law. The third part (8–11) explains how this line of thought is the basis of the justification of the duty to promote the happiness of others. It will not be possible to do full justice to this complex line of thought here. Its core idea seems to be that the moral obligation to promote the happiness of others cannot be grounded in sympathy for others, because we cannot presuppose that everyone has this sentiment. Nor can it be grounded in the recognition that we all pursue our own happiness; since my maxim would differ from everyone else’s in respect of having me as its subject, it cannot serve as a basis for a universal law. Therefore, the obligation to promote others’ happiness can only be grounded in the twofold recognition that (i) I seek my own happiness and (ii) everyone else likewise seeks their own happiness. In order to be able to think of my maxim (to

58

Florian Marwede

promote my own happiness) as a universal law, it is not sufficient that I grant to others the possibility of promoting their own happiness. Rather, I need to formulate a maxim that can hold equally for everyone, which is the maxim of promoting universal happiness (including my own).¹⁵ In the following, I will analyse more closely how this last step of the argument works. 4. The demand to include the happiness of others in the matter of my maxim is clearly based on the categorical imperative.¹⁶ Nevertheless, Kant’s procedure here differs from the one described in the Groundwork. The latter asks us to universalize a certain maxim; if the maxim fails the test, we know that it is morally impermissible. If the maxim passes the test, however, it seems that we are permitted to act on it. The passage from the second Critique also asks us to universalize a certain maxim (that of promoting only my own happiness); unlike the procedure from the Groundwork, however, where our original maxim fails to be universalizable, we then replace it with its universalized counterpart (that of promoting universal happiness). How can these two versions of the universalizability test be reconciled? This question becomes more pressing once we recognize that Kant’s argument in the above passage is perfectly general. It holds not only for the ends of individual and universal happiness, but for any end one might have. Even if the end could be presupposed to hold for everyone (which, according to Kant is true only for happiness), this in itself would not make a maxim of promoting it an “objective practical law”, since the holding of this end is compatible with each person’s promoting only their own happiness. For example, even if everyone were interested in physical fitness, this would not make the maxim “to exercise in order to become, or stay, physically fit” a universal law; in acting on it, people would be pursuing only their own fitness, which means that their maxims would hold for them alone and not universally. Assuming that it is morally permissible to exercise in order to remain physically fit, how then are we to derive the relevant permission on the model of the above argument from KpV 5:34? 5. What is more, is it morally permissible to exercise in the first place? Kant’s point seems to be that it depends. Consider someone who exercises while being aware that this involves refusing to help others in need. Presumably, in  Already in , Kant had the idea of a virtuous agent, who “would certainly love and value even himself, but only in so far as he is one among all to whom his widespread and noble feeling extends himself” (Beobachtungen :).  This is indicated by the criterion this argument tries to fulfil. Cf. “be fit for a law” in step  and “become an objective practical law” in step .

Kant on Happiness and the Duty to Promote the Highest Good

59

this case exercising will be morally impermissible. And indeed, the maxim “I will exercise whenever it is convenient for me” could hardly be willed as a universal law (given that this could result in the breaking of promises, a failure to help those in need, etc.). Obviously, this does not mean that exercising is morally impermissible. It only means that the maxim in question has to be specified further by including one’s own moral obligations and other people’s concerns and interests. Before we take a closer look at this specification, let us briefly consider how this relates to the categorical imperative procedure outlined in the Groundwork. Practical inconsistency (that is, failure to satisfy the universalizability test) is a sufficient condition for moral impermissibility. And the practical opposite of an impermissible maxim defines a moral duty. But this in itself does not tell us how to promote a contingent (not morally necessary) end in a morally permissible way. If exercising whenever convenient is morally impermissible, there will be instances where, although it would be convenient for me, I am under a duty not to exercise. But this does not tell me when (if at all) it is permissible for me to exercise in order to improve my fitness. In other words, rather than telling me which maxim to adopt, it tells me only which maxim not to adopt. This is why Kant’s derivation of duties from the categorical imperative in the Groundwork is not accompanied by positive examples of universalizable maxims (GMS 4:421 ff., 429 ff.; also cf. KpV 5:27 f.).¹⁷ The thought that must be added in order to arrive at universalizable maxims is contained in the passage currently under discussion. According to this passage, the egoistic pursuit of individual happiness, which is Kant’s starting point in step (6), must be transformed into a pursuit of universal happiness, since only in this way can it become the matter of a universalizable maxim – that is, one in which the promotion of one’s own happiness is permissible. 6. How should we conceive this transformation? The answer to this question must take into account Kant’s distinction between narrow and wide duties. Narrow duties apply only to highly specific situations. If you are about to commit suicide in despair, the corresponding duty tells you very specifically what

 Furthermore, Kant speaks only in a negative and limiting way when it comes to the selection of maxims. Cf. e. g.: “In accordance with this principle all maxims are repudiated that are inconsistent with the will’s own giving of universal law” (GMS :). Affirmative expressions are used with regard not to maxims but to duties, which arise from repudiated maxims. Cf. e. g.: “These are a few of the many actual duties, or at least of what we take to be such, whose derivation from the one principle cited above is clear” (GMS : f.).

60

Florian Marwede

(not) to do (GMS 4:421 f.). But if you do not contemplate suicide, this duty does not affect you in any way. By contrast, wide duties can affect your course of action to different degrees. For example, the duty of beneficence does not always require you to seek out drowning people and risk your life to save them. For the most part, what it requires you to do are things like offering advice to a friend or, more subtly, holding back on the giving of advice because you notice that it would help your friend even more to work it out herself.¹⁸ However, wide duties are not limited to specific situations. The specification “wide” is not to be taken as permission to make exceptions whenever it is convenient (MS 6:390). This is the other side of the coin when it comes to wide duties: they are “wide” not only in the sense of allowing for discretion with regards to how and to what extent we obey them, but also in the sense of being relevant always and everywhere, not just in specific situations, thus pervading all our decisions and actions.¹⁹ Given that the duty to promote other people’s happiness is a wide duty, it follows that the described transformation of my original maxim into a maxim with universal validity, and thus the inclusion of the happiness of others in it, has to affect every maxim I act on. But since it is a wide duty, the extent to which one’s maxims are affected is not objectively circumscribed. There are two different ways in which this might be spelled out. On the one hand, some passages suggest that this is only a duty to help or assist others in need or hardship²⁰ and that “a maxim of promoting others’ happiness at the sacrifice of one’s own happiness, one’s true needs, would conflict with itself if it were made a universal law” (MS 6:393). These passages suggest that there is a difference between the way a virtuous person promotes her own happiness and the way she promotes the happiness of others. The former is limited by the moral law, but not at the price of sacrificing “one’s true needs”; the latter consists in helping others, but only where they cannot help themselves.

 In regard to wide duties, there is “a playroom (latitudo) for free choice in following (complying with) the law, that is, that the law cannot specify precisely in what way one is to act and how much one is to do by the action for an end that is also a duty” (MS :).  Of course, this does not mean that every action must contribute to the happiness of others. Kant’s theory of duties is first and foremost located at the level of maxims. For this reason, it is at least misleading to claim that wide duties are only to be carried out “sometimes”, as Hill does (:  ff.).  It is about helping others who “have to contend with great hardships”, because one cannot will to rob oneself “of all hope of the assistance” in times of personal hardship (GMS :; cf. MS :, ).

Kant on Happiness and the Duty to Promote the Highest Good

61

On the other hand, in the passage from the second Critique under discussion (KpV 5:34), Kant does not limit the inclusion of the happiness of others in one’s maxim in any way. In the Groundwork, too, Kant states that “the ends of a subject who is an end in itself must as far as possible be also my ends” (GMS 4:430). Furthermore, in the Metaphysics of Morals Kant explicitly claims that the only limitation that comes with a wide duty is the permission to “limit one maxim of duty by another (e. g., love of one’s neighbour in general by love of one’s parents)” (MS 6:390). These passages suggest that the virtuous person does not make any distinction between her own happiness and the happiness of others. According to these passages, the end of one’s own happiness converges completely with the end of universal happiness. How can these two notions be reconciled? I suggest that they actually function on different levels. On a general level, I have to respect another person as an end in herself and as my equal (GMS 4:429). I therefore have to make her ends my own ends as far as possible. However, this is limited not only by other duties but also by the fact that everyone has to promote their own happiness fundamentally on their own. For example, I can only encourage a friend to exercise, but I cannot do it for her. It follows that, at the level of choice of concrete ends and actions, I consider my own ends differently from those of others – not because I’m morally entitled to prefer my own ends, but because of the very nature of human agency and the pursuit of happiness.²¹ 7. If we now return to the original problem, we can use the results of this analysis to advance one step ahead on the way to a solution, for Kant’s way of deriving the “transformed” universally valid maxim (which includes the happiness of others in it) is to show that the opposite maxim does not qualify as a universal law. In order to see this, let us take a closer look at steps 6 and 7 of Kant’s argument. Kant first considers a maxim that has “my own happiness” as its matter. Kant does not explicitly formulate this maxim, but he gives us a hint. He says that the condition for becoming an objective practical law, which is not satisfied by this maxim, is the inclusion of the happiness of others. Thus, the original maxim does not include the happiness of others. Taking this into account, we can formulate the maxim as follows: I shall promote my own happiness without considering the happiness of others.

 Cf. Engstrom :  f. for a similar account.

62

Florian Marwede

This maxim does not pass the test of the categorical imperative. This can be shown by applying the procedure laid out in the Groundwork. The maxim entails that I judge my happiness but not the happiness of others to be good. For if I were not to consider the happiness of others, I would harm them whenever necessary for the promotion of my own happiness. Others’ happiness would have no value to me. If this maxim were a universal law, others would judge their own happiness (but not mine) to be good. If I were to agree to this and maintain my former judgment, I would and would not deem everyone’s happiness good at the same time, which would amount to a practical contradiction.²² Because the maxim does not pass the test, we can derive a duty to act on the opposite maxim: Promote your own happiness only in conjunction with promoting the happiness of others. This formulation is ambiguous, however, because it does not clarify the precise way in which my efforts to promote others’ happiness must relate to my efforts to promote my own happiness. I suggested in the previous subsection that we have to value the ends of others as being equal to our own on the more general level, whereas the nature of human agency implies a mere duty to help others in need at the level of concrete actions. Since we are dealing with morality on a very general level here, this duty must be read in terms of treating others’ ends as equal to our own. This argument can be further confirmed by the following thought. The maxim “I judge my happiness to be more important than others’ happiness” does not necessarily seem to conflict with the duty as formulated above. However, I cannot agree to a universalized version of this maxim, because this would involve my judging that the happiness of everyone is more important than the happiness of everyone. This reading would thus result in a practical contradiction as well (cf. Engstrom 2009: 212 f.). If we want to prevent this misunderstanding, we might reformulate the duty as follows:

 This is Stephen Engstrom’s reconstruction of Kant’s argument for the duty of benevolence in GMS : (Engstrom : ). Two comments are worth mentioning here: (i) In my opinion, this reconstruction fits better with Kant’s argumentation in the second Critique because the agent in the Groundwork thinks “I shall take nothing from him nor even envy him” (GMS :). This implies that the agent judges the happiness of others to be good to at least some degree in contrast to a person who promotes his own happiness whatever the harm to others. (ii) Note that this reconstruction is not uncontroversial. Cf. e.g. O’Neill :  ff., where it is argued that a certain form of means-end reasoning is necessary to demonstrate the contradiction.

Kant on Happiness and the Duty to Promote the Highest Good

63

Promote your own happiness only as part of universal happiness. Here, we have almost achieved our aim. We have shown that Kant develops, in addition to the duty to promote the happiness of others, a further duty that is closely connected to it. This gives the personal happiness of the agent a proper place in moral conduct. We have also shown that this amounts to more than a demonstration of permissibility (the permissibility of promoting personal happiness as part of universal happiness); because the practical opposite of this principle is forbidden, it is a genuine duty. 8. One part of the solution is still missing, though. For it may seem possible to comply with the duty to promote one’s own happiness only as part of universal happiness by refraining from promoting one’s own happiness altogether. After all, the logical structure of this duty is conditional: If you promote your own happiness, then do so only as part of universal happiness. If we compare the duty at issue with the prominent examples of duties derived in accordance with the universalizability test, it is striking that the duties which arise are also conditional: “Do not increase your wealth by every safe means”, “Do not borrow money by means of promises you cannot keep”, etc. These duties might still be valid for someone who is not in need of money, etc., but they do not affect him. We can, for example, reformulate the second duty as follows: “Either do not borrow money or keep the related promises”. We can reformulate the duty developed here in a similar way: Either do not promote your own happiness or promote your own happiness as part of universal happiness. However, it is impossible not to promote one’s own happiness at all. For according to Kant, the pursuit of personal happiness is “unavoidable” for human beings (KpV 5:25; cf. RGV 6:45n.; MS 6:387). He claims that this orientation towards one’s own happiness can be traced back to the fact that the human being is a “rational but finite” being.²³ This is “a problem imposed upon him by his finite nature itself” (KpV 5:25).²⁴ The expression “finite nature” alludes to the fact that human beings are continually confronted with needs that cause pain as long as they are unsatisfied. And nature is not constituted in such a way that satisfaction  “To be happy is necessarily the demand of every rational but finite being and therefore an unavoidable determining ground of its faculty of desire” (KpV :; cf. GMS : f.).  Cf. the connotation of happiness as a “natural end that all human beings have” (GMS :) and as “the end assigned us by nature itself” (Gemeinspruch :).

64

Florian Marwede

is always granted. Moreover, these needs might change or come into conflict with each other. For this reason, human beings must deal with unsatisfied needs all the time. Prudence requires us to systematize our needs such that our actions contribute to their maximum satisfaction in the given circumstances. Of course, Kant does not have a particular conception of happiness in mind; he refers only to the idea of a satisfaction of needs that is as complete as possible, regardless of what these needs might be.²⁵ Kant assumes that these needs can be traced back to the sensibility of human beings. Thus the concept of happiness is defined by means of the concept of inclination in all three Critiques and in the Groundwork. ²⁶ According to these definitions, happiness is the satisfaction of all inclinations, where an inclination is a “[h]abitual sensible desire” (Anth 7:251; cf. GMS 4:413n.). However, Kant does not reduce the human striving for happiness to the demand of physically experienced pleasure. As Kant points out in the second Critique, what matters is not whether a need is based on a sensual or an intellectual representation, but whether the realization of the end provides pleasure. In this manner, he can also count activities such as reading an interesting book, having a lively conversation, and giving to charity among those done for the sake of happiness – assuming they are done to provide pleasure.²⁷ All ends which we actually promote can therefore be subsumed under the pursuit of personal happiness.²⁸ But if it is impossible not to seek one’s own hap-

 Cf. e. g.: “[…] [I]n what each has to put his happiness comes down to the particular feeling of pleasure and displeasure in each and, even within one and the same subject, to needs that differ as this feeling changes […]” (KpV :).  Cf. KrV A/B; KpV :; KU :n.; GMS :.  “The same human being can return unread an instructive book that he cannot again obtain, in order not to miss a hunt; he can leave in the middle of a fine speech in order not to be late for a meal; he can leave an intellectual conversation, such as he otherwise values highly, in order to take his place at the gaming table; he can even repulse a poor man whom at other times it is a joy for him to benefit because he now has only enough money in his pocket to pay for his admission to the theatre. If the determination of his will rests on the feeling of agreeableness or disagreeableness that he expects from some cause, it is all the same to him by what kind of representation he is affected” (KpV :). For the complex connections Kant develops between pleasure, desire, and reason, cf. e. g. Reath , Engstrom , and Höwing .  Note that actions that are only done from duty and that go against all inclinations are not counterexamples. For, as we have seen, narrow duties do not provide ends to promote but rather tell us what not to do. And wide duties are already balanced – that is, they already consider our own interests. While it might be the case that a virtuous agent sometimes acts in a way he does not find at all enjoyable, the underlying maxim can only be one that promotes universal happiness, including his own.

Kant on Happiness and the Duty to Promote the Highest Good

65

piness, we can move from the conditional duty formulated above to the following unconditional duty: Do promote your own happiness as part of universal happiness.

III In section II, I argued that personal happiness can be part of an end the pursuit of which is a duty and that this need not conflict with Kant’s claim that there cannot be a direct duty to promote one’s own happiness. This is possible because the moral law requires us to transform the pursuit of personal happiness into a pursuit of universal happiness. We can further illustrate this point by considering the example of the maxim “to exercise in order to become, or stay, physically fit”. Let us assume that it has become my habit to exercise after work. One day, a co-worker asks me to stay at work longer to help him with some urgent task he has been assigned but is running behind on. While the pursuit of personal happiness favours a solution that maximizes my own well-being, a pursuit of universal happiness favours a solution that considers all interests equally as far as possible. Therefore, a possible solution would be to stay for some limited time (say, an hour) to help my co-worker, and to exercise afterwards.²⁹ A clarification is in place here. On various occasions, Kant accompanies his claim that there cannot be a direct duty to promote one’s own happiness with the claim that there is an indirect duty to promote it (GMS 4:399; KpV 5:93; MS 6:388). This indirect duty is based on the idea that some degree of happiness is necessary in order to be able to do one’s moral duty. For instance, dire poverty may make it difficult, or even impossible, to respect the moral law. One might ask how the duty developed here relates to this indirect duty. This can be answered by further considering the example of helping the co-worker. In general, I have a duty to find a balance between the interests of the parties concerned. This is the duty developed here. But there are moral criteria beyond consideration of these interests that need to be satisfied. One of these is, obviously, that my maxim must pass the universalizability test. Another criterion is that, in the long term, my “share of happiness” in the chosen course of action must not be so small that it tempts me to transgress my duties. For instance, if it turns out  Cf. Wood :  ff. for a detailed example of a housing reformer who develops an idea of universal happiness, which is nevertheless rooted in his own interests. This example might be a valid interpretation of KpV :, but it would be misleading to assume that the end of one’s own happiness must converge completely with the end of universal happiness, as in this case.

66

Florian Marwede

that I am helping my co-worker night after night, and as a result never find the time to exercise, this might in the end undermine my motivation, or even my ability, to help him. Thus, there is a duty not to forget my own concerns and needs. And this is the indirect duty Kant refers to in the passages indicated above. So this indirect duty simply articulates one specific aspect of what it means to promote one’s own happiness as part of universal happiness. Furthermore, one might object that, alongside denying that we have a direct duty to promote our own happiness, Kant sometimes explicitly claims that we are only permitted, and not obligated, to promote our own happiness (or, as Kant calls it, to be benevolent to ourselves).³⁰ We can answer this objection as follows: When Kant says that there is no direct duty to promote our own happiness, or that we are not obligated to be benevolent towards ourselves, what he wants to say is that such a duty or obligation would be pointless, since according to Kant one can’t be obligated to do something one cannot avoid doing anyway. This is not to deny, however, that one can be obligated to do in a specific way what one cannot avoid doing – in this case, pursuing one’s happiness as part of universal happiness. Now we might worry that this obligation concerns only the way in which one promotes one’s own happiness rather than the pursuit itself. But this underestimates the transformation that happens to the pursuit of personal happiness when one includes other people’s happiness in it. Pursuing my own happiness as part of universal happiness involves more than pursuing my happiness under some external restriction. Rather, it transforms the content of my pursuit. It means that I make other people’s ends my own, thus finding part of my own happiness in their achievement of their ends and in their becoming happy. In this sense, when it comes to those who promote their own happiness as part of universal happiness, we cannot specify what they will pursue in the name of their own happiness independently of the way in which this is done (namely as part of universal happiness), and vice versa. (This is a lesson from our description of what it means to obey a wide duty.) But if this is true, it does not make any sense to say that our duty concerns only the way in which we are to promote our happiness and not the pursuit

 “[…] But since all others with the exception of myself would not be all […] the law making benevolence a duty will include myself […]. This does not mean that I am thereby under obligation to love myself (for this happens unavoidably, apart from any command, so there is no obligation to it); it means instead that lawgiving reason […] includes me as giving universal law along with all others in the duty of mutual benevolence, in accordance with the principle of equality, and permits you to be benevolent to yourself on the condition of your being benevolent to every other as well; for it is only in this way that your maxim (of beneficence) qualifies for a giving of universal law […]” (MS :).

Kant on Happiness and the Duty to Promote the Highest Good

67

itself. We can conclude that the duty to “promote your own happiness as a part of universal happiness” is indeed part of Kant’s ethics. We can now clarify how this insight relates to our question concerning the highest good. We have seen that the procedure outlined at KpV 5:34 is the only way to derive positive moral ends and maxims we can act on. Without it, we can only determine which maxims are morally impermissible. We can say the following about each of these moral ends: First, they aim at universal happiness. Second, my own happiness is part of universal happiness and thus must be included in this end, not just because I am one of those others who must be taken into account, but because this new end is based on an end that I developed to satisfy a personal need. Third, because it is a duty to transform each and every one of my maxims so as to include universal happiness, every single end promoted by the virtuous agent combines individual and universal happiness. The highest good can be understood as the totality of morally good ends to be promoted by virtuous agents (e. g. KpV 5:108). On this picture, it is plausible to say that to comply with the duty to promote one’s own happiness as a part of universal happiness just is to promote the highest good.³¹ When compared with the standard interpretation, which views the categorical imperative as a merely negative criterion that “forbids” some maxims and “permits” others, our reading has a distinct advantage. For the standard view must assume that a virtuous agent promotes two different projects. On the one hand, there is the “moral project” that consists in carrying out the duties that arise on the basis of forbidden maxims. On the other hand, there is the “project of one’s own well-being”, which consists in acting on permissible maxims. When it comes to the highest good, this interpretation is forced to conclude that it combines two different kinds of ends, “systematizing” them and “subordinating” the second kind to the first.³² There are two problems with this view. First, Kant’s theory of the highest good does not introduce any new rules for systemizing these ends. The one rule that we might appeal to – that morality always tops any other end I

 It also sheds some light on passages where Kant seems to claim that universal happiness is the aim of morality (cf. KrV A f./B f., A/B; Refl , :; Refl , :; Refl , : f.; Met-L :).  “A conception of the Highest Good is constructed by systematizing different kinds of ends, but as they are structured by the Moral Law. The Moral Law combines these two kinds of ends into a single scheme by subordinating the natural to the moral” (Reath : ). If I understand him correctly, Reath has a two-step procedure in mind: (i) the moral law authorizes two kinds of ends, demanded or moral ends and permitted or natural ends; (ii) the moral law combines and systematizes both kinds of ends within the highest good.

68

Florian Marwede

might have – leaves entirely unclear when I am allowed to do something for my own well-being at all. Second, it is not clear how there can be a necessary connection between virtue and happiness in the highest good since, so described, what I am morally obligated to do and what I am permitted to do are two totally separate things.³³ By contrast, the reading developed here has the consequence that a virtuous agent promotes only one project. The subordination that the standard reading locates in the concept of the highest good has already taken place in the determination of the will of that agent by the moral law. To the extent that they are virtuous, agents already intend to bring about universal happiness, including their own. This means that there is already a necessary connection between virtue and happiness in the will (and, potentially, in every single end) of the virtuous agent.³⁴ Of course, virtuous agents can fail in their attempts to bring about universal happiness due to external factors beyond their control. For this reason, the proportionality between virtue and happiness within the highest good (and a necessary connection between virtue and happiness in the world) is not yet explained by the above considerations (e.g. KpV 5:110). This is no flaw of this proposal, however, for it merely points the way toward further discussion of important aspects of the highest good, including the antinomy of practical reason (KpV 5:113 f.), the postulates (KpV 5:122ff.), and the notion, hinted at by Kant, that there is something in the highest good that exceeds the moral law (e.g. RGV 6:6n.).³⁵

 Consequently, Reath rejects such a connection: “It implies no necessary connection between virtue and happiness, but instead describes the Highest Good as a union of two distinct ends, one of which is subordinate to the other. The first would be the moral perfection of all individuals, and the second the satisfaction of their permissible ends” (Reath : ).  There are several similarities to Kleingeld’s account (:  ff.). First, she also notes that the categorical imperative is often seen as only a “maxim-filtering device” and that the passage analysed here (KpV :) changes this picture: “it would also follow that one’s own happiness and the happiness of others are interdependent”. But she does not draw the conclusion that this is also a solution to the problem of the somewhat artificial separation of “permissible” and “morally required” ends. Guyer (:  ff.) works out an account of a duty to promote universal happiness that is similar to the line of thought presented here. However, he does not relate it to the passage in KpV :. In contrast, he states that the “distinction between the selfish end of one’s own happiness and the unselfish end of happiness in general is not clearly stated in the Critique of Practical Reason […]” (Guyer :  note ). Cf. also Denis (:  f.).  Versions of this essay were presented at the conference “The Highest Good in Kant’s Philosophy” at Frankfurt University in September of  and to the Department of Philosophy at UCSD in May of , and I thank members of these audiences for helpful comments and discussion, especially Craig Agule, Claudia Blöser, Michael Hardimon, Thomas Höwing, Andrews Reath, and Marcus Willaschek. The work on this essay was supported by the DFG research project “Der Begriff des höchsten Guts in der Philosophie Kants”. This help is gratefully acknowledged.

Kant on Happiness and the Duty to Promote the Highest Good

69

References Allison, Henry 2011, Kant’s Groundwork for the Metaphysics of Morals: A Commentary, Oxford: Oxford University Press Beck, Lewis White 1960, A Commentary on Kant’s Critique of Practical Reason, Chicago: The University of Chicago Press Denis, Lara 2005, “Autonomy and the Highest Good”, Kantian Review 10, 33–59 Engstrom, Stephen 1992, “The Concept of the Highest Good in Kant’s Moral Theory”, Philosophy and Phenomenological Research 52, 747–80 Engstrom, Stephen 2007, “Kant on the Agreeable and the Good”, Poznan Studies in the Philosophy of the Sciences and the Humanities 94, 111–60 Engstrom, Stephen 2009, The Form of Practical Knowledge: A Study of the Categorical Imperative, Cambridge/Mass.: Harvard University Press Guyer, Paul 2000, Kant on Freedom, Law, and Happiness, Cambridge: Cambridge University Press Hill, Thomas E., Jr. 1992, Dignity and Practical Reason in Kant’s Moral Theory, Ithaca/New York: Cornell University Press Horn, Christoph/Mieth, Corinna/Scarano, Nico (eds.) 2007, Immanuel Kant: Grundlegung zur Metaphysik der Sitten, Berlin: Suhrkamp Höwing, Thomas 2013, Praktische Lust: Kant über das Verhältnis von Fühlen, Begehren und praktischer Vernunft, Berlin/Boston: de Gruyter Kleingeld, Pauline 1995, “What Do the Virtuous Hope For? Re-reading Kant’s Doctrine of the Highest Good”, in: Hoke Robinson (ed.) 1995, Proceedings of the Eighth International Kant Congress, Volume I, Part I, Sections 1–2, Milwaukee: Marquette University Press, 91–112 Korsgaard, Christine M. 1996, Creating the Kingdom of Ends, Cambridge: Cambridge University Press Mariña, Jacqueline 2000, “Making Sense of Kant’s Highest Good”, Kant-Studien 91, 329–55 O’Neill, Onora 1989, Constructions of Reason: Explorations of Kant’s Practical Philosophy, Cambridge: Cambridge University Press Reath, Andrews 1988, “Two Conceptions of the Highest Good in Kant”, Journal of the History of Philosophy 26, 593–619 Reath, Andrews 2006, Agency and Autonomy in Kant’s Moral Theory, Oxford: Oxford University Press Reath, Andrews 2010, “Formal Principles and the Form of Law”, in: Andrews Reath/Jens Timmermann (eds.) 2010, Kant’s Critique of Practical Reason: A Critical Guide, Cambridge: Cambridge University Press, 31–54 Timmons, Mark 2006, “The Categorical Imperative and Universalizability”, in: Christoph Horn/Dieter Schönecker (eds.) 2006, Groundwork for the Metaphysics of Morals, Berlin: de Gruyter, 158–99 Wentscher, Max 1900, “War Kant Pessimist?”, Kant-Studien 4, 32–49 Wood, Allen W. 1970, Kant’s Moral Religion, Ithaca, NY: Cornell University Press

Birgit Recki

“Mixtum Compositum”: On the Persistence of Kant’s Dualism in the Doctrine of the Highest Good In the texts of his critical philosophy, Kant does not have a univocal understanding of the concept of the highest good. In the second section of the Groundwork to the Metaphysics of Morals, for example, we encounter the “concept of God as the highest good” (GMS 4:408 f.; my emphasis), which corresponds to the discussion of the “highest original good” in the Canon of the Critique of Pure Reason (KrV A814/B842; my emphasis).¹ Alongside this characterization is the conception of the highest good as a mixtum compositum – a mixed compound – which Kant develops in the second chapter of the Dialectic of Pure Practical Reason in the Critique of Practical Reason. In this paper, I shall be concerned with the latter conception of the highest good. With this said, since the highest good is conceived both here and in the Canon of the first Critique as a harmony between happiness and the worthiness to be happy (as the explanatory term for the concept of virtue), and since such harmony is described in both texts as being dependent on God, occasional reference to the Canon will nonetheless be helpful. This paper aims, first, to determine the legitimacy of the human claim to happiness in Kant’s doctrine of the highest good, and thereby also to bring to light the continuity between the view expressed here and key claims on the same topic voiced both earlier and later in his moral theory.² Second, this paper brings to light a systematic problem: that and how the dualism pres-

 Kant also speaks here of the “ideal of the highest good”, by which he means “the idea of such an intelligence, in which the morally most perfect will, combined with the highest blessedness, is the cause of all happiness in the world, insofar as it stands in exact relation with morality (as the worthiness to be happy)” (KrV A/B; my emphasis). Once again, he here refers to God. With reference to this last specification of an ideal of the highest good, Kant excludes – not at first glance but upon further reflection – a possible way of reading a genitivus subjectivus. With sufficient clarity, he distinguishes between the ideal of the highest good, captured by the concept of God (or more specifically the kind of state that we should attribute to God) and the highest good itself. This is specified here just as it is in the relevant section of the Critique of Practical Reason: although the concept of God plays a decisive role in the solution to the problem that Kant addresses with his concept of the highest good as a mixtum compositum, I do not wish to deal with the concept of God as the highest good in any further detail.  On this, see the contextualization of Kant’s critique of knowledge and aesthetics in Recki .

72

Birgit Recki

ent in the doctrine of the highest good and in the related doctrine of the postulates solves one problem only at the expense of creating another.

1 Virtue and Individual Happiness As Kant writes in the Canon of Pure Reason: “Thus happiness in exact proportion with the morality of rational beings, through which they are worthy of it, alone constitutes the highest good of a world into which we must without exception transpose ourselves in accordance with the precepts of pure but practical reason” (KrV A814/B842). At the level of content, Kant defines the highest good in the same way in the Critique of Practical Reason. There, we learn that “the highest” can refer to either that which is supreme (supremum) or that which is complete (consummatum). In accordance with the clarifications made in the Analytic of the second Critique, he then goes on to claim “[t]hat virtue (as worthiness to be happy) is the supreme condition of whatever can even seem to us desirable and hence of all our pursuit of happiness, and that it is therefore the supreme good […]. But it is not yet, on that account, the whole and complete good as the object of the faculty of desire of rational finite beings; for this, happiness is also required”. Briefly stated, according to this conception “virtue and happiness together constitute possession of the highest good in a person” (KpV 5:110). But how does Kant understand the two elements of the proposed synthesis: virtue and happiness? Taking a realistic view of potential temptation, Kant describes virtue as the “moral disposition in conflict” (KpV 5:84). Less dramatically, he formulates virtue as “the firmly grounded disposition to fulfil one’s duty” (RGV 6:23n.), or as “the strength of a human being’s maxims in fulfilling his duty” (MS 6:394).³ Happiness is defined as “the state of a rational being in the world in the whole of whose existence everything goes according to his wish and will” (KpV 5:124). In the Canon of Pure Reason, happiness is described as “the satisfaction of all of our inclinations (extensive, with regard to their manifoldness, intensive, with regard to degree, and also protensive, with regard to duration)” (KrV A806/B834; first emphasis mine). Both the appeal to everything’s going according to “wish and will” and the description of happiness as the satisfaction of all our inclinations (along with the fact that Kant incorporates familiar

 See Esser .

“Mixtum Compositum”

73

everyday intuitions into his concept of happiness) seem to confirm that, as with virtue, Kant is referring to the happiness of the individual. ⁴ As we have already seen, “virtue and happiness together constitute possession of the highest good in a person” (KpV 5:110; my emphasis). It is this formulation in particular that contains the decisive clue to answering the controversial question of whether Kant has the happiness of the individual or the happiness of all in mind in the doctrine of the highest good. With reference to Kant’s position in the Groundwork of the Metaphysics of Morals, one might suggest that to promote one’s own happiness under the condition of worthiness to be happy is to promote it on the condition that one also promotes the happiness of others – for according to what Kant says in the Groundwork, this is precisely what the duty of the rational agent consists in.⁵ But from this it cannot be concluded that Kant’s concept of the highest good is concerned only with collective rather than individual happiness. Indeed, personal happiness (and its pursuit) does not cease to be individual simply because it is rendered legitimate only on the condition that we are committed to promoting the happiness of others. Kant’s use of the phrase “in a person” reveals the individual dimension of the demand for perfection that pertains to the concept of the highest good. Moreover, occasional formulations referring to the realization of the highest good (and therefore also to the proportion of happiness) in the world or in the human condition do not signal Kant’s abandonment of this systematic interest in the happiness of the individual. If the highest good consists of the supreme good and happiness in correspondence to it, it is indeed difficult to accept that a reduction in the demand for perfection might be allowed on one of the two sides because the idea of humanity replaces the individual with an overall balance, as it were. Quite the contrary, it would seem that if a reduction with regard to the overall condition of the world is not acceptable, then the happiness belonging to the collective concept of the highest good cannot be understood without referring to every individual member of the collective.⁶  See Himmelmann .  In fact, the hierarchical relationship between the two supreme maxims that Kant presents in the Religion in order to determine the moral disposition can be reformulated in such a way that the advancement of the happiness of others, which according to the second formulation of the categorical imperative is in agreement with respect for others as ends in themselves, is to be regarded as a limiting condition for the advancement of one’s own happiness.  See also the contribution by Frederica Basaglia, who, with recourse to God in the doctrine of the postulates, draws attention to the fact that when it comes to happiness we are concerned with the empirical happiness of the individual: if we were concerned with general well-being, it would be more plausible to conceive of the balancing-out of happiness as an aim of both society and historical progress rather than referring to God.

74

Birgit Recki

2 “Everything Goes According to Wish and Will”: A Morally Restricted Concept of Happiness? At the same time, it is a foregone conclusion for Kant that the demand that we integrate the concepts of virtue (as the idea of a moral disposition) and happiness (as the idea of the satisfaction of needs resulting from inclination) into a common concept hinges on proving the possibility of overcoming a continually pressing discrepancy. Maxims of happiness and maxims of virtue differ fundamentally, and Kant is far from regarding their relation as analytic, as if one concept were contained in the other. Rather, he conceives of the relation as synthetic, such that both irreducibly make up the complex constitution of the highest good. It remains to be shown, however, exactly how this synthetic relation is to be conceived. Kant’s concept of the highest good contains two potentially opposed concepts as constitutive elements. As such, he speaks of an antinomy that must be solved.⁷ Kant is obviously aware that his claim contains something surprising. The way in which he emphasizes the synthetic character of the highest good as a mixtum compositum is quite striking. With rhetorical redundancy, he writes: “Now, inasmuch as virtue and happiness together constitute possession of the highest good in a person, and happiness distributed in exact proportion to morality (as the worth of a person and his worthiness to be happy) constitutes the highest good of a possible world, the latter means the whole, the complete good” (KpV 5:110 f.). With this concept of the highest good, he seems to want to demonstrate that his critique of reason involves the combination of transcendental idealism and empirical realism in the domain of practical reason as well. In the synthesis of virtue (as worthiness to be happy, with regards to which the transcendental condition of the autonomous will must be thought of as efficacious) and happiness (with regards to which dependence on empirical reality is implied through the satisfaction of sensible inclinations), Kant puts forward two self-standing claims that are not reducible to each other and that both con-

 Kant focuses on the burden of argumentation that is bound up with this synthetic unification in the solution to the antinomy, in that he shows that the unification of the virtuous disposition and happiness is thinkable – via the illustration of a possible “satisfaction with oneself”, a rationallyeffected feeling of self-enjoyment, “which cannot be called happiness”, but which “nevertheless resembles the latter” (KpV :). In my view, his reflection on the two postulates, which follows the section on the primacy of practical over theoretical reason, belongs with the program of the solution of the antinomy as far as content is concerned because it is here that the mere possibility of thinking of the synthetic connection of happiness and virtue is precisely determined, and in such a way that, according to the succession of cause and effect, happiness follows upon virtue, in the intelligible world under the condition of immortality and through God.

“Mixtum Compositum”

75

cern one’s orientation in action: “that happiness and morality are two specifically quite different elements of the highest good and that, accordingly, their combination cannot be cognized analytically (as if someone who seeks his own happiness should find, by mere resolution of his concepts, that in so acting he is virtuous, or as if someone who follows virtue should in the consciousness of such conduct find that he is already happy ipso facto)” (KpV 5:112 f.). In emphasising the fact that we are concerned with “two specifically quite different elements” when it comes to virtue and happiness, Kant rejects a view that has occasionally been attributed to him by his commentators. According to this interpretation, the conception of happiness developed by Kant in the doctrine of the highest good and the postulates is circumscribed by the antecedent demands of virtue. For example, there have been attempts to regard the notion of satisfaction with oneself (Selbstzufriedenheit) as significant in this regard (a notion that is addressed in the solution to the practical antinomy). Contrary to such interpretations, however, Kant’s intention in discussing satisfaction with oneself was not in fact to provide a morally transformed conception of happiness from a systematic perspective. Rather, Kant can be seen as pursuing a methodological-propaedeutic aim in introducing the concept of self-satisfaction. By appealing to a rationally reflective feeling in general, Kant wants to ensure that the main object of his argument – the reconciliation of happiness and virtue – is at least conceivable.⁸ His description of “satisfaction with oneself” as a rationally effected feeling of self-enjoyment, “which […] cannot be called happiness”, but which “nevertheless resembles” it (KpV 5:118), should be treated as seriously as the inclusion of wish in the determination of happiness (KpV 5:124).

 According to the view defended by Dirk Setton in his commentary on my paper at the conference “The Highest Good in Kant’s Philosophy” (Frankfurt, / September ), the Dialectic of Pure Practical Reason deals with the transformation of the claim to happiness via a “revolution in the way of thinking” of the sort that Kant describes as being available to the good will of human beings in the Religion. This revolution consists in the radical decision to recognize the priority of the maxims of morality and to subordinate the maxims of self-love (sensibility) to them. This “revolution in the way of thinking” does not alter the fact that the claim to happiness as such must be subordinated to the demands of morality. Reference to the appropriateness of including a speculative thought does not in the least suggest the proper interpretation: if Kant is concerned with an already morally modified understanding of happiness in his doctrine of the highest good, then the expectation of just compensation from God would no longer be required.

76

Birgit Recki

3 The Highest Good as Mixtum Compositum: A Radical Break with Past Conceptions By rejecting the idea of an analytic connection between happiness and virtue, Kant critically refers to the “attempts at coalition that have hitherto been made” (KpV 5:112), in which virtue and happiness are to be made congruent with one another by identification or deduction. Kant’s targets here are the Epicurean and the Stoic accounts. He criticises these accounts on two fronts: namely, for their plainly misguided attempt to analytically deduce one concept from the other, and for conceiving of the connection between the two elements as a relation of cause and effect – a relation that is supposed to be given in the physical world of appearances, and hence real. As Kant explains, while Epicurus identifies virtue with happiness (or, more specifically, conceives of virtue as an effect of happiness), the Stoics identify happiness with virtue (or, more specifically, conceive of happiness as a consequence of virtue) (KpV 5:112, 115). The particular accounts that Kant has in mind here are those of Epicurus and Seneca. However, in line with the present paper’s focus on the recognition of the genuine claim to happiness, it should be remembered that, before the emergence of Stoic ethics, as exemplified by Seneca, it was Aristotelian ethics that succinctly defended the conception of the highest good that Kant would ultimately reject. It would be worth asking why, at this important point in his critical ethics, Kant looks to Stoicism rather than to Aristotle for the model that is to serve as a background for his own solution. When it comes to specifying the nature of the highest good, both Aristotle and Seneca reveal a similar liberality in relation to the real wishes and claims of human beings when – in a broadly equivalent way – they reduce happiness to virtue. Aristotle takes pride in distinguishing his concept of happiness from false, competing conceptions of happiness (such as a life that aims at pleasure or political honour). For him, happiness consists in leading one’s life under the direction of reason, which makes it possible for human beings to achieve what they strive for in an optimal way. In turn, the constitution that makes it possible for human beings to realize such happiness is itself an optimal state – namely, the kind of excellence that is included in the concept of virtue. In other words, happiness is “an activity of the soul in accord with virtue” (Nicomachean Ethics 1098a18). This means that happiness consists in virtue, insofar as “the virtue of a human being will […] be the state that makes a human being good and makes him perform his function well” (Nicomachean Ethics 1106a23), a state that manifests itself in the fact that we take pleasure in the right things. Happiness is virtue. Indeed, we are also familiar with the famous concession, which reveals what we so esteem in Aristotle as the founding

“Mixtum Compositum”

77

father of context-sensitive realism:⁹ “For we do not altogether have the character of happiness if we look utterly repulsive or are ill-born, solitary, or childless; and we have it even less, presumably, if our children or friends are totally bad, or were good but have died” (Nicomachean Ethics 1099b3–6). The identification thesis should therefore be modified. In essence, happiness is virtue, according to Aristotle. But external goods also contribute to happiness as “resources”, or as something “added” (Nicomachean Ethics 1099b1, 1100b10). The Stoic Seneca recognized true happiness in the good that cannot be taken away from us against our will because it is solely in our control. Omnia mea mecum porto, it is affirmatively stated in De constantia sapientis – a statement that refers to all external goods of life,¹⁰ which the person speaking here regards merely as items loaned to him from destiny, and which he is ready at any time to return without complaint. The highest good, the only long-lasting good, consists in the self-determined constitution of the soul, which in this way has made itself fundamentally independent of external goods. Therefore, Seneca does not seem to be committed to the asceticism that is often sweepingly and falsely attributed to the Stoics. He explicitly endorses uninhibited contact with all possible enjoyments of life, as long as it is assured that one does not make one’s happiness dependent upon them. Indeed, Seneca’s famous “carpe diem” refers not only to using our limited lifetime above all to lead a virtuous life, but also to a speedy gathering of the moments of happiness while “on the run”.¹¹ This is simply what life is for the brave soldiers fighting the blows of destiny, according to the Stoic conception – which is made evident by the concession that Seneca also makes to adiaphora, i. e. to morally indifferent things. What is decisive for the self-determined, i. e. virtuous, disposition is the consciousness of losing nothing of importance at the moment that something external is taken away. With these accounts in mind, the fact that Kant conceives of the summum bonum as a mixtum compositum of virtue and happiness, and conceives of the relation between these concepts as synthetic, seems at first glance to mark a radical break with the above positions. Indeed, the view according to which happiness fol-

 Compare here the view of Ernst Tugendhat, which is representative of many interpreters: “This is where the concept of the practical capacity to judge (phronēsis), which is so important for Aristotle, enters. Only the person who has sound judgement, who can judge well, can recognize in an individual case when and how generously one ought to act” (Tugendhat : ).  Seneca, De constantia sapientis, –.  “Unhappy one, you don’t understand how to live on the run!” Seneca makes this objection to those who have not grasped that happiness involving external goods is not permanent, unlike the happiness of virtue, and is therefore a momentary phenomenon (Seneca, Ad Marciam de consolatione, ).

78

Birgit Recki

lows from virtue seems painfully and indefensibly to neglect the real claims to happiness of acting subjects. Although it may seem evident that morality requires the strict fulfilment of duty and that ethical thought therefore boils down to being oriented towards the supreme good of virtue, it does not follow that we must therefore view our happiness in terms of virtue (nor would this be acceptable). No matter how much one may wish to grant Kant regarding his take on joyous “satisfaction with oneself” both in the solution to his practical antinomy and in the Doctrine of Method of the Critique of Practical Reason (i.e. on the limited mode of happiness, in which the intrinsic gratification that arises as the reliable effect of leading a virtuous life makes itself manifest), the claim that our entire happiness (and hence the “complete good”) should therefore lie in virtue hardly seems plausible. To expect this would be an affront to both our bodily constitution and the worldly constitution of our lives. Methodologically speaking, such an expectation is extremely anti-intuitive. It is simply an unreasonable demand. Against the background of Seneca’s influential position (outlined above), what is liberating about Kant’s doctrine of the highest good as a mixtum compositum becomes apparent (where “liberating” also refers to the sense of overcoming the problematic and systematic ‘placelessness’ of the needs of inclination). Without questioning the primacy of morality,¹² external goods are no longer treated here as an afterthought. The recognition of their participation in human happiness no longer has the character of a nolens volens concession, which carries a certain subsequent or unreal status. From a systematic point of view, what for Aristotle belongs precariously to the concept of a resource or something added, and for Seneca to the concept of adiaphora, is now determined as equally constitutive of the highest good in Kant’s explicit description of the “proper proportion” between happiness and worthiness to be happy. What is even more decisive about this radical break with the traditions outlined above is the fact that it abandons the Aristotelian-Stoic identification of happiness with virtue in the concept of the highest good, in which, from the perspective of human reality, a reconciliation of the claims of duty and inclination, of

 Kant already made this clear in the Canon of Pure Reason in the first Critique, where he articulates the second of his questions concerning the determination of human beings: What may I hope? The complete thought here is: “If I do what I should, what may I then hope?” (KrV A/ B). This formulation of the question reveals the systematic assumption of the entire reflection, which is the virtuous life. With the claim “all hope concerns happiness” (KrV A/B), Kant anticipates the connection between happiness and virtue.

“Mixtum Compositum”

79

the soul and of the body, is achieved only via a kind of deception.¹³ To the extent that it includes sensible rational beings’ claims to happiness, and to the extent that it incorporates the “faculty of desire of rational finite beings” (KpV 5:110), Kant’s doctrine of the highest good gives the impression of taking the reality of human needs seriously.¹⁴ This revaluation of human happiness is expressed not least in Kant’s use of the term “worthiness to be happy” (Glückswürdigkeit) as an equivalent to “virtue”, an expression which indicates that Kant views happiness as the gratification of virtue, the latter being the most supreme good.

A Variant of the Analytic Connection Between Virtue and Happiness: An Excursus on Hegel In the interest of systematic completeness, it is worth noting another variant of the analytic connection between happiness and virtue, which is also misguided. This variant, which is formulated in terms of a theory of action, is presented in Hegel’s famous objection, in the Phenomenology of Spirit, to Kant’s doctrine of the highest good as a mixtum compositum and to the related doctrine of the postulates. Here, Hegel completely overlooks the internal structure of Kant’s conception of morality. What is more, he simultaneously ignores the realistic intuition that Kant regards so highly in his synthesis. Hegel claims that, by presupposing that moral action will not result in a positive balance of happiness in an empirical world that is not geared towards happiness, Kant ignores the fact that the subject realizes his ends with the actions in question, i. e. that agency already intrinsically contains its own reward: “[T]he action is nothing other than the actualization of the inner moral purpose, nothing other than the production of an

 The same is true of Epicurus’s identification of virtue with happiness. I do not specifically discuss this here, since my focus is Kant’s departure from previous conceptions of this relation and, in particular, his rehabilitation of the claim to happiness.  If one focuses on the various places in Kant’s published and unpublished writings where a negative balance of pleasure is bound up with a pessimistic view of human existence as a whole, one might suppose at first glance that the opposite is the case. For example, Kant writes in his notes on metaphysics that, where the prospects of becoming happy are very low, a sensible rational being cannot rationally agree to an existence in which the essential point of view is that of morality. Existence as a condition of morality is “as far as my wish is concerned impossible” (:). In more than one place, Kant stresses that only in a “fairy world” could a human being rationally wish to repeat his or her existence. But a view of this kind speaks not to a low estimation of (the claim to) happiness, but rather to the opposite; it reveals the importance Kant ascribes to human contentment with one’s condition and the prospect of appropriate compensation in the realization of the highest good.

80

Birgit Recki

actuality determined by the purpose, or of the harmony of the moral purpose and actuality itself. […] and because, in the accomplished deed, consciousness knows itself to be actualized as this particular consciousness, or beholds existence returned into itself – and enjoyment consists in this – there is also contained in the actuality of the moral purpose that form of actuality which is called enjoyment and happiness” (Hegel, Phenomenology of Spirit, 375). This point about the theory of action boils down to the following objection: why expect an additional reward, even in cases where we successfully realize our own ends? With reference to the actualization of the inner moral purpose, Hegel both assumes and attributes to Kant the view that success with regards to an action is the same as the realization of its end. If this assumption were compelling, Kant would have to grant this objection. However, against consequentialism, Kant makes it unmistakeably clear that the moral value of an action depends not on the actualization of the inner moral purpose, but rather on the universalizability of the maxim on which one acts. Being oriented towards such a criterion is what the good will consists in, and the same goes for the moral disposition, in which Kantian virtue consists – “let the result be what it may” (GMS 4:416). Hegel’s objection is successful only under two assumptions: first, under the assumption of consequentialism, which is precisely not Kant’s position, and second, under the assumption that moral action is guaranteed to succeed.

4 Sensible Rational Beings and Their Claims to Happiness: A Common Thread in Kant’s Ethical Thought To be sure, it is not as though the conception of human happiness contained in the synthetic conception of the highest good suddenly appeared in the Critique of Practical Reason. Indeed, in 1764 Kant had already articulated what is noteworthy and systematically essential to the view: “Since a human being finds himself happy only insofar as he satisfies an inclination, the feeling that makes him capable of enjoying a great gratification without requiring exceptional talents is certainly no small matter” (Beobachtungen 2:207). Twenty years later, the line of thought in the Groundwork of the Metaphysics of Morals makes it clear that happiness is not fit to serve as a principle and is therefore not something on which morality can be properly grounded. However, this does not prevent the critic of reason from re-establishing the human pursuit of happiness as legitimate, and indeed we are not speaking here of a morally informed concept of happiness. In the Groundwork, in the methodological explanation of the concept of happiness as an ideal of the imagination, Kant articulates in passing the criterion for the fulfilment of sensible inclinations, which confirms that we are deal-

“Mixtum Compositum”

81

ing with an understanding of happiness that includes the interest we intuitively take in our empirical happiness (i. e. prior to ethical reflection on ‘true happiness’ according to Aristotle or Seneca). The idea that every rational being naturally strives for its own happiness is also already documented here (GMS 4:399, 415). And not least, the determination of the duty to promote the happiness of others supports the idea that happiness is recognized as a good. In sum: the sensible rational being’s claim to happiness is regarded as indispensable.¹⁵ After all, we are told here that “to assure one’s own happiness is a duty (at least indirectly); for, want of satisfaction with one’s condition, under pressure from many anxieties and amid unsatisfied needs, could easily become a great temptation to transgression of duty” (GMS 4:399). Although what is articulated here is simply the licence to be happy within the bounds of the moral law, such a consideration, characterized explicitly as a rational recognition of the finitude of sensible rational beings, always provides occasion to be indignant about the presumptuousness of a rational position in contrast to the human claim to happiness. Even worse than Kantian rigorism with regard to duty is the permission to obtain happiness that comes along with it. Indeed, the imposed duty to be happy in the interest of morality is for many precisely where Kant misjudges humanity under the absolutism of reason (as if, within the framework of a philosophical grounding of morality, one could expect to find a reasonable position defending an unrestricted claim to happiness). The licence granted here already implies the doctrine of the necessary subordination of the two supreme maxims, which Kant presents in Religion within the Boundaries of Mere Reason; the above passage from the Groundwork presents the distinction from the Religion even less sharply. Kant assumes that the sensible rational subject of action and self-determination faces two supreme maxims in his consciousness (essentially as a kind of fact of reason) – the maxim of sensibility (to act according to the demands of personal happiness) and the maxim of morality (to act according to the moral law). Neither of the two can be removed or ignored, and nor can one be reduced to the other; rather, they must be placed in the proper relation. If I subordinate the maxim of morality to the maxim of sensibility, this means that I only intend to follow the moral law insofar as this does not prevent me from indulging my claim to happiness. On the other hand, if I subordinate the maxim of sensibility to the maxim of morality, this means that I shall only indulge my claim to happiness insofar as it can be united with the moral law:

 See Gerhardt .

82

Birgit Recki

“Hence the difference, whether the human being is good or evil, must not lie in the difference between the incentives that he incorporates into his maxim (not in the material of the maxim) but in their subordination (in the form of the maxim): which of the two he makes the condition of the other” (RGV 6:36).

With this doctrine in the first section of the Religion, Kant not only gives order, retrospectively, to the relationship between his theory of action and his ethics, but also expresses in a methodological way the idea that the sensibility of rational beings is good in itself (RGV 6:26 ff.) and is not, as Kant’s critics have continuously and falsely suspected, the genuine source of immoral action. In essence, this doctrine endorses the legitimacy of the claim to happiness, which Kant takes up throughout the Critique of Practical Reason and which he attempts to make clear with recourse to the central role that needs play in human nature: “To be happy is necessarily the demand of every rational but finite being and therefore an unavoidable determining ground of its faculty of desire. For, satisfaction with one’s whole existence is not, as it were, an original possession […] but is instead a problem imposed upon him by his finite nature itself, because he is needy” (KpV 5:25). The character of the analysis shows that Kant does not restrict himself to grasping the legitimacy of the human claim to happiness by focusing on the primacy of the fulfilment of duty, i. e. he does not merely justify the legitimacy of this claim by appealing to its necessity, which is itself deduced from the problematic status of the existential contentment of human beings. Rather, he goes beyond this by making the rational claim to happiness comprehensible by transforming the insight with which it is bound into a claim placed on reason: “The human being is a being with needs, insofar as he belongs to the sensible world, and to this extent his reason certainly has a commission from the side of his sensibility which it cannot refuse, to attend to its interest and to form practical maxims with a view to happiness in this life and, where possible, in a future life as well” (KpV 5:61; my emphasis). Kant therefore argues for the recognition of striving for happiness not only indirectly, as was the case in the Groundwork with respect to the untainted disposition to fulfil duty, but also directly, with reference to our sensible neediness, which makes itself positively recognizable as an undisputed fact in the disturbance caused by unsatisfied needs. Against this background, it is in the name of consistency that Kant claims that in the concept of the highest good as a mixtum compositum of “happiness and morality two specifically quite different elements” (KpV 5:112 f.) should be integrated with one another. Above all, this doctrine illustrates the development of the idea that reason must fulfil the demands of sensibility, which are here rendered valid.

“Mixtum Compositum”

83

5 Dualism as the Signature of a Problem Upon closer inspection, it is clear that the doctrine of the highest good accomplishes the above not as a principal aim but alongside other key tasks. The question then becomes how well, i. e. how consistently, it accomplishes this. Just as obvious as Kant’s emphasis on the claims of sensibility that cannot be refused is the puzzling formulation from the same passage, according to which this demand on reason is directed towards maxims that have not only the happiness of this life in view, but also, and where possible, the happiness of a future life. It is this puzzling formulation that finds its explanation in the doctrine of the highest good. And this explanation accords with what Kant presents as the solution to the antinomy. It therefore appears that the legitimacy of the claim to worldly happiness, which is not reducible to the achievement of virtue, does not have the last word in this doctrine, insofar as Kant’s solution to the antinomy attempts to achieve something that amounts to rendering the claim to happiness intelligible. But this is based on the very dualism that Kant regarded as central to the solution of the antinomies of the Critique of Pure Reason (a dualism that is present in the solution to the problem of the freedom of the will in the third antinomy in particular). In the Dialectic of the second Critique, with regard to the case where “one and the same acting being” is regarded “as appearance” and “at the same time as noumenon”, he states: “It is just the same with the foregoing antinomy of pure practical reason” (KpV 5:114). Let us take another look at Kant’s criticism of the “attempts at coalition that have hitherto been made”, which he takes to be mistaken insofar as they conceive of the relation between virtue and happiness as one of cause and effect in the empirical world. By contrast, Kant claims that “the possibility of such a connection of the conditioned with its condition belongs wholly to the supersensible relation of things” (KpV 5:119) and indeed must be thought to take place “by means of an intelligible author of nature” (KpV 5:115). Whereas, according to Kant, Epicurus and the Stoics made the mistake of thinking that virtue follows from happiness and happiness from virtue in the world of appearances, Kant works under the assumption of a ‘lexical order’ when it comes to happiness and worthiness to be happy, an assumption that is grounded in the unconditional validity of the moral law (KpV 5:119, 122). We may (and must) think of happiness as a consequence of the worthiness to be happy only in the sense of an intelligible relation, by means of an intelligible creator of the world. In fact, this means that Kant once again uses the tried and tested remedy of dualism as a means of solving this antinomy: he distinguishes between the perspective of the understanding with regard to the world of appearance and the perspective

84

Birgit Recki

of reason with regard to the intelligible world. He can thus claim to have avoided the problem of a cause-effect relation that is neither provable nor achievable in the empirical world, to the extent that he views this relation as established in the intelligible world by the highest authority of reason. In this way, Kant locates the solution to the discrepancy in our lives between virtue and happiness in the doctrine of the postulates, in which the rational ideas of God and immortality of the soul are reformulated, with reference to a “need of reason”, as having a new methodological status, i. e. as postulates.¹⁶ The idea that can be derived from the character of this Kantian reflection is perfectly understandable: the finite rational being has a legitimate claim to happiness. In this world, the appropriate proportion between achieved happiness and our effort to be moral is not guaranteed. Only a rational being – who, by virtue of its threefold role as creator, legislator and judge, is answerable for the entire structure of the world – is capable of achieving this appropriate proportion. Indeed, this is achieved not only to the extent to which human beings are part of the world of appearances, but also, in conformity with a complete concept of the world, with respect to the entirety of the world as such: “Therefore the highest good in the world is possible only insofar as a supreme cause of nature having a causality in keeping with the moral disposition is assumed” (KpV 5:125). This idea is conceivable as a consequence of two-world dualism, and it implies mindbody dualism: “Thus happiness in exact proportion with the morality of rational beings, through which they are worthy of it, alone constitutes the highest good of a world into which we must without exception transpose ourselves in accordance with the precepts of pure but practical reason, and which, of course, is only an intelligible world” (KrV A814/B842).¹⁷ As free, as a person, as a possessor of pure practical reason – and now, in the doctrine of the postulates, as an immortal soul – I transfer myself to this intelligible world, which is apparently the exact same kingdom of ends into which, according to the Groundwork of the Metaphysics of Morals, I place myself in thought as a legislating member. The representation of this intelligible world equips me with the perspective of an “endless progress” (KpV 5:122), the supposition of which renders conceivable both an un-

 “[A] postulate of pure practical reason (by which I understand a theoretical proposition, though one not demonstrable as such, insofar as it is attached inseparably to an a priori unconditionally valid practical law)” (KpV :).  The formulation “the highest good of a world” must be read alongside the following specification in the relative clause: the highest good of a world, which is only an intelligible world. This makes it clear that this passage cannot be used to turn the individual’s claim to happiness into a requirement of the general condition of the world; rather, it can only be used to support the Kantian claim that the highest good can be fulfilled not in this world, but only in an intelligible world.

“Mixtum Compositum”

85

ending approximation to the ideal of holiness and the enjoyment of the problematic proper proportion between virtue and happiness. But is the delegation of this most urgent existential problem to the supersensible, to the mere world of thought, not a depressing result compared to the spirit of optimism engendered by the concept of the highest good as a mixtum compositum? What happens to the extremely real claim to happiness with which we, with Kant, started once we locate the solution to the antinomy in the supersensible world? Under the assumption of God and the immortality of the soul, the claim to happiness that was initially taken seriously becomes a promise that threatens to render the nature of its fulfilment unrecognizable. Or is the gratification which shall conceivably be granted to the immortal soul on condition of a virtuous disposition achieved over the course of a supposedly endless progress precisely the same happiness we rejected to the extent that it was identified with virtue? The conception of the highest good as a mixtum compositum is supposed to do justice to the necessary claim that sensibility makes on practical reason. Kant can take this claim into account as he draws the consequence of his dualism: the insight into the two components of the highest good contains the methodological recognition that, when it comes to the subjects of action and morality, we are concerned with sensible rational beings. Kant conceives of these complex sensible rational beings according to his metaphysical conception of the two worlds: on the one hand they are homo phenomenon, and on the other they are homo noumenon. Here, he utilizes the same dualistic conception of two worlds that forms the basis of this application of transcendental idealism: mundus sensibilis – mundus intelligibilis. It is clear from the commitment to two-world dualism in the doctrine of the postulates that the price of recognizing sensibility’s claim on reason is the ratification of mind-body dualism, upon which the acceptance of the delegation of the claim to happiness to the intelligible world depends. With reference to the persistence of this dualism, the question concerns whether it can exist together with the conclusion that this doctrine aims to offer a turning point that breaks with past conceptions. The problem facing Kant’s doctrine of the postulates as a possible solution to the antinomy of pure practical reason is not that it renders his ethics a theological morality, as one particular ‘Kantian gone wild’ has claimed. According to Schopenhauer, “[t]he complete incomprehensibility and absurdity” of Kant’s “unconditional ought” comes to light in the doctrine of the postulates in the Critique of Practical Reason: “Precisely that unconditional ought postulates a condition, indeed more than one, namely a reward, as well as the immortality of the person to be rewarded and a person who rewards” (Schopenhauer, Preisschrift §4, trans. Walschots). Schopenhauer could have read Kant a bit more carefully: Kant makes it unmistakably clear that the doctrine of the postulates concerns

86

Birgit Recki

not a condition of morality after the fact, but rather a consequence of morality. However, even if this objection is not accurate, the problem facing the doctrine of the postulates is still a serious one. For what functions as the recognition of the finite rational being’s claim to happiness leads in this way to infinity – and therefore into a mere world of thought. We console ourselves about the derogation of sensibility, which runs against not only our inclinations individually but also our claim to happiness as a whole; we guard against defeatism (as the effect of the unprovable insight that, against Kant’s suggestive repetition of the formulation “in the sensible world”, the highest good is unachievable in this world) as we ultimately delegate this claim to happiness to the intelligible world. Could we not have achieved this disappointing result more easily? We should be fair here, however. It may be helpful to consider the troublesome question of what an alternative to Kant’s view would look like, i. e. what a defensible alternative would consist in (assuming one wishes neither to be unrealistic about the claim to happiness and lose sight of the real condition of the world, nor to lose sight of the demands of morality). One can also understand the reflection that Kant presents in his doctrine of the highest good as a more detailed explanation of the function of faith from a moral point of view – an explanation that makes faith comprehensible in its dynamic, reduces it to its basic elements, and assigns it a legitimate place in a critique of reason.¹⁸ In the end, in the Canon of Pure Reason and with reference to the ideas of God and immortality (which function here as postulates), Kant speaks as if he were discussing articles of belief or faith,¹⁹ and his speculation in the Dialectic of the Critique of Practical Reason explicitly refers to a practical belief of reason. Everything that is presented in the doctrine of the highest good, including the doctrine of the postulates, therefore only has validity under the assumption that we believe or have faith.²⁰ But of course this brings to light what is perhaps the most essential problem facing this doctrine. If we do not presuppose this belief, the elements of which (and their function) have been reconstructed here, one might object that such deductions neither involve nor lead to what one normally understands by belief – an objection that can similarly be raised in relation to the speculations on probability in Pascal’s wager. What use is it, if one does not believe? In other words: does this thought fulfil its function with reference to what the finite, needy, rational being has hoped for with respect to the claims of happiness? Kant himself

 See Recki .  See Recki .  See Gerhardt .

“Mixtum Compositum”

87

connects his thoughts on securing the highest good to the second question in the Canon of Pure Reason, which asks of the determination of man: what may I hope? And it appears as if the doctrine of the highest good answers this question the following way: I can hope for the fulfilment of my claim to happiness in precisely that intelligible world, in which the proper proportion of virtue and happiness is guaranteed by an intelligible being. Presented as simply as possible, this is guaranteed in thought. Accordingly, what is considered perhaps also as a consolation (a reflection on the preservation of practical optimism, which is conceived of as a motivational-psychological element of Kantian ethics) can fulfil its function if and only if the condition of belief as a mode of thought is fulfilled. Is this ultimately the point of Kant’s peculiar formulation, which calls to mind the concept of duty – namely that it is “morally necessary to assume the existence of God” (KpV 5:125)?²¹

6 Conclusion As unconvincing as the doctrine of the highest good may be as a hermeneutic instrument, i.e. as a rational explanation of a presupposed belief, it does have a considerable advantage. The true consolation that can be gained from this necessarily unsatisfying attempt at a solution might lie in the fact that, in contrast to the conceptual harmony of the analytic connection associated with Aristotle and the Stoics, it nonetheless allows for consciousness of a difference and, moreover, for consciousness of a discrepancy. The sting of the unfulfilled claim to happiness is preserved nonetheless. According to the doctrine, and according to the reception of the difficulties that are bound up with the mixtum compositum of the highest good, we at least know how much effort would be involved in deceiving ourselves.²² Translation: Michael H. Walschots

 What is “morally necessary” supposed to mean here, assuming this expression does not have to do with an explanation of the concept of duty? (Compare also KpV :.)  What still remains to be investigated is whether something changes (and if so, what this is) when the results of this problem are revisited in the explicit attempt to overcome dualism in the third Critique.

88

Birgit Recki

References Aristotle 1999, Nicomachean Ethics, second edition, translated by Terence Irwin, Indianapolis: Hackett Esser, Andrea Marlen 2004, Eine Ethik für Endliche. Kants Tugendlehre in der Gegenwart, Stuttgart/Bad Cannstatt: Frommann-Holzboog Gerhardt, Volker 2002, Immanuel Kant. Vernunft und Leben, Stuttgart: Reclam Gerhardt, Volker 2014, Der Sinn des Sinns. Versuch über das Göttliche, München: C. H. Beck Hegel, Georg Wilhelm Friedrich 1977, Phenomenology of Spirit, translated by A.V. Miller, Oxford: Oxford University Press Himmelmann, Beatrix 2003, Kants Begriff des Glücks, Berlin/New York: Walter de Gruyter Langthaler, Rudolf 2014, ‘Zwischen skeptischer Hoffnungslosigkeit und dogmatischem Trotz’. Geschichte, Ethik und Religion im Anschluss an Kant, Berlin/Boston: Walter de Gruyter Recki, Birgit 1998, “Der Kanon der reinen Vernunft (A795/B823-A831/B859)”, in: Georg Mohr/Marcus Willaschek (eds.) 1998, Kritik der reinen Vernunft, Berlin: Akademie Verlag, 597–616 Recki, Birgit 2001, Ästhetik der Sitten. Die Affinität von ästhetischem Gefühl und praktischer Vernunft bei Kant, Frankfurt/Main: Klostermann Recki, Birgit 2009, “‘Allein die Vernunft fing alsbald an sich zu regen’: Kants Kabinettstück zu einer aufgeklärten Mythologie”, in: Heiner F. Klemme (ed.) 2009, Kant und die Zukunft der europäischen Aufklärung, Berlin/New York: de Gruyter, 232–50 Recki, Birgit 2012, “Kant über das Glück”, in: Ozren Zunec/Petar Segedin (eds.) 2012, Zblizavanja, Festschrift for Damir Barbaric, Zagreb: Matica hrvatska, 103–21 Schopenhauer, Arthur 1977, Preisschrift über die Grundlage der Moral, in: Schopenhauer, Arthur 1977, Werke, Volume VI, Zürich: Diogenes Seneca, Lucius Annaeus 1980, Philosophische Schriften, Vol. 1, ed. Manfred Rosenbach, Darmstadt: Wissenschaftliche Buchgesellschaft Tugendhat, Ernst 1993, Vorlesungen über Ethik, Frankfurt/Main: Suhrkamp

Stephen Engstrom

The Determination of the Concept of the Highest Good When Kant criticized his predecessors for relying on a conception of the will as heteronomous in their accounts of the principle of morality, he identified the ancient philosophers as openly exhibiting this error. These philosophers, he claimed, based their moral investigations entirely on the problem of determining the concept of the highest good, the problem, that is, of specifying what this good consists in. In taking this problem to be primary, he held, the ancients sought to gain knowledge of an object that could be antecedently recognized to be of ultimate or highest worth, by reference to which an effective principle of conduct could then be determined. Such an approach, he maintained, opens the door to misconceptions of morality’s principle and cannot accommodate the intrinsic, unconditioned goodness of moral conduct. In opposition to such accounts, Kant argued that the concept of the good should be determined, not before the principle of morality, but only after it and through it. Moral philosophy’s first task, he held, is to provide a clear articulation of the principle of moral conduct, in which it is exhibited as representing such conduct as unconditionally required. Kant’s own pursuit of this recommended approach is plainly reflected in the organization of his Critique of Practical Reason. Only after first expounding the supreme principle of practical reason and identifying it as the principle of morality does he turn to the concept of the good and then finally undertake to determine the concept of the highest good. Faced with Kant’s presentation of these starkly opposing approaches, we may reasonably wonder whether the whole truth must lie simply on one side or the other. It may be granted that there is justice in Kant’s criticism as applied to certain ancient doctrines. The Epicurean school, which identifies the good with pleasure or the absence of pain, will have difficulty accommodating the categorical character of moral obligation. But it would be an overstatement to say that an approach that begins with the problem of determining the concept of the highest good is doomed to founder on such a rock. If a philosopher’s antecedent understanding of the highest good is ethically informed, no misconception of the standard of moral judgment and of the goodness of moral conduct need result. Aristotle, for example, with whose practical thought Kant seems to have been largely unacquainted, articulated a conception of the highest good that contains

90

Stephen Engstrom

from the start an ethical dimension, reflected in his famous determination of it as “activity of soul in accordance with virtue”.¹ Moreover, the ancients’ approach has an appealing feature, which Kant’s alternative, at least as so far described, seems not to share. By starting their ethical investigation with the problem of the highest good – where ‘highest’ is understood in such a way as to imply that the good in question is not just the best among an array of goods, but rather the complete and final good, the good suited to be the ultimate object of all conduct in life – the ancients were well positioned to ensure that their account of ethical conduct and virtue would be situated within a unified understanding of the complete good, an understanding that would make intelligible how a life of ethical conduct would or could be a complete and happy life. This unity of conception finds striking expression in Aristotle’s wellknown remark that the highest good, eudaimonia, can be characterized as a certain living well, consisting in doing well.² By comparison, Kant’s alternative, as described above, may well seem deficient. By requiring that the problem of determining the concept of the good not be taken up until after the completion of an analysis of obligation and duty to reveal their principle, Kant appears to open the door to the problem of accounting for how a life of conduct in accordance with duty can or may be a happy life. Even if the ancients’ attention to the connection between ethical virtue and the highest good may in certain instances lead them astray, it is nevertheless a deeply appealing feature of their approach, and one that Kant’s procedure appears to close off. The appearance is heightened by the impression that may be conveyed by the account of the highest good that Kant offers when he finally takes up the problem of determining its concept. In approaching this problem, Kant characterizes the highest good as “the unconditioned totality of the object of pure practical reason” (KpV 5:109), thereby indicating that he does aspire to achieve a unified conception of the good in its entirety. But doubts are triggered when he proceeds to determine this concept. After stating that the highest good includes two ingredients, the first being virtue, identified as the worthiness of happiness, and the second being the happiness commensurate with it, Kant adds that in the highest good these elements must stand in a causal relation, and he describes the relation as a distribution of happiness in proportion to virtue. Since this description invites the thought that the causal relation is conceived as depending on the intervention of some mediating agency, it may encourage us to imagine  A more extended defense of Aristotle from the criticism can be found in Engstrom b.  Nicomachean Ethics I. b–. At b–a Aristotle specifies that the relevant sense of ‘living well’ (euzōia) is to be fixed by ‘doing well’ (eupraxia) (cf. Eudemian Ethics I. a–).

The Determination of the Concept of the Highest Good

91

that Kant introduces his account within an eschatological setting, and that he must do so because his approach leaves him with no recourse but to invoke a deus ex machina. Only a god could restore to unity what has been so radically sundered by the philosopher’s analysis. We thus reach a familiar outcome, the appearance of dualism in the representation of Kant’s philosophy. But just as a closer scrutiny of the ancients’ approach can reveal that it need not inevitably implicate a heteronomous conception of the will, so a closer scrutiny of Kant’s alternative can show that it does not lead to the dualism just noted. Indeed, as I will attempt to show, closer attention to Kant’s engagement with the problem of the highest good and with the ancients’ accounts can not only bring to light that he was himself aware that the ancients’ approach need not lead to the problem of heteronomy, but can also help us reach an understanding of his alternative that frees it from the appearance of dualism and exposes a deep kinship it shares with the best of the ancient accounts.

I The Highest Good as Starting Point 1. We may begin by observing that, just before presenting his account of the highest good in the Critique, Kant notes that it is plainly possible to conceive of this good in such a way that the moral law is included from the start (KpV 5:109–10). This is the way in which he himself understands the concept, claiming that the good it represents includes not only happiness but also virtue, or the will’s determination to act in accordance with the moral law. In mentioning this obvious possibility, Kant acknowledges in effect the point noted above, that it is possible in principle for philosophers to base their ethical reflections on the problem of determining the concept of the highest good without making themselves vulnerable to his criticism. It is striking in this connection that in his lectures on ethics, where the ancients’ views of the highest good are discussed in some detail, Kant observes that although these accounts differ in notable respects, they share a fundamentally sound core understanding of this ideal. Student records indicate that he explicitly credits the ancients generally with the insight that the highest good includes not only happiness but also the worthiness of happiness, and the further insight that this worthiness is based in the disposition of a morally good will, which is the condition under which reason allows the inclusion of happiness in the first place (Mo-Collins 27:247–8). Equally striking, this review of the ancient accounts figures in an extended discussion of the highest good that is located at the outset of the lectures, prior to an investigation of morality’s principle, and moreover occupies a position within the sequence of topics addressed in the lectures that

92

Stephen Engstrom

corresponds to the position that Kant’s famous account of the good will occupies in the argument of the Groundwork for the Metaphysics of Morals. ³ 2. When we consider the Groundwork’s discussion of the good will more closely and in the light of the student notes, we find that Kant himself begins his ethics by addressing the question of the highest good, and that, like Aristotle, he conceives of the highest good in a way that ensures that virtue, or the disposition of the will lying at its basis, is included from the start.⁴ Of course, Kant nowhere announces in the Groundwork’s opening paragraphs that his discussion of the good will is animated by a concern with the highest good, and despite a few hints that ancient authors are on his mind, this concern is easy to miss in the glare of his celebration of the good will. The concern becomes apparent, however, when those paragraphs are read in context and in the light of a distinction noted in the Critique between two senses of ‘highest good’. According to that distinction, the expression can signify either the supreme good, the good that is unconditioned and as such the condition of any other good’s goodness, or the complete good, which includes, along with the unconditioned good, any good whose goodness the unconditioned good conditions (KpV 5:110). The Groundwork’s opening three paragraphs (GMS 4:393–4) are dedicated to showing that the good will is the supreme good, or the highest good in the first sense. But this argument, though complete in itself, also constitutes the first stage of a more extended discussion, which contains, in its second and concluding part (GMS 4:395–6), further reflections intended to confirm and to supplement the conclusion reached in the first. There Kant considers a doubt that can be expected to arise in the thoughts to someone who, like the proponents of the Enlightenment eudaimonism popular among many of his empirically-minded contemporaries, takes happiness to be the end that nature has ordained for the practical use of our reason. In response, Kant argues that both the natural-teleological assumptions underlying such a doubt and the judgment of mature experience re-

 These points concerning Kant’s lectures are spelled out in Engstrom a. The lectures in question are recorded in numerous sets of notes thought to date from the mid-to-late s, including the notes published in Menzer , which have long been available in English translation (Infield ), the Collins notes, which are contained in volume  of the Akademie edition of Kant’s writings and also in English translation in Heath/Schneewind , and the Kaehler notes, recently published in Stark .  Similarities between Aristotle’s account of the highest good and Kant’s account of the good will have been noted by a number of authors, including Herman , Hill , and Broadie . In Engstrom b the similarity between Aristotle and Kant is attributed to their appreciation of an insight articulated in the work of Plato.

The Determination of the Concept of the Highest Good

93

garding the life devoted to the rational cultivation of enjoyment support the conclusion that the supreme good lies, not in happiness, but in the good will, the unconditioned good, even though the complete good may include also happiness as a conditioned good. So it turns out that the account of the complete good that in the Critique is not set out until the analysis has been concluded is outlined in the Groundwork at the point at which the analysis begins. Whereas the Critique first considers morality in isolation from happiness and then undertakes to represent the two as combined in a whole, the Groundwork, following its analytic approach, begins with the highest good, seeking to isolate from among the elements belonging to it the unconditioned good and its principle.⁵ To the extent that the Groundwork’s search for the principle of duty takes place within the context of an initial reflection on the highest good, there is reason to reconsider whether the Critique’s method does after all entail a dualism of the sort described earlier. Moreover, once we see that the Groundwork’s opening argument is concerned with the highest good, we can draw on it to appreciate aspects of Kant’s understanding of the highest good that are not highlighted in the account offered in the Critique. What is emphasized in the Critique is that the highest (i. e. complete) good contains two elements, virtue and happiness, and that these differ in that the latter is merely a conditioned good, whereas the former is the unconditioned good and condition of the conditioned good’s goodness. But little is said about the relation between them. No more is provided than the abstract description noted earlier, the depiction of it as a causal relation in which happiness is distributed in proportion to virtue. Consideration of the Groundwork’s argument can help fill in the picture. 3. Before turning to that argument, however, we need to clear the ground by setting aside a common misinterpretation of the account offered in the Critique, encouraged by Kant’s talk of a distribution of happiness in proportion to virtue. According to this interpretation, the distribution Kant envisages is conceived from the start as to be based on an assessment of actual character and disposition. The interpretation can be spelled out in different ways, according as the distri-

 It bears noting that in the Groundwork too the analysis culminates in a representation of what could fairly be characterized in terms of the Critique’s definition of the highest good as “the unconditioned totality of the object of pure practical reason”. But since this complete object is there framed in accordance with the formula of autonomy, it is represented, not through the idea of causality, or in accordance with the asymmetrical relation of ground and consequent, as it is at the beginning of the Groundwork and in the Dialectic of the Critique, but rather through the idea of community, or as a whole of reciprocally interacting members, under the name of a “kingdom of ends” (GMS :–).

94

Stephen Engstrom

bution is conceived as to be effected through human actions and institutions or through the administration of divine justice. But these differences need not concern us here. What calls for scrutiny is the central idea, namely that the concept of the highest good contains in its original determination the specification that this good is achievable only through a retrospective or in any case theoretical knowledge of the actual dispositions of persons’ wills. The only ground for conceiving of the distribution along such lines is a judgment that persons’ wills are or may be deficient, and any such judgment must have an empirical or in any case extra-practical source. As we noted, however, the good that Kant has in view is the unconditioned totality of the object of pure practical reason (KpV 5:109), and he makes clear that an object of pure practical reason is not determined empirically, in accordance with any extra-practical judgment (cf. KpV 5:57–8). Any attempt to determine what is good given certain deficiencies will result in a derivative conception; for in the recognition of deficiency an antecedently possessed conception of the good is presupposed. Only the presupposed conception could be adequate to the idea of the highest good. The determination of the concept of the highest good must therefore be carried out within the original practical point of view, which is wholly prospective.⁶ This point is reflected in Kant’s statement that the moral law commands that the highest good be made “the ultimate object of all conduct” (KpV 5:129). As such an object, the highest good must be represented in the first material principle of practical reason, the first act in which pure practical reason represents an end, an act that is presupposed in any exercise of wish in the will of a particular person.  None of this is to impugn Kant’s subsequent consideration of whether and how the highest good is possible, a consideration that leads first to the antinomy of practical reason and then to its resolution and to the postulates of the immortality of the soul and of the existence of God. In that subsequent discussion, God is represented as an omniscient being who possesses intuitive knowledge of rational beings’ dispositions (KpV :, ) and whose distribution of happiness is limited solely by deficiency in the conformity of those dispositions with the moral law (KpV :). What needs to be noted here is that all such limitation presupposes and hence is posterior to the original representation of the highest good. The limitation is at bottom no different from the limitation that figures in our understanding of the duty of beneficence. Beneficence requires in the first instance that we help others in need, that is to say, that we help them without first making an assessment of their character (just as our expectation that others help us when we ourselves are in need is not contingent upon such assessment), yet our obligation is limited in that it cannot require that we provide assistance that would directly contribute to wrongful action on the part of the recipient (MS :). Whether carried out by a finite or an infinite rational being, the distribution of happiness in accordance with the moral law is originally and essentially practical; it is limited only incidentally, namely in cases of cognizance either of wrongful conduct (in the case of beneficence) or of deficiency of disposition (in the case of divine goodness).

The Determination of the Concept of the Highest Good

95

According to the account outlined in the Critique, that first act has two moments. In the first, the recognition that virtue, the determination of the will by the moral law, is unconditionally good necessitates virtue’s inclusion in the complete good; in the second, the recognition that the complete good is not exhausted by virtue but requires proportional happiness in addition grounds the inclusion of happiness, represented as virtue’s rationally necessary consequence. Although this account is sufficiently determinate to make clear the inadequacy of the retrospective interpretation, it may still leave us with little sense of what else Kant could have in mind. So long as there is no available alternative reading, the mention of proportionality will continue to invite interpretations that mistakenly place Kant’s account of the highest good in an eschatological setting or at least introduce an illicit retrospective point of view. Here the Groundwork’s argument can be of help.

II The Good Will and the Highest Good 1. In contrast to the Critique’s abstract description of the relation between the two elements in the highest good, the Groundwork’s argument brings to light the practical relation the good will bears to the other items with which it is contrasted. Kant calls attention to this relation, speaking of the will’s use of the other items, and a brief review of his argument suffices to reveal that appreciating this practical relation is crucial to understanding how the argument supports its conclusion that only the good will is intrinsically and unconditionally good. In presenting his argument, Kant classifies all of the items he initially contrasts with the good will under the single general heading of gifts, dividing them into gifts of nature, including talents of mind and qualities of temperament, and gifts of fortune, such as power, wealth, honor, even health and happiness.⁷ In employing this single heading, he represents all such things as given to the will, available to it for use. They can accordingly be described as material conditions of the exercise of the will. Indeed, they are requisites of that exercise, in that every action of the will depends on gifts of nature, and every outer action depends as well on gifts of fortune. Every action of the will, then, is a use of gifts, and every such use is an action. The argument’s conclusion that these gifts, un To streamline discussion, I shall focus on the initial paragraph of Kant’s argument, which concerns these gifts, and leave to the side the second paragraph’s treatment of what might be called qualities of character, which implicate the gifts of nature in a developed form and count as genuine virtues provided that they belong to a will that is good. A more detailed analysis of the argument can be found in Engstrom .

96

Stephen Engstrom

like a good will, are not unconditionally good is reached on the strength of the consideration that they may be bad or harmful if the will that is to make use of them is not good. The thought here, evidently, is that if the will is not good, there is no guarantee that the gifts will be put to good use, no assurance that they will be used well or correctly.⁸ The argument thus turns on the idea that the good will differs from all of these gifts in that it cannot be bad or harmful: the good will puts to good use the gifts at its disposal, using them well or correctly, and so is the only thing that can neither have nor in any way contribute to a bad or harmful effect. This is an extraordinary proposition, and it would betray a misunderstanding of it to react to its seeming strangeness by supposing it to be at best an overstatement, in need of qualification or limitation in order to be rendered plausible. All that is called for are certain clarifications to forestall misunderstanding. The principal point that needs to be appreciated is that the argument does not attempt to show, in a posteriori fashion, from a consideration of cases in which a good will makes use of things, that the actions in such cases are not bad or harmful. Actions with unhappy results can easily be found, as critics have often enough pointed out.⁹ Rather, the argument merely directs attention to our at least implicit recognition that a good will is unconditionally good, a recognition contained in the “common moral rational knowledge” under consideration in the Groundwork’s first section. That recognition positions us to know in advance that in cases of the sort just mentioned, whatever harm may issue from the action must be ascribed to some factor other than the willing, perhaps to a deficiency in natural endowments, perhaps to limitations in their development, as for instance a lack of knowledge, judgment, or skill, perhaps again to some misfortunate circumstance. The unconditioned goodness of the good will is recognized a priori so far as it is recognized at all, and it is on this recognition that knowledge of the impossibility of the good will’s having bad effects is grounded. The argument merely points out, by way of contrast, that nothing falling under the heading of a gift, nothing whose goodness depends on use, can possibly share this strange and unique characteristic. 2. What must the good will consist in if it is possible for us to recognize a priori that it has the distinctive goodness to which this argument directs attention? Two factors  Although this thought figures only tacitly in the Groundwork argument, it is made explicit at other points where Kant is arguing along similar lines, e. g. Mo-Collins :–, –; MoMrong II :. It is also explicit in Plato’s versions of this argument, which appear to be a source of inspiration for Kant’s; see e. g. Meno d–a.  E. g. Ross : –.

The Determination of the Concept of the Highest Good

97

seem plainly to be required, one cognitive, the other causal. First, a good will must rest on some sort of knowledge or understanding on which correct use of the gifts depends. Second, a good will must be disposed to use such gifts in accordance with that knowledge. But we need to specify these factors more precisely. Here two points are of particular importance, one concerning the knowledge itself, the other concerning its relation to the disposition. If, as I just suggested, one way of making intelligible how an action expressing a good will may in a given case nevertheless be harmful is by citing a lack of knowledge, then there must be a difference between the knowledge essential to the good will – knowledge on which the good will’s good use of things constitutively depends – and knowledge that is not essential, even though it may be necessary in certain conditions or in this or that case if bad or harmful results are to be avoided. In addition, the knowledge on which the good use of things essentially depends cannot be knowledge whose goodness itself depends on good use. If it were, it would not be essential, and there would need to be some other knowledge, distinct from it, on which the good use of it depended. This requirement rules out all theoretical knowledge, or knowledge of “what is”, and all productive knowledge, or know-how, whether crude and commonplace or refined and sophisticated. The only knowledge that satisfies the requirement is knowledge of the final, complete good, the good suited to be the ultimate object of all conduct, regardless of whether, as the Stoics held, this knowledge consists exclusively in knowledge of an unconditioned good, or whether, as Kant will eventually maintain, it also includes knowledge of a conditioned good. The use of things that expresses a good will accordingly expresses knowledge of the ultimate end. It may also exhibit excellence in the choice of actions undertaken for the sake of that end, the excellence Aristotle calls phronēsis. But it need not, in that choices exhibiting such excellence depend in part on knowledge that, like the things used in those choices, depends for its goodness on good use. The knowledge that is essential specifically to the excellent choice of actions undertaken for the sake of the end is integral to virtue in its full perfection, but it presupposes knowledge of the end, growing out of it through the exercise of deliberation and choice in practice. Second, the knowledge must be the inner ground of the good will’s disposition. If the knowledge lay outside the will, among the things available to it for use, it would depend for its goodness on good use, rather than being knowledge on which good use itself constitutively depends. As the ground of the good will’s disposition, this knowledge must have a certain efficacy in respect of that disposition, whereby it can make actual the object it represents. And since all knowledge is selfconsciously constituted, this efficacy must itself be self-conscious in the sense that it must reside in the cognition’s understanding itself to be efficacious. Now such self-

98

Stephen Engstrom

conscious efficacy is what Kant identifies as the differentia that distinguishes practical knowledge from theoretical (KrV Bix–x; KpV 5:46), so the knowledge we are considering must be practical: rather than being anything that can be given to the will and thereby available to it for use, it is knowledge that can determine the will. And since the knowledge is the inner ground of the good will’s disposition, its determination of the will is nothing distinct from the will’s own self-determination. This is no surprise, for the capacity for practical knowledge is practical reason, which Kant famously identifies with the will (GMS 4:412). Two factors, then, one cognitive, one dispositional, related along the lines just described, make up the good will. Whereas the first resides in the will, or practical reason, what is distinctive of the good will is provided by the second, its disposition to use the gifts at its disposal in accordance with practical knowledge. And whereas the second factor is accordingly requisite for the distinctive goodness that constitutes the good will as the unconditioned or supreme good, we have seen that the first lies in knowledge of the complete good. Together, the two factors make up what Kant, with a nod to the ancients, identifies as wisdom. According to the characterization offered in the Critique, “wisdom theoretically regarded signifies the knowledge of the highest good and practically the conformity of the will to the highest good” (KpV 5:130–1; cf. 5:108). Thus the good will consists in wisdom so understood. Our consideration of the practical relation in which the good will stands to the other items with which Kant’s argument contrasts it has brought into view that the good will bears an intimate relation to the highest good, a relation with a cognitive and a causal aspect. It also positions us to see that this relation has implications for how the highest good must be constituted. The good will depends essentially on knowledge of the complete good. Since this knowledge is practical, it includes an understanding of itself as efficacious in respect of its object. The good will’s knowledge of the complete good accordingly includes knowledge of its own causality in respect of that good, and should it turn out that this good includes not only the unconditioned good, or the good will itself, as the Stoics held, but also some further, conditioned good, then the good will’s knowledge will represent itself not only as self-productive but also as the cause of that conditioned good. It follows, then, from the practical-cognitive relation the good will bears to the highest good that if that good contains the two elements Kant identifies, then those elements must stand in a causal relation, as stated in the Critique. 3. Before proceeding further, I should address a complication that I have so far ignored. In order to throw light on the way virtue and happiness are related in the Critique’s account of the highest good, I have been considering the practical relation

The Determination of the Concept of the Highest Good

99

of use that the Groundwork’s argument represents the will as bearing to the gifts available to it, the material conditions of its exercise. Now the gift that is of direct relevance for our understanding of Kant’s account of the highest good is obviously happiness. In addition to being the only gift mentioned in that account, it is the only gift Kant discusses that is unqualifiedly an end and therefore even so much as a candidate for inclusion in a good suited to be “the ultimate object of all conduct”. Precisely for that reason, however, happiness is not among the things we normally speak of as used. Everyone recognizes that happiness is not a mere instrument or means we employ to attain some further good. Yet while Kant’s inclusion of happiness in the highest good shows that he shares this recognition, the Groundwork argument seems nevertheless to regard happiness as among the things available to the will for use. Although Kant does not, in that argument, say in so many words that happiness can be used, he does seem evidently to intend something of the sort when he remarks that happiness can lead to boldness and thereby to overboldness or haughtiness without the presence of a good will to correct and to make universally purposive its influence on the mind.¹⁰ But we can make sense of this remark, provided that we understand the practical relation of use not in a narrow way, modeled on the use of a tool or instrument or on the use of materials in the making of an artifact, but in a broad sense, suitable for application across the entire range of the will’s exercise. Use so understood embraces all management of and coping with the material conditions of action, whether favorable or unfavorable, and it includes in particular one’s mode of dealing with one’s good fortune, one’s way of enjoying or appreciating it. As we noted, all human action can be described as use in this broad sense. Such a description serves in part to express the finitude of human volition and action, its self-conscious reliance on conditions outside itself for its actualization.

III Happiness and the Highest Good 1. So far, we have seen that the Groundwork’s opening argument implicitly involves the idea that the good will is based in knowledge of the ultimate end, the highest good. Reflection on the practical relation the good will bears to the gifts available to it has brought into view the practical-cognitive relation in which it stands to that end, revealing that the good will lies in wisdom, and that the object of this wis GMS :; cf. MS :. Student notes record Kant as having said, “satisfaction [Vergnügen], well-being [Wohlbefinden], constantly cheerful heart [stets fröhliches Herz] are good only under the condition that man has a good will, in order to make use even of them” (Mo-Mrong II :).

100

Stephen Engstrom

dom, the highest good, must, if it contains more than the good will, be constituted as stated in the Critique, namely as a whole consisting of two elements, one unconditioned, the other conditioned, standing in a causal relation, a relation we are now in a position to recognize as practical-cognitive in character. But we have yet to consider how, according to the account presented in the Critique, happiness enters into the highest good as the second of those elements. In what follows, we shall approach this issue by following the path traversed in the Critique’s Analytic of Pure Practical Reason, which begins with practical reason’s principles and proceeds to its object. In doing so, however, we shall situate Kant’s analysis within the context of his practical-cognitivist understanding of the good will as grounded in knowledge of the highest good. For we now have reason to anticipate that when the analysis is seen in this context, it will become apparent that no dualism of the sort described earlier is entailed. Practical principles, as Kant explains them, are universal determinations of the will. Now as was mentioned Kant identifies the will with practical reason, and since reason is a cognitive power, indeed the highest cognitive power, a self-conscious and self-determining capacity, it follows that so far as reason itself – pure reason – can be practical, the will too is a cognitive capacity, the capacity for practical knowledge. A practical principle, then, can be a principle of practical knowledge, provided that it is objectively grounded and universally valid, qualifying it as a practical law. As practical, it is distinguished from theoretical knowledge by its self-conscious efficacy. It is important to note that the knowledge in question here is discursive, or conceptual, and as such finite in character: although as knowledge it lies in selfdetermination, it is nevertheless limited in that it consciously depends for its actuality on certain enabling conditions, namely affections of consciousness through receptivity, or sensibility, and so on conditions outside itself. Kant captures these two aspects of discursive knowledge through an analysis of it in terms of form and matter, and this analysis accordingly structures his account of practical knowledge and practical principles. Thus, in his analysis of practical principles he undertakes to show that, although “all willing must also have an object and therefore a matter” (KpV 5:34), the practical principles constituting the will’s acts of self-determination can be objective, or practical laws, or, what comes to the same, practical cognitions, only if they are determined not by their material condition but by the formal condition of knowledge, namely the condition of universality – suitability for universal legislation (KpV 5:27) – which is just the form of discursive knowledge in its practical application. In carrying out his analysis, Kant aims to reach an articulation of this formal condition as the fundamental, formal law of pure practical reason, an articulation that reveals it to be identical with the moral law. Our purpose here, however,

The Determination of the Concept of the Highest Good

101

is to understand the relation happiness bears to the highest good, and for this it will be necessary to focus on the contrasting material condition. In preparation for doing so, it will be useful to describe briefly the psychological setting in which Kant situates his argument.¹¹ 2. Kant’s analysis is developed in the context of an understanding of the will of a human person as materially conditioned by the natural constitution of a human being. That constitution, however, is considered abstractly, merely through the concept of a human being as a rational animal. As an animal, a human is constituted by a faculty of desire as well as by a faculty of perception. Kant identifies self-production as the mode of existence peculiar to organisms generally,¹² and in the case of an animal this self-production is through representation, namely through the reciprocal cooperation of desire and perception, the two basic ways the power of representation can operate. Thus, the perception of an object that agrees with the conditions of animal existence determines inwardly – sensibly, through the feeling of pleasure – the faculty of desire to desire such an object. The resulting desire, which consists in an efficacious representation of the object, works to make what it represents actual, and so far as the ensuing action is successful, perception of that object follows, giving rise again to the feeling of pleasure, reinforcing the sensible desire. In this way, animal being is self-productive through representation, and the faculty of desire is precisely its essential capacity for such self-production. As rational, however, humans are subjects, distinguished from nonrational animals in that their basic life-power, their basic capacity for self-production through representation, is a capacity for such self-production through knowledge. One respect in which this knowledge enters into such self-production is through the deliberative role that theoretical knowledge of the material conditions of choice and action (knowledge of one’s powers and of their extent in one’s situation) can play in specifying a practicable course of action sufficient to actualize a wished-for object of sensible desire. So far as the ensuing action is successful, the object can be perceived or experienced, and as before the experience may issue in a feeling of pleasure. In this case, however, pleasure can result not only from the perception of the object produced, but also from the perception of the object’s production, that is, from perception of the action itself, one’s success in realizing the object. Pleasure  Pointing out that his analysis presupposes abstract, untendentious explanations of certain concepts drawn from psychology, Kant offers explanations of the concepts life, faculty of desire, and pleasure, all of which are implicated in the concept of animal existence (KpV :n.). The description to follow relies on these explanations and also on MS :– and RGV :–.  See KU §§–.

102

Stephen Engstrom

arising in the first way reinforces, as before, the initial desire for the pleasing object, but pleasure resulting in the second way gives rise to a new sensible desire, namely to engage in such action. But since in this deliberative role knowledge contributes to the choice of means rather than to the wish for the end (except indirectly, as a condition of desires arising in the second of the two ways just noted), we need not consider it further here.¹³ The cognitive capacity in a human being also makes possible comparisons between the pleasing perceptions and experiences that register one’s needs as a living being, comparisons that enable one to frame a concept of pleasurable experience in general and the notion of the totality of such experience – the complete object of sensible desire – under the name of happiness. The happiness represented here is one’s own happiness, in that the pleasure attending the experiences is in every case one’s own. Since the pleasure is just the sensible determination of one’s faculty of desire, this concept of happiness may itself have an influence on that faculty, in that in a rational animal the faculty of desire is an application of a power of representation that embraces not only the imagination and its sensible representations but also the capacity of thought and its concepts. It should be noted that the concept of happiness is framed independently of theoretical knowledge of the material conditions of choice and action. It depends only on the pleasurable experiences, registering the needs of life that make up the material conditions of wish, or of desire for the end. Thus, in a human being, a desire for happiness naturally arises, through the exercise of the cognitive capacity in respect of these material conditions. And because this desire is general, containing under it all particular sensible desires, it constitutes the primary human sensible desire, embracing the totality of the material conditions of wish. Moreover, since this concept is understood by a human being to represent the complete object of its own faculty of desire, it figures as a representation of its own complete existence, or complete life, and as such constitutes a conception of living well that stands as the ground of a human being’s employment of its reason in its self-production.

 This is not to deny that in the life of a rational animal desires arising in the second way have a special significance. Because these knowledge-dependent desires take as their objects the exercise of the powers of rational animal existence rather than the satisfaction of its natural needs and dependencies, their objects come in the end to be assigned a preeminent place in the happiness of a rational animal, tempting many philosophers to attribute them to a higher faculty of desire (cf. KpV :–), despite the fact that, considered just as desires, they share the same sensible form as desires that arise in the first way.

The Determination of the Concept of the Highest Good

103

3. In a human person – that is, a human subject in which reason itself, pure reason, is practical – we reach the possibility of self-production not just by means of knowledge but from knowledge. This brings us back to Kant’s analysis of practical principles. In the will of a human person, the self-determination in which a practical principle consists is an act composed of form and matter. Although the principle depends on its material condition, namely the desire for happiness, the will is not determined by that condition, and to the extent that the principle is objective, or practical knowledge, this material condition is registered in the principle’s content just so far as is consistent with the formal condition of universality, which lies in the formal principle of pure practical reason, the form of practical knowledge. This hylomorphic analysis has a significance that bears underscoring. Just as in his theoretical philosophy Kant counters not only the empiricism of Locke but also the rationalism of Leibniz in describing theoretical knowledge as requiring, in addition to concepts that provide such knowledge with its intellectual form, also sensible intuition to furnish its matter, so in his analysis of the practical principles that constitute practical knowledge he counters not only the empiricism of Epicurus but also the rationalism of the Stoics in saying, as we noted, that all such willing must have, in addition to the form of universality, also an object and therefore a matter, the possibility of which depends on sensible desire. On its own, pure practical reason represents no object whatsoever and so can yield no material principles, no practical knowledge, at all. It can do so only through determining some sensible desiderative matter. Since the matter is provided by the rationally-framed concept of one’s own happiness, pure reason in an individual human person depends for its exercise on that same individual’s employment of its reason in its own self-production. Even though the form determines this matter, it nevertheless depends on it.

IV Proportionality: the Highest Good as Good Conduct 1. We have now to bring the foregoing consideration of Kant’s hylomorphic analysis of practical principles to bear on his account of practical reason’s complete object, under the name of the highest good. First, however, a few observations are called for relating to the second chapter of the Analytic, in which Kant advances from practical principles to the object a practical principle represents. Attention to Kant’s account of this object, and in particular to the grounds on which he relies in identifying it with the good, will confirm that, notwithstanding his methodological requirement that the concept of the good be determined only after and through the moral law, the analysis of principles carried out in the first

104

Stephen Engstrom

chapter does not dualistically sever his account of morality and its principle from the good and the highest good. We begin with a few points concerning the meaning of ‘good’ (Gutes). Kant employs this term in accordance with a long tradition of established usage going back through the Scholastics to Plato and Aristotle. In that tradition, a distinction is drawn between rational and sensible desire, and ‘good’ and ‘pleasant’ (or ‘agreeable’) are used, respectively, to refer to their objects.¹⁴ Thus, Kant observes, the use of language distinguishes ‘good’ and ‘agreeable’, demanding that good “always be judged through reason, hence through concepts, which allow of being universally communicated, and not through mere sensation, which is limited to individual subjects and their receptivity” (KpV 5:58). And since the usage implies not only this relation to reason but also a reference to desire, he also says, “what we call good must be, in the judgment of every rational human being, an object of the faculty of desire”, and “the good […] always signifies a relation to the will, so far as this is determined through the law of reason” (KpV 5:60–1). As these passages reflect, then, Kant follows the traditional usage, identifying the good as what is represented in desire- or will-determining judgments that have a law-like character, involving the objective and subjective universal validity that he elsewhere identifies, along with necessity, as features characteristic of the exercise of reason. In accordance with this explication of the traditional usage, Kant offers a characterization of ‘good’ that is formal in the sense that it articulates the concept solely by reference to the capacity through whose exercise its content is to be determined: “a necessary object of the faculty of desire […] in accordance with a principle of reason” (KpV 5:58). Relying on that understanding of the term, he can straightforwardly identify the object of practical reason with the good, in that he identifies practical reason with the will, which is just what the tradition recognizes as the higher, rational faculty of desire. So understood, then, the good is just what is known in practical reason’s act of practical knowledge, or in the exercise of will that constitutes an objective practical principle. Thus Kant’s analysis of practical principles, far from severing them from all relation to the good, is nothing but an analysis of the act of knowing the good.

 Aristotle, for instance, holds that all animals have sensible desire (epithumia), which is for the pleasant, but only animals that have reason have a form of desire – rational desire, or wish (boulēsis) – whose object is regarded as good. See e. g. De Anima II. b–, III. b–; Nicomachean Ethics III. b–, III.. A similar distinction can be found in Plato (e. g. Charmides e). Kant’s well-known discussion of how properly to distinguish the higher and the lower faculties of desire is in effect a defense of this traditional distinction between forms of desire (KpV :–).

The Determination of the Concept of the Highest Good

105

2. If the good is the object of practical knowledge, and the highest good is the unconditioned totality of that object, then the determination of the concept of the highest good must lie in the objective self-determination of the will, given that the will is practical reason, the capacity for practical knowledge. And if, as Kant states, the moral law commands that the highest good be made the ultimate object of all conduct, the highest good must be represented in the will’s original act of objective material self-determination, in which it first sets itself an end in an act of practical knowledge. In our discussion of Kant’s hylomorphic analysis of practical knowledge, we saw that this knowledge depends not only on the formal condition of universality, but also on a material condition, the concept of one’s own happiness. We have now to consider how in this knowledge, depending as it does on this material condition, an object is represented. Insofar as the concept of one’s happiness is the material condition of the act of practical knowledge, the act of the will that this concept makes possible constitutes a use of that concept, a use integral to the will’s representation of its object in an act of practical self-determination: the will attaches a concept of its object to itself in a judgment that, as practical, regards and therein constitutes itself as the cause of the object and therein also constitutes the object as an end (cf. KU 5:219–20). Kant speaks of this use of the material condition as an “adding” of matter to the will (KpV 5:34). But in order for one’s concept of one’s happiness to figure in a practical judgment that qualifies as practical knowledge, as is required if the end represented is to be good, this use of the concept, this adding of matter to the will, must be in conformity with the form of universality. The addition must be universal. Since universality has two sides, objective and subjective, to add matter to the will in conformity with this form is to add it, not just in conformity with the practical representation of every will’s addition of happiness to itself, but in conformity with the practical representation of every will’s addition of happiness to every will.¹⁵ Here it should be noted that the will’s original addition of happiness to itself is not an act of choice, but of wish. It is the origin rather than the outcome of deliberation with regard to practicability. Now so far as this wish for one’s happiness is an act of practical knowledge, its conformity with such cognition’s form

 In the following passage, Kant illustrates how matter and form figure in an objective practical principle: “Let the matter be, for example, my own happiness. This, if I attribute it to each (as, in the case of finite beings, I may in fact do), can become an objective practical law only if I include in it the happiness of others. Thus the law to promote the happiness of others arises […] merely from this: that the form of universality, which reason requires as the condition of giving to a maxim of self-love the objective validity of a law, becomes the determining ground of the will […]” (KpV :).

106

Stephen Engstrom

of universality is not a mere accident. As practical knowledge, it self-consciously maintains that conformity, in the at least implicit recognition that such maintenance is constitutive of all practical knowledge, and it thereby self-consciously depends on the practical representation of everyone’s wishing for everyone’s happiness, just as the conclusion of a syllogism self-consciously depends on its principle, its major premise, for its possibility and its necessity alike.¹⁶ Moreover, when considered at a fundamental level, this wish’s maintenance of its own conformity with form can be seen to be grounded in a basic wish for maintenance of such conformity in general. For since in the exercise of the capacity for practical knowledge maintenance of conformity with form is self-consciously recognized to be constitutive of practical knowledge in general (in that the form of universality is a condition of the self-agreement essential to knowledge), such maintenance is an unconditionally necessary object of such knowledge and as such the supreme good, so that every wish, as an exercise of that capacity, contains the at least implicit formal wish that every wish maintain conformity with form, or in other words the wish that wishing in general be good, a wish that grounds the disposition of a good will and virtue. In contrast, the happiness represented in the wish that maintains this conformity is a conditionally necessary object of practical knowledge, in that it depends for its standing as an object of such knowledge – and hence for its goodness – on that same wish’s cognition-constituting maintenance of conformity with cognitive form. Now the wish for the second of these objects (one’s happiness) maintains conformity with form through grounding itself in a wish for the first (every wish’s maintenance of such conformity).¹⁷ So insofar as a person’s wish for happiness maintains conformity with cognitive form, it includes within itself a wish within which it is itself in a sense contained (in the way a conclusion is contained in its principle), namely a wish that everyone’s happiness be realized through everyone’s wishing – out of a common basic formal wish to maintain cognition-constituting conformity with form – for everyone’s happiness. This is as much as to say that in this original wish virtue and happiness are represented as so related that universal happiness follows as the collective effect of universal virtue. We

 In other words, the will’s addition of happiness to itself in a way that maintains conformity with the practical representation of every will’s addition of happiness to every will is nothing other than its addition of happiness to itself through that universal practical representation; such addition is thus knowledge from a principle, or knowledge of “the particular in the universal” (KrV A/B). As Kant remarks, the mere form of a law, which limits the matter, must also serve as a ground of its addition to the will (KpV :).  The talk of objects in this sentence and in the preceding two will need to be qualified, for reasons that will become clear in §IV..

The Determination of the Concept of the Highest Good

107

thus arrive at the determination of the concept of the highest good presented in the Critique. 3. In conclusion, I want to return to Kant’s discussion of the concept of the good in order to consider a remark that may seem to conflict with his account of the highest good. Although the apparent discrepancy may at first look like another symptom of dualism, the remark will in the end confirm the interpretation I have outlined and will also help draw our attention to the unity of conception achieved in Kant’s account, a unity strikingly reminiscent of the unity secured in the best of the ancient accounts. We have been speaking, with Kant, of happiness as included in the highest good, and this inclusion might seem to imply that such happiness is a good, albeit a conditioned good. But happiness lies in the totality of agreeable experience making up the complete object of sensible desire, and Kant explicitly denies that agreeable states count as good. He says that good “relates properly to actions, not to the state of sensation of the person” and praises the Stoics for appreciating this point (KpV 5:60). Nor does he mention happiness at all when summing up his discussion of the good: he cites only action in accordance with the moral law (as good in itself) and a will whose maxim is always in conformity with this law (as absolutely, in all respects good and the supreme condition of all good) (KpV 5:62). Readers are often surprised, then, when they find Kant later including happiness in the highest good. This apparent discrepancy vanishes, however, when we recall Kant’s hylomorphic analysis of practical principles and knowledge. Happiness is included in the highest good, but only as the matter. Considered in itself, the matter is not an object of practical knowledge and hence not anything good. When Kant speaks of the “adding” of matter to the will, he is thinking of incorporation, in which the matter, in being taken up into the will, is informed by the form of practical knowledge. The concept of one’s happiness that figures as the material condition of the good will’s exercise represents happiness as agreeable experience; but in the adding of this matter to the will, that concept is used in an act of practical knowledge, which as self-consciously efficacious represents as its own effect the happiness it works to make actual. Hence, the object of this knowledge is not happiness considered in itself, but happiness resulting through that very cognition’s production of it. Such production, however, is nothing other than good conduct, which in the original universal representation of it is sufficient to achieve its effect and constitutes the distribution of happiness in accordance with, or in proportion to, good willing. This is tantamount to saying that the highest good is a unity, lying in the single activity of good conduct itself, in that such conduct is just the causal relation that the form of the highest good bears to

108

Stephen Engstrom

its matter – or, what comes to the same, the determining, form-giving relation in which the supreme good stands to the complete good. If this is correct, then in Kant’s understanding of it as well as in Aristotle’s, the highest good can be characterized simply as a certain living well, consisting in doing well. This is the object of the good will’s wisdom and ultimate end of all its conduct.¹⁸

References Broadie, Sarah 2005, “On the Idea of the Summum Bonum”, in: Christopher Gill (ed.) 2005, Virtue, Norms, and Objectivity, Oxford: Clarendon Press, 41–58 Engstrom, Stephen 1997, “Kant’s Conception of Practical Wisdom”, Kant-Studien 88, 16–43 Engstrom, Stephen 2015a, “Ancient Insights in Kant’s Conception of the Highest Good”, in: Lara Denis/Oliver Sensen (eds.) 2015, Kant’s Lectures on Ethics: A Critical Guide, Cambridge: Cambridge University Press, 103–19 Engstrom, Stephen 2015b, “The Complete Object of Practical Knowledge”, in: Ralf Bader/Joachim Aufderheide (eds.) 2015, The Highest Good in Aristotle and Kant, Oxford: Oxford University Press, 129–56 Heath, Peter/Schneewind, Jerome (eds.) 1997, Immanuel Kant: Lectures on Ethics, Cambridge: Cambridge University Press Herman, Barbara 1993, “Leaving Deontology Behind”, in: Barbara Herman 1993, The Practice of Moral Judgment, Cambridge: Harvard University Press, 208–40 Hill, Thomas E., Jr. 2002, “Is a Good Will Overrated?”, in: Thomas E. Hill, Jr. 2002, Human Welfare and Moral Worth, Oxford: Clarendon Press, 37–60 Infield, Louis (tr.) 1930, Immanuel Kant: Lectures on Ethics, London: Methuen Menzer, Paul (ed.) 1924, Eine Vorlesung Kants über Ethik, Berlin: Rolf Heise Ross, David 1954, Kant’s Ethical Theory, Oxford: Clarendon Press Stark, Werner (ed.) 2004, Immanuel Kant: Vorlesung zur Moralphilosophie, Berlin: de Gruyter

 I thank Thomas Höwing for helpful comments.

II Kant’s Moral Arguments and the Postulates of Pure Practical Reason

Gabriele Tomasi

God, the Highest Good, and the Rationality of Faith: Reflections on Kant’s Moral Proof of the Existence of God In his Groundwork for the Metaphysics of Morals (1785), Kant dismisses the notion that belief in God’s existence in any way constitutes the motivational force of morality. Kant claims that “the pure representation of duty and the moral law in general […] has, by way of reason alone […] an influence on the human heart so much more powerful than all other incentives” (GMS 4:410). This socalled ‘purity thesis’ is reasserted in the Critique of Practical Reason (1788), in which Kant states that “respect for the moral law is the sole and indisputable moral incentive” (KpV 5:78). Kant was convinced that people somehow recognize that they have moral duties and that these moral duties can ‘move’ them to act, simply by virtue of being moral agents; he further argued that people are moral agents simply by being the kind of rational creatures they are.¹ However, despite his acknowledgement that reason can be practical in and of itself and can determine a person’s will, Kant did not sever any ties between morality and religion. Kant did not forget the notion that people are embodied moral agents who have claims to happiness in the natural world; indeed, he establishes that people are essentially obligated not only to do their duty but also (though not in terms of duty) to pursue their own happiness. Despite the overriding nature of the moral law, it is not plausible to assume that people can fully renounce their claim to happiness, as it is a fundamental component of what it means to be an embodied moral agent. Thus, Kant conceives of the necessary object of the moral will in terms of a “highest good” – that is, of a good somehow ‘higher’ than the ‘supreme’ and unconditioned good of virtue, as it, according to a principle of distributive justice, also contains happiness: the happiness that people have made themselves worthy of by their virtue. Therefore, the highest good is the perfect, complete, or entire good (cf. KpV 5:110).² In the Dialectic of the second Critique, Kant argues that, if we are to have assurance that the highest good can be achieved, we must postulate that God exists

 Cf. Pinkard : .  In a nice formulation by Kant, the highest good is “a happiness of rational beings harmoniously coinciding with conformity to the moral law” (KU :). On the notion of the highest good, cf. Watkins , O’Connell  and Pasternack : –.

112

Gabriele Tomasi

and that the soul is immortal. Dismissed from the motivational aspect of morality, the reference to God thus reappears on its teleological side. The notion that it is somehow morality that gives rise to religion, and not the other way around, is further pursued by Kant in the Critique of the Power of Judgment (1790),³ again in relation to the possibility of the highest good. In §87 of this work, he presents what he calls a “moral proof” for the existence of God, through which he shows that, as embodied (imperfect) moral beings, it is rational to assume that God exists. The aim of this paper is to examine this proof and to assess the doxastic status of the belief that it grounds, in light of a series of qualifications that Kant introduces after having formulated the proof. Ultimately, the proof is based on one’s practical commitment and has an effect on conviction only in relation to one’s self-awareness as a moral being. This allows the claim that the rationality of the belief that God exists is in line with one’s rational efforts to make sense of the world on the basis of one’s moral commitments. Kant’s view, as developed in the final sections of the third Critique, is understood to be that a belief in God’s existence is an essential ingredient in consistent moral activity and a self-conscious commitment to the demands of the moral law.⁴ A more commonly used expression these days, a ‘belief in God’, refers to an attitudinal faith as a faith in some being; attitudinal faith is “partly a state of the will”.⁵ Having faith in God entails taking as true that God exists and certain other propositions about Him (e. g., as in Kant’s case, that He is an omnipotent and omnibenevolent, and at the same time a just, eternal, and omnipresent, original being, cf. KU 5:444).⁶ Attitudinal faith, as Robert Audi notes, implies instances of propositional faith – faith that – concerning the same object, but faith is not reducible to belief. Propositional faith often does not embody a belief in the proposition that expresses its content. Audi states that “it may be non-doxastic”.⁷ I will

 That, according to Kant, ethics does not inevitably lead to religion is shown by Dörflinger . For a short and clear reconstruction of Kant’s ideas on the topic, cf. Förster : – and Tafani : –.  In this paper, I do not consider the other ‘ingredient’ of fully motivated moral conduct, namely the belief that our soul is immortal. However, Kant himself scarcely mentions it in this context (cf. KU :, ). It is worth recalling that, in the Critique of Practical Reason, the postulate of immortality is directly connected not to the commitment to the highest good but to the complete fulfillment of the moral law (cf. KpV :), which, on the other hand, is a necessary condition of the highest good.  Audi : . I borrow the expression ‘attitudinal faith’ from Audi : , .  I use the expression ‘taking a proposition as true’ in the sense of accepting it. In this sense, it does not imply the belief that the proposition is true, but rather something like forming an intention to take it “as a basis for conduct or to use it as a basis of inference” (Audi : ).  In such cases, Audi refers to it as fiducial faith (Audi : , ).

God, the Highest Good, and the Rationality of Faith

113

try to show that Kant’s view is akin to this. While having faith in God presupposes His existence, according to Kant, this kind of presupposition, although a ‘taking as true’, does not entail a belief that God exists in the sense of a truth-valued or fullblown doxastic attitude. In spite of the fact that they usually go together, Kant distinguishes between the mental state of holding that ‘God exists’ is true and the mental state of taking ‘God exists’ to be true in one’s practical reasoning when the question whether God exists becomes salient. Having faith that God exists and has settled the world so that the unification of nature with the final end, which the moral law imposes on rational beings, is possible, is to have a positive disposition towards the obtaining of that state of affairs: that is, towards the truth of the proposition stating it. However, Kant’s moral proof is not an argument for the truth of the proposition that God exists (and has settled the world in a certain way); these matters are “inaccessible for theoretical cognition” (KU 5:471). This, however, does not mean that Kant denies that there are also theoretical considerations that support assent to that proposition and make it rationally acceptable. Kant opposes the physico-teleological proof of the existence of God (cf. KrV A620–30/B648–58; KpV 5:138–40), now referred to as ‘the argument from design’; however, he thinks that the organization exhibited by the world offers ample material for physical teleology and invites the notion of a divine order (cf. KrV A685–6/B713–14; KU §75, §78). In his view, one’s self-understanding as a moral being intertwines with a vision of the world as purposively structured – to such an extent, he claims, that “the inner moral vocation” of our existence directs us “to conceive of the supreme cause, for the final end of the existence of all things” (KU 5:447). It is no coincidence that §87 opens, as if to frame the moral argument within a world-view that includes nature in the moral order of things, with observations on moral teleology and physical teleology and on how these combine into a view that fosters the presupposition of an intelligent cause of the world. The assumption is that these teleological considerations are meant to support assent to the existence of God, though they offer no objective ground for it. As we will see, according to Kant, it is only in the epistemic attitude of moral faith that we hold as true the proposition that God exists. This paper will begin with some observations on moral teleology and physical teleology because a consideration of both types of teleology and of their relation to theology helps us to understand and evaluate Kant’s moral proof in §87. The proof will then be presented, and a discussion of Kant’s qualification of its scope and aim will follow. Finally, the paper will clarify the nature of the belief grounded by the proof.

114

Gabriele Tomasi

1 On Physical Teleology, Moral Teleology, and Theology At the very beginning of §87, Kant compares physical teleology and moral teleology in relation to theology. Physical teleology and moral teleology are different kinds of teleology and have different domains. Physical teleology concerns the natural world. According to Kant, physical teleology might direct us to consider nature as if it were the product of a supersensible intelligence (cf. KU §§73–5); additionally, it might give our theoretically reflecting power of judgment a sufficient basis to assume the existence of “an intelligent world-cause” (KU 5:447). However, physical teleology cannot produce a theology in the strictest sense (cf. KU 5:440); that is, it cannot raise the notion of an intelligent cause of the world up to the concept of God.⁸ As for moral teleology, Kant notes that it is a teleology that we find “in ourselves, and even more in the concept of a rational being endowed with freedom (of its causality) in general”. Moral teleology concerns us “as beings in the world”; more precisely, as he adds, it concerns “the relation of our own causality to ends and even to a final end that must be aimed at by us in the world, and thus the reciprocal relation of the world to that moral end and the external possibility of its accomplishment” (KU 5:447–8). I will return to this characterization of moral teleology; for now, it may be enough to recall that, according to Kant, it is this kind of teleology that “compels” our rational judging to seek “an intelligent supreme principle” (KU 5:447–8). He further suggests that even in a world that gives no trace of organization and offers no hint of a physical teleology, reason would still find in its moral ideas sufficient ground to produce a theology (cf. KU 5:478–9). Kant claims that in a world of this kind, moral teleology would be adequate “by itself” to refer “the world to a supreme cause, as a deity” (KU 5:444–5) because it does not argue from given natural facts to that supreme cause but from those moral ‘facts’ that are the moral ends of rational beings (cf. KU 5:436). However, in Kant’s view, the world we come across does not simply reveal the “effects of a mere mechanism of raw matter” but also exhibits an organization that offers ample material for physical teleology (KU 5:478) and even for connecting this teleology and moral teleology. Indeed, Kant intersects these two forms of teleology when he shifts from the consideration of organized beings to the notion of nature as a purposive whole

 Kant speaks of physicotheology as “a misunderstood physical teleology”, admitting that it is usable only as a “propaedeutic” to genuine theology (cf. KU :). About physical teleology, he claims that it could guide the mind “on the path of ends in the contemplation of the world, and thereby to an intelligent author of the world” (KU :, ).

God, the Highest Good, and the Rationality of Faith

115

(cf. KU 5:379, 454). In fact, to consider nature in these terms, he needs to identify its ultimate end – that is, something that can be at the top of a hierarchy of ends, so to speak (cf. KU 5:429). Kant claims that the only eligible candidate for this position is the human being because humans are the only beings capable of forming a concept of ends for themselves and making a system of ends out of an aggregate of purposively formed things (cf. KU 5:426–7).⁹ Nonetheless, Kant also argues that human beings are entitled to keep this position only on the condition that they somehow go beyond nature (cf. KU 5:431). This is because in nature as nature, there is no end that is not also a means to something else. In order to be the ultimate end of nature, human beings must provide some kind of barrier against a possible regress into the context of the means-ends relation. Kant expresses this condition, claiming that human beings can be the ultimate end of nature (letzter Zweck) only if they give to nature and to themselves a relation to “a final end [Endzweck]”. Furthermore, since he conceives of the final end as an end that “can be sufficient for itself independently of nature” (KU 5:431) and that needs no other end “as the condition of its possibility” (KU 5:434), he could identify it with the morally good (cf. KU 5:454). By making the idea of an ultimate end dependent on the notion of a final end, which is not “an end of nature (within it)” (KU 5:443) but is provided by practical reason, Kant ends up intertwining physical teleology and the moral order of ends. In his view, these purposeful orders correspond with different concepts of an end: specifically, with the concept of the end of nature (and of the theoretical reflecting power of judgment) on the one hand, and, on the other, with the concept of the end of freedom or of pure, practical reason.¹⁰ However, Kant also claims that these teleological orders converge in the human being, who is both the ultimate end of nature and its final end, though no longer as a member of nature, but “considered as noumenon” or as “the only natural being” in which we can cognize “a supersensible faculty (freedom)” (KU 5:435). In §87, Kant recalls these reflections on the position of the human being within nature, though he interestingly rephrases them in terms of value. He claims that a world with no rational beings “would have no value at all, because there would exist in it no being that has the slightest concept of a value” (KU 5:449). Thus, he offers a background for projecting moral teleology, which is, in any case, independent of physical teleology, onto a wider vision of the world as a purposive whole. This is, after all, required by the very concept of a moral teleology. In fact, according to the passage quoted above, moral teleol-

 For a clear reconstruction of Kant’s argumentation for this claim, cf. Höffe .  Cf. Brandt : –.

116

Gabriele Tomasi

ogy has two aspects: the first concerns the relation of our own causality to both the conditioned ends and a final end; the second concerns the relation of the world to the final end and its accomplishment. As for the first aspect of moral teleology, the double reference to the conditioned ends – which all, in Kant’s view, ultimately refer to happiness – and to the final end roughly corresponds with the double role of the moral law in relation to the will. The moral law concerns the suitability of the subjective principle of the will to be recognized as objective or as valid for the will of every rational being; since every act of the will must have an object or be directed towards an end, however, the law also affects the ‘material’ aspect of the will. With regard to this, the form of universality functions both as a limiting or restricting condition and as a reason to add an object to the will, which Kant identifies as the highest good in the world (cf. KpV 5:122).¹¹ This means that the moral law not only commits practical judgment to subordinating conditioned ends, related to the desire for happiness, to moral conditions, but also promotes, through the notion of the highest good, the creation of an order or a systematic unity among the ends of a rational, finite being that is consistent with its moral vocation. One is thereby provided with a teleological context within which one can orient and understand one’s morally motivated activity.¹² As we have seen, the second aspect of moral teleology concerns the relation of the world to the final end and its accomplishment. While morality depends on us, the possibility of happiness “is empirically conditioned, i. e., dependent on the constitution of nature” (KU 5:453). Therefore, we need to conceive of the world as a place in which the highest good can be attained. It becomes clear why there seems to be no other way to cope with this question than to resort to theology if we consider how Kant conceives of the relationship between the elements of the highest good. In fact, he views the dependence of happiness on morality as constitutive of the highest good, but he conceives of happiness and morality as heterogeneous factors with heterogeneous principles, which precludes their relationship’s being either conceptual on the one hand, or synthetic and causal (at least in an empirical sense) on the other. In his view, virtue and happiness are independent concepts that are “contained in another concept” in such a way that they are parts of the whole that contains them – that is, the highest good – and not identical with it (KpV 5:112).¹³ He also denies that happiness is  On the reasons why the highest good must be an object of the moral will, cf. Watkins :  ff.  On this, cf. Watkins : – and O’Connell .  This is the reason why he rejects both the Stoic and the Epicurean conceptions of the highest good (cf. KpV :–).

God, the Highest Good, and the Rationality of Faith

117

an effect of virtuous actions in the familiar physical sense of the word ‘effect’, since virtue belongs to the reasons for, not to the (empirical) causes of, our actions.¹⁴ However, though the relationship between virtue and happiness is not causal, since happiness depends on the constitution of nature, any alignment between them, as Adrian Moore notes, “must have something to do with the actual working of the world”.¹⁵ As things stand, it seems that either we consider the relationship between virtue and happiness completely accidental, thereby rendering the highest good a matter of chance rather than the final end of our actions, or we go “beyond the world” and seek an intelligent, supreme principle “in order to represent nature as purposive even in relation to the morally internal legislation and its possible execution” (KU 5:447–8). Since the former is a perspective that an impartial reason cannot accept (cf. KpV 5:110), moral teleology takes the latter path. The scene is thus set for the proof of the existence of God. However, before shifting to Kant’s argument, it is worth briefly pausing on the notion of a final end. As we will see, the commitment to the final end is the true premise of the proof, but there seems to be an ambiguity in that notion.

 This may be disputed since Kant seems to suggest that the relation between virtue and happiness is causal: “In the highest good […] virtue and happiness are thought as necessarily combined […]. Now, this combination […] must be thought synthetically and, indeed, as the connection of cause and effect, because it concerns a practical good, that is, one that is possible through action” (KpV :). As a practical good, the highest good should be “made real” through action; it should be a possible effect of our actions. However, it is not easy to conceive of the relation between virtue and happiness as a causal connection. Kant’s claim that happiness depends on virtue certainly means that it somehow has to be a consequence of virtue, but that the dependence in question is causal seems hard to believe. To assume that the virtue-happiness relation is one of causal dependence would, very roughly, be to hold the conditional that if a person is virtuous then she is happy, and if she is not virtuous then she is not happy. Unfortunately, things do not go this way, and Kant does not seem to assume the contrary (cf. footnote ). Maybe there is room for not interpreting literally the “must be thought synthetically and, indeed, as the connection of cause and effect”. According to Kant, the relation between virtue and happiness is a relation of “being worthy of”: virtue does not make a person happy but rather makes her worthy of happiness. In a sense, it is relevant to, but not responsible for, the happiness of a person since it does not generate the factual conditional ‘if a person is virtuous, then she is happy’. If the connection between virtue and happiness is not stricto sensu a causal relation, Kant’s suggestion that we should conceive of it according to the model of a causal relation may be meant to convey the idea of a kind of necessary connection. This does not take causality out of the scene. Causality is indirectly involved in the connection between virtue and happiness. The idea of a dependence of happiness on virtue might be expressed by the thought of a condition of which virtue is a part that is sufficient for happiness while virtue by itself is not. Another part of this condition is a noumenal cause, namely God, by which, as Wilhelm Vossenkuhl glosses, “virtue is indirectly connected with happiness” (Vossenkuhl : ).  Moore : .

118

Gabriele Tomasi

2 On the Final End and the Existence of God I have mentioned that Kant introduces the notion of a final end in the discussion of nature as a purposive whole. In that context, he considers the human being “the final end of creation” (KU 5:435, 443). However, he also conceives of the final end of creation as “that constitution of the world” which corresponds with “the final end of our pure practical reason, insofar as it is to be practical” (KU 5:454–5) – that is, as a constitution of the world in which the human being’s final end, namely the highest good (cf. KU 5:450), is realized. These two conceptions seem to be different: according to the first, a particular being in the world is considered its final end; according to the second, however, a condition or a state of the world is considered its final end.¹⁶ Nonetheless, they are not inconsistent. A world in which the human being (as a moral being) is the final end should also be a world in which that being’s final end can be realized. Be that as it may, it is worth noticing that, in the eyes of Kant, the notion of the final end seems to have theological implications. This is evident, for instance, in the title given to a section devoted to that notion (§84): “On the final end of the existence of a world, i. e., of creation [Schöpfung] itself” (KU 5:434). In the Critique of the Power of Judgment, Kant does not commonly refer to the products of nature as creatures, or to the world as creation.¹⁷ The reference to ‘creation’ in this title is therefore significant. Still more significant is the fact that, in the body of §84, Kant claims that if there is a “productive understanding”, then “the final end” for which the world exists should also be “the objective ground” that could have determined an intelligent cause to create it (KU 5:435). Since Kant had previously argued that the human being, as a moral being, is the final end and that, through what one does “purposively and independently of nature”, one can give life a value such that “even the existence of nature can be an end” (KU 5:434n.), he can now claim that, at least insofar as the human being qualifies as a moral being, his existence is also the ground that “the highest reason would require” for the creation of nature (KU 5:436n.).¹⁸ But if the existence of a moral being is the objective ground for the creation

 I owe this point to Cunico : –.  These terms occur, e. g., in KU :, , , .  In § he restates the point, claiming that “only of the human being under moral laws can we say, without overstepping the limits of our insight, that his existence constitutes the final end of the world” (KU :n.).

God, the Highest Good, and the Rationality of Faith

119

of a world, than the world cannot but have a constitution which corresponds to the final end of pure, practical reason.¹⁹ Obviously, Kant does not assume here that such a “highest reason” exists. Kant’s reasoning is hypothetical. He simply suggests that if there is an intelligent cause of the world, or a God, the existence of a moral being, namely of an objective supreme end, is the ground that He would require for creating the world. Given the traditional concept of God, this claim might be justified even in the eyes of someone who thinks that there is no God. That such an author of the world exists is what the moral proof proposed at the core of §87 tries to make rationally acceptable. As we will now see, what Kant proposes is an argument that, rather than constituting a proof in the strictest sense of the word, is better interpreted as part and parcel of a rational attempt at making ethical sense of things from the perspective of a morally committed person.

Kant’s Moral Proof of the Existence of God Kant had already hinted at there being a question of a moral author of the world in a passage in the “Remark” at the end of §86, offering a preview of the proof. In that passage, he hinted at the paradoxical situation in which the moral agent finds herself, since she feels forced by the moral law to strive for the highest good, while “at the same time” she judges herself and nature to be incapable of attaining that end – that is, a synthesis of virtue and happiness. In light of this disconcerting situation, Kant maintains that we have “a pure moral ground of practical reason” to assume an intelligent world-cause (KU 5:446), thereby hinting that the real premise of the moral proof of the existence of God is the commitment to the highest good on the one hand and the recognition that we

 I use the terms ‘world’ and ‘nature’ in a rather loose way, thereby glossing over their possible double reference to both the empirical and the noumenal world. Kant, in fact, seems to presuppose this double meaning of the terms. In a sense, he needs it because, while on the one hand he states that the “final end […] can become actual only in nature and in accord with its laws” (KU :), on the other he observes that nature does not pay attention to a person’s worthiness to be happy (cf. KU :). This suggests that the accord between nature and the moral order, required for the connection between virtue and happiness, is possible only in the noumenal world. However, if this is the case, the question arises how we can consider the highest good as a practical idea, namely as a possible aim of our actions (for otherwise we cannot be committed to it). As we will now see, Kant resorts to God as the moral author of the world to answer precisely this question. Cf. Ricken : –.

120

Gabriele Tomasi

are not capable of attaining it on the other. Roughly, the proof as he formulates it in §87 runs as follows.²⁰ Kant starts from the double assumption that the moral law determines for us, a priori, a final end, namely “the highest good in the world possible through freedom”, and that it makes it “obligatory for us” to strive after that end (KU 5:450). This double assumption articulates the first premise of the proof. Kant then unpacks the content of the final end and distinguishes within it both a subjective and an objective condition, corresponding to the two elements of the highest good. He states that the subjective condition under which we can set a final end for ourselves under moral law is happiness. Hence the highest (physical) good that we have to promote as a final end, “as far as it is up to us”, is happiness, although only “under the objective condition of the concordance of humans with the law of morality, as the worthiness to be happy” (KU 5:450). As a second premise of the proof, Kant reiterates what he had already claimed in §86: “Given all of the capacities of our reason”, it is impossible for us to represent the two requirements of the final end “as both connected by merely natural causes and adequate to the idea of the final end as so conceived” (KU 5:450). Given this impossibility and our moral commitment to the highest good, he concludes that, in order to set this end before ourselves, we must assume a moral cause of the world (an author of the world) that guarantees its attainability.²¹ In the next section, we will consider more closely how Kant qualifies his proof. First, however, two points are worth noting. The first concerns the second premise of the proof, namely the non-representability claim. Interestingly, Kant does not make an empirical point in this regard: he does not merely observe, for instance, that morally good people often suffer and evildoers often enjoy undeserved happiness. Rather, his point seems to have both an epistemic and a metaphysical character. His claim is that we cannot understand how the two conditions of the highest good, namely virtue and happiness, can be connected in the context of the causality of nature. Furthermore, Kant establishes a sort of incongruity between the concept of the moral necessity of the final end and the concept of the (physical) possibility of producing it. This incongruity, which somehow also concerns the fabric of the world, must be resolved if we are to

 For a more detailed interpretation, cf. Cunico : –.  Kant’s argument goes along the same lines as that for the postulates of God and immortality in the Critique of Practical Reason. As we will see, though Kant makes little use of the term ‘postulate’ in these sections of the third Critique (cf. KU :, ), the assent based on the moral proof clearly has a doxastic status akin to that of the postulates. On Kant’s postulates of practical reason, cf. Ricken , and, for an excellent discussion of their meaning and epistemic standing, Willaschek .

God, the Highest Good, and the Rationality of Faith

121

conceive of the world as a place in which the final end is to be realized. As we have seen, it is precisely the need to deal with this question that drives reason to shift from moral teleology to theology. Secondly, it is worth noting that Kant refers to his argument as a ‘moral proof’. Presumably, the reason for this is that the proof requires a moral point of view, which also indicates that the necessity of the conclusion is of a moral kind. This latter statement should be qualified. The argument actually operates under the premise that it is morally necessary for us to aim at realizing the final end in the world. However, the sense of ‘must’ in the statement ‘we must assume a moral cause of the world’ is not the sense in which we must do our duty. That the proof is established from a moral premise does not entail that the conclusion also has this character, namely that it is morally necessary to assume that “there is a God” (KU 5:450). This would turn the affirmation of God, which, as an assumption of existence, is a theoretical proposition, into a moral duty, but it does not seem that we can take something to be true at will even if practical reason would seem to require it (cf. KU 5:455–6). What, then, is the source of the necessity expressed by the ‘must’ that leads to the conclusion that God exists? The series of observations made by Kant in the second part of §87 and in §88 (commenting on his proof) help us to answer this question and to clarify the kind of validity he attributes to his argument. As we will now see, the existence of God is assented to in view of a coherent and functioning commitment to morality.

3 The Reasonable Moral Person and the ‘Incoherent’ Atheist Having formulated his moral proof, as if he were distancing himself from it, Kant goes on to contemplate the possibility of a person who is firmly convinced that there is no God. At first glance, this might appear strange. Since the aim of proofs is to convince, to consider such a figure is to somehow allow for imperviousness to the moral proof that God exists. However, Kant is not suggesting that the proof falls short of the rigor required for convincingness; rather, he is making clear that it is not the aim of his argument to produce a conviction or, more precisely, a logical or theoretical conviction (cf. Logik 9:72, 73n.). The moral proof, he claims, “is not meant to provide any objectively valid proof of the existence of God, nor meant to prove to the doubter that there is a God”; what it means to prove is that “if his moral thinking is to be consistent, he must include the assumption of this proposition among the maxims of his practical reason” (KU 5:450–1n.). Even if the moral argument cannot have an effect on the theoretical convictions of a doubter, it might well have it on her practical convictions.

122

Gabriele Tomasi

Kant’s admission that the moral proof cannot prove the existence of God to the doubter comes as no surprise. A (theoretical) conviction is an assent to a proposition for which the subject has sufficient objective grounds, namely, grounds that indicate “with a moderate-to-high-degree of confidence” that the proposition is true.²² Therefore, a proof can have an effect on (theoretical) conviction only if its basis is an objectively valid ground or what Kant calls “a logical ground for cognition” (KU 5:461).²³ As a matter of fact, the moral proof, resting on our commitment to the final end, lacks such a basis, and therefore it cannot have an effect on theoretical conviction.²⁴ However, this weakness does not particularly affect it since it aims not at knowledge or theoretical conviction but at practical conviction.²⁵ These considerations suggest that the necessity expressed by the ‘must’ that grounds the conclusion of the moral proof, though connected to the form of the argument, does not have a merely logical character. I postulate that it originates from the feeling that we have no alternative but to assume that “there is a God” (KU 5:450) when making sense of the moral necessity of aiming at the highest good. However, before I corroborate this claim, there is a further point related to the necessity of the conclusion of Kant’s argument that is worth noticing. Kant endorses two ideas. On the one hand, he states that, insofar as this final end is necessary, it is also necessary, “in the same degree and for the same reason”, to assume that there is a God. On the other hand, he is careful to distinguish the necessity of acknowledging the validity of the moral law from that of assuming the existence of God. Kant clearly wants to avoid the argument that “whoever cannot convince himself” that God exists might judge himself “to be free from the obligations” of the moral law. All that would have to be surrendered, he claims, is “the aim of realizing the final end in the world” (KU 5:451). This statement is revealing. It suggests that the moral necessity of the highest good does not have “the same degree” as that of the moral law. Modifying the conception that he endorsed in the Critique of Practical Reason, Kant has now loosened the direct connection between

 Chignell : . According to Chignell, this is just one of the features of an objective ground.  “We have conviction about a thing when we cognize it as true with the consciousness that our judgment is objective” (Log-DW :).  This is not a specific fault of such a proof. According to Kant, no logical ground is possible in the case of the existence of God. It is no coincidence that, in his illustration of the person firmly convinced that there is no God, Kant connects that person’s conviction partly to “the weakness of all the speculative arguments” aimed at proving that God exists (KU :).  According to Kant, a practical conviction is what “can be called a belief in the proper sense” (Logik :). As we will see in the last section of this paper, it is precisely as a Glaubenssache for pure practical reason that the moral argument proves the existence of God.

God, the Highest Good, and the Rationality of Faith

123

the moral law and the highest good. He still claims that the highest good is an end imposed upon us by the moral law; however, he does not deem a commitment to it to be as practically necessary as duty itself. Therefore, he has no need to postulate God in order to assure us that the object imposed by the moral law is attainable and the law that prescribes it is not chimerical (cf. KpV 5:114, 119, 122). By turning the highest good into an object of the practical power of reflecting judgment,²⁶ Kant can reaffirm that we are forced by the moral law to strive for the highest good while also claiming that the moral law is binding, even if one is agnostic about the realizability of the highest good or, in the worst case, considers it an impossible state of affairs. Why, then, doesn’t he leave God outside of the moral scene? Kant discusses this possibility, evoking the case of “a righteous man (like Spinoza)” who takes himself to be firmly convinced that there is no God but who will nonetheless unselfishly “establish the good” to which the moral law “directs all his powers” (KU 5:452). This seems a perfectly rational position in the sense that it is consistent with reason. Indeed, there is nothing contradictory in the idea of the highest good: it is a logically possible state of affairs according to the course of nature. The point is that our reason is not able to conceive of its real possibility without assuming divine intervention. Kant claims that if morality and its rules “can very well exist without theology”, surely the final purpose imposed on us by the moral law “cannot exist without theology, for then reason would be at a loss with regard to that final end” (KU 5:485). If the highest good is to be not only logically but also practically possible, we must be able to make it an end of our actions, which, as Marcus Willaschek notices, in turn requires “that we can understand how we might be able to realize” it through our actions.²⁷ And since the only way we can understand this is by appealing to God, the Spinozian attitude may appear, to borrow an expression from Audi, not completely consonant with reason.²⁸ Therefore, while Kant does not consider atheism to be fatal for morality, as the motivational force of the moral law does not depend on theological assumptions, he views a Spinoza-like moral attitude as untenable; it is not the kind of attitude that might be expected from a reasonable person – a person governed by reason – if consistency, or at least the attempt to avoid incoherence, is a condition of having reason. A person committed to the moral law, Kant argues, must assume the proposition ‘God exists’ to be among the guiding principles of her practical reason if her moral thinking “is to be consistent” (KU 5:451n.) – that is, if she wants to avoid the embarrassing

 Cf. Brandt : –.  Cf. Willaschek : –.  Cf. Audi : –.

124

Gabriele Tomasi

situation of holding two incompatible beliefs, namely that the moral law imposes upon her a final end as a duty, and that the world in which that final end should become real is such that its realization is impossible (cf. KU 5:458). Actually, the situation of the righteous atheist, as Kant depicts it, is rather paradoxical. As a well-intentioned person who conforms to the moral law, she has the final end in view and aims at it in her moral conduct. However, to the extent that she is convinced that God does not exist, hers is a standpoint from which it is impossible rationally to conceive of the attainability of the highest good – that is, of the aim pursued in moral actions, the possibility of which she might already doubt on the basis of a consideration of the injustice of the world and the contingent way in which events harmonize with that end. Consequently, as Kant has already noted, she should refrain from consciously aiming at it. Why shouldn’t we make this move and settle for a more modest aim? This appears to be a possible (and also rational) move. However, it is worth recalling that the highest good is not really a particular state of affairs that we can make the end of our actions, as we do with other objects. As an object of the practical power of reflecting judgment, it is somehow the unifying aim of our moral life and, furthermore, the aim which also gives expression to our sense of justice, namely to the feeling that “it could not in the end make no difference if a person has conducted himself honestly or falsely, fairly or violently” (KU 5:458). Though Kant’s doubt about the Spinoza-like moral attitude may have a pragmatic ground,²⁹ I am inclined to think that it is deeply related to this feeling. Kant might have viewed a Spinoza-like moral attitude as untenable because it requires us to give up the idea, deeply connected to our reflection “on right and wrong”, of an “inner vocation” of our mind. In fact, if the “ordinary course of the world” is regarded “as the only order of things”, it is hard to reconcile that vocation with an “outcome” of unrewarded virtue and unpunished wrongdoing. Kant is convinced that reason resists this view, and the way it “straightened out” that “irregularity” and “thought up” a principle of the possibility of the unification of nature with the “inner moral law” of human beings was the representation of a “supreme cause ruling the world in accordance with moral laws” (KU 5:458). Kant thinks that his moral proof simply offers a logical articulation of an argument laid down “in the human faculty of reason” (KU 5:458), which suggests that, in his view, it may be difficult for a person committed to morality to make  Kant could have thought that, since we are not pure moral beings but embodied moral beings, giving up the final end as impossible would inevitably weaken the moral disposition. In Kant’s view, given the kind of beings we are and the nature of our reason, faith that God exists has a “reactive relationship” (Insole : ) with our sense of obligation and moral motivation. On this cf. also Pasternack : –.

God, the Highest Good, and the Rationality of Faith

125

ethical sense of herself and of the world without assuming “an author and ruler of the world who is at the same time a moral legislator” (KU 5:455), namely a God, or without at least articulating her worldview in religious language. Kant seems to think that the set of beliefs of a righteous person should also include the belief that God exists. But is he really thinking of what we now call a belief in this case? Beliefs are a kind of cognitive commitment and are truth-valued. I propose that Kant is presumably thinking of what we could more appropriately consider a kind of faith: a faith that has a cognitive or propositional content, to be sure, but one that does not entail belief. With regard to this, it is revealing that Kant qualifies as subjective the necessity of assenting to the conclusion of the moral proof.

4 Moral Proof, Belief, and Faith As we have seen, Kant claims that his moral argument for the existence of God cannot prove to the doubter that there is a God. To emphasize that this argument has a limited kind of validity, he repeatedly points out that the reality of God is “adequately established merely for the practical use of our reason” or “for the practical power of reflecting judgment” (KU 5:456). Kant therefore makes clear that the moral proof establishes the existence of God from a practical point of view only, namely, as he explains, “in order to form a concept of at least the possibility of the final end” that is prescribed to us by morality (KU 5:453). He also hints at the fact that such an assumption, made in the interest of consistent moral thinking, does not determine anything with regard to the existence of God from a theoretical point of view. The moral proof is not a performance of the determining power of judgment, and the ground that it provides for the conclusion that God exists, though it affords us something necessary for comprehension, has a non-epistemic character; that is, it does not directly indicate that this conclusion is true. This is made clear by Kant’s claim that his argument “proves the existence of God only as a matter of faith [als Glaubenssache] for practical pure reason” (KU 5:475). Faith (Glaube) is a kind of “holding-for-true” that he qualifies as assent to a content that is only subjectively sufficient and at the same time held to be objectively insufficient (cf. KrV A822/B850). To approximate, in Kant’s view a subject’s assent to a proposition is objectively sufficient if and only if the subject has sufficient objective grounds for assent, namely, grounds that provide evidence based on experience or on reason about the object of the proposition (cf. Logik 9:70). On the contrary, the ground relevant to faith is non-epistemic. Faith is assent to a proposition that is only subjectively sufficient, that is,

126

Gabriele Tomasi

based on a ground that has sufficient non-epistemic merit for the subject who assents to the proposition.³⁰ Therefore, by claiming that his moral argument proves the existence of God only as a matter of faith, Kant states that it offers only a subjectively sufficient ground for assenting to the existence of God. If he might still call it a “proof”, it is because, offering such a ground with regard to an aim in accordance with the moral law, it can count as subjectively sufficient for every moral agent (cf. KU 5:451), and, as such, it might make “a sufficient claim of moral conviction” – that is, conviction from the point of view of the interests of each subject committed to the moral law (KU 5:463). It is in this sense, and not in any idiosyncratic sense of the word ‘subjective’, that the moral proof “is a subjective argument” (KU 5:451), which incidentally means that such a qualification does not deprive the belief/faith that God exists of objectivity. It is objective, at least, in the broader sense of ‘objective’ that, as Andrew Chignell points out, “applies to any assent that is rationally acceptable for someone in the assenting subject’s position”.³¹ In the light of these considerations, it might be argued that, though the moral proof does not offer evidence for the existence of God, it nevertheless establishes the rational acceptability of the belief that God exists. It does this by showing that, for everyone who is committed to the demands of the moral law and is interested in having a coherent moral self-understanding, assenting to the proposition that God exists has the non-epistemic value of allowing him or her to avoid a sense of practical incoherence.³² This incoherence can take the form of a chasm in the use of reason between the practical necessity of the highest good and the epistemic impossibility of reason’s representing morality and happiness as both connected by merely natural causes and adequate to the idea of that final end. With regard to this, there is still another point I wish to mention. In enumerating the final end as a commanded effect among the matters of faith (res fidei) (cf. KU 5:469–71), Kant states that faith (Glaube) “(simply so called) is trust [ein Vertrauen] in the attainment of an aim the promotion of which is a duty but the possibility of the realization of which it is not possible for us to have insight into” (KU 5:472). Trust

 I have followed Chignell  both in characterizing objective sufficiency (cf. Chignell : ) and in using ‘merit’ to designate a property of assent “that makes it valuable or desirable for a particular subject given his or her goals, interests, and needs” (Chignell : ). On the epistemic standing of belief/faith cf. also Ameriks .  Chignell : . Obviously, the grounds that justify faith that God exists are subjective (and not objective) because, though they are valid for every moral being, “they still do not provide insight into reality” (Beiser : ).  Cf. Chignell : –.

God, the Highest Good, and the Rationality of Faith

127

is an important element in faith. We cannot have faith in something we do not trust, but we may have faith that something is so while doubting it to a degree that prevents us from believing that it is so. Usually, the closer we are to being sure that something is so, the less appropriate it is to say ‘I trust that’.³³ Kant speaks of “trusting the promise of the moral law”. However, he does not mean that the moral law itself somehow contains a promise. Rather, it is the moral subject that ‘put the promise’ into the law on the grounds that a final end cannot be commanded by any law of reason “without reason simultaneously promising its attainability” (KU 5:471n.). The trust Kant speaks about is primarily trust in reason itself (cf. Logik 9:69n.). The way reason promises the attainability of the highest good is by permitting assent to the only conditions under which it can conceive of it, namely the existence of God and the immortality of the soul. If this is correct, then to have faith that God exists is to trust reason itself and its coherence. The fiducial element present in faith, emphasized by Kant’s equation of faith with trust, is above all a trust in reason. Therefore, faith that God exists is not only rational but also consonant with reason. And this is what the moral proof ultimately shows.

5 Conclusion: Faith, Belief, Acceptance In this paper, I have considered the context, aim, and reach of the moral proof of the existence of God as formulated by Kant in §87 of the Critique of Judgment. We have seen that Kant’s moral proof offers a ground for a subjectively sufficient holding-for-true of the proposition that God exists. What Kant does, in a sense, is to put in the form of an argument the idea that, if we do not assent to this proposition, it may be hard for us to conceive of ourselves as rational agents self-consciously committed to the end imposed on us by the moral law. To Kant, there is apparently no other way for our reason to conceive of a constitution of the world which corresponds to the final end imposed upon us by the moral law than to turn to the idea of a moral author of the world. Therefore, he thinks that we must believe that God exists if we are to be consistent in our self-conscious commitment to the moral law. Such a faith plays what could be called a “sense-conferring” role in our life.³⁴ The strength of Kant’s moral proof is clearly at one with the epistemic fact that all other possibilities for conceiving of the attainability of the highest good seem out

 On trust and faith, cf. Audi : –.  Cf. on this Moore : .

128

Gabriele Tomasi

of reach for our reflecting judgment. However, our apparently having no alternative but to affirm that God exists when it comes to conceiving of the attainability of the highest good does not entail the truth of the proposition ‘God exists’. As Moore observes, it might be that the absence of alternatives prevents us from converting into agnosticism the recognition that the truth of that proposition is theoretically indeterminate,³⁵ but the possibility that it is false remains. According to the interpretation suggested in this paper, the moral proof, as a subjective argument, is not an argument for the truth of the belief that God exists. It does not help us to close the gap between our having to assent to the proposition that God exists and the truth of the assertion of His existence. In Kant’s view, this does not affect the rationality of the holding-for-true grounded by the moral argument. The attitude here is that of moral faith rather than knowledge; therefore, the justification for the assent is different in kind from that required by knowledge. Kant claims the rationality of a theistic worldview without presupposing its truth. That God exists is not an incoherent thing for us to believe (in the Kantian sense), and this not only because ‘God exists’ is a logically possible proposition that we cannot theoretically refute. Rather, the point is that, for a subject committed to morality, belief in God may have non-epistemic merit since such a belief plays a role in her efforts at making sense (or a certain sense) of her engagement with the demands of practical reason. It is required by reason on pain of practical incoherence. Accordingly, although belief in God is not strongly justified, since the assent is only subjectively sufficient, it is nevertheless supported by reason. It has appropriate grounds: grounds for faith, that is, non-epistemic advantages, which by their nature do not dissolve theoretical uncertainty about the objective reality of the concept of God. Therefore, although it is rational, moral faith in God is compatible with a degree of doubt; “dubiety [Zweifelglaube]” (KU 5:472) can be reconciled with a positive attitude towards, or fiducial acceptance of, the proposition that God exists.³⁶

References Ameriks, Karl 2008, “Status des Glaubens (§§ 90–91) und Allgemeine Anmerkung über Teleologie”, in: Otfried Höffe (ed.) 2008, Immanuel Kant. Kritik der Urteilskraft, Berlin: Akademie Verlag, 331–49 Audi, Robert 2011, Rationality and Religious Commitment, Oxford: Clarendon Press

 Cf. Moore : .  I am grateful to Carolyn Benson, Thomas Höwing and Wilhelm Vossenkuhl for comments that greatly helped me to improve my paper.

God, the Highest Good, and the Rationality of Faith

129

Beiser, Frederick C. 2006, “Moral Faith and the Highest Good”, in: Paul Guyer (ed.) 2006, The Cambridge Companion to Kant and Modern Philosophy, Cambridge: Cambridge University Press, 588–629 Brandt, Reinhard 2007, Die Bestimmung des Menschen bei Kant, Hamburg: Meiner Chignell, Andrew 2007, “Belief in Kant”, Philosophical Review 116, 323–60 Cunico, Gerardo 2008, “Erklärungen für das Übersinnliche: Physikotheologie und moralischer Gottesbeweis (§§ 85–89)”, in: Otfried Höffe (ed.) 2008, Immanuel Kant. Kritik der Urteilskraft, Berlin: Akademie Verlag, 309–29 Dörflingler, Bernd 2004, “Führt Moral unausbleiblich zur Religion? Überlegungen zu einer These Kants”, in: Norbert Fischer (ed.) 2004, Kants Metaphysik und Religionsphilosophie, Hamburg: Meiner, 207–23 Förster, Eckart 2000, Kant’s Final Synthesis. An Essay on the Opus postumum, Cambridge (Mass.)/London: Harvard University Press Höffe, Otfried 2008, “Der Mensch als Endzweck (§§ 82–84)”, in: Otfried Höffe (ed.) 2008, Immanuel Kant. Kritik der Urteilskraft, Berlin: Akademie Verlag, 289–308 Insole, Christopher 2008, “The Irreducible Importance of Religious Hope in Kant’s Conception of the Highest Good”, Philosophy 83, 333–51 Moore, Adrian W. 2003, Noble in Reason, Infinite in Faculty. Themes and Variations in Kant’s Moral and Religious Philosophy, London/New York: Routledge O’Connell, Eoin 2012, “Happiness Proportioned to Virtue: Kant and the Highest Good”, Kantian Review 17, 257–79 Pasternack, Lawrence R. 2014, Kant on Religion within the Boundaries of Mere Reason, London/New York: Routledge Pinkard, Terry 2011, German Philosophy 1760–1860. The Legacy of Idealism, Cambridge: Cambridge University Press Ricken, Friedo 2004, “Die Postulate der reinen praktischen Vernunft”, in: Norbert Fischer (ed.) 2004, Kants Metaphysik und Religionsphilosophie, Hamburg: Meiner, 161–77 Tafani, Daniela 2006, Virtù e felicità in Kant, Firenze: Leo S. Olschki Editore Vossenkuhl, Wilhelm 1987-88, “The Paradox in Kant’s Rational Religion”, Proceedings of the Aristotelian Society, New Series 88, 179-92 Watkins, Eric 2010, “The Antinomy of Practical Reason: Reason, the Unconditioned and the Highest Good”, in: Andrews Reath/Jens Timmermann (eds.) 2010, Kant’s Critique of Practical Reason. A Critical Guide, Cambridge: Cambridge University Press, 145-67 Willaschek, Marcus 2010, “The Primacy of Practical Reason and the Idea of a Practical Postulate”, in: Andrews Reath/Jens Timmermann (eds.) 2010, Kant’s Critique of Practical Reason. A Critical Guide, Cambridge: Cambridge University Press, 168–215

Stephan Zimmermann

Kant on “Moral Arguments”: What Does the Objectivity of a Postulate of Pure Practical Reason Consist In?¹ The epoch which Kant in the Critique of Pure Reason describes as the “age of criticism” (KrV Axin.) was, amongst much else, a high point of criticism of religion. The Enlightenment movement of the modern era and its methodical conscience, which permitted only that which can prove itself before the tribunal of human reason, battled with the tradition of the Church’s orthodoxy, i. e. its dogmatism, ethics, and interpretation of the Bible. Divine revelation, as the ultimate explanatory source (having itself earlier prevailed over the telling and passing down of myth), was brought from this point on into sharp opposition with reason, which relied simply on itself and, furthermore, had confidence in the scientificity of its efforts to engage with fundamental insights into all areas of human concern. But it was especially “liberation from superstition” (KU 5:294), as Kant notes, that truly earned the illustrious title of enlightenment. Critical engagement with religion in no way amounted to sheer iconoclasm; it did not aim to dispose of it as a whole. Rather, according to Kant, the ruling maxim of all enlightenment must be that the credibility of religious contents depends on that authority which human reason grants them with a cool head. The natural theology which Kant, from the Critique of Practical Reason onward, outlines as a doctrine of so-called postulates, aims to separate true faith, which is fitting for mankind, from superstition, by seeking to reveal the origin of its fundamental principles in the universal condition of human rationality. The move towards transcendence is supposed to be included in the structure of our subjectivity, thus constituting an unshakeable aspect of our orientation to the world. As such, it clearly points beyond the narrow limits of a political state or organized religious community. The legitimate elements belonging to the traditions and sacred texts of religious creeds reveal themselves, according to Kant, via reasons held inherently by every finite rational being. That the human soul is immortal, our will free, and God is are the first principles of philosophically enlightened faith.

 For helpful suggestions and hints I thank Benedikt Grimm, Theo Kobusch, and Marcus Willaschek.

132

Stephan Zimmermann

Two years later, in the Critique of the Power of Judgment, Kant again discusses these subjects. At the end of the “Methodology” of the second part, which deals with questions of teleology, he describes the search for and justification of these necessary propositions of faith, the postulates, as “moral arguments” (KU 5:459)² or “moral proofs” (KU 5:447), both for the first time and repeatedly.³ The final paragraphs of the entire work, paragraphs 86 to 91, are especially relevant in this regard. Kant uses the expressions “moral argument” and “moral proof” terminologically. That is to say, not everything that is dealt with in moral philosophy is already part of such an argument; not every proof that it presents is to be designated as a moral one in the proper sense. Only particular proofs are labeled as such, namely those that motivate the tenets of the true common human religiousness. Yet Kant’s talk of moral arguments is far more than a linguistic innovation. He does not retroactively assign to a kind of justification a particular name, with which the fastidious reader is already familiar from the second Critique.⁴ Indeed, a number of Kant’s comments, which are now gathered together under the title of moral proof and illuminate different aspects of this type of proof, can already be found in previous writings. In addition, however, the explanations of the third Critique do represent progress in at least one important respect, for they give cause for clarification of a tricky question, which until then remained open or had at any rate not been satisfactorily answered by Kant. This question is: in what exactly does the “increment” (KpV 5:134), as Kant puts it in the Critique of Practical Reason,⁵ consist when the immortality of the human soul, the freedom of our will, and the existence of God are postulated? How does such postulating, which is said to have a peculiar objective content unfathomable to theoretical reason, surpass the sheer subjectivity of the underlying ideas of the soul, freedom, and God? I therefore want to pose the question: what does the objectivity of a postulate of pure practical reason consist in? And I want to show that Kant’s remarks on moral arguments entail an interesting answer to this question.

 See, for example, KU :, , .  See, for example, KU :, , .  Or from the Critique of Pure Reason where the corresponding argument for the existence of God is called “transcendental” (KrV A/B).  See also KpV :, .

Kant on “Moral Arguments”

133

I Let us first turn our attention to the crucial premise with which Kant supports moral proofs. This premise not only carries the whole burden of proof, but also, thanks to its distinctive character, determines in advance what can and cannot be concluded on the basis of it. By ‘argument’ one generally understands a certain connection of judgments. This connection is such that from one or several antecedent propositions a concluding proposition, as Kant’s Logic compendium edited by B. G. Jäsche puts it, is “derived”.⁶ In this context, derivation means that, beyond the mere formal correctness of the connection between the propositions, the assent that one gives to the content of the former in successful cases is transmitted to that of the latter. Whoever accepts the premises cannot dismiss the conclusion. Normally, we associate such a relation of justification with a special sort of judgment, namely judgments of experience in the widest sense. Kant himself speaks, mainly in the Critique of the Power of Judgment, of “theoretical” propositions, differentiating them from practical, aesthetic, and teleological ones.⁷ Judgments of experience are judgments that are accompanied by the claim that the spatiotemporal circumstances to which they refer are such, could be such, or must be such as they claim to be. They are truth-apt statements. Kant calls them “theoretical” because they refer to the particular circumstances via sensible intuition (gr. theōrós). Thus, an argument consisting of such propositions, if it is successful, derives the truth of a judgment from the truth of one or several other presupposed judgments. It induces us to believe something to be true, only because we are already convinced of the truth of something else. A moral argument, however, with which Kant’s religious criticism operates according to its own self-understanding, is not an argument of this type. What serves as antecedent proposition here is anything but a judgment whose truth value can be decided with the help of empirical evidence. Consequently, it is not a question of the possible, actual, or necessary cognition of an object in the sense of experiencing it. Kant puts religious faith on another basis. As is well known, the starting point for deriving the postulates is, for Kant, the highest principle of morality, the moral law – more precisely, the categorical imperative. The articles of faith of the religiousness intrinsic to every human being, which moral arguments are thought to exhibit and prove, are considered to lie within  “By inferring is to be understood that function of thought whereby one judgment is derived from another. An inference is thus in general the derivation of one judgment from the other” (Logik :).  Cf. KU :, , , ; EEKU :, .

134

Stephan Zimmermann

our moral-practical subjectivity. As such, it is not only the nature of our reason that advances as the sole basis of a rationalizing reading of biblical texts, which Kant will later elaborate on in his writing on Religion. Above all, the Christian relationship between dogmatics and ethics reverses itself completely: it is not by virtue of divine revelation that certain behavioral norms are compulsory, but, quite the opposite, it is the conviction of an individual about the binding force of the morality of reason that makes a certain way of being religious, determined by reason, unavoidable.⁸ We must be clear on this: the moral law is at base neither a true judgment nor even capable of truth – i.e., it is not a theoretical judgment at all. It does not state anything, and it cannot be confirmed or refuted by anything that is or is not the case in the sensible world. Note as well that it is not a statement about the moral law that stands at the beginning of a moral proof, but the law itself. The postulates of that proper faith, in which all human beings can know themselves to be connected despite all possible differences at the level of denomination and mode of worship, have their rational criterion in the moral ‘ought’ common to all human beings. Moral arguments begin with a judgment of duty, which does not say at all what there is, was, or can be, but which prescribes, against any possible opposition from sensible interests, what every rational being, e.g. the human being, should always do. Precisely for this reason, Kant briefly labels them “moral” arguments. But what, then, can be deduced from this kind of premise? Hardly the truth or falsity of a statement. One may hear the stern voice of the categorical imperative clearly within oneself – but this does not permit one to make any conclusions about facts in the empirical world. Kant himself stresses the importance of this in different formulations. Already in the Critique of Pure Reason he says, among other things, that the moral law’s imposing itself on us does not lead to any “cognition” or “knowledge”⁹ and that it does not extend pure reason for “speculative” or “theoretical” purposes.¹⁰ In addition to this, Kant explains in the second Critique that postulates possess no “theoretical objective reality”, as he there puts it.¹¹ These and other expressions can subsequently be found in the Critique of the Power of Judgment as well. However, the claim that they express is merely a negative one, namely that postulates are judgments whose objects, as a matter of principle, are not accessible via perception. The concluding proposition of a moral argument is thus also not a judgment of experience with a

 On the difference between revealed and natural religion, see RGV : f.  Cf. KrV A/B, A f./B f.; KpV :; KU :.  Cf. GMS :n.; KpV :, , ; KU :, .  Cf. KpV :, , , ; KU :, .

Kant on “Moral Arguments”

135

possible truth value: a postulate is not a theoretical judgment and therefore does not possess the objectivity typical of theoretical judgments.¹² Nevertheless, genuine faith does not belong to any extent to the realm of free-playing imagination, for it does not lack real certainty. A moral argument is, as Kant often remarks, an “argument, sufficient for moral beings” (KU 5:450n.).¹³ Acceptance of the postulates is not left to the discretion of the individual or a cultural community. What serves as evidence here, the moral determination of the subject by the moral law, is said to lead necessarily to the concepts of the soul, freedom, and God. Moral proofs infer, generally speaking, from the highest principle of morality to that which is entailed by its accepted validity. A judgment of faith identifies a consequence of the factual obligatory force of the moral law for the thinking of the subject concerned.¹⁴ And this consequence is considered to be an unavoidable one: the moral law determines the content of a judgment of faith with necessity. ¹⁵ I shall here pass over the fact that the postulate of freedom is directly connected to the categorical imperative, whereas the postulates of the soul and of God come into play only via the concept of the highest good (according to which the degree of my lived virtue is supposed to bring about a proportional degree of happiness).¹⁶ What is crucial is that, from the fact of morality, as the way in which the categorical imperative makes itself known to me, it follows that I must believe in the freedom of my will, the immortality of my soul, and the existence of God.¹⁷ And yet in another and for us significant regard, the religious articles of faith do not remain within the boundaries of the subjective. If we follow Kant, the postulates of pure practical reason, unlike the ideas of pure theoretical reason, are not without any correspondence. That is, they are not empty representations, but instead possess an objective counterpart. Kant describes this distinctive feature of the conclusion of a moral proof – the special objectivity of a judgment of faith – with the help of diverse positive characterizations. I shall here pick out the

 Indeed, Kant defines a “postulate of pure practical reason” at one point as “a theoretical proposition”, but he adds immediately “though one not demonstrable as such” (KpV :). The question is then what “theoretical” actually means here.  Cf. KU :, n., . See also an earlier articulation at KrV A f./B f.  See KrV A/B, A f./B f., A/B, A/B, A/B; KpV : f., , , , , , ; KU :, ,  f., , .  Kant calls the necessity with which moral arguments prove a “moral” or “practical necessity”. Cf. KpV :, , .  Cf. KpV :, , , ,  f.,  f., , .  Whether there are postulates other than the immortality of the human soul and the existence of God which may be compatible with the highest good remains a concern that cannot be pursued here. See Wood :  ff.

136

Stephan Zimmermann

three that seem to be the most common and promising in all his writings. According to the first, the principle of morality lays the foundation for an attitude that Kant introduces under the label of moral belief; according to the second, postulates have a genuine practical objective reality; and, according to the third, they present an extension of pure reason for practical purposes. In what follows, I will explain these notions in further detail.

II To do so, I must return to the Critique of Pure Reason, for that in relation to which the postulates postulate something comes directly from Kant’s theoretical philosophy. We must begin here in order to paint a better picture of the entire problem. Like the pure concepts of the understanding, or the categories, the pure concepts of reason, or ideas, are described by Kant in the first Critique as subjective conditions of possible experience. “All our cognition starts”, he writes, “from the senses, goes from there to the understanding, and ends with reason, beyond which there is nothing higher to be found in us to work on the matter of intuition and bring it under the highest unity of thinking” (KrV A298 f./B355). The contribution that reason makes to the construction of experience consists in bringing the manifold of experiences, which are each already conceptualized and ordered by the understanding, into a general unity. The concepts that provide this “highest unity of thinking” are the ideas of soul, freedom, and God. They fulfill a unifying function with regard to the understanding just as the categories do with regard to sensibility.¹⁸ In contrast to the categories of the understanding, however, the ideas of reason do not have an object meaning. When the intellect under the name of reason refers to its own activity as understanding and establishes unity among the various objective cognitions of the latter (which does not take place in a temporally subsequent act), the intellect adds no new predicates to the cognized objects. Kant therefore describes the ideas as “subjective”, as “a subjective law of economy for the provision of our understanding” (KrV A306/B362).¹⁹ For the sake of

 Cf. KrV A/B, A f./B, A/B, A f./B f., A/B.  Cf. KrV A/B, A/B, A/B. Kant, however, sometimes expresses himself in such a way that the ideas indeed do not refer to phenomena “directly [zunächst or geradezu]” (KrV A/B, A f./B, A/B, A/B) and have no “immediate [unmittelbare] reference” (KrV A/B), whilst they do nonetheless possess “indirect [indirecte or mittelbare]” (KrV A/B; Prol :) or “some objective validity” (KrV A/B). This seems to me to be more exact, for the pure concepts of reason are, as they relate to judgments of the

Kant on “Moral Arguments”

137

this mere subjective validity of the ideas, Kant calls their share in the carrying out of our experience a “regulative” one and sets it apart from the “constitutive” role of the objectively valid categories.²⁰ That ideas are subjective means that our reason synthesizes the judgments of the understanding into an absolute whole in accordance with them. This whole transcends the bounds of possible experience, for that which is thought in an idea and serves as a common reference point – the unconditional subject of empirical states of mind, the unconditional ground of empirical effects, and the unconditional ground of the division of all empirical reality – cannot itself become present via sensible intuition. The soul, freedom, and God are not phenomena in space and time. Hence, the activity of speculative reason is not accompanied by any claim to truth. Its judgments are not truth-apt statements about something.²¹ The content of these theoretical concepts of reason now reappears in the articles of faith of moral-practical reason. And this promotion of the ideas in the propositional content of the postulates means nothing other than the acquisition of objectivity, for as Kant’s three aforementioned characterizations show, the postulates exceed the ideas in a certain way. Kant wants to say: with respect to objectivity, to be a postulate is to be more than merely an idea. In contrast to the latter, the former have a certain objective validity. In the Critique of Pure Reason, which does not yet give priority to the concept of a postulate in the sense that concerns us here,²² Kant writes the following on the topic of so-called moral faith: if a judgment is “only subjectively sufficient and is at the same time held to be objectively insufficient, then it is called believing” (KrV A822/B850). “Objectively insufficient” means that the proposition does not refer to any corresponding spatiotemporal fact. As becomes clear from what follows, however, “subjectively sufficient” means that the proposition is instead derived from the moral law.²³ The second Critique follows suit by introducing the concept of a postulate as a technical term. According to Kant, a postulate is to be understood as a thought insofar as it “is attached inseparably to an a priori ununderstanding, indirectly objective because they also link the spatiotemporal objects of those judgments to a whole.  Cf. KrV A ff./B ff.  Kant talks about judgments of reason as being opposed to those of the understanding in KrV A; KU :; EEKU :. And these judgments can still be sensibly acknowledged as ‘theoretical’ because the ideas are by their very nature indirectly objective, as they also, together with the judgments of the understanding, synthesize their spatiotemporal objects into a whole.  Here, Kant briefly develops the concept of a postulate with respect to its essential features but does not make further use of this. Cf. KrV A f./B f.  “[…] that this rational belief is grounded on the presupposition of moral disposition” (KrV A/B).

138

Stephan Zimmermann

conditionally valid practical law” (KpV 5:122). Finally, in the Critique of the Power of Judgment, Kant says, among much else: “Objects that must be conceived a priori in relation to the use of pure practical reason in accordance with duty […] are mere matters of faith” (KU 5:469). What these explanations have in common is that they all claim that being convinced of the obligating validity of the moral law is the sufficient ground for a moral attitude of faith with specific content. Whoever recognizes the obligating force of the categorical imperative, therefore, cannot avoid postulating certain things. We already know this. But what does faith mean here? What does postulating mean? And what kind of thinking is at stake here? The notion of a practical objective reality promises an answer to this. In Kant’s first Critique, ‘objective reality’ is an expression for the object reference of representations in a judgment of experience: they do not have an arbitrary content, but represent what belongs to the determinateness (realitas) of an object. And this, their objectivity, comes about by means of perception. So a judgment is objectively valid insofar as it is fundamentally possible for it to be assessed as true or false in light of a perceptible fact in space and time. The Critique of Practical Reason then distinguishes this now retrospectively labeled “theoretical” objective reality from a practical one. In the final two sections of “Chapter One” in particular, Kant points out with respect to reason’s concept of freedom that it has “objective though only practical reality” (KpV 5:48).²⁴ And Kant speaks in a similar fashion elsewhere in relation to the two remaining contents of faith. Indeed, such practical objective reality is also explicitly characteristic of the postulates of immortality and God. It applies to all of them: the postulates “give objective reality to the ideas of speculative reason” (KpV 5:132).²⁵ It remains open to question, however, what this practical objectivity could be. It is certainly practical, as it is justified by the moral law; this law to a certain extent takes the place occupied by sensible intuition in the realm of the theoretical. But then what does the moral law justify? We cannot infer from the abovecited and similar passages what kind of a strange objective validity this might be – a validity that Kant allots without a single doubt to moral faith from the second Critique onward. That we actually have to speak of a certain kind of objectivity with regard to the postulates is witnessed at last by Kant’s articulation of an extension of reason for practical purposes.²⁶ This extension is indeed also one of the predicative  Cf. KpV : f., , , , , , , , ; KU :, , ; RGV :, n.; Gemeinspruch :, ; MS :.  Cf. KpV : f., , , , , ; KU :,  f.  Cf. KpV : ff.,  ff.; KU :, .

Kant on “Moral Arguments”

139

determinations of the ideas. For instance, in the case of God, in connection with considerations about the highest good, the characteristics of omniscience and omnipotence (among others) are added. On the one hand God must know my innermost moral attitude and therefore know about my worthiness to be happy; on the other hand he must be able to bestow upon me, in a future life, the correct measure of happiness in proportion to my virtue. Neither characteristic is already part of the idea of God, as it is rooted in speculative reason.²⁷ Yet the question of what additional properties are to be attributed in detail to the soul, freedom, and God shall not be my concern here. Rather, the question is: what could it mean to say that God (together with these and other properties) is, or that the human soul is immortal and our will free? I take Kant’s talk of an extension to be exclusively about the determination of objectivity that falls to the ideas when postulating. A particularly expressive passage in this regard can be found in the Critique of Practical Reason: “The above three ideas of speculative reason are themselves still not cognitions […]. Now they receive objective reality through an apodictic practical law […], that is, we are instructed by it that they have objects, although we are not able to show how their concept refers to an object, and this is not yet cognition of these objects […]. Nevertheless, theoretical cognition […] of reason […] is extended by this insofar as objects were given to those ideas by the practical postulates […]” (KpV 5:135; my emphasis).

The ideas of theoretical reason by themselves establish no knowledge of an object at all; in the wake of the moral law, however, they attain objective reality. In the postulates of practical reason they “have” what they previously lacked: i. e., as Kant unequivocally states, “objects”. Nonetheless, and Kant is once more explicit here, they still do not provide any cognition of these objects. We cannot even say that a postulate “refers” to its object, according to Kant. We can thus in no way surmise that judgments of faith, like judgments of experience, have object validity understood as object reference. To have an object and to refer to an object are not the same for Kant: postulates indeed do not refer to an object which then determines their truth, but they do have an object nonetheless. And such a having means, whatever else might be said about it, the attainment of objectivity in contrast to the plain not-having of the ideas on the one hand, but something other than a full-blown reference of statements on the other hand. In this way, Kant delineates the tenets of the universal religion of reason on

 Cf. KrV A f./B f.; KpV : ff.; KU :, ,  f.

140

Stephan Zimmermann

two fronts in relation to the variety of validity claims of human thinking, thus locating them.²⁸ Let us recap: the concepts of soul, freedom, and God in the postulates shed the subjectivity that adheres to them qua ideas in our experience, and they extend themselves expresso verbo to a certain objectivity by virtue of our being morally determined. Difficulties remain, however, when it comes to explaining this objectivity. For how are we to understand the notion that objects are “given” to the ideas through the categorical imperative, that the judgments of faith which are founded in human rationality have an object without referring to it? In the second Critique, Kant does not answer this question. It is first explored in the Critique of the Power of Judgment.

III Before we turn to the text of the third Critique, some prior observations are called for in order to sharpen what we are talking about when we speak of the postulates of practical reason and their having objects. In the Critique of Practical Reason, Kant more carefully distinguishes what it is, exactly, that the postulates postulate. Indeed, he determines the objects of the judgments of faith according to their modality. A very early passage in the “Preface” is instructive in this context. There, Kant announces to the reader that with the “concept of freedom, insofar as its reality is proved by an apodictic law of practical reason”, “all other concepts (those of God and immortality)” lose their position as “mere ideas” and “acquire objective reality”. He adds by way of explanation: “that is, their possibility is proved by this: that freedom is actual [wirklich]” (KpV 5:3 f.).²⁹ Here, three notions appear alongside each other: possibility, actuality, and objective reality. Kant uses these terms as follows: because of the moral law, which pronounces its own validity, the ideas are said to attain “objective reality” – the idea of freedom directly, and the ideas of soul and of God indirectly. Kant goes on, however, to paraphrase the “objective reality” of the postulates of the soul and God in terms of their “possibility”, which pos-

 Kant does not stick to this definition completely all the way through. He once writes in the Critique of the Power of Judgment: “The faith, therefore, which is related to particular objects that are not objects of possible knowledge or opinion […], is entirely moral” (KU :; my emphasis).  The translation of the Critique in the Cambridge Edition of the Works of Immanuel Kant obscures an important distinction in the German text between reality (Realität) and actuality (Wirklichkeit). For Kant, reality is a matter of the quality of a judgment, whereas actuality belongs to its modality. In this respect, I will correct the Cambridge translation.

Kant on “Moral Arguments”

141

sibility is proved, according to Kant, by the fact that freedom is “actual”. This means that while Kant identifies the objective reality of the concepts of the soul and God with their possibility (of whatever kind), he identifies, in the same sentence, the objective reality of the concept of freedom with its actuality (of whatever kind). To begin with, “possibility”, as it is introduced here, cannot simply mean logical possibility. It is not the consistency of ideas, be it internal or external, that is at stake. Kant does not merely want to say that these can only be consistently thought in the wake of the categorical imperative. It was already shown in the first Critique that the concepts of pure reason do not stand in conflict with those of the pure understanding per se – that the correct use of the ideas is compatible with that of the categories.³⁰ The second Critique declares that it wants to surpass this. Nonetheless, the conception of possibility in question cannot correspond to objective possibility either, at least not if one interprets it in the sense of the modal category of the understanding with the same name. For by that category something is determined as a possible object of experience, and thus as something which can be given by perception.³¹ In the Critique of Practical Reason, the soul, freedom, and God remain subject to the restrictions of cognition established in the Critique of Pure Reason. The ideas of reason do not acquire, as we know, objects that could grant them the status of an empirical cognition. Kant must have something else in mind. We have to pay attention to the fact that, both in the quoted passage from the “Preface” and elsewhere in the second Critique, he explains “objective reality” in terms of “possibility” several times, and vice versa, not only when discussing the ideas of the soul and God, as in the “Preface”, but also when discussing the idea of freedom.³² Hence, we are allowed to and even must concede, with Kant, a certain “possibility” to all concepts of pure reason where they appear in the propositional content of postulates. And this in turn amounts to conceding “objective reality” to them, which in the context of the second Critique is practical objective reality. What the ideas of speculative reason attain through their use in judgments of faith would thus have to be described as practical objective possibility. However, what this means remains, of course, far

 It also recurs in the Dialectic of Pure Practical Reason: “That which is required for the possibility of any use of reason as such, namely, that its principles and affirmations must not contradict one another, constitutes no part of its interest but is instead the condition of having reason at all” (KpV :).  Cf. KrV A/B, A ff./B ff.  Cf. KpV :, , .

142

Stephan Zimmermann

from clear. Only one thing is certain: that this must be a possibility that is not subject to the conditions of space and time.³³ Similar things have to be said about the conception of actuality with which Kant operates in this context. For Kant does not speak of actuality only with respect to the idea of freedom, but also with respect to the two other ideas. Indeed, when describing what moral faith believes in, he usually refers to the “being [Dasein]” of God,³⁴ the “existence [Existenz]” of the human soul,³⁵ and the “actuality [Wirklichkeit]” of the freedom of our will.³⁶ Yet he deviates now and then from these almost terminological formulations and also says, for example, that God is postulated as actual³⁷ or existing,³⁸ and the soul as being.³⁹ This suggests that he uses the expressions “being”, “existence”, and “actuality” interchangeably. Accordingly, it is also the case that we must here grant every pure concept of reason a certain “actuality”. According to the “Preface”, however, this is supposed to be the same as granting them “objective reality”: more precisely, practical objective reality. What the ideas of speculative reason acquire as part of the propositional content of judgments of faith would consequently be practical objective actuality. Yet it remains once again unclear what this could consist in. Certainly, one is not allowed to foist the meaning of the corresponding modal category of the understanding onto this conception of actuality – which Kant also alternately labels “being”, “existence” or “actuality” in the Critique of Pure Reason. ⁴⁰ Through it, something is represented as an actual object of experience, i. e. as something which is given by perception.⁴¹ In the case of the

 Kant himself speaks once explicitly of “practical possibility” (KpV :), although in regard to the concept of the highest good. However, he often says of this too that it is “possible” and means by this that it likewise posesses practical objective reality (cf. KpV :, , , , , , ). More generally, Kant refers several times to the highest good as a postulate (cf. KpV : f., ), and this with good reason, for that my virtue constitutes an entitlement to happiness, and that this is allocated to me in another life, belongs “wholly to the supersensible relation of things” (KpV :) and is a question of faith. The nature of the relationship between the postulate of the highest good and the other postulates – above all those of the soul and God – and whether it implicates them or is a separate fourth postulate are questions that I shall not pursue further here.  Cf. KpV :, , ; KU :, , , , , .  Cf. KpV :, , , , .  Cf. KpV :, , , .  Cf. KpV :; KU :.  Cf. KpV :, , ; KU :.  Cf. KpV :, .  Cf. KrV A ff./B ff., A/B, B, B f., A/B f., A/B, A/B, A/B; Prol :.  Cf. KrV A/B, A ff./B ff.

Kant on “Moral Arguments”

143

soul, freedom, and God, however, this is completely out of the question from the start. It must therefore concern a form of being, existing, and being actual which does not take place in spatiotemporal circumstances.⁴² Despite these divergent comments here and elsewhere in the Critique of Practical Reason, we would be mistaken to view Kant as indecisive with regard to the modality of the objects of the postulates. Indeed, he writes for the most part that it is merely “the possibility of those objects of pure speculative reason” (KpV 5:134) which is believed in.⁴³ However, if one reads attentively and takes the text seriously, it becomes clear that Kant repeatedly and explicitly advocates the stronger version, according to which it is the “actuality” (KpV 5:125) of the soul, freedom, and God which is postulated.⁴⁴ What is actual must for this reason also be possible, yet conversely, that which is possible must for this reason not already be actual. Consciousness of the obliging character of the highest principle of morality, however, shall safeguard not only mere objectivity or objective possibility with regard to the ideas of speculative reason, but also objective actuality. Therefore, judgments of faith are a type of judgment of being:⁴⁵ they represent, as Kant unequivocally articulates, that “there are such objects” (KpV 5:135) and that reason has “to assume that existence” (KpV 5:146), that “real objects belong” (KpV 5:134)⁴⁶ to the ideas and that they “really [wirklich] have their (possible) objects”. But what does possibility mean here? What does actuality mean? This cannot be said with certainty on the basis of the second Critique. ⁴⁷

 See Winter :  ff.  Cf. KpV :,  f., , n., , , , , , , .  Cf. KpV :, , , , , ; KU :.  See also Ricken : ; Guyer speaks of “existential propositions [existenzielle Sätze]” (Guyer : ). Beck is to a certain extent mistaken when he states without differentiating that Kant in no way derives a judgment of existence from a judgment of duty (cf. Beck : ). Already in the first Critique, Kant himself says that belief “finally comes down to the inference that something is […] because something ought to happen” (KrV A/B).  See also KpV :.  Even the “Introduction” to the Jäsche-Logic, published only in , leaves things undetermined in this respect. Here we read: “Believing […] relates to objects in regard to which we […] can only be certain that it is not contradictory to think of such objects as one does think of them. What remains here is a […] holding-to-be-true of what I accept on moral grounds, and in such a way that I am certain that the opposite can never be proved” (Logik :; my emphasis). Therefore, belief refers to objects whose concepts can consistently be thought and are thus logically possible, on the one hand, and whose “opposite can never be proved”, on the other hand, by which Kant must mean that no reasons can be displayed for the objective impossibility of the concepts soul, freedom, and God. Just what kind of objectivity this could be, the impossibility of which is not provable, remains unclear, however.

144

Stephan Zimmermann

IV Let us therefore proceed to the Critique of the Power of Judgment. Under the title “On the kind of affirmation involved in a moral proof of the being of God”, Kant gives a meaningful indication at §90, which hints at how one might have to spell out the objective content of philosophically revised articles of faith. We can understand this in a general way. Kant here restricts himself thematically to the postulate of the being of God – a postulate with which he occupies himself predominantly on account of the teleological problematic in the last paragraphs of the work – but there is no reason for us to limit ourselves in the same way. Indeed, what Kant says applies to the other postulates in exactly the same way. In the aforementioned paragraph, Kant grapples with the notion that the contents of religious faith cannot be addressed argumentatively. He dismisses altogether those “grounds for proofs tending to theoretical conviction” (KU 5:463). One such argument is the “inference from analogy”.⁴⁸ When it comes to an analogy, we are dealing with something that is identical and at the same time different. Traditionally, an analogy names an identity of relations: α is to β as γ is to δ. Things that are related to each other are different – α and β on the one side, γ and δ on the other – but the way in which they belong together is in both cases the same – for instance, each pair might relate as cause and effect. “An analogy”, Kant explains in a footnote, “is the identity of the relation […], insofar as that identity obtains in spite of the specific difference between the things […]” (KU 5:464n.). Thus, α behaves as the cause of β, and γ does exactly the same in relation to δ.⁴⁹ It is important to differentiate this from an inference from analogy, which Kant discusses here, among other things, as a form of proof that makes use of theoretical judgments. An inference from analogy consists in reaching a conclusion based on an analogy: the analogy functions as the ground for the proof.

 What is not important, for Kant, is the exact logical form of a moral argument: “This proof, which one could easily adapt to the form of logical precision […]” (KU :).  The notion of a postulate comes originally from Pythagorean mathematics: from Euclid’s Elements, to be more precise. Kant was acquainted with it from Baumgarten’s Acroasis logica in Christianum L. B. de Wolff () and Wolff’s Philosophia rationalis sive logica II (), among other works. In the first Critique, he adapts it to his undertaking of a transcendental-philosophical foundation of metaphysics in the case of the postulates of empirical thinking. Already at this point, he writes: “In philosophy analogies signify something very different from what they represent in mathematics. In the latter they are formulas that assert the identity of two relations of magnitude […]. In philosophy, however, analogy is […] the identity […] of two qualitative relations” (KrV A f./B). And in the Critique of Practical Reason, Kant again uses the term ‘postulate’ for lack of a better one, as he himself explains. Cf. KpV :n.

Kant on “Moral Arguments”

145

From the known identity of two relations, conclusions are inferred regarding possible commonalities between the respective, but for present purposes undetermined, relata. Although this is a classical model of argumentation, its structure has been interpreted quite differently. In a more common version, one starts from the assertion of a certain similarity between two things and derives a conclusion regarding another respect in which they might be similar: α has something in common with β, β exhibits the characteristic γ, thus α possibly also possesses the characteristic γ.⁵⁰ Such arguments are commonly used in the modern, experimental sciences in particular, less as an instrument for guaranteeing the truth of discoveries than as a tried and tested method for coming up with research hypotheses which guide further procedures and still call for empirical confirmation or refutation. Kant, however, resolutely opposes diluting the concept of an inference from analogy in this way. He considers an analogy to be not only a similarity between two things but, exclusively, an identity of two relations.⁵¹ Kant gives several examples of an inference from analogy. One draws attention to the comparison of human beings and certain animals that, like the beaver, are also able to set up structures through their work, albeit of a limited nature and size. According to Kant, we are entitled to deduce from these kinds of activities, which resemble each other, the notion that “the animals also act in accordance with representations” (KU 5:464n.). This is only so, however, on the condition that both are similar to each other, that we count them “as members of the same genus”, namely as “living beings”. This, as Kant declares, is the “principle that authorizes such an inference”. However, we are not allowed to conclude that because humans act with reason, the beaver or another animal “must have the same sort of thing”. In the background lies Kant’s notion of life, to which one must appeal for an explanation and according to which life includes, among other things, the capability to entertain certain representations and the causal capacity to act in accordance with these representations.⁵² Intelligence, according to Kant, is not a feature of the living as such and not limited to finite beings. Supersensible beings, such as angels or God, are also rational beings and therefore have a share in a noumenal order, although they are neither living nor mortal.⁵³ For this reason, the conceptual overlap between man and animal, which is necessary for an inference from analogy, does not extend any fur-

   

See, for instance, Spree : . Cf. Prol :. Cf. KpV :n.; MAN :. See also the earlier Träume :n. See, for instance, KpV :; GMS :. See also Zimmermann :  ff.

146

Stephan Zimmermann

ther. A sideways glance on human beings is not substantive enough to support the conclusion that animals are gifted with reason.⁵⁴ Natural theology also cannot deduce its central tenets in this way.⁵⁵ For here we are dealing, according to Kant’s justification, with “two dissimilar things” (KU 5:464). Indeed, an object corresponds to our usual empirical concepts; drawn from spatiotemporal objects, these concepts can at any time rightfully be applied to them again in judgments of experience. But the ideas of pure reason are different. Like the categories of pure understanding, they have their seat in ipsa natura intellectus pari. In contrast to these, however, which are at least constitutive of the objectivity of spatiotemporal objects and therefore relate a priori to appearances, the ideas do not have – let alone refer to – any objects at all in fulfilling their regulative function. Consequently, we are not allowed to “draw an inference by means of the analogy, i. e., transfer this characteristic of the specific difference from the one to the other”. To conclude that the concepts of pure reason are objective simply because concepts of experience have an object meaning is to go against the strict condition of every inference from analogy, since they are not “members of the same genus”. That the latter have objectivity is in no way a proof of the objectivity of the former. In the course of his remarks about the inference from analogy, Kant now makes an important distinction. He differentiates between inferring from an analogy (nach der Analogie schließen) and thinking by means of an analogy (nach der Analogie denken). “One can, of course, think of one of two dissimilar things, even on the very point of their dissimilarity, by means of an analogy with the other; but from that respect in which they are dissimilar we cannot draw an inference by means of the analogy, i. e., transfer this characteristic of the specific difference from the one to the other” (KU 5:464).

Kant presents four examples in connection to this, each developing this contrast. Something that cannot be inferred from analogy can nonetheless always be thought by means of analogy. This is also the case with regard to the “artistic actions of animals” (KU 5:464n.). It is completely appropriate to say here, as Kant acknowledges, that what instinct is to an animal, say to a beaver, is what intellect is to human beings. This is not to claim that an animal is equipped with reason, but merely that “the ground of the artistic capacity in animals, designated as instinct, is in fact specifically different from reason, but yet has a sim-

 Kant also makes such a stipulation regarding the inference from analogy in Logik :. For this, see the detailed account in Maly :  ff.  Kant himself speaks in this context of a “deduction” (KpV :, , ).

Kant on “Moral Arguments”

147

ilar relation to the effect (comparing, say, construction by beavers with that by humans)”. In other words, it is an analogy in the true sense of the word – an identity of relations – which is assumed instead. Instinct and intellect are similar in the sense that, when it comes to the activity of building, they play a comparable causal role. What animals manage by natural instinct, human beings achieve by virtue of the power of their intellect.⁵⁶ In my opinion, this is the crucial point for a proper understanding of the objective content of the enlightened, rational religiousness with which Kant is concerned. Could it not be that the preceding considerations about the inference from analogy concern more than the limited range of a certain model of argumentation? Could it not be that they reach beyond this and positively engage with the question of the objectivity of moral judgments of faith? For if one cannot infer and thus cognize the continued existence of the soul after death, the actuality of the freedom of the human will, and the being of God, can these be thought by means of an analogy and in this way explained? And even further: does the sought objectivity of the postulates, understood in terms of their having objects, lie in such analogical thinking?

V This is really what Kant has in mind. Much textual evidence can be gathered from his writings. One finds the first directly in the aforementioned passages, for one of the other examples that Kant draws on in §90 deals with the idea of God. The structure of this example is, with regard to its content, due to the specifically teleological theme of the second part of the Critique. Given the existence of so-called natural

 The same account can be found in Kant’s writing on Religion. In a footnote, Kant remarks “we always need a certain analogy with natural beings in order to make supersensible characteristics comprehensible to us” (RGV :n.). He calls this the “schematism of analogy (as means of elucidation)” and differentiates it from the “schematism of object-determination (as means for expanding our cognition)”. But if we look closer, this is nothing other than the aforementioned difference of the third Critique between inferring from and thinking by means of an analogy. For as Kant warns: “in the ascent from the sensible to the supersensible, we can indeed schematize (render a concept comprehensible through analogy with something of the senses) but in no way infer by analogy that what pertains to the sensible must also be attributed to the supersensible (thus expanding the concept of the latter)”. Kant’s example at this point is that “I cannot make the cause of a plant comprehensible to me (or the cause of any organic creature, or in general of the purposive world) in any other way than on the analogy of an artificer in relation to his work (a clock), namely by attributing understanding to the cause”; but I am not allowed to conclude: “so too must the cause itself (of the plant, of the world in general) have understanding”.

148

Stephan Zimmermann

purposes – organized living beings – we can, according to Kant, think of God “in analogy with an intelligence” (KU 5:465) as the creator of such organisms. God’s causality with regard to purposive products in nature is then represented as corresponding to human causality with regard to the intentional creations “that we call artworks”. However, we must not “infer by an analogy”, as Kant warns once more, that God is literally a being with an intellect like us. For we have knowledge of such a practical faculty, which can be exercised in accordance with concepts of purposes, i.e. a will, simply in relation to finite beings – ourselves, namely – whereas God as the original being is outside of the sphere of finitude. On account of this fundamental difference, the tertium comparationis – the indispensable precondition for an inference from analogy – is missing.⁵⁷ Admittedly, the question for Kant, here, is not one about the being of God. Rather, he demonstrates how we can attribute certain properties to God, in particular those of intellect and will, without going as far as to make statements capable of truth and violating the restrictions of cognition that the Critique of Pure Reason had conclusively laid out. And yet: in Kant’s example it is at least clearly revealed that the via analogiae is available, in principle, also beyond the world of animals and humans, in that realm which absolutely transcends all sensible intuition and therefore empirical cognition.⁵⁸ Analogical thinking may be legitimately used both in the empirical world and in our endeavors to transcend that world and to deal with the sphere of the noumenal: in making the noumenal sphere understandable despite its being non-cognizable. This path is in principle also open to philosophical reflection on the absolute.⁵⁹ Furthermore, this way of thinking is in no way limited to reflection on God. Only a few pages prior, at §88, Kant expresses himself in the same way with regards to the soul. Here, he first of all writes generally that we can “name a cause after the concept that we have of its effect (though only with regard to its relation to the latter)”, without “thereby meaning to determine its internal constitution intrinsically by means of the properties that are all that we know about such causes and which must be given to us by experience” (KU 5:457). Clearly, what is already at issue here is a kind of thinking that operates with analogies, even if Kant does not describe it this way. He continues by developing this

 Cf. KU :n.  See Pieper :  f.,  f., .  This is already to be found in the Prolegomena, where Kant argues for a “cognition” of God “according to analogy” (Prol :), just as we encounter it later in the Critique of the Power of Judgment, although it is there based on the factual consciousness of our moral determinedness. See also RGV :n.

Kant on “Moral Arguments”

149

down to the particular and by illustrating it at first with the example of the soul, before finally turning to God: “[…] e. g., just as we would attribute to the soul among its other properties a vim locomotivam, because real movements of the body arise whose causes lie in the soul’s representations, without thereby meaning to ascribe to the soul the only ways in which we are acquainted with moving forces (namely, through attraction, impact, repulsion, hence movement that always presupposes an extended being)” (KU 5:457).

What we are unable to cognize – the human soul and its power to intentionally put our body into action – is brought to mind by an analogy with physical bodies and their effects on each other, which is familiar to us from experience. But we must be careful not to uphold as an alleged insight into the nature of the soul that which, for lack of anything better, really comes merely from a comparison with material things: the soul is only conceived as if it were like those things. And the very same, we can assume, will also hold true for freedom. It is true of any unconditional represented through one of the ideas of speculative reason that we indeed cannot make conclusions about it per analogiam, yet this does not prevent us from thinking it in such a way.⁶⁰ And that is not all. It is at §88 that Kant, in wonderfully clear terms, discloses what our whole questioning aims at. Under the title “Restriction of the validity of the moral proof”, which alludes to the proof of the existence of God that the previous paragraph had unfolded, Kant addresses the peculiarity of the result of a moral argument. He makes clear that, when postulating, we are not meant to think directly of God: “Now in order to avoid a misunderstanding that can easily arise, it is most necessary to mention here, […] that we can think these properties of the highest being only by means of an analogy. For how would we investigate its nature, nothing similar to which can be shown to us by experience?” (KU 5:456)

Indeed, it is the term “think” which Kant puts the stress on and which he sets in spaced letters. In light of the later considerations at §90, however, it is really the

 This concerns, then, those properties through which we think the order under moral laws: it is a “world of the understanding” or “supersensible nature” (KpV :), an “intelligible order of things” (KpV :) or an “enduring natural order” (KpV :), it includes a “law for all rational beings” (KpV :) and thus also regulates the morals of purely spiritual beings, and so on. For what it means for something to exhibit the character of an order, a nature or world, or to be enduring and what praxis is, etc. is familiar to us from human existence and can only be transposed from there to the noumenal in analogical thinking.

150

Stephan Zimmermann

whole phrase that ought to be emphasized: to “think […] by means of an analogy”. It is no coincidence that Kant uses exactly this phrase several times only a few pages later. It holds the answer we have been looking for. We have to represent the being of God in such a way that this representing does not refer to God as an object but nevertheless has God as its object: to postulate God’s being means to “think” of God’s being “by means of an analogy”. And the same holds for the remaining judgments of faith. One must also interpret the other postulates accordingly, although in the Critique of the Power of Judgment Kant does not explicitly make this observation. Anyone who, given his receptiveness to the demands of the categorical imperative, represents the objects of these ideas indirectly, i. e. per analogiam, believes in the actuality of the freedom of the will and the existence of an immortal soul. In Kant’s adaptation of the concept of a postulate, undertaken from the standpoint of a critical philosophy of religion, this concept bears witness, in its innermost core, to an act of analogical thinking. If one looks back from this point to the Critique of Practical Reason, one encounters another passage which is otherwise rather inconspicuous but says exactly the same thing. Appropriately, in the chapter titled “On the Warrant of Pure Reason in its Practical Use to an Extension which Is not Possible to it in its Speculative Use”, Kant writes towards the end, but nonetheless with a conscious glance towards the later passages of the “Dialectic”, that we “shall find in the sequel” that we have the “warrant” to “admit and presuppose” “properties”, “even where supersensible beings (such as God) are thought by means of an analogy, that is, by a purely rational relation of which we make a practical use with respect to what is sensible” (KpV 5:56 f.; my emphasis). Indeed, the verb in the italicized part of the above quotation is missing in the original German text. Without such a verb, however, this part of the text would not make sense for purely grammatical reasons. The German Academy Edition conjugates this missing verb as “annimmt”, which the Cambridge Edition of the Works of Immanuel Kant accordingly translates as “assumed”. I suggest, however, in light of the wording to be found in the third Critique, that it instead be translated as “thought”. Either way, what Kant states is that we are entitled to think of nonempirical beings – and God serves for him here as a well-chosen example – in analogy with empirical beings and in this way to attach diverse characteristics to them.⁶¹

 In this light one should also read a passage in the chapter about the “Critical Resolution of the Antinomy of Practical Reason”. There, Kant talks about the phenomenon of moral “contentment with one’s person” (KpV :). This feeling, which can be found among finite beings such

Kant on “Moral Arguments”

151

VI But what kind of an analogy is it, we must finally ask, that gives objectivity to the ideas?⁶² It can only be one with empirical concepts and their objects. Kant’s idea is that, just as empirical concepts relate to their objects, ideas also stand in relation to something.⁶³ But how ought this to be possible given that both are so different? Given that they do not count “as members of the same genus”? If we ask these questions, we misconceive the problem. For the fact that concepts of experience and concepts of reason are not “members of the same genus” precludes only an inference from analogy. But this does not take place at all here. Empirical concepts, which have their origin in perception, are applied to spatiotemporal objects in theoretical judgments. If, on the other hand, the pure concepts of soul, freedom, and God in judgments of faith move forward in a new dogmatic use, break out of their subjectivity, and extend themselves for practical purposes, they do indeed acquire objects. The one set of concepts, however, is not at all assumed to have objects because the other set does. The claim is therefore not that their objects are of the same kind as the objects of the others. The objects of pure concepts of reason are instead thought by means of an analogy with those of empirical concepts. In doing so, it is not a mere similarity which is supposed here, as was customary in the theological tradition of analogical thinking about God before Kant, which aimed to render comprehensible the infinite being of God in the face of the finite nature of all human attempts at knowledge.⁶⁴ Every example of analogical thinking which deserves the name consequently involves an identity of relations – which Kant insists on, as observed earlier. And might we not then say that this identity of relations in our case consists in the identity of the relations of the respective concepts to an object initially reduced to its conceptual core? That empirical concepts in a judgment of experience and pure concepts of reason in a postulate have one thing as humans, needs to be distinguished from (among others) “beatitude”, which falls especially to God, for its concept contains no “inclinations and needs” at all. Regardless of this, Kant continues, the moral self-contentment of man is “analogous” to this “self-sufficiency that can be ascribed only to the supreme being”.  Specht does not go into this question, though he does otherwise see correctly that Kant thinks the objective reality of postulates by the via analogiae. Cf. Specht :  ff.  For this, see §, according to which every act of analogical thinking “performs a double task, first applying the concept to the object of a sensible intuition, and then, second, applying the mere rule of reflection on that intuition to an entirely different object”, wherefore “empirical intuitions are also employed” (KU :).  See Kluxen : col.  ff.; Pannenberg .

152

Stephan Zimmermann

in common, namely that in both cases the human intellect sets something against itself and its activity? That the self-conscious thinking puts something before itself, drawing it out as an other? This is, after all, the core meaning of the word ‘object’: whoever is concerned with objects is thereby concerned with something that he himself throws before and against himself.⁶⁵ Having objects of articles of faith is in this way initially defined in a merely one-sided and abstract manner because subsequent questions concerning the particular nature and accessibility of these objects are still open. Nonetheless, the practical objective reality, mediated by a judgment of duty, with which postulating supplies the ideas can be explained such that, for the first time, we are now allowed and even compelled to assume that something in fact falls under the ideas, that something corresponds to them. The Critique of Pure Reason had not been able to give any warrant for this; only the Critique of Practical Reason puts us in the position of placing the pure concepts of reason in relation to something that is what it is independently of its being thought by us, and of therefore providing objects for them. And it is solely this identity of the relation to something standing against me, something self-dependent, which judgments of faith can claim for themselves in analogy with judgments of experience. If, in addition, we also take into account the difference between the relata, that is to say the nature and accessibility of the objects, two comprehensive though not equivalent concepts of objectivity are to be distinguished. There is on the one hand the phenomenal and explanatorily primary objectivity, which adheres to judgments of experience and is the focus of the first Critique. For Kant, the access conditions of such objects are the conditions of our sensibility: space and time as forms of intuition. Since all intuited objects necessarily accord with them, these are always, as Kant puts it, appearances in space and time.⁶⁶ On the other hand, and only in analogy to this, there is the objectivity of judgments of faith, which in this way is a merely derived objectivity and does not run counter to the critically restricted cognizability of objects of experience. For the soul, freedom, and God are precisely not represented as existing in spatiotemporally ordered relationships or as accessible in possible, actual, or necessary experience. In Kant’s own language, one can perhaps speak here of a noumenal objectivity.⁶⁷ The objects of the ideas remain, according to their nature,

 ‘Object’, a foreign word of Latin origin, is found in German from the fourteenth century onwards. ‘Obiectum’ is the substantiated neuter noun made from the perfect participle of ‘obi(a)cere’, which literally translated means to throw in front, or throw against. An object is hence actually that which has been thrown in front or against. Cf. Georges : cols.  f.  For this, see KrV A ff./B ff., A ff./B ff.  Cf. KpV :, , ,  f., , , , , .

Kant on “Moral Arguments”

153

beyond our access conditions for phenomena: they are completely sensibly undetermined and undeterminable entia rationis (thought things), bare noumena.⁶⁸ And this is still not all. For the objectivity that the ideas acquire is not a completely undifferentiated one, as we have seen. Rather, the ideas have a certain modal character; what postulates postulate, according to Kant, is not the mere possibility of the soul, freedom, and God, but their actuality. It now becomes clear that Kant’s conception of a practical objective reality is to be interpreted in a strictly analogical way. That the immortality of the human soul, the freedom of our will, and the existence of God are actual for natural faith means that, given the obligatory character of the moral law, it is not the case that they can only be represented as objects in general, which theoretical reason is incapable of doing where it is left to itself; they are instead brought into analogy with objects that – according to their modality – are not merely possible but even actual objects of empirical cognition. In comparison with sensible objects, which the concepts of experience refer to in statements and which are really given by perception, we think of the noumenal objects of concepts of reason in the postulates as if they existed like them – that is to say, as if they were what they are not only possibly but in fact independently of their being thought by us. Against this background, we must retroactively reassess some of Kant’s statements in the Critique of Pure Reason. There, in the “Appendix to the Transcendental Dialectic”, Kant already says not only that the ideas allow for a regulative use (in contrast to the constitutive use of the categories), but also that this use would consist in thinking of their objects as if they were objects of possible experience; even more, Kant explains this explicitly and repeatedly in such a way that the objects of the ideas would thereby be thought as analogues of objects of experience.⁶⁹ In this respect, Kant’s discussion in the Critique of the Power of

 A noumenon “in the negative sense”, by which Kant understands in the Critique of Pure Reason “a thing […], insofar as it is not object of our sensible intuition” (KrV B). Kant also talks there of “merely intelligible objects” and states that their concept is “entirely devoid” (KrV A/B). That only changes with the Critique of Practical Reason: “There was therefore no extension of the cognition of given supersensible objects, but there was nevertheless an extension of theoretical reason and of its cognition with respect to the supersensible in general, inasmuch as theoretical reason was forced to grant that there are such objects, though it cannot determine them more closely and so cannot itself extend this cognition of the objects” (KpV :). And Kant now qualifies, for instance, the existence of the human soul several times as “supersensible” or “intelligible”. Cf. KpV :, , , , .  Cf. KrV A/B, A/B, A/B. Kant explicitly uses the phrase “thought […] by means of an analogy”; he writes that God can be “thought […] by means of an analogy with an intelligence (an empirical concept)” (KrV A/B). Further: “Thus by means of an analogy of realities in the world, of substances, causality, and necessity, I will think of a

154

Stephan Zimmermann

Judgment must therefore count as a silent self-correction: what Kant had previously believed he was entitled to say can in fact only be said following the Critique of Practical Reason, or at least following the search for and establishment of the moral law. The substance of Kant’s remarks in the first Critique on thinking of the objects of ideas in the ‘as if’ mode, and hence in analogy with objects of possible experience, is not thereby rendered invalid; rather, what the later Kant of the third Critique apparently has in mind is simply that this is not their proper place. The thesis remains, but its justification changes. No matter how we might have to characterize in further detail the regulative use of the ideas (in opposition to the constitutive use of the categories), on no account can it involve an ‘as if’ in the sense of an analogy.⁷⁰ If this is correct, then having an object in the full sense involves, on account of the obligating effectiveness of the categorical imperative, something’s being represented in an act of analogical thinking as really standing before or against the concepts of pure speculative reason – something which, however, remains inevitably hidden from sensible intuition. For Kant, this is the rational meaning of an enlightened religious attitude, which indeed remains transcendent for modern scientific knowledge (since this form of knowledge owes itself to the methodologically disciplined arrangement of experiment and observation), but which retains its every right against it. And the content of the belief central to this religion of reason encompasses an absolute subject of all empirical states of mind, an absolute ground of empirical effects, and an absolute ground of the division of all empirical reality. It is this noumenal objectivity that constitutes the superiority held by postulates of pure practical reason over ideas. Translation: William Pimlott

being that possesses all of these in their highest perfection” (KrV A/B). Finally: “Finally, if the question is third whether we may not at least think this being different from the world by means of an analogy with objects of experience, then the answer is: by all means” (KrV A/ B).  This self-correction is not the only one. Similarly invalidated by the Critique of the Power of Judgment are other statements by Kant in the first Critique, according to which the regulative use of the idea of God already entitles us to ascribe ends to nature both as a whole and with regard to each of its products (cf. KrV A ff./B ff.). This notion also changes with the third Critique because, according to this later text, it is in truth only the teleologically reflecting power of judgment in its engagement with products of nature that are organisms which is capable of motivating such ascriptions.

Kant on “Moral Arguments”

155

References Baumgarten, Alexander G. 1973, Acroasis logica in Christianum L. B. de Wolff (1725), in: Christian Wolff, Gesammelte Werke, Abt. 3, Bd. 5, Hildesheim/New York: Olms Beck, Lewis W. 1960, A Commentary on Kant’s Critique of Practical Reason, Chicago: Chicago University Press Euklid 1933, Die Elemente, 1. Teil: Buch I–III, trans. and ed. by Clemens Thaer, Leipzig: Akademische Verlagsgesellschaft Georges, Karl E. 81918, Ausführliches lateinisch-deutsches Handwörterbuch, Vol. 2, Hannover/Leipzig: Hahn Guyer, Paul 1997, “In praktischer Absicht: Kants Begriff der Postulate der reinen praktischen Vernunft”, Philosophisches Jahrbuch 104, 1–18 Kluxen, Wolfgang 1971, “Analogie”, in: Joachim Ritter (ed.) 1971, Historisches Wörterbuch der Philosophie, Vol. 1: A–C, Darmstadt: Wissenschaftliche Buchgesellschaft, cols. 214–27 Maly, Sebastian 2012, Kant über die symbolische Erkenntnis Gottes, Berlin/Boston: de Gruyter Pannenberg, Wolfhart 2007, Analogie und Offenbarung. Eine kritische Untersuchung zur Geschichte des Analogiebegriffs in der Lehre von der Gotteserkenntnis, Göttingen: Vandenhoeck & Ruprecht Pieper, Annemarie 1996, “Kant und die Methode der Analogie”, in: Gerhard Schönrich/Yasushi Kato (eds.) 1996, Kant in der Diskussion der Moderne, Frankfurt/Main: Suhrkamp, 92–112 Ricken, Friedo 2002, “Die Postulate der reinen praktischen Vernunft (122–148)”, in: Otfried Höffe (ed.) 2002, Immanuel Kant. Kritik der praktischen Vernunft, Berlin: Akademie Verlag, 187–202 Specht, Ernst K. 1952, Der Analogiebegriff bei Kant und Hegel, Köln: Kölner Universitätsverlag Spree, Axel 2003, “Analogieschluss”, in: Wulff D. Rehfus (ed.) 2003, Handwörterbuch Philosophie, Göttingen: Vandenhoeck & Ruprecht, 245 Winter, Aloysius 2000, Der andere Kant. Zur philosophischen Theologie Immanuel Kants, Hildesheim/Zürich/New York: Olms Wolff, Christian 1983, Philosophia rationalis sive logica II (1740), in: Christian Wolff, Gesammelte Werke, Abt. 2, Bd. 1, Hildesheim/Zürich/New York: Olms Wood, Allen W. 1970, Kant’s Moral Religion, Ithaca/London: Cornell University Press Zimmermann, Stephan 2014, “Praktische Kontingenz. Kant über Verbindlichkeit aus reiner praktischer Vernunft”, in: Simon Bunke/Katerina Mihaylova/Daniela Ringkamp (eds.) 2014, Das Band der Gesellschaft. Verbindlichkeitsdiskurse im 18. Jahrhundert, Tübingen: Mohr Siebeck, 81–98

Paul Guyer

Kant, Mendelssohn, and Immortality In the revised version of the Paralogisms of Pure Reason in the second edition of the Critique of Pure Reason, Kant criticized Moses Mendelssohn’s argument for the immortality of the soul in the second dialogue of his Phaedon (1767), an argument that turns on the supposed impossibility of a continuous change from existence to non-existence. This criticism of the argument is Kant’s only explicit reference to Phaedon or any other work by Mendelssohn in the first Critique. ¹ But throughout his career, Kant was far more engaged with Mendelssohn’s philosophy than his occasional references to him suggest, and in particular two central ideas of the first Critique are reminiscent of ideas that Mendelssohn expounds in Phaedon even if not demonstrably due to Mendelssohn. One striking idea in Mendelssohn’s work is his conception of the soul as the source of all combination, a lemma in his proof of the simplicity of the soul that foreshadows Kant’s view that all synthesis is due to the understanding although it does not identify synthesis with judgment and thereby pave the way for Kant’s proof of the objective validity of the categories. Another striking feature of Mendelssohn’s work is his argument in its third dialogue that the soul must be immortal because the full development of all its potential or capacities, thus its capacities for both happiness and virtue, would take forever and God would not have given the soul such potential without also giving it adequate time for its development. This argument foreshadows Kant’s treatment of the postulate of immortality in the Canon of Pure Reason in the Critique of Pure Reason, which turns on the need for immortality to perfect happiness, as well as that in the Critique of Practical Reason, where Kant shifts the ground for the postulate of immortality from the necessity of our being granted adequate time for the realization of our happiness to our being granted adequate time for the perfection of our virtue. This shift is striking. However, in the 1790s, Kant sometimes seems to omit any argument for the postulate of personal immortality altogether, often reducing the postulates of pure practical reason to the single postulation of God as the ground

 In the “Elucidation” of the section on time in the Transcendental Aesthetic, Kant responds to several “insightful men” who had criticized his transcendental idealist theory of time as originally expounded in the inaugural dissertation of . The “insightful men” who had communicated their objections to Kant by letter included Mendelssohn (letter of December , , :–), as well as Johann Heinrich Lambert (letter of October , , :–), and Johann Georg Sulzer (letter of December , , :–), but Kant does not mention Mendelssohn (or the others) by name in his response.

158

Paul Guyer

of the possibility of the realization of morality in nature, while at other times he seems to revert to an argument closer to that of the first Critique than to that of the second, arguing that we need to believe in immortality in order to believe in the possibility of the complete realization of the highest good but not specifically in order to believe in the possibility of the perfection of our virtue. One point I want to make in this paper is that Kant has good reason for this second shift, because the conception of freedom of the will that he finally reached in the 1793 Religion within the Boundaries of Mere Reason undercuts any thought that the perfection of our virtue or at least of our will need take any time at all, let alone eternity. The second point I will make is that in many works of the 1790s – although this view had been anticipated in the 1784 essay on “The Idea of Universal History from a Cosmopolitan Point of View” – Kant argues that the perfection of morality can be expected only in the complete history of the human species precisely because human individuals are mortal rather than immortal. This argument would make most sense if Kant had simply dropped the postulate of personal immortality. In the third section of the 1793 essay “On the common saying: That may be correct in theory but it is of no use in practice”, Kant makes another explicit reference to Mendelssohn, presenting his theory that the human species is capable of perpetual moral progress as an alternative to what he takes to be Mendelssohn’s view that human history is an endless cycle of progress and regress. Ironically, in what appears to be Kant’s most enthusiastic endorsement of the inevitability of the moral progress of the human species, or at least the inevitability of its progress towards the external condition of justice, namely the 1795 pamphlet Towards Perpetual Peace, Kant adds an afterthought that, likewise in light of his account of human freedom in the Religion, undercuts any thought of a guarantee of moral progress, namely the necessity of “moral politicians” for the establishment of republics and of their peaceful international federation, who however can never be guaranteed to do the right thing because no human being can ever be guaranteed to do the right thing. Thus Kant ends up by coming closer to the position of Mendelssohn that he had rejected in “Theory and Practice” as “abderitism”, the position that humans may simply oscillate between moral progress and regress, at least insofar as he gives up any idea that the moral progress of the human species can ever be regarded as guaranteed. The revolution in Kant’s thinking about freedom in the 1790s thus creates problems not only for his long-standing commitment to the postulate of personal immortality but also for his faith in the inevitable progress of the human species as well.

Kant, Mendelssohn, and Immortality

159

1 The Argument of Phaedon I begin with a brief account of Phaedon, which was not only the sole work by Mendelssohn to which Kant explicitly refers in the Critique of Pure Reason but also one of his most popular and successful works altogether. Mendelssohn presents his work as an updating of Socrates’s deathbed argument for the immortality of the soul in Plato’s Phaedo that will employ the best resources of recent philosophy. The work begins with an encomium to Socrates, which is followed by three sections that, following Plato’s model, report the discussion that took place on Socrates’s final day of life. In the first section, where Mendelssohn still follows Plato closely, Socrates tries to reassure his friends that death is nothing to be feared because in death the soul will be liberated from the body and the obfuscations of the senses, but his auditors will not be comforted by this thought until Socrates proves that the soul actually survives the dissolution of the body. At that point Mendelssohn starts putting his own arguments into the mouth of Socrates. The remainder of the first dialogue is taken up with the argument that nothing simple can really go out of existence, because that would require a discontinuous change from being to non-being although all change in nature except for the mere rearrangement of parts in something complex is continuous rather than discontinuous. Thus “[t]he soul cannot eternally perish, for the last step, no matter how long it might be postponed, would always be a jump from being to nothingness, which can be grounded neither in the essence of an individual thing nor in its whole interconnection” (Mendelssohn, Phaedon, I: 390). But this argument is sound only if the soul truly is simple, and the second dialogue is devoted to the proof of this crucial lemma. The central argument is that there can be no combination or unification of any manifold without an act of unification by thought, and that this itself cannot be the product of anything composite, on pain of an infinite regress. Thus Mendelssohn’s Socrates argues that “[o]rder, balance, harmony, regularity, in general all relations that require an apprehension and comprehension [Zusammennehmen und Gegeineinanderhalten] of the manifold are effects of the faculty of thought. […] And once this is conceded, this faculty of thought itself, this cause of all comparison and comprehension, cannot possibly arise from these its own actions […] thus I cannot place the origin of this faculty of representation in a whole that consists of […] separate parts” (Mendelssohn, Phaedon, I: 403). Thus among our parts “there must be at least a single one [ein einziges] which unifies and comprehends all these cognitions, desires and inclinations, everything that is to be encountered in our soul” (Mendelssohn, Phaedon, I: 405), and “[t]here is therefore in our body at least a single substance that is not extended, not composite, but is simple, has a power of representation, and unites in itself all our concepts, desires

160

Paul Guyer

and inclinations”. This completes Mendelssohn’s argument for the simplicity and therefore the indestructability of the soul. In his comment on Mendelssohn in the second edition of the Paralogisms, Kant objects that Mendelssohn’s proof turns on assigning to the soul only extensive rather than intensive magnitude, which could be destroyed not by the separation of parts but “by gradual remission of all its powers (hence, if I may be allowed to use this expression, by elanguescence)” (KrV B414). This seems not to address Mendelssohn’s claim that there can be no discontinuous change from being to non-being, which is supposed to apply precisely to those cases of change that are not mere dispersions of parts of extensive magnitudes. Although Mendelssohn had not used the term “intensive quantity”, Kant is nevertheless simply denying what Mendelssohn had asserted. However, Kant’s larger argument in the Paralogisms, that the formal unity of mental acts, and in particular the logical simplicity of the representation of the self through the unanalyzable expression “I”, does not imply the simplicity of the underlying source of thought, whatever that might be, bears precisely on Mendelssohn’s overall argument: that the act of thought is simple, the point on which Mendelssohn insists, does not imply that its source is simple. In Kant’s words, “it is not at all possible through this simple self-consciousness to determine the way I exist, whether as substance or as accident” (KrV B420; see also A381–2, A399–400, and B406–10). Yet it might also seem that Mendelssohn’s insistence that all combination must be due to thought could have made a deep impression on Kant, for that is certainly presented without demur as the primary premise of the Transcendental Deduction in the second edition of the Critique (KrV §15, B129–30), and in a text as late as the 1792–95 drafts on the Real Progress of Metaphysics Kant argues, in Mendelssohnian terms, that “[t]hat [the human being] is not solely a body can […] be rigorously proved, since the unity of consciousness, which must necessarily be met with in every cognition (and so likewise in that of himself) makes it impossible that representations distributed among many subjects should constitute unity of thought” (Fortschritte 20:308). Thus we might think of the larger argument of the Transcendental Deduction and the Paralogisms together as both an assimilation and a critique of Mendelssohn: Mendelssohn was right to recognize that all combination or synthesis is due to thought, indeed to a unitary act of thought, but he failed to recognize that thought always takes the form of judgment, thus he failed to deduce the categories, and he was wrong to think that the simplicity of the act of thought – or at least the simplicity of the concept “I” that Kant was prepared to recognize – implies the simplicity of the thing that thinks and thus wrong to think that it could be used as a premise in a proof of the immortality of the soul.

Kant, Mendelssohn, and Immortality

161

Of course, the conclusion that Kant draws from his critique of Mendelssohn’s and all “rationalist” arguments from the simplicity to the immortality of the soul, just like the conclusion he draws from his subsequent critique of all rationalist arguments for the existence of God, is not a disproof of the point in question, but rather the impossibility of either proof or disproof, thus the impossibility of deciding the question on any theoretical grounds. Kant claims that “the strictness of critique, by proving the impossibility of settling anything dogmatically about an object of experience beyond the bounds of experience, performs a not unimportant service for reason regarding this interest, in securing it likewise against all possible assertions of the contrary” (KrV B424). And the interest to which he refers is the practical interest of belief in immortality as a condition of the possibility of a human being fulfilling the demands of morality: the “refusal of our reason to give an answer to those curious questions”, whether the soul is material and mortal or immaterial and immortal, is “reason’s hint that we should turn our self-knowledge away from fruitless and extravagant speculation toward fruitful practical uses” (KrV B421). My suggestion is then that in spite of Kant’s rejection of his argument from simplicity to immortality, Mendelssohn’s further argument in the third Dialogue of Phaedon actually provides the model for Kant’s account of the practical belief in immortality, indeed for the two different accounts of such belief that Kant offers in the first and second critiques. Having proved in the second Dialogue that the soul is immortal, in the third Socrates is pressed by his friends to prove that “God has not destined us to eternal misery”, that is, that eternal existence promises eternal happiness and not eternal misery (Mendelssohn, Phaedon, I: 409). He responds with an argument (not to be found in Plato) that he says will appeal to God’s benevolence rather than to (his) cognition. The argument is that we cannot have been given powers that we are not destined to develop fully, or that are destined to disappear without a trace, like “foam upon water or the flight of an arrow through the air” (I: 412). Rather, Socrates argues, we can be confident that our powers are destined to reach their goals or at least to make unending progress toward them, thus “we can assume with good ground that this progress toward perfection, this increase, this growth in inner perfection, is the destiny [Bestimmung] of rational beings, thus also the highest final goal [Endzweck] of creation” (I: 417). But foremost among our capacities is our capacity for happiness, foremost among our goals happiness itself, and therefore we can be sure that we are destined not to eternal misery but to eternal happiness or at least unending progress toward it. Socrates waxes eloquent: “Is it fitting for wisdom, to bring forth a world in which the spirits that it places there […] could be happy and […] yet to withdraw eternally from these spirits the capacity for contemplation and happiness? […] Oh no, my friends! providence has not given us a longing

162

Paul Guyer

for eternal happiness in vain; it can and will be satisfied. […] Thus here below we serve the regent of the world by developing our capacities; thus we will also in that [eternal] life continue under his divine protection to practice in virtue and wisdom continually making ourselves more perfect and industrious in fulfilling the series of divine aims that stretch before us into the infinite” (Mendelssohn, Phaedon, I: 418).

In this passage, Mendelssohn’s Socrates goes beyond the challenge posed by his friends of proving that eternal existence would not be eternal misery, for he here sets not one but two goals for us, the perfection of our happiness and the perfection of our virtue, the latter understood as the fulfillment of our duties to God, above all (following Christian Wolff as well as Jewish tradition) the admiration of God’s own perfection, thus that he has argued that the soul needs and has infinite time to realize both goals or to come ever closer to them.² This is why I earlier suggested that in his first treatment of the postulate of immortality in the first Critique Kant adopts Mendelssohn’s argument for the case of happiness, while in the second Critique he rather adopts it for the case of virtue. But in work after the second Critique, I will argue, while sometimes continuing to base an argument for immortality on its necessity for the realization of happiness, Kant abandons what is basically his adaptation of Mendelssohn’s suggestion that we need to believe in immortality in order to believe in the possibility of perfecting our virtue. As I suggested, this is because the revision of his conception of freedom that Kant undertakes in the Religion obviates the need for such a doctrine.

2 Kant’s Initial Assimilation of Mendelssohn’s Conception of Immortality The highest good and the doctrine of postulates of pure practical reason grounded upon it are in many ways the culmination of Kant’s philosophy. They are treated at the conclusion of each of his three critiques, in the Canon of Pure Reason in the first Critique, in the Dialectic of Pure Practical Reason in the second, and in the Doctrine of Method in the Critique of the Teleological Power of Judgment

 In my allusion to Wolff, I am thinking of his argument in his teleology that “[t]he chief intention for the world is that from that we [human beings] should recognize the perfection of God”, and that everything else in the world is meant to support us in this activity; see Vernünfftige Gedancken von den Absichten der natürlichen Dinge, Chapter II, §, p. ; in my reference to Jewish tradition, I am thinking not of the obligation of Jews to fulfill God’s numerous commandments about observances but of the contents of such prayers as even the mourner’s Kaddish, which consists of little but praise of the glory of God.

Kant, Mendelssohn, and Immortality

163

in the third. Yet Kant’s conception of the highest good and certainly his argument for its necessity are sufficiently obscure that even after those three discussions he still had to return to it in the first section of the 1793 essay on “Theory and Practice”, in the contemporaneous Preface to Religion within the Boundaries of Mere Reason, in the 1794 essay on “The End of All Things”, and in the drafts for the essay competition on the real progress of metaphysics. For present purposes I will suggest only that at least some of the obscurity surrounding Kant’s idea is occasioned by the fact that he suggests two different derivations of it (although often in the same work). On one approach, it would seem that the natural end of practical reason, call it prudential or empirical practical reason, is one’s own happiness, while everyone’s pure practical reason calls for observance of the moral law by both themselves and others, so the combination of those two aspects of practical reason naturally proposes and morally allows the complete realization of one’s own happiness within the boundaries of the moral law, a condition which is equated with worthiness to be happy. This approach would leave any one agent’s concern for the happiness of others to follow from his use of pure rather than empirical practical reason, that is, to follow from the moral law, and it is not immediately obvious how the moral law imposes on each a concern for the happiness of all – although part of what Kant suggests on this approach is that the natural interest of each in his own happiness must be satisfied so that one’s resolve to be moral not be weakened (KrV A813/ B841), which might lead, although Kant does not explicitly say this, not only to an indirect duty to promote one’s own happiness in order to reduce temptations to immorality for the benefit of one’s own morality (MS 6:388) but likewise to an indirect duty to promote the happiness of others so that their resolve to be moral will not be weakened. Were there to be such a duty towards others it would of course be in addition to a direct duty to assist others in need of help in order to realize their ends, as explicated in the illustrations of the first two formulations of the categorical imperative in the Groundwork (GMS 4:423, 430) or in the form of the imperfect duty of virtue to promote the happiness of others asserted in the Metaphysics of Morals (MS 6:385, 387–8). On the other approach, at which Kant already hints in the Canon of the first Critique when he says that in a “moral world”, that is, the natural world as it would be if we “abstracted from all hindrances to morality (of inclinations)”, “a system of happiness proportionately combined with morality” would be “necessary” because “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 and at the same time that of others” (KrV A809/B837). The connection between freedom and happiness that is presupposed here may be under-

164

Paul Guyer

stood if we think of freedom as the freedom to set ends and of happiness as the condition that follows from the realization of ends: if we conceive of morality as prescribing the maximally intra- and interpersonally consistent use of the freedom to set ends, as Kant suggests we should in his Lectures on Ethics (Mo-Collins 27:346; Mo-Kaehler 178–80), then what would result from freedom so exercised under ideal conditions, in which all hindrances to morality, at least those coming from human inclinations themselves, were removed, and in which all hindrances to the realization of ends freely set, at least those stemming from humans themselves, would likewise be removed, would be the greatest possible realization of ends and thus the greatest happiness possible, again at least as far as the effects of human motivations, or “plagues” that mankind invents “for itself”, such as “the oppression of domination” and the “barbarism of war”, rather than non-human forces such as hurricanes or pestilence are concerned (KU §83, 5:430).³ Although Kant already suggests the second approach in the first Critique, he does not unequivocally adhere to it in later works, suggesting the first approach, which separates the ends of empirical and pure practical reason but recombines them in the composite idea of the highest good, in the second Critique and even in the Preface to the Religion, where he implies that our interest in happiness as part of the highest good is a “natural need” rather than a need of pure reason (see KpV 5:110–11, and RGV 6:5). However, my concern here is not the basis for Kant’s doctrine but rather its application in the form of the postulates of pure practical reason, in particular the postulate of immortality, although it would be nice if one could argue that it was Kant’s eventual resolution of his ambivalence about the foundation of his own doctrine that led to his eventual de-emphasis of this postulate: that is, if Kant had ever made it completely clear that the need for the happiness component of the highest good comes not from a promise of one’s own eventual happiness as a condition of one’s continuing resolve to be moral but rather from a need for the possibility of the eventual happiness of the human species as a con In some of his earliest remarks on moral philosophy, the notes he wrote in his own copy of the  Observations on the Feeling of the Beautiful and Sublime, Kant suggests that it is only the domination of our wills by the wills of others and not the restriction of our wills by non-human nature that we find morally offensive (e. g., Bemerkungen :). It similarly seems to be compensation for the obstacle to individual happiness created by the immorality of other humans that happiness in personal immortality is meant to provide in the Critique of Pure Reason (cf. KrV A/B). Perhaps by means of his argument throughout his mature works that we must postulate God as the author of nature – without any restriction – in order to conceive of the possibility of the highest good Kant means that we must believe in the eventual remission of non-human as well as human obstacles to happiness (individual or collective), but he never says this explicitly.

Kant, Mendelssohn, and Immortality

165

dition of the rationality of working to promote the freedom of members of the species to set their own ends, then it would also have become self-evident that there would be no need for a promise of personal immortality. The texts in which Kant emphasizes only the necessary possibility of the moral progress of the human species as a whole are certainly compatible with this way of grounding the idea of the highest good as the complete object of morality, although what Kant most clearly retracts is only the second Critique’s argument that personal immortality must be postulated as the condition of the possibility of the perfection of the perfection of personal virtue. Once he does that, however, he has moved away from Mendelssohn’s position even once that had been transformed into a mere postulate of pure practical reason. Be their later fates what they may, as I suggested we might regard Kant’s diverging arguments for the postulate of immortality in the first and second critiques as his separation and adaptation of the two capacities to be realized in eternity according to Mendelssohn’s Socrates in Phaedon. In the Canon of Pure Reason in the first Critique, Kant most clearly suggests that the individual agent needs to be able to believe in the eventuality of his own happiness in spite of the immorality of others around him and the damage that their immorality might do to his own prospects for happiness in order to maintain his resolve to be moral. What Kant argues is that there does not appear to be a causal connection between the individual’s worthiness to be happy in the “world of appearances” and his actual happiness, which depends in good part not on himself but on the actions of others, so that we must postulate that this connection will be established in a “world that is future for us”. Both “God and a future life are two presuppositions that are not to be separated from the obligation that pure reason imposes on us” (KrV A810–11/B838–9), although God seems to be necessary in order to ensure that even in that future life happiness will indeed be the consequence of virtue, that is presumably virtue in the present life or “world of appearances”. “Without a God and a world that is now not visible to us but is hoped for, the majestic ideas of morality” would be “objects of approbation and admiration but not incentives for resolve and realization because they would not fulfill the whole end that is natural for every human being and determined a priori and necessarily through the very same pure reason” (KrV A813/B841). It is not part of Mendelssohn’s argument in Phaedon that the promise of eventual realization of our natural happiness is a necessary condition for maintaining our resolve to be moral, but it is part of his argument that because our capacity for happiness is part of our natural endowment and we must believe that our natural endowment can eventually be fulfilled, even if not in the natural world, we must believe that Providence or God intends that capacity to be fulfilled. Kant’s argument in the Canon and Mendelssohn’s inference to our realization of happiness in eternity

166

Paul Guyer

share the assumption that our interest in happiness is both natural and properly accommodated by morality. In the second Critique, by contrast, Kant argues that we must believe in immortality in order to believe in the possibility of perfecting our virtue and worthiness to be happy, not our happiness. His argument is that “complete conformity of dispositions with the moral law is the supreme condition of the highest good” and required of us, but that such complete conformity, or “holiness”, is “a perfection of which no rational being in the sensible world is capable at any moment of his existence”, so we must instead assume the possibility of “an endless progress toward that complete conformity”, which is in turn “possible only on the presupposition of the existence and personality of the same rational being continuing endlessly (which is called the immortality of the soul)” (KpV 5:122). Kant’s argument here is that we must have adequate opportunity to perfect our moral vocation, and this seems similar to Mendelssohn’s claim that God would not have given us the capacity for virtue unless he has also given us a future life with the possibility of “continuing to practice” and perfect “virtue and wisdom under his divine guidance” (Mendelssohn, Phaedon, I: 418), a future life in which “all conflicts of obligations, all collisions of duties, which could send a limited being into doubt and uncertainty, will find their irrevocable resolution” (I: 420). Of course, endless existence cannot be observed in the sensible world, so presumably this endless or immortal existence must be possible for us in another or future world, as the first Critique had said, although it would no doubt take some fancy footwork to assign this endless existence to our noumenal selves, which are supposed to be altogether atemporal. Indeed, in the essay on “The End of All Things”, which followed Religion within the Boundaries of Mere Reason by one year, Kant will suggest that the image of endless time is only our way of representing something that is not temporal at all.⁴ But there are other problems with Kant’s adoption of Mendelssohn’s idea that we have an infinite life in which to perfect our virtue. One problem is immediately apparent, namely that Kant’s argument for the postulate of the existence of God turns on the necessity of believing in “a supreme cause of nature having a causality in keeping with the moral disposition”, a “cause of all of nature” that “is to contain the ground of the correspondence of nature not merely with a law of the will of rational beings but with the representation of this law, so far as they make it the supreme deter See Kant, Das Ende aller Dinge, e. g., :. It might seem as if endlessness follows directly from the atemporality of things in themselves. But all that Kant would be entitled to argue is that we cannot represent any cessation of existence of a thing in itself temporally, because we cannot represent anything about things in themselves temporally, and not that things in themselves might not cease to exist in some way that we cannot represent.

Kant, Mendelssohn, and Immortality

167

mining ground of the will” (KpV 5:125). The problem is that after Kant’s initial worry that there is no apparent causal connection between virtue and happiness, he introduces belief in God as “a cause of all nature, distinct from nature”, to explain how we can believe in the possibility of a causal connection between virtue and happiness within nature when he just has argued that the perfection of the moral will or of virtue is not something that we can expect within nature. So there is no reason why we should have to believe that happiness can be the consequence of virtue within nature, thus no reason to postulate a God who is the cause of harmony between virtue and happiness within nature. It would even seem that on this account the realization of happiness, as something that is to take place within nature, would have to run ahead of the perfection of virtue, as something that can take place only in an endless, clearly non-sensible, non-phenomenal, and at least in that sense non-natural life, that is, a world that is not a Sinnenwelt. This seems incoherent, putting the cart before the horse. But an even more serious problem lies in Kant’s assumption that the perfection of human virtue or our worthiness to be happy is equivalent to holiness, a condition toward which we could only endlessly progress and thus attain in no natural life. We might well think that what morality requires of us is unremitting commitment to the moral law no matter what conflicting temptations we experience, not a kind of holiness that consists in never experiencing any temptations contrary to morality at all. At least in the Religion, Kant defines the kinds of holiness possible for human beings simply as “the universal and pure maxim of the agreement of conduct with the law, as the germ from which all good is to proceed”, a “change of heart which must be possible because it is a duty” and which Kant does not say requires the humanly impossible elimination of temptation in the form of inclination (RGV 6:66–7). And Kant gives us no reason to think that human beings are incapable of fulfilling this demand in their natural or normal lives, on the contrary his transcendental idealist theory of the will seems intended to explain precisely how human beings can really be free, thus free to subordinate all inclination and self-love to the moral law, while nevertheless appearing to be in time and therefore appearing to be causally determined. Transcendental idealism is not a promise of the freedom to be truly moral in some future life, but an account of how, despite appearances, we are already so free in our natural life. Kant’s postulate of an unnatural immortality in the second Critique thus seems to rest on a demand for an unnatural holiness that is not otherwise part of his account of the demand of morality and our freedom to fulfill it, that is indeed in tension with it. I would suggest that Kant came to recognize this point after the publication of the Critique of Practical Reason, and thus retracted its Mendelssohnian suggestion that belief in immortality is necessary for striving to perfect our virtue. I

168

Paul Guyer

would also suggest that his recognition of this point combined with his enduring faith in the possibility of moral progress may well have led him to increased emphasis on the possibility of moral progress in the human species rather than human individuals in the final decade of his work, although he had already introduced that theme in the critical decade. I will discuss this point in the final section of this paper.

3 Kant’s Diminution of the Postulate of Personal Immortality Kant’s final decade of work began with the Critique of the Power of Judgment. Here Kant clearly enunciated the premise of the doctrine of the postulates of pure practical reason as he had developed it in the preceding two critiques, stating that “a final end cannot be commanded by any law of reason without reason simultaneously promising its attainability, even if uncertainly” (KU §91, 5:471n.) even though by the entire argument of the book he essentially reassigned the conception of nature as a realm in which we can realize our moral goals and which we must believe to have been authored by God to make this possible from reason to reflective judgment. But even though in the third Critique Kant continues to mention immortality along with the existence of God “as the conditions under which alone we can, given the constitution of our (human) reason, conceive of the possibility” of the highest good as the “effect of the lawful use of our freedom” (KU §91, 5:470), he offers no explicit reason why we must postulate immortality in addition to the divine authorship of nature in order to conceive of the possibility of the highest good. Even more striking, in Religion within the Boundaries of Mere Reason, published three years after the third Critique, the work in which we might most naturally expect to find Kant defending the rationality of belief in immortality as well as in the existence of God, we find that his treatment of the highest good is restricted to its Preface and makes no reference to immortality at all, while in the body of the work, although Kant does not explicitly reject the possibility of personal immortality, he undercuts any argument that personal immortality is necessary for the perfection of our virtue.⁵ Instead, he argues that because of our sensible thus temporal nature, it is natural for us to represent even our successful moral conversion only as progress, but he specifically says that even though this progress can be cut off at any time, by the  It is striking that James DiCenso’s recent commentary on the Religion (DiCenso ) contains no reference to the concept of immortality at all. DiCenso does not discuss the passages from Part Two of the Religion I am about to discuss, which might seem to suggest Kant’s continuing commitment to the postulate of immortality but which, as I will argue, do not.

Kant, Mendelssohn, and Immortality

169

natural fact of death, God can still recognize the perfection of our moral conversion in our natural, finite life spans though we ourselves cannot. Indeed, he goes even further and suggests at least once that our image of eternity is really only our way of representing the fact that we are faced with the choice between evil and good as long as live, however long that happens to be. Kant thus essentially reverses his own and Mendelssohn’s earlier representation of immortality as the necessary condition for the realization of our moral perfection into a representation of our inescapable freedom to choose good or evil that can be exercised in our normal lives even though we ourselves are not in an epistemic position to be certain of our success or failure in moral choice. In a footnote to the Preface to Religion that is almost as long as the Preface itself, Kant presents the challenge presented by the concept of the highest good as explaining how “The proposition, ‘There is a God, there is a highest good in the world,’ […] is to proceed (as proposition of faith) simply from morality”, as “a synthetic a priori proposition” (RGV 6:6n.). He then proceeds to explain in what sense the command to seek to realize the highest good can be regarded as a synthetic a priori proposition: he argues that “One’s own happiness is the subjective ultimate end of rational beings belonging to the world (they each have this end by virtue of their nature which is dependent upon sensible objects)”, and thus, though Kant does not explicitly state this, the proposition that any rational agent does seek its own happiness would be a synthetic a posteriori proposition; “But that every human being ought to make the highest possible good in the world his own ultimate end is a synthetic practical proposition a priori, that is, an objective-practical proposition given through pure reason” (RGV 6:6–7n.) precisely because it commands something that goes beyond merely natural desire or inclination. So far, this explains the synthetic a priori status only of the command to seek the realization of the highest good, not of the proposition linking that commandment to the conditions of its possibility, and it does not explain precisely why reason commands that each be concerned with the happiness of all. Kant says nothing to address the latter omission, but he does conclude his note by adding that “But now, if the strictest observance of the moral laws is to be thought of as the cause of the ushering in of the highest good (as end), then, since human capacity does not suffice to effect 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 would come about, i. e., morality leads inevitably to religion” (RGV 6:7–8n.).

Kant does not explicitly say that the proposition that there is a sufficient cause for the realization of the highest good is synthetic a priori, but it is a causal claim, therefore synthetic, and it is obviously not a posteriori, so he must be assuming that it will be equally obvious that it is synthetic a priori, even if it is a

170

Paul Guyer

matter of rational faith rather than theoretical speculation, and so his initial question will have been answered. But the point I want to emphasize is that in this account of necessary belief in the conditions of possibility of realization of the highest good, there is no reference to personal immortality at all. Indeed, as I suggested, Kant actually undermines any need for belief in personal immortality in the Religion. This takes place in several steps. First, although Kant does not abjure holiness of will as the perfection of the moral vocation of human beings, as already mentioned he subtly revises his conception of holiness in a way that disconnects it from immortality. While he may previously have conceived of holiness of will as the absence of all inclination, something that can never be expected to occur in phenomenal human existence, in Religion he carefully defines holiness as “holiness of maxims in the compliance to one’s duty”, “recovery” (because we are all born with a disposition to the good, even though we also have a tendency to evil and have apparently chosen to act upon it) “of the purity of the law, as the supreme ground of all our maxims, according to which the law itself to be incorporated into the power of choice, not merely bound to other incentives, nor indeed subordinated to them (to inclinations)” (RGV 6:46). That is, Kant does not define holiness as consisting in the elimination of inclinations, but only as the thoroughgoing subordination of them or any maxims they would suggest to the moral law as one’s fundamental maxim – and this is something that, because of the transcendental freedom of the human will, any human being can achieve at any time, not something that has to be deferred to some future life. With this account of holiness combined with the conception of radical freedom to which Kant has committed himself in the Religion, there is no need of immortality for the perfection of virtue (or worthiness to be happy), nor does Kant suggest that there is. In the second part of the Religion, however, which describes “the battle of the good against the evil principle for dominion over the human being” (RGV 6:57), Kant introduces what might look like a reference to personal immortality after all. In one passage, Kant clearly refers to personal immortality only in the subjunctive. He argues that the radical character of human freedom allows for complete conversion from good to evil within the natural life span of human beings, which allows the inference that if there is an afterlife that moral conversion will continue to hold. He writes here: “For a human being who from the time of his adoption of the principles of the good and throughout a sufficiently long life henceforth has perceived the efficacy of these principles on what he does, i.e., on the conduct of his life as it steadily improves, and from that has cause to infer, but only by way of conjecture, a fundamental improvement in his disposition, can yet also reasonably hope that in this life he will no longer forsake his present course but will rather press on with it with ever greater courage, since his advances, provided that their principle is good, will always increase his strength for future ones; nay, if after this life another

Kant, Mendelssohn, and Immortality

171

awaits him, that he will persevere in it (in all appearances under different circumstances, yet according to the very same principle) and come ever closer to his goal of perfection, though it is unattainable; for, on the basis of what he has perceived in himself so far, he can legitimately assume that his disposition is fundamentally improved” (RGV 6:68).

This passage emphasizes, first, that humans can convert to pure morality during their natural lives although they cannot be sure that they have done so, an uncertainty that Kant sometimes presents as due to the empirical and thus contingent fact that one can often do the morally right thing out of the motive of self-interest which is very good at hiding itself (see GMS 4:407), but which of course also follows necessarily from the transcendental idealist supposition that the choice of fundamental maxim is noumenal rather than phenomenal. Kant then goes on merely to state that if there is an afterlife, such a person might have the same degree of confidence that his moral conversion will continue there too, but this is hardly an assertion, even on practical grounds, that there is an afterlife. (It might be objected that the last sentence in this passage implies only the possibility of continued progress toward conversion, not of the completion of conversion; I would rather suggest that the final reference to a fundamental improvement of disposition implies the possibility of complete conversion, although Kant may also be worrying about the possibility of relapse even from complete conversion.) A preceding passage, however, seems actually to assert that there is an afterlife. Carefully read, however, this passage continues Kant’s epistemic theme, actually arguing only that because of our uncertainty about our own moral conversion, we represent our lives as endless, but that God, who is not subject to our epistemic limits, can know that we really have undertaken moral conversion in our natural lives: Kant writes here: “According to our mode of estimation, [to us] who are unavoidably restricted to temporal conditions in our conceptions of the relationship of cause to effect, the deed, as a continuous advance in infinitum from a defective good to something better, always remains defective, so that we are bound to consider the good as it appears in us, i. e., according to the deed, as at each instant inadequate to a holy law. But because of the disposition from which it derives and which transcends the senses, we can think of the infinite progression of the good towards toward conformity to the law as being judged by him who scrutinizes the heart (through his pure intellectual intuition) to be a perfected whole even with respect to the deed (the life conduct). And so notwithstanding his permanent deficiency, a human being can still expect to be generally well-pleasing to God, at whatever point in time his existence be cut short” (RGV 6:67).

In this intricate passage, Kant makes two different suggestions. One is the general point that because of our restriction to phenomenal self-knowledge we can never be certain of our complete conversion from evil to good because of the

172

Paul Guyer

noumenal and thus hidden character of that conversion. The other is that because outwardly good actions could always be performed from self-love rather than respect for the moral law, even if our conversion from the evil to the good fundamental maxim were complete we could not tell this from even the best actions, and we might thus represent our conversion only as progress in infinitum, not as ever complete. But either way, Kant goes on to imply that our conversion or choice of good fundamental maxim can be completed in our natural or phenomenal lifetime, before whatever point in time it is cut short, and that God can know what our choice was even if we cannot. There is thus no argument for a postulate of personal immortality here, rather there is an argument for the possibility of complete moral conversion within the natural, finite human life span. Even if he does not reject belief in immortality itself, Kant has thus rejected the connection between moral perfection and immortality suggested by Mendelssohn and then adopted by himself in the second Critique. Kant makes a further suggestion about the significance of the idea of immortality for us. If transcendental freedom means that a human is free to convert from evil to good at any point in his life, such a being must also be free to relapse or convert back from good to evil at any point as well – even though we cannot cognize noumenal choice temporally, it would be a mistake to infer that we get to make only one noumenal choice, for that itself would be an illegitimate importation of temporality into our conception of the noumenal.⁶ In fact, it is already part of Kant’s model of choice in Religion that every human being can make at least two free choices – first the free choice to be evil, for as imputable evil must be the product of choice (RGV 6:21n. and 6:37), and then (or “then”, in scare-quotes), the free choice to be good. But if any human gets at least two noumenal choices, in spite of the non-temporality of the noumenal, then there is no reason why she cannot have more than two; in fact, a human retains freedom of choice at every moment of her life (whatever non-representable and “inscrutable” form this must take at the noumenal level), and thus always has the power to choose good even though she has previously chosen evil, or evil even if she has previously chosen good. In a way, our moral conversion is imperfect precisely because our freedom is, on Kant’s account, perfect. Kant puts both this point and the previous point that he is not actually affirming immortality by adding that “without any necessity to presuppose dogmatically, as an item of doctrine, that an eternity of good or evil is the human lot also objectively” (RGV  This mistake would be of the same sort as assuming that because we cannot count things in themselves using our ordinary spatio-temporal forms of differentiation, there can only be one thing in itself, a mistake that Schopenhauer makes but that Kant does not. See for example Arthur Schopenhauer, On the Basis of Morality, §, p. .

Kant, Mendelssohn, and Immortality

173

6:69). “Eternity” might seem to imply immortality, but Kant’s remark that we do not need to presuppose it dogmatically means that he is not actually asserting it. Thus there is no reason to think that Kant is taking back his previous remark that human life might be cut short at any time without the individual knowing of his own conversion while God does. Instead, this just seems to be a way of describing how we represent our moral condition to ourselves: as long as we live we are confronted with the choice between good and evil, we are never relieved of that choice no matter how well we have chosen in the past, and perhaps given our notorious inability to imagine our own deaths we represent this condition to ourselves as eternal. But in Religion Kant has not argued that human life is immortal, only that human freedom is inescapable. I think that Kant is making a similar point in the opening pages of the essay on “The End of All Things”, which was published in the Berlinische Monatsschrift, by now moved to Saxon Jena to escape Prussian censorship, two years after the first part of the Religion. Here Kant begins by arguing that we cannot think of eternity as a continuation of ordinary time and thus cannot think of change as possible if we have once passed through the judgment day to eternity; so we must suppose that the moral character that we have adopted by the end of our actual life, good or evil, will remain fixed for eternity. But all this, I believe, is not meant as an argument that we actually will or even must believe that we will somehow endure in a changeless eternity; it is rather meant to suggest simply that we must make every effort to attain moral perfection within our earthly lives, “[f]or we see nothing before us now that could teach us about our fate in a future world except the judgment of our own conscience, i.e., what our present moral state, as far as we are acquainted with it, lets us judge rationally concerning it” (Ende 8:330). When it comes to moral improvement, there’s no time like the present.

4 The Immortality of the Species rather than the Person Even in the Religion Kant continued to hold that belief in immortality is both natural and a mark of true religion. For example, that Judaism neither promises nor commands a belief in immortality is part of Kant’s argument that in its original form it was the statutory law for the civil condition of a particular people, not a religion based on universally valid ethical law – even though “it can hardly be doubted that the Jews subsequently produced, each for himself, some sort of religious faith which they added to the articles of their statutory faith” (RGV

174

Paul Guyer

6:126).⁷ But he had reduced the endless progress of human individuals to immortality to a mere way of representing both our imperfect knowledge of our own conversion and even more importantly our need to choose between evil and good as long as we live. Yet even in his final decade Kant found it difficult to give up entirely on a more substantive idea of endless progress, so he transferred this possibility from the human individual to the human species. To be sure, Kant had been concerned with the moral progress of the species while he was still concerned also to defend the rationality of belief in personal immortality: notably, the Second Proposition of the 1784 essay “Idea for a universal history from a cosmopolitan point of view” had stated that “[i]n the human being (as the only rational creature on earth), those predispositions whose goal is the use of his reason were to develop completely only in the species, but not in the individual” (Idee 8:18). Even though this piece was composed between the first and second critiques, thus while Kant continued to emphasize the postulate of personal immortality, Kant’s rationale for it emphasized the mortality of individual human beings: “Reason itself does not operative instinctively, but rather needs attempts, practice and instruction in order gradually to progress from one stage of insight to another. Hence every human being would have to live exceedingly long in order to learn how he is to make a complete use of all his predispositions; or, if nature has only set the term of his life as short (as has actually happened), then nature perhaps needs an immense series of generations, each of which transmits its enlightenment to the next, in order finally to propel its germs in our species to that stage of development which is completely suited to its aim” (Idee 8:19).

Perhaps Kant had in mind some subtle difference between the perfection of human reason and the perfection of human virtue. Perhaps he had in mind the perfection of theoretical and prudential reason rather than the perfection of pure practical reason, which as we have just seen he would argue a decade later in the Religion is available to everyone at all times. Perhaps in this popular essay he was only talking

 Upon hearing a first version of this paper, Stephen Palmquist suggested that in a note to the second edition of RGV at :, Kant defended Mendelssohn against the calls for his conversion to Christianity, originally made by Johann Caspar Lavater in  and then repeated in a review of the first edition of Jerusalem by August Friederich Cranz, on the ground that Mendelssohn’s personal addition of belief in immortality, attested precisely by Phaedon, to the articles of statutory Jewish faith, obviated the need for any such conversion. That seems to me a reasonable implication of the whole argument of the Religion, but not the point of the note at :, which says only that Mendelssohn thought that if the “yoke of external observances” were removed from Judaism, the “yoke of a profession of faith in sacred history” would remain. Kant does not explicitly say that the second yoke includes belief in immortality.

Kant, Mendelssohn, and Immortality

175

about the phenomenal human being, not the noumenal human being. Be all that as it may, in this essay he was clearly focused on progress in the human species because of the obvious mortality of individual human beings. After what I have suggested is the turning point of the third Critique, Kant returned to the question of the possibility of progress on the part of humanity as a whole. One revealing passage occurs in the third part of the Religion, where Kant argues that “[t]he human being ought to leave the ethical state of nature in order to become a member of an ethical community” (RGV 6:96). Whether his argument here is that the realization of the highest good can never be brought about by a single individual, but only by the human species as a whole, or only that an individual’s efforts at virtue and toward the realization of the highest good can always be weakened by the influence of others unless all human beings are members of an ethical community, there is in either case no need to postulate personal immortality although there is need to posit the existence of God as the condition of the possibility of human cooperation and thus of an ethical community itself: “Now, we have here a duty sui generis, not of human beings toward human beings but of the human race toward itself. For every species of rational beings is objectively – in the idea of reason – destined to a common end, the promotion of the highest good as a good common to all. But, since this highest good will not be brought about solely through the striving of one individual person for his own moral perfection but requires rather a union of such persons into a whole toward that very end, toward a system of well-disposed human beings in which, and through the unity of which alone, the highest moral good can come to pass […] – We can already anticipate that this duty will need the presupposition of another idea, namely, of a higher moral being through whose universal organization the forces of single individuals, insufficient on their own, are united for a common effect” (RGV 6:98).

There is no hint of a postulate of personal immortality here, only an argument for faith in a divinely-grounded possibility of the human cooperation that would be necessary for human kind ultimately to realize its highest good. Another text from the 1790s in which the postulate of personal immortality gives way to faith in the possibility of the progress of the human species is the essay on “Theory and Practice”, published at the same time as the Religion. This essay begins with a response to Christian Garve’s objection that Kant’s doctrine of the highest good has undermined the purity of his moral philosophy itself by making the promise of personal happiness the motivation to morality. Kant responds that the highest good is only meant to be the object of morality, not its ground, although as the object of morality its realization must be possible if the attempt to be moral is to be rational, and further that the happiness with which the highest good is concerned is not individual or “selfish” happiness but “univer-

176

Paul Guyer

sal happiness combined with and in conformity with the purest morality throughout the world” (Gemeinspruch 8:279–80). In the course of this response to Garve, Kant says, perhaps not entirely disingenuously, that “[i]t is not as if the universal concept of duty first gets ‘support and stability’ only on the presupposition” of both God and a future life, “that is, gets a sure basis and the requisite strength of an incentive, but rather that only in the ideal of pure reason does it also get an object” (Gemeinspruch 8:279); his point here is that if morality ultimately commands us to seek universal happiness, then we must also be able to believe that universal happiness is ultimately possible – but we do not need to have a promise of our own happiness and thus of our own immortality in order to retain our incentive to be moral, although that is precisely what the first Critique had maintained. The second part of “Theory and Practice” provides a preliminary statement of Kant’s philosophy of right, to be more fully expounded four years later in the first half of the Metaphysics of Morals, and the third part raises the question “[a]re there in human nature predispositions from which one can gather that the race will always progress toward what is better and that the evil of present and past times will disappear in the good of future times?” (Gemeinspruch 8:307) Kant argues for an affirmative answer to this question, and explicitly presents his position as an alternative to the position he ascribes to Mendelssohn, not in Phaedon but in Jerusalem, which is that while individuals may make moral progress, “the human race as a whole make[s] small oscillations, and it never takes a few steps forward without soon afterward sliding back twice as fast into its former state” (Gemeinspruch 8:307).⁸ The heart of Kant’s position is that it is the duty of each human being not to realize the highest good in his own lifetime but to make the best contribution that he can to the moral progress of the species. His explication of his position clearly assumes human mortality: “I shall therefore be allowed to assume that, since the human race is constantly advancing with respect to culture (as its natural end) it is also to be conceived as progressing toward what is better with respect to the moral end of its existence, and that this will indeed be interrupted from time to time but will never be broken off. I do not need to prove this presupposition; it is up to its adversary to prove [his] case. For I rest my case on my innate duty, the duty of every member of the series of generations – to which I (as a human being in general) belong and am yet not so good in the moral character required of me as I ought to be and hence could be – so as to influence posterity that it becomes always better (the possibility of this must, ac-

 Kant quotes from Jerusalem, Section II, pp. –; the passage may be found in Mendelssohn, Jerusalem, or on Religious Power and Judaism, pp. –. Mendelssohn seems to be arguing that the moral oscillation of the human race is precisely the condition that gives human individuals the freedom to be moral on their own, with the “appropriate means” to so doing that God has granted to every human at every time and place (p. ).

Kant, Mendelssohn, and Immortality

177

cordingly, also be assumed), and to do it in such a way that this duty may be legitimately handed down from one member [in the series of] generations to another […] however uncertain I may always be and remain as to whether something better is to be hoped for the human race, this cannot infringe upon the maxim, and hence upon its presupposition, necessary for practicable purposes, that it is practicable” (Gemeinspruch 8:308–9).

This passage is important because it clearly distinguishes between the natural and the moral end of human beings, so that the claim that it makes upon the assumption of human mortality explicitly applies to the latter and not to the former. With equal clarity Kant then applies his standard premise for a postulate of pure practical reason – ought implies can, i. e., what is morally necessary must be possible or practicable – to the moral duty of each human being “so to influence posterity that it becomes always better”. What we must assume to be possible is that each individual can be motivated to make his proper contribution to the morality of future generations and that human nature allows for the influence of past and present generations upon future ones. This is very different from any assumption of the possibility of personal immortality; it is precisely what must be assumed to make the moral progress of the species possible in the absence of personal immortality. In this passage Kant thus clearly argues that all that we must postulate in order to make possible the moral progress of the species is the possibility of the moral influence of human individuals on their successors; this replaces the postulate of personal immortality. Kant takes the burden of proof to be on (the late) Mendelssohn to disprove this possibility, not to be on himself to prove it. Does he go even further in his response to Mendelssohn’s position by arguing that the moral progress of the human race is not just possible but necessary? He seems to do precisely this in several places, one the essay on the “Idea of Universal History” previously discussed and another the 1795 pamphlet Toward Perpetual Peace. In the culminating Ninth Proposition of the former Kant speaks of a “consoling prospect into the future […] in which the human species is represented in the remote distance as finally working itself upward toward the condition in which all germs nature has fully placed in it can be fully developed and its vocation here on earth can be fulfilled” (Idee 8:30). In the latter Kant speaks, for example, of the “guarantee (surety)” offered by “the great artist nature […] from whose mechanical course purposiveness shines forth visibly, letting concord arise by means of the discord between human beings even against their will” (Frieden 8:360–1). However, Kant cannot mean by such remarks more than that nature, including human nature, contains mechanisms that make moral progress possible, and that can be used as means for moral progress by properly motivated human beings. This is clear from his statement in the essay on history that since the mas-

178

Paul Guyer

ters of human beings who will devise and enforce the laws that will make their subjects behave, or make crooked timber grow straight, are themselves only human, the problem of right can be solved only if such rulers have “correct concepts of the nature of a possible constitution, great experience practiced through many courses of life and beyond this a good will that is prepared to accept” the former: all the mechanisms of nature are only necessary conditions for justice, and there is never a sufficient condition without a good will (Idee 8:23). It is likewise clear in Towards Perpetual Peace from the First Appendix, which argues that “the concept of right would be an empty thought” without a “moral politician”, one who “will make it his principle that, once defects that could not have been prevented are found within the constitution of a state or in the relation of states, it is a duty, especially for heads of state, to be concerned about how they can be improved as soon as possible and brought into conformity with natural right” (Frieden 8:372). Again, without a good will on the part of those in power there is no sufficient condition for justice (which I am assuming is part of morality). And while it may be difficult to reconcile this conclusion with Kant’s talk of a guarantee of right from nature in Perpetual Peace, it is entirely consistent with Kant’s claim in the history essay that we must be able to have a prospect of a natural “condition in which all germs nature has fully placed in it can be fully developed and its vocation here on earth can be fulfilled” (emphasis added), for this is a straightforward application of Kant’s principle that ought implies can, which is always consistent with his conception of the radical freedom of human beings. And nothing less than the possibility that human beings can always employ or subvert the mechanisms of nature would be.⁹ Does this conclusion bring Kant’s position closer towards Mendelssohn’s after all? It does not bring Kant back toward the argument for the practical postulate of personal immortality as a condition of the perfection of personal virtue into which he had transformed Mendelssohn’s metaphysical conviction in the second Critique; he is still claiming only that it must be possible for mortal human beings to make their individual contributions to the progress of the human race. But it might bring Kant closer to Mendelssohn’s position that human history is an “oscillation”, for he cannot say more than that it must be possible for human politicians to play their special role in realizing human progress. It would hardly be compatible with his theory of human freedom as well as with his or anyone else’s observation of human history to hold that human politicians always will play that role well. In this way, the distance between Kant’s philosophy of history and Mendelssohn’s may not be as great as it seems to Kant

 I have argued this point at greater length in Guyer  and Guyer .

Kant, Mendelssohn, and Immortality

179

himself, even though the difference in their metaphysics of immortality is ultimately very great indeed.

References DiCenso, James J. 2012, Kant’s Religion within the Boundaries of Mere Reason: A Commentary, Cambridge: Cambridge University Press Guyer, Paul 2006, “The Possibility of Perpetual Peace”, in: Luigi Caranti (ed.) 2006, Kant’s Perpetual Peace: New Interpretative Essays, Rome: LUISS University Press, 161–81 Guyer, Paul 2009, “The Crooked Timber of Humankind”, in: Amelie Rorty/James Schmidt (eds.) 2009, Kant’s ‘Idea for a Universal History with a Cosmopolitan Aim’: A Critical Guide, Cambridge: Cambridge University Press, 129–49 Mendelssohn, Moses 1983, Jerusalem, or on Religious Power and Judaism, translated by Allen Arkush, with introduction and commentary by Alexander Altmann, Hanover: University Presses of New England for Brandeis University Press Mendelssohn, Moses 2009, Phaedon oder über die Unsterblichkeit der Seele, in drey Gesprächen, in: Moses Mendelssohn, Ausgewählte Werke: Studienausgabe, Band I: Schriften zur Metaphysik und Ästhetik 1755–1772, edited by Christoph Schulte, Andreas Kennecke, and Grażyna Jurewicz, Darmstadt: Wissenschaftliche Buchgesellschaft Schopenhauer, Arthur 2009, “On the Basis of Morality”, in: Arthur Schopenhauer 2009, The Two Fundamental Problems of Ethics, translated and edited by Christopher Janaway, Cambridge: Cambridge University Press Wolff, Christian 1726, Vernünfftige Gedancken von den Absichten der natürlichen Dinge, second edition, Frankfurt and Leipzig: Renger

Jochen Bojanowski

Life without Death: Why Kantian Agents Are Committed to the Belief in Their Own Immortality Whether it is good or bad that there is an end to our lives is controversial. One might think that death is bad because it deprives us of some of the good things a human life has to offer. Death, so understood, might be a deprivation of future happiness (Nagel 1970). On the other hand, an endless life might become tedious and boring, in which case it would seem to be a good thing for our lives to end after all, if not early on, then at least at some point (Williams 1973). If immortality makes a life worse rather than better, it is neither something we should wish for nor constitutive of a good human life. Hence, in a fully enlightened secular moral philosophy, the idea of immortality seems to suffer the same fate as the metaphysical ideas of God and libertarian free will; they have no place in moral philosophy. In his argument for immortality, Kant takes a radically different point of departure. His argument is not driven by the question whether the end of life deprives us of future happiness. On the contrary, the idea of immortality only comes into play for Kant because it is a necessary presupposition of the search for moral perfection. If we want to do what we ought to do, to perfect ourselves morally, we have to believe in our immortality. In other words, believing in our immortality makes possible an action that we cognize as practically necessary (i.e. obligatory), namely the action of perfecting our moral character. For this reason, Kant calls the belief in our immortality a “postulate of pure practical reason” (KpV 5:122). It is not particularly hard to understand why Kant’s argument for the immortality of the soul has received little attention. Kant wants to derive the belief in immortality from the consciousness of our moral obligation to bring about the highest good. But if, for example, one doesn’t accept the claim that this obligation is categorical or that moral obligation is given in our consciousness as a “fact of pure reason” (KpV 5:31), one won’t see any need to address Kant’s argument for the claim that our immortality is a postulate of pure practical reason. However, it is indeed surprising that even those who are willing to endorse the main tenets of Kant’s moral philosophy do not extend this willingness to the postulate of our immortality. In this paper, I would like to revisit Kant’s argument for the postulate of the immortality of the soul as he presents it in the second Critique. The argument is to be found in the fourth section of the second chapter of the Dialectic (KpV 5:122). It is only two paragraphs long, but it is dense. Most commentators believe that the argument is untenable (Beck 1960: 268 f.; Allison 1990: 172; Kleingeld

182

Jochen Bojanowski

1995: 156; cf. Surprenant 2008). I believe that we do not yet fully understand why moral perfection, even in the case of human beings, requires not just virtue as a “moral disposition in battle” (KpV 5:84; translation modified) but also the idea of holiness. Moreover, we need to give a coherent account of why Kant claims both that human beings cannot be holy and that they have an obligation to become holy. I am going to argue that if we are indeed morally required to promote the highest good, as Kant claims, and if one sense of the highest good is the demand for moral perfection, this demand also requires us to “postulate”, in a specifically Kantian sense, our immortality. In order to show that this practical belief in our immortality does not have its source in some kind of mysticism, but is instead based on an act of cognition, I will need to bring Kant’s notion of a postulate into focus. If the argument as I reconstruct it is correct, not only should we be more appreciative of Kant’s argument for the postulate of immortality, but, as I would like to suggest in closing, we also have good reason to revisit Kant’s account of moral virtue. I shall now assess Kant’s argument for the postulate of the immortality of the soul in detail. This argument’s grounding premise is established in the Analytic of the second Critique. I am not going to question the truth of this premise here, but I will spell out its significance. (1) “The production of the highest good in the world is the necessary object [of our will]” (KpV 5:122). (P) The phrase “necessary object of our will” could be taken to mean that we cannot but will the production of the highest good. That is, it is impossible for us to will otherwise than to will the production of the highest good. This reading is clearly incompatible with Kant’s view of human imperfection, however, which entails that it is possible to will in ways contrary to our highest good. Hence, we have to read “necessary” in a practical rather than a theoretical sense. “x is practically necessary for us” means that we are (morally) obliged to do or will x. We can then reformulate the first premise as follows: (1)* We are morally obliged to produce the highest good in the world. Yet even this revised premise, as it stands, seems to require too much. Since “ought” implies “can”, (1)* suggests that we have sufficient power to bring about the highest good in the world through our own actions. This would make us responsible not only for the goodness of our own dispositions but, since Kant speaks here of “world”, for everyone else’s dispositions as well. Moreover, a component of the highest good also consists in being happy. If “produc-

Life without Death

183

tion” is taken to mean that we have sufficient power to realize a state of affairs in which morality and happiness are proportionately distributed, we would be required to bring about happiness. But this requirement goes beyond our powers, since we cannot foresee what will make us happy (KpV 5:36). Since “cannot” implies “not-ought” (as the contrapositive of the entailment from ought to can), it turns out that the requirement stated in (1)* leads to absurdity. However, we do not need to read “production” here in a strong sense. Production can also be read in the sense of contributing to a joint effort. In this sense of “production”, the individual agent does not have sufficient power to realize x, but she partakes in the collective enterprise of producing x. This weak reading of “production” is corroborated by two sentences later in the text, where Kant instead uses the term “promote” (KpV 5:122). Each individual promotes the highest good, even if no single individual produces it. In other passages, Kant explicitly states that we have to promote the highest good “with all my powers” (KpV 5:142, see also KpV 5:122, 125, 130). So in order to disambiguate (1)* we can say: (1)** We are morally obliged to promote the highest good in the world as much as it is in our power to do so. The second premise makes explicit what the promotion of the highest good requires on the part of the moral agent: (2)

The “complete conformity of dispositions with the moral law is the supreme condition of the highest good” (KpV 5:122). (P)

Since the conformity of dispositions with the moral law is what in the human case is called virtue, and since Kant even explicitly holds that virtue is the supreme condition of our happiness (KpV 5:122), it is tempting to simplify this premise in the following manner: (2)* Virtue is the supreme condition of the highest good. However, Kant speaks here not of mere conformity but of “complete conformity” (KpV 5:122). Complete conformity is a state of perfection that goes beyond virtue, since virtue is “moral disposition in battle” (KpV 5:84; translation modified). The battle signifies an imperfection. As long as we still have to fight against obstacles, our moral conviction is not complete. Complete conformity is instead what Kant in this context explicitly calls “holiness” (KpV 5:122). I will come back to this crucial distinction in a moment when we get to the conclusion in (3). For now let me just reformulate premise (2) more accurately:

184

Jochen Bojanowski

(2)** Holiness is the supreme condition of the highest good. Kant distinguishes between two senses of “highest” (KpV 5:110). In one sense (the sense he mentions in premise (2)), it means “supreme (supremum)”. That which is supreme is the “condition which is itself unconditioned, that is, not subordinated to any other (originarium)”. In the second sense it means “complete (consummatum)”. That which is complete is that “whole which is not part of a still greater whole of the same kind (perfectissimum)” (KpV 5:110). To say that holiness is the highest good in the sense of being supremum means that it is good in itself, i. e. its goodness does not depend on something other than itself. Conversely, all other goodness depends on the conformity of our dispositions with the moral law. This claim sounds controversial because it seems to rule out a kind of goodness that is merely instrumental or morally neutral. However, to say that x is merely instrumentally good is to say that its goodness is not fully determined by its instrumental value. It is good only under some condition. This condition, Kant claims, is the complete conformity of our dispositions with the moral law. Even in the case of so-called morally neutral goods, their goodness depends on their moral permissibility. If a choice between different means, x and y, does not make a moral difference, it is still a necessary condition for their goodness that both options qualify as morally permissible. In this sense, x and y are not morally neutral at all. Hence, Kant holds that all goodness depends on the conformity of our dispositions with the moral law. The claim that morality is the highest good in the supremum sense of “highest” was already established in the Analytic. The consummatum sense of “highest” is the driving force of the Dialectic. For being moral does not analytically contain happiness, yet “to need happiness, to be also worthy of it, and yet not to participate in it cannot be consistent with the perfect volition of a rational being that would at the same time have all power, even if we think of such a being only for the sake of the experiment” (KpV 5:110). Hence, if it were impossible for the moral agent to also be happy, the highest good (in both senses of the term “highest”) could not be realized. However, it is important to note that with regard to the postulate of immortality, the consummatum sense of “highest” can be disregarded. The postulate of immortality is supposed to articulate a rational belief to which we are committed only insofar as we act morally. The consummatum sense of “highest”, on the other hand, commits us to a belief in God rather than immortality. Thus I take it to be misleading to suggest that the conjunction of both components of the highest good is of fundamental importance for the postulate of immortality (cf. Surprenant 2008: 90). The driving force of the immortality postulate is the supremum sense of “highest”. I will come back to this point below.

Life without Death

185

Kant moves on by drawing the following conclusion from premises (1) and (2): (3) “This [complete] conformity must therefore be […] possible […]” (KpV 5:122). (1, 2) This conclusion rests on the implicit inference that “ought” implies “can”. Since we are obliged to promote the highest good, the highest good presupposes our holiness, and “ought” implies “can”, it must be possible for us to become holy. We can reformulate (3) such that the main point for the immortality postulate comes to the fore: (3)* Holiness must therefore be possible for us. Premise (4) might look like a plain contradiction of premise (3)*: (4) “[H]oliness [is] a perfection of which no rational being of the sensible world is capable at any moment of his existence” (KpV 5:122). (P) If no rational being of the sensible world is capable of holiness, how can holiness be possible for us? Human beings always remain susceptible to sensible inclinations. Holiness, by contrast, can only be fully ascribed to a “self-sufficient intelligence [whose] choice is […] represented as incapable of any maxim that could not at the same time be objectively a law” (KpV 5:32). In other words, only if we were able to eradicate our susceptibility to sensible inclinations would we be able to be holy. But this cannot be done. “[A]ll the moral perfection that a human being can attain is […] only virtue, that is, a disposition conformed with law from respect for law, and thus consciousness of a continuing propensity to transgression or at least impurity […]” (KpV 5:128). Virtue, Kant claims, is all human beings can actually attain; holiness, by contrast, is out of reach. Since “cannot” implies “not-ought”, Kant must either drop premise (1) and give up on our obligation to promote the highest good or spell out the supremum sense of “highest” in terms other than those of holiness. If he were to spell out the supremum sense only in terms of virtue, his whole argument for the immortality of the soul could not get off the ground. For why shouldn’t a mortal soul acquire virtue? Indeed, many critics believe that, in his argument for the immortality of the soul, Kant moves illegitimately from the claim that moral perfection is virtue to the claim that moral perfection is holiness. Since human beings can at best be virtuous, their highest good in the supremum sense is virtue, not holiness (Beck 1960: 268 f.; Allison 1990: 172; Kleingeld 1995: 156).

186

Jochen Bojanowski

Let us first consider whether Kant’s move to holiness is indeed unnecessary. The objection seems to downplay the imperfection of virtue on the whole. Even if we manage to win in “battle”, the fact that we have to battle means that acting against the moral law is still possible. Virtue is merely the “firmness of our maxim in following our duty”. It “always contains the temptation of the principle of evil to violate the law, which distinguishes it from holiness” (23:260; my translation). So victory in battle is not sufficient for true perfection of our moral disposition. Virtue is in this sense only a partial victory. In a perfectly moral being, a being that is in full agreement with the moral law, any inclinations which are not in line with the moral law would be irreversibly silenced. Still, even if one is willing to concede that virtue is not full moral perfection, it might just be the most perfect state human beings can possibly attain, because human beings are essentially imperfect. In claiming that holiness is a “perfection of which no rational being of the sensible world is capable” Kant himself seems to claim that it is impossible for us to win this battle satisfactorily (KpV 5:122). So the next move in Kant’s argument remains puzzling: (5) We are obliged to become holy. (1, 2) This conclusion already follows from premises (1) and (2); our obligation to promote the highest good and the fact that to become holy is part of the highest good also make it obligatory to become holy. Yet again, if holiness is “a perfection of which no rational being of the sensible world is capable”, as Kant claimed in the previous step, and if “cannot” implies “not-ought”, how can we be obliged to become holy? If we look at premise (4) more closely, we can see that it comes with a restriction: A rational being of the sensible world is only incapable of perfecting himself “at any moment of his existence” (KpV 5:122; my emphasis). But this restriction does not seem to help Kant’s argument, for if we don’t exist, how can we continue to promote the highest good? The achievement of holiness is only possible, if it is possible at all, within our lifetime. In (6), Kant seems to acknowledge the problem. Since it is impossible for a rational being of the sensible world to be holy at a particular point in time, there has to be an alternative way in which she can be holy, for otherwise we would arrive at the absurdity that she ought to x but cannot x. This alternative way, Kant claims, is an “endless progress” (KpV 5:122) toward holiness. (6) Our holiness (human holiness) “can only be found in an endless progress toward […] complete conformity [of our dispositions with the moral law]”, i. e. the holiness of a supersensible being (KpV 5:122).

Life without Death

187

I believe we don’t lose anything substantial if we reformulate (6) as follows: (6)* Our holiness is only possible “in an endless progress” toward the holiness of a supersensible being. With this claim, Kant seems to fall even deeper into the same absurdity. We ought to promote the highest good; to promote the highest good is to become holy; therefore we ought to become holy. But as finite rational beings we cannot become holy; therefore, we can only approach holiness. Kant seems to have changed the inference rules here: “Ought” does not necessarily imply “can” after all; sometimes, as in the case of holiness, it merely implies “almost can”. So the objection that Kant illegitimately moves from virtue to holiness still seems live (Beck 1960; Allison 1990; Kleingeld 1995). I don’t think that Chris Surprenant’s recent suggestion helps to divert these worries (cf. Surprenant 2008). Surprenant claims that the objection overlooks the important distinction between the highest good for the person and the highest good for the world (KpV 5:110 f.). According to Surprenant’s reading, only the highest good for the person is concerned with virtue, while the highest good for the world consists in holiness. Since the first premise of Kant’s argument for immortality holds that the production of the “highest good in the world is the necessary object [of our will]” (KpV 5:122; my emphasis), Surprenant infers that “virtue never enters into this discussion [i. e. discussion of immortality], and so no unwarranted shift [i. e. a shift from virtue to holiness] is ever made” (Surprenant 2008: 91). However, as I have argued above, there are good reasons to believe that the postulate of immortality is concerned with the highest good in the supremum sense. This sense of highest is in a broad sense the moral goodness of our moral disposition. Now, when Kant introduces the distinction between highest good for the world and highest good for the person, he explicitly holds that in both cases virtue is the supreme condition (KpV 5:111). The question rightly raised by Kant’s critics is why the highest good in the supremum sense is sometimes couched in terms of virtue and sometimes, as in the context of the immortality postulate, in terms of holiness. And they rightly ask for a justification of why Kant’s argument for immortality speaks of holiness, even though holiness seems to be something of which finite rational beings are incapable. Moreover, Surprenant’s interpretation of the first premise of Kant’s argument for the immortality postulate, and with it the textual evidence for his claim, is unconvincing. Kant’s claim that the promotion of the “highest good in the world is the necessary object of [our] will” (KpV 5:122) does not imply that the postulate of immortality is only concerned with holiness and not with virtue. First, in the passage where Kant introduces the distinction between highest good of the

188

Jochen Bojanowski

world and highest good of a person, he speaks of “virtue” in the context of the highest good of a person, and he speaks of “morality” (Sittlichkeit) in the context of the highest good in the world (KpV 5:110 f.). Surprenant seems to equate “morality” with “holiness”, but he does not give an argument for this equation. Second, Surprenant’s explanation of the distinction between highest good for the person and highest good for the world is at best misleading (Surprenant 2008: 90). According to Surprenant, the highest good for a person is attained “when he always acts in accordance with the moral law and has things go his way in the world”. The highest good for the world, by contrast, “is attained when a perfect relationship between virtue and happiness is exhibited”. However, the contrast between the highest good for a person and the highest good for the world is not that, in the one case, the relation between virtue and happiness is “perfect”; it is rather that the same “perfect” relationship is not merely limited to the individual but extended to the whole moral community. I can see why Surprenant wants to say that the highest good of the world, if attained, would be more perfect than the highest good of a particular individual. But the postulate of immortality only relates to the perfection of our moral disposition and not to the extension of the highest good from one to all members of the moral community. In short: I don’t think that Surprenant successfully responds to the objection that Kant’s move from virtue to holiness is illegitimate. The worry that Kant illegitimately infers holiness from virtue in his immortality argument still remains. However, I believe there is a viable interpretation that can respond to this worry. What needs to be challenged here is the presupposition, shared by critics of Kant’s immortality argument, that we cannot become holy. Kant’s claim that we ought to become holy does not need to be taken to imply that we can in fact achieve holiness (cf. Allison 1990: 172). To be governed by the idea of holiness as an end, it must be possible not to attain holiness but to endlessly approach it. Even if holiness (that enjoyed by a supersensible being) cannot be attained, it can still guide us in our efforts to become perfect. So to hold that we have an obligation to become holy implies only that we can engage in an infinite progress towards our moral perfection (RGV 6:60 f.). Perfection for us consists in an infinite journey. The idea that guides our moral conduct in general is not merely virtue, but the absence of all resistance to morality in us, which absence consists in holiness. To be virtuous once or twice or several times is to have won a battle, yet it is not to have won the war. Yet for finite rational beings, the war cannot ever be entirely put to a close. Kant can therefore hold on to claim (6); holiness for us is possible only as endless progress. If we can indeed make this concession to Kant, it follows that this endless progress toward the complete conformity

Life without Death

189

of our dispositions with the moral law is what we must be aiming at when we aim at the highest good. (7) Thus, in promoting the highest good, endless progress toward holiness is “the real object of our will” (KpV 5:122). (1, 2, 5, 6) The notion of endless progress makes room for Kant’s postulate of immortality. For if it is true that we are obliged to become holy, and this becoming is in our case an endless progress, then we must think of our existence as also “continuing endlessly”. To think of ourselves as capable of the sort of perfection that consists in an endless journey, we must believe in our immortality. For we cannot think of ourselves as capable not just of a finite but of an infinite journey if we do not think of ourselves as existing throughout the journey. This endless existence of our personality is what Kant calls “immortality of the soul” (KpV 5:122). (8) An endless progress toward holiness is only possible if “the existence and personality of the same rational being continu[es] endlessly”, i. e. if the soul is immortal (KpV 5:122). (P) For the sake of simplicity we can reformulate (8) as follows: (8)* An endless progress toward holiness is only possible if our soul is immortal. With this premise in hand, Kant can draw the inference that our soul is immortal. For if we are obliged to promote the highest good, which is possible only if we become holy, and if our holiness is itself only possible as an endless progress toward holiness, which in turn requires an immortal soul, then Kant is entitled to conclude that, insofar as we are obliged to promote the highest good, we consider our soul to be immortal. (9) Promoting the “highest good is practically possible only on the presupposition of the immortality of the soul” (KpV 5:122). (1, 2, 5, 6, 7, 8) Let me first try to get a better grip on the notion of practical possibility here. Why does Kant speak of practical rather than theoretical possibility? If Kant had shown that the highest good is only theoretically possible on the presupposition of the immortality of the soul, he would have shown that the immortality of the soul is a condition of the possibility of experience of the promotion of the highest good. However, since it is opaque to us whether actions are done merely in

190

Jochen Bojanowski

accordance with duty or from duty, the highest good in the supremum sense, let alone the highest good on the whole, cannot be an object of possible experience. Alternatively, “practical possibility” could be taken to be the counter-notion of “practical necessity”. Since practical necessity means obligation, practical possibility would mean permissibility. However, the moral law, and not the immortality of the soul, is what makes our actions right or lawful. Hence it does not even make sense to say that the promotion of the highest good is morally permissible only on the presupposition of the immortality of the soul. On yet another reading, “practically possible” could be taken to refer to our faculty of volition. In other words, willing the promotion of the highest good is only (rationally) possible on the presupposition of the immortality of the soul. Or again, if we do not presuppose that our soul is immortal, we cannot will the promotion of the highest good without thereby contradicting ourselves. However, we have to be careful with the term “contradiction” here. Contradiction in willing is the broad criterion of immoral action (GMS 4:424). One might think that if it is morally obligatory to promote the highest good, and if to aim at the highest good without believing in the immortality of the soul entails a contradiction in our will, then believing in the mortality of the soul is morally impermissible. However, Kant explicitly holds that the proposition “the soul is immortal” is merely, as he says, a “rational belief” (Vernunftglaube, KpV 5:146). Since “a belief that is commanded is an absurdity” (KpV 5:144), it does not make sense to claim that we are morally obliged to believe in the immortality of the soul. Thus, if there is some sort of contradiction in willing with respect to believing in the immortality of the soul, it must be of a different kind than that associated with immoral action. Let me elaborate on this difference here. This will help us to see more clearly what “practically possible” in (9) means. A contradiction in willing is morally impermissible if the end cannot be willed or agreed to by every rational cognizer, insofar as there is a contradiction involved in willing both the maxim and that this maxim should become a law. For example, to the extent that I will both the maxim of telling a lie to secure a loan and that this principle should become a universal law, I will both that I should receive a loan and, at the same time, that no one should lend money to anybody. This is effectively like willing to Φ and willing that it be impossible to Φ. In the case of the immortality of the soul, by contrast, the contradiction arises not between a maxim and the maxim as a law, but between our end (the highest good) and our belief in the mortality of our soul. In other words, if I want to Φ and it is constitutive of my Φ-ing that I believe that p, it would be contradictory (irrational) to will to Φ and at the same time to believe not-p. So, although Kant is right that a “belief that is commanded is an absurdity”, there is still a sense in which I am logically bound to believe that p. To put this more concretely, willing a maxim of promot-

Life without Death

191

ing the highest good whilst believing in the mortality of the soul is like willing a maxim of running a marathon whilst believing that one will necessarily cease to exist at the 25-mile mark. Should I find myself in such a situation, it would still be possible for me to achieve my end, i. e. running the marathon (or promoting the highest good) without explicitly or consciously believing in the constitutive features of willing my end. However, in committing myself to these ends I also implicitly commit myself to the belief that the soul is immortal (or that my running farther than 25 miles is indeed possible). And if I explicitly believe in the mortality of the soul (or in the impossibility of my running more than 25 miles), then I cannot rationally will to promote the highest good (or to run the marathon). So in this sense, the relevant belief is a necessary presupposition of the possibility of the act of volition. That is, willing the highest good (or running the marathon) is “possible” only if one holds the relevant belief. This is what I take Kant’s phrase “practically possible” to mean. There is of course an important difference between the immortality belief and the marathon belief. And this difference will help us to get a better grip on the peculiarities of Kant’s notion of a postulate, which he defines in the next step of his immortality argument. (10) A “postulate of pure practical reason” is a “theoretical proposition” which is not theoretically “demonstrable” but is “attached inseparably to an a priori unconditionally valid practical law” (KpV 5:122). (P) The “unconditionally valid practical law” is obviously the moral law. Kant had argued in premise (1) that we are morally obliged to promote the highest good. At (9), he arrived at the conclusion that this promotion of the highest good is practically possible only if we presuppose that the soul is immortal. Now since Kant has shown in the Paralogisms in the first Critique that there is no substantive theoretical knowledge to be had with respect to the human soul, he is entitled to the claim that proving the immortality of the soul is theoretically impossible. (11) “The soul is immortal” is not a theoretically demonstrable proposition. From here we can move directly to Kant’s final conclusion: (12) Therefore, “The soul is immortal” is a postulate of pure practical reason (KpV 5:122). (1, 9, 10, 11)

192

Jochen Bojanowski

We must now bring Kant’s notion of a postulate into focus. For only if we understand what a postulate of pure practical reason is do we know what kind of belief our belief in our immortality is. One of the essential features of a practical postulate is that our reason for believing it is not a theoretical reason. If we transfer this back to our examples of the marathon and the highest good, one might think that in both cases I have a non-epistemic reason to believe that I am immortal (or that I might possibly run more than 25 miles). It would be irrational to will the realization of some end whilst believing that some of the conditions of its realization are not met. Insofar as I have reason to be rational, it seems that, in both the marathon case and the highest good case, I have a reason to believe that p just in virtue of willing that I Φ. Andrew Chignell and Marcus Willaschek have attempted to read Kant’s notion of a practical postulate in terms of non-epistemic reasons (Chignell 2007 and Willaschek 2011). I believe talk of “non-epistemic reasons” in the immortality case is appropriate if it means that we do not have theoretical reasons for our belief. However, the will to promote the highest good ought not to be understood in a voluntaristic sense. A moral agent does not merely “set an absolutely necessary moral end” (Chignell 2007: 356) but rather cognizes this end as absolutely necessary. Cognition of the highest good is a synthetic practical cognition a priori. Kant thinks of his practical philosophy as a practical epistemology. Both the first and the second Critiques are critiques of our faculty of cognition. Acts of practical reason (the will) are no less acts of cognition than are acts of theoretical reason. Hence a more accurate way of drawing the distinction between different kinds of reasons for belief would be to say that in the case of the highest good we have a practical epistemic reason to believe that we are immortal. The relation between the belief in immortality and the highest good is not one between a belief and an external, empirical fact about the world but is instead a relation between a belief and an internal cognitive demand of reason. My belief that I can run more than 25 miles is implied in my will to run a marathon, but since my will to run a marathon is not an act of practical cognition, I cannot claim that willing gives me justification to believe that I can run more than 25 miles. The justification for this belief has to come from experience. That I will something to be true does not give me evidence for thinking it is true. To be sure, both the propositions “I can run more than 25 miles” and “the soul is immortal” are, as Kant says, “theoretical proposition[s]” (cf. KpV 5:122). They are both propositions of what is the case rather than what ought to be the case. Yet Kant also holds that “the soul is immortal” is a “postulate of pure practical reason” (KpV 5:122, my emphasis). How do these two claims go together? How can a belief be a theoretical proposition and yet in another sense be practical? What Kant attempts to establish here is that there are some theoretical

Life without Death

193

propositions we are entitled to hold only because they have a justificatory ground in pure practical reason. In other words, pure practical reason gives these beliefs their justification, which is what makes them beliefs of pure practical reason. They are, as Kant says, “attached inseparably” to the moral law (KpV 5:122). Hence the belief “the soul is immortal” should be understood as a practical belief (KrV A823/B851). The belief “I can run more than 25 miles”, by contrast, is not a practical belief at all. For in practically believing something, the reason for believing it to be the case is a practical idea, a concept of pure practical reason, e. g. the highest good. Instead, this case comes closer to what Kant in the first Critique calls “pragmatic belief” (KrV A824 f./B852 f.), where the reason for holding this belief to be true is some empirical fact. Thus the certainty associated with pragmatic beliefs is fundamentally different to that associated with practical beliefs. In the pragmatic case, the belief comes in degrees, and the subjective certainty can be tested with a bet. The more money I am willing to bet, the higher my certainty (KrV A824 f./B852 f.). By contrast, the practical belief – the belief in our immortality – is absolute and does not come in degrees at all, for it follows analytically from a practical cognition a priori. Beliefs of both kinds cannot bring a claim to “certainty” (KrV A822/B850), however. For even if we are subjectively fully convinced that the theoretical proposition is true, as long as we cannot prove in intuition that our judgment corresponds to its object, we have belief rather than knowledge. For this reason, Kant distinguishes sharply between a practical postulate and a theoretical postulate in mathematics, i.e. a postulate that is a theoretical proposition but that also has its justificatory ground in theoretical cognition. In the latter case, the postulate “postulate[s] the possibility of an action, the object of which has been previously theoretically cognized a priori with complete certitude as possible” (KpV 5:11). To make Kant’s thought more transparent here, let’s take a look at the first postulate of Euclidian geometry. It reads as follows: “Let it be postulated to draw a straight line from any point to any point” (Euclid, Elements, Book 1, Postulates). The line in this postulate is the “object” of which Kant speaks in the passage above. The line is cognized as possible in pure intuition. Yet the postulate of Euclidian geometry does not postulate the possibility of the line (the object). Instead, what the postulate itself declares as possible is, as Kant says, the “action” of drawing the line from one point to the other. In the case of the practical postulate, by contrast, “the possibility of an object itself ([…] the immortality of the soul)” is postulated (KpV 5:11). And this possibility is not, as in the theoretical case, “cognized with respect to the object”, which means that the possibility of the immortality of the soul is not known in pure intuition a priori. Instead, the possibility of this object is an “assumption necessary [nothwendige Annehmung] with respect to the subject’s observance of its objective but practical

194

Jochen Bojanowski

laws” (KpV 5:11). So only under this assumption is doing one’s duty possible. Since the immortality of the soul cannot be directly cognized in theoretical cognition and is only a necessary presupposition of the realization of the highest good, the immortality of the soul is, from a theoretical standpoint, “merely a necessary hypothesis” (KpV 5:11). Kant defines a hypothesis as “a holding-to-be-true of the judgment of the truth of a ground for the sake of its sufficiency for given consequences” (Logik 9:84). This is a complicated way of saying that a hypothesis is a judgment that we hold to be true because it can sufficiently explain certain given consequences. We cannot directly verify the hypothesis, but the fact that the consequences are explained sufficiently by it gives us some justification to assert it. This last claim can help us to see the theoretical significance of the practical postulate more clearly. Whatever one might think from a practical perspective, the proposition that the soul is immortal sounds rather far-fetched from a theoretical standpoint. If Kant is entitled to such a hypothesis, why not hypothesize that there are ghosts or elves, or other supersensible beings? What entitles Kant to such a hypothesis even from a theoretical perspective? If the answer were just that it is impossible to prove that the soul is mortal, which is in fact the result of the Paralogisms, it seems that the Dialectic in the first Critique does indeed make room for all sorts of supersensible entities. Why stop with immortality and God? Here is Kant’s answer: “If the imagination is not simply to enthuse but is, under the strict oversight of reason, to invent [dichten], something must always first be fully certain and not invented, or a mere opinion, and that is the possibility of the object itself. In that case it is permissible to take refuge in opinion concerning the actuality of the object, which opinion, however, in order not to be groundless, must be connected as a ground of explanation with that which is actually given and consequently certain, and it is then called an hypothesis” (KrV A770/B798).

Here is what I take to be the main claim of this passage. There are two conditions that have to be fulfilled in order for something to count as a hypothesis in the Kantian sense. First, it has to be certain that the object is logically possible. If it is logically possible, we are entitled to an opinion (Meinung) but not to the belief or even to the knowledge claim that the object is actual. This opinion as such is still arbitrary and without justificatory ground. This brings us to the second condition: The opinion has to “be connected as a ground of explanation with that which is actually given and consequently certain”. Now, it is obvious how in the postulate of immortality both of these conditions are fulfilled. First, the belief “the soul is immortal” is logically possible. Second, it follows from the cognition of the moral law, which is “given” to us as a “fact of pure reason”

Life without Death

195

(KpV 5:31). So the proposition “the soul is immortal” is, from a purely theoretical standpoint, merely an opinion. It only becomes a hypothesis because pure practical reason gives us a reason to assert it. It is a hypothesis because only under this presupposition can we explain the obligation to promote the highest good. And this explanation is not like explaining an earthquake by making the “earth out to be an animal, in which the circulation of the inner fluids produces warmth” (Logik 9:85), for the object, the highest good, has been cognized as practically necessary and as such it must also be possible. This is precisely not the case in the earthquake explanation. Yet the proposition “the soul is immortal” is more than just a hypothesis; it is not merely a presupposition that best explains some given consequences. Instead, it is a presupposition we absolutely must believe in if we want to do what we ought to do – to become holy. In other words, the practical belief makes possible an action that we cognize as practically necessary. This is what it means for a belief in the soul’s immortality to be a postulate of pure practical reason.

Conclusion Kant’s argument for the immortality of the soul is part of his larger project of holding on to the core ideas of traditional metaphysics by showing that their real employment lies in practical cognition. Kant believes that his moral philosophy as a practical metaphysics can give cognitive content to fundamental ideas that cannot be theoretically asserted. For this reason, Kant claims that practical reason has primacy over theoretical reason. As we have seen, critics of Kant’s argument for immortality don’t challenge this general argumentative strategy. What they find most suspicious is that Kant’s argument for immortality is inconsistent with what they take to be his more considered view of moral virtue. They believe that since it is impossible for us to be holy, Kant’s considered view is that moral perfection is virtue, not holiness. In this paper I have attempted to show why virtue is not equivalent to the ultimate or supreme end of moral agency, which is in fact holiness, even if virtue is the best we can achieve with regards to each of our particular actions. This idea of holiness as an infinite end guides our attempts to perfect our moral character. If we are indeed morally required to promote our moral perfection, then, in order to do what we ought to do, we need to “postulate” our immortality. We need not have intuitive knowledge of our own immortality in order to postulate it in the Kantian sense; rather, our belief in our own immortality makes possible an action that we practically cognize as obligatory. If what I have been arguing in this paper is correct, then not only should we think of the postulate of immortality as a necessary consequence of Kant’s moral phi-

196

Jochen Bojanowski

losophy, but we are also thereby given good reason to reconsider our account of Kantian virtue. Let me outline this in closing. Dating back at least to Schiller, Kant has been accused of confusing virtue with mere continence. According to the Aristotelian alternative, the virtuous agent does not engage in battle. Instead, everything that could speak against the moral good in him is fully “silenced” (McDowell 1978: 26). In recent years, attempts have been made to bring Kant closer to Aristotle by embracing the Aristotelian notion and showing that Kant’s conception of virtue is, despite some rather unrepresentative passages, compatible with Aristotle’s (Baxley 2003). I believe that reflecting on Kant’s notion of holiness can partly accommodate Aristotle’s conception of virtue, but at the same time it might also help to clarify the differences between Kant and Aristotle. Kant’s notion of holiness can only “partially” accommodate the Aristotelian notion because, for Kant, the state in which the morally virtuous agent finds herself is always imperfect. According to Kant, even the most cultivated virtuous agent still struggles with being or remaining virtuous. The Aristotelian, by contrast, holds that virtue is the “lack of struggle” (McDowell 1978: 27). Yet Kant does make room for this Aristotelian conception of virtue by transforming it into an idea of pure practical reason. The idea of complete agreement between one’s moral dispositions and the moral law is the supremum sense of the highest good. This idea guides the virtuous agent to further perfect herself. In the second chapter of the Religion, Kant will lay out in more detail what he only alludes to in a footnote in the second Critique (KpV 5:127n.): namely how this idea gets its motivational force by schematizing it as an ideal. An ideal is a concept which represents an “individual thing which is determinable, or even determined, through the idea alone” (KrV A568/B596). This ideal is not the Stoic sage but the “Son of God” (RGV 6:62). The son of God (Jesus) represents an individual human being whose superior goal is the promotion of the highest good. The idea of moral perfection thus precedes this ideal. Hence the ideal is not an ideal of the imagination, as with the Stoic sage, but is instead “presented to us by reason for emulation” (RGV 6:61; my emphasis). This ideal is the prototype (Urbild) on the basis of which we are supposed to model our own moral disposition (Ectypus) (RGV 6:61; cf. KrV A578/B606). But since the full realization of this ideal is impossible for us, we can only approach holiness through an endless progression. This idea of an endless progression of our existence, which is our immortality, is constitutive of our undertaking to promote the highest good. Hence, in sharp contrast to Aristotle, we cannot morally perfect ourselves, according to Kant, without postulating immortality.

Life without Death

197

References Allison, Henry E. 1990, Kant’s Theory of Freedom, Cambridge: Cambridge University Press Baxley, Anne Margaret 2003, “Does Kantian Virtue Amount to More than Continence?”, Review of Metaphysics 56, 559–86 Beck, Lewis White 1960, A Commentary on Kant’s Critique of Practical Reason, Chicago: University of Chicago Press Chignell, Andrew 2007, “Belief in Kant”, Philosophical Review 116, 323–60 Euclid 1956, The Thirteen Books of Euclid’s Elements, translation and commentary by Sir T.L. Heath, New York: Dover Publications Kleingeld, Pauline 1995, Fortschritt und Vernunft: Zur Geschichtsphilosophie Kants, Würzburg: Königshausen und Neumann McDowell, John 1978, “Are Moral Requirements Hypothetical Imperatives?”, Proceedings of the Aristotelian Society, Supplementary Volumes 52, 13–29 Nagel, Thomas 1970, “Death”, Noûs 4, 73–80 Surprenant, Chris W. 2008, “Kant’s Postulate of the Immortality of the Soul”, International Philosophical Quarterly 48, 85–98 Willaschek, Marcus 2010, “The Primacy of Practical Reason and the Idea of a Practical Postulate”, in: Andrews Reath/Jens Timmermann (eds.) 2010, Kant’s Critique of Practical Reason. A Critical Guide, Cambridge: Cambridge University Press, 168–96 Williams, Bernard A.O. 1973, Problems of the Self: Philosophical Papers 1956–1972, Cambridge: Cambridge University Press

III Epistemology, Science, and Metaphysics

Thomas Höwing

Kant on Opinion, Belief, and Knowledge Kant’s doctrine of the highest good is part of a broader philosophical agenda that is expressed by his famous slogan “I had to deny knowledge in order to make room for faith” (KrV Bxxx). As Kant argues in the first Critique, any attempt to gain knowledge of God, the human soul, and freedom of the will is doomed to fail. What is more, since these objects transcend the limits of possible experience, we are not even entitled to form scientific hypotheses or opinions about them.¹ However, as moral agents, we are justified in holding the Belief that God exists, that the human soul is immortal, and that we have free will. And what justifies this moral Belief is a genuinely practical consideration – a consideration about the way in which we might realize the highest good.² Kant’s agenda is based on a novel account of the traditional distinction between opining (Meinen), Believing (Glauben), and knowing (Wissen).³ According to Kant’s rationalist predecessors, the three attitudes present different grades on an ascending scale of epistemic justification. Roughly speaking, the rationalists associate opinion with probabilistic justification, Belief with testimony, and knowledge with demonstrative reasoning.⁴ By contrast, Kant’s agenda presupposes that the attitude of Believing differs in a crucial respect from both knowledge and opinion: a Belief in the Kantian sense requires non-epistemic justification. As Kant puts it in the first Critique, “[o]nly in a practical relation, […] can a theoretically insufficient holding-to-be-true be called believing” (KrV A823/B851; my trans.). In other words, a Belief is justified not by virtue of evidence but by virtue of its relation to action. This departure from the rationalist tradition at least partly accounts for the impression of imbalance conveyed by Kant’s discussion of the tripartite division

 See KrV A–/B–, A/B; KU : f.  See KrV A–/B–, A/B; KpV : ff.; KU : ff.  The term Glauben, as Kant uses it, is notoriously difficult to translate. ‘Faith’ is too narrow since, in principle, Glauben is not restricted to matters of faith. ‘Belief’ is problematic as well. For one thing, it is controversial whether Glauben, as Kant understands it, involves outright belief, as opposed to what some recent philosophers have called ‘acceptance’ (see Chignell a: ; Chignell b: ). Moreover, Kant thinks that Wissen and Glauben are mutually exclusive. If you know that p, you cannot glauben that p, and vice versa. Thus to avoid misunderstanding, I shall follow Chignell (Chignell b: n.) and write “Belief” with a capital B.  See Wolff, German Logic, Vorbericht § and Chapter , §§, ,  f.; Wolff, Latin Logic §§, , ; Baumgarten, Acroasis Logica §§,  f.,  f.; Meier, Vernunftlehre §§, , , . See also Theis .

202

Thomas Höwing

in the first Critique. Rather than carefully laying out his version of the division, Kant devotes less than half a page to it and rushes on to argue that Belief, as he understands it, differs in quality from the other two attitudes. Indeed, it has already been noted that Kant’s official formulation of the distinction is incredibly dense and obscure: “Having an opinion is taking something to be true with the consciousness that it is subjectively as well as objectively insufficient. If taking something to be true is only subjectively sufficient and is at the same time held to be objectively insufficient, then it is called believing. Finally, when taking something to be true is both subjectively and objectively sufficient it is called knowing” (KrV A822/B850).

What seems to be clear is that all three attitudes involve assent, or what Kant calls ‘holding to be true’ (Fürwahrhalten). In other words, if an agent opines, Believes or knows that p, she takes p to be true. What is less clear is how the three attitudes differ from one another. A major reason for this obscurity is that Kant never really explains the underlying distinction between the subjective/objective insufficiency/sufficiency of an assent (or, in short, the SOIS-distinction). What is more, the way in which the SOIS-distinction figures in Kant’s formulation creates a major exegetical puzzle. Consider a more schematic version of Kant’s formulation: (i) Opinion is subjectively insufficient and objectively insufficient assent. (ii) Belief is subjectively sufficient and objectively insufficient assent. (iii) Knowledge is subjectively sufficient and objectively sufficient assent. To see the puzzle, consider Kant’s description of Belief. Kant describes Belief in terms of two features, namely, a positive feature that it shares with knowledge (subjective sufficiency) and a negative feature that it shares with opinion (objective insufficiency). In other words, Kant suggests that Belief is something like a ‘mixed’ assent that shares its essential features with knowledge and opinion. However, this seems to run counter to Kant’s central claim that Belief has at least one unique feature – it requires non-epistemic justification. My aim in this paper is to explore this puzzle in further detail. In the first section, I will argue that the two most recent interpretations of Kant’s distinction – those offered by Andrew Chignell and Lawrence Pasternack – fail to provide an adequate solution to the puzzle. In the second section, I shall outline an alternative reading of Kant’s tripartite division that emerges from a much-neglected passage in the late Jäsche Logik. As we will see, however, my reading will not yield a full-blown analysis of the three attitudes. Rather, it will show that Kant’s division is based on a more elementary classification of the different

Kant on Opinion, Belief, and Knowledge

203

ways in which a ground (Grund) might make it rationally appropriate for the agent to assent to some judgment. In the third and final section, I shall argue that, once we spell out Kant’s SOIS-distinction in terms of this classification, the above-mentioned exegetical puzzle dissolves.

1 Two Recent Interpretations To begin with, one might wonder whether there is a puzzle at all. Kant’s formulation, one might say, concerns not different modes of justification but different kinds of assent. Thus it might seem that Kant’s formulation is simply not intended to capture epistemological differences between types of justification but is instead meant to capture psychological differences between types of attitude. In other words, it might seem that there is simply no need for Kant to mention the fact that Belief requires non-epistemic justification. However, other passages clearly suggest that Kant’s formulation is meant to capture differences at the level of justification. For instance, in the late Jäsche Logik, Kant presents the following version of the formulation: (i)

Opinion is “based on a ground of cognition that is neither subjectively nor objectively sufficient” (Logik 9:66). (ii) Belief is “based on a ground that is objectively insufficient but subjectively sufficient” (Logik 9:67). (iii) Knowledge is “based on a ground of cognition that is objectively as well as subjectively sufficient” (Logik 9:70). In this modified formulation, Kant applies the SOIS-distinction not to different types of assent but to different types of ground. This suggests that, in distinguishing the three attitudes, Kant is concerned with different modes of justification after all. Kant’s focus seems to be the different ways in which a ground might make it rationally appropriate for the agent to assent to some judgment.⁵ In light of this suggestion, we may articulate the puzzle a bit more clearly. Kant describes the grounds required for Belief in terms of two features, namely, a positive feature shared with grounds required for knowledge (subjective suffi This assumption seems to be supported by the fact that, in the case of opinion and knowledge, Kant speaks of an underlying “ground of cognition” (Erkenntnisgrund). This term is of course familiar from Kant’s famous claim that the consciousness of the moral law justifies our holding that we have free will – or, as Kant puts it, that the moral law serves as the ‘ground of cognition’ (ratio cognoscendi) of the possibility of freedom (KpV :n.).

204

Thomas Höwing

ciency), and a negative feature shared with grounds required for opinion (objective insufficiency). And this seems to run counter to Kant’s claim that the grounds of Belief have at least one unique feature: they count as non-epistemic or as “practical” grounds. Now one might still object that this is not a real puzzle. Kant, one might say, is simply drawing a distinction between the three kinds of ground, and to do so he does not need to refer to the unique features of these grounds. To take a somewhat artificial example, suppose I were to draw a distinction between a snake, a house and an elephant. To distinguish these items, I obviously don’t need to refer to the fact that the house has some unique feature (such as, for instance, the fact that the house is not a living being). I might also refer to a set of non-unique features that, taken together, are sufficient to distinguish the house from both the snake and the elephant. Thus I might say that, just like the elephant and unlike the snake, the house is around four meters high, and just like the snake and unlike the elephant, the house does not have a trunk. The objection shows that there is no need to assume that Kant’s tripartite division provides us with full-blown descriptions or with self-standing definitions of the three kinds of ground. Kant might just as well refer to those features of the three grounds that allow him to draw a distinction between the corresponding attitudes. However, even if this is granted, Kant’s silence on the supposedly ‘non-epistemic nature’ of grounds of Belief remains puzzling. As we have seen, the claim that Belief requires non-epistemic justification lies at the very centre of Kant’s overall agenda of ‘denying knowledge in order to make room for faith’. Moreover, this claim must have been quite controversial among eighteenth-century German philosophers. After all, as we have also seen, it departs from the prevailing rationalist account of the three attitudes. In other words, although Kant’s official formulation of the tripartite division might not entail a straightforward inconsistency, it is just not clear why it should remain silent on the non-epistemic nature of grounds of Belief. That there is something puzzling about Kant’s formulation is also suggested by the different readings that recent commentators have proposed. To begin with, Andrew Chignell claims that Kant is operating with two different notions of subjective sufficiency. The first notion denotes an internalist condition on the justification required for knowledge. Roughly speaking, the notion requires that the agent be in a position to cite, upon reflection, what she takes to be the objectively sufficient ground of her assent to some proposition.⁶ By contrast, the second notion denotes a pragmatic condition on the justification required for

 Chignell a: , –; Chignell b: –.

Kant on Opinion, Belief, and Knowledge

205

Belief. Again putting things very roughly, the second notion requires that a firm assent to some proposition have sufficient ‘non-epistemic merits’ for the agent, i.e. it allows the agent to meet some non-epistemic goal or interest.⁷ Chignell’s reading has an obvious advantage: it preserves the idea that Kant’s description of Belief makes reference to non-epistemic justification. However, it seems that the reading also has an obvious disadvantage: it assumes that Kant’s formulation rests on an ambiguity. To be sure, it is a sad fact that Kant often uses the same term in different ways, sometimes doing so even though he is perfectly aware that the term has different meanings. When he does this, however, there is typically some other passage that accounts for the ambiguity, in which he distinguishes between the different senses of the term.⁸ In the case before us, however, we lack any such passage. As Lawrence Pasternack has already pointed out, nowhere does Kant indicate that he uses ‘subjective sufficiency’ in different senses.⁹ Moreover, the SOIS-distinction appears not only in the first Critique but also in Kant’s late writings and lecture notes, and always in roughly the same way.¹⁰ This suggests that Kant did not feel the need to revise or even clarify the distinction in a substantial way. In short, I think it is implausible to assume that Kant’s division rests on an ambiguity. More recently, Lawrence Pasternack has proposed an alternative interpretation of Kant’s tripartite division (see Pasternack 2011a, 2011b, 2014). In response to the puzzle, Pasternack adopts a somewhat different strategy – he simply denies that Kant’s description of Belief makes reference to non-epistemic justification. Instead, Pasternack suggests that Kant describes Belief in largely psychological terms, by referring to the high degree of firmness involved in this kind of assent.¹¹ More specifically, Pasternack argues that Kant describes Belief in terms of two features: (i) a high degree of firmness, which it shares with knowl-

 Chignell b: –; Chignell a: –, esp. .  To take a thematically related case, in the first Critique Kant introduces the concept of conviction, and he suggests that the latter refers to objectively sufficient assent (KrV A–/B–). However, Kant then moves on to claim that “[s]ubjective sufficiency is called conviction (for myself)” (KrV A/B). As Chignell suggests (correctly I think), Kant is working with different concepts of conviction in the passage. But this suggestion is supported by the fact that, in other passages, Kant clearly distinguishes between different senses of the term (see Logik :; Chignell a:  note ).  Pasternack :  f. note .  See, for example, Logik Busolt (), : f.; Logik Dohna Wundlacken (), :; Jäsche Logik (), :, , .  As Pasternack suggests, the Kantian term ‘firmness’ (Festigkeit) is to be taken in an absolute sense, as denoting “both the stability of one’s commitment to the proposition and the feeling of surety one has when reflecting upon it” (Pasternack : –; see : f. note).

206

Thomas Höwing

edge (subjective sufficiency), and (ii) a lack of strong epistemic support, which it shares with opinion (objective insufficiency). Pasternack’s reading thus preserves the consistency of Kant’s formulation – it shows that the two terms Kant uses in his description of Belief can be consistently applied to knowledge and opinion, respectively. However, Pasternack’s reading also has an obvious disadvantage: it gives us a largely psychological description of Belief. One worry is that this runs counter to the philosophical importance that Kant attaches to the non-epistemic justification of Belief. Another worry is that it is doubtful whether Pasternack is right in suggesting that a Belief always involves a high degree of psychological firmness.¹² Although there is evidence pointing in this direction,¹³ at least in the first Critique Kant suggests something different. Shortly after presenting his formulation of the division, Kant introduces the notion of pragmatic Belief, i. e. an assent that is justified by virtue of instrumental considerations. And Kant goes on to suggest that pragmatic Belief has “only a degree, which can be large or small according to the difference of the interest that is at stake” (KrV A825/B853). If Pasternack were correct, Kant would be committing something like a blatant inconsistency here: he would be claiming, in the very same passage, that Belief involves a high degree of firmness by definition and that certain Beliefs may involve a small degree of firmness.¹⁴

 Note that a similar assumption is made by Chignell, who argues that subjective sufficiency, as it appears in the description of Belief, denotes the non-epistemic merits of firm assent.  In the first Critique, Kant introduces the notion of doctrinal Belief, i.e. Belief that concerns thoroughly theoretical issues (KrV A–/B–; Chignell b: –; Pasternack a). He goes on to explain that the “expression of belief is in such cases an expression of modesty from an objective point of view, but at the same time of the firmness of confidence in a subjective one” (KrV A/B; see Chignell b:  f. note ). Similarly, in the very same passage, Kant points out that the degree of firmness or confidence with which a proposition is held by an agent can be determined by introducing an imaginary betting scenario. As Kant puts it, “[t]he usual touchstone of whether what someone asserts is mere persuasion, or at least subjective conviction, i.e., firm belief, is betting” (KrV A/B). However, in these passages, ‘firmness’ and related expressions may also be taken in a relative sense. In the first passage, Kant compares doctrinal Beliefs and scientific hypotheses. And his claim seems to be that, while the former do not require the theoretical support of the latter, they are at least as firmly held by the agent. In the second passage, Kant is talking about the degree of firmness that can be detected by introducing imaginary betting scenarios.  Pasternack recognizes this, but he argues that there is a tendency in Kant’s later works to restrict Belief to moral Belief, the latter being always firmly held by the agent. As Pasternack suggests, this more restrictive view is already present in the formulation of the division in the first Critique, but not in the subsequent treatment of pragmatic Belief (Pasternack b:

Kant on Opinion, Belief, and Knowledge

207

In this section, I have argued that the two recent attempts to deal with the puzzle face a variety of exegetical difficulties. But even if we ignore these difficulties for the moment, it seems that both readings share a more general disadvantage. Rather than dissolving the puzzle, they are somehow built around it. Both Chignell and Pasternack seem to assume that, in interpreting Kant’s formulation of the division, we have two options: either we show that Kant’s description of Belief makes reference to a justificatory feature not shared by knowledge or opinion or we show that the two terms used by Kant to describe Belief can be consistently applied to knowledge and opinion, respectively. We simply cannot have both. And this, of course, is just what the puzzle suggests. In other words, if we want to dissolve the puzzle, we need something different, namely, an interpretation that shows that both of these seemingly conflicting tasks can be met. As I want to suggest in the remainder of the paper, such an interpretation can be given.

2 An Alternative Interpretation To develop this interpretation, it will be necessary to consider another passage from the late Jäsche Logik in somewhat more detail. In this passage, Kant provides an alternative formulation of his tripartite division that does not rely on the SOIS-distinction. Let me begin by quoting the whole passage: “Opining is problematic judging, believing is assertoric judging, and knowing is apodeictic judging. For what I merely opine I hold in judging, with consciousness, only to be problematic; what I believe I hold to be assertoric, but not as objectively necessary, only as subjectively so (holding only for me); what I know, finally, I hold to be apodeictically certain, i.e., to be universally and objectively necessary (holding for all), even granted that the object to which this certain holding-to-be-true relates should be a merely empirical truth” (Logik 9:66).

In this passage, Kant formulates the division of justified assent in terms of the three modal modes of judgment, i. e. in terms of his distinction between problematic, assertoric, and apodictic judging.¹⁵ Moreover, Kant suggests that, at least partly, the three attitudes differ with regard to the intersubjective validity of the corresponding judgments. As Kant puts it, what I know holds “for all” judging agents, but what I Believe holds “only for me”.  f., –). It seems implausible to me to suppose, however, that Kant should have mixed up two inconsistent accounts within the span of only three pages.  It is remarkable that this passage has received no attention in the more recent debate on Kant’s account of the tripartite division. The most detailed discussion of which I am aware is that offered by Mattey (see Mattey ; see also Motta : –).

208

Thomas Höwing

Unpacking this passage would require a detailed discussion of central parts of Kant’s epistemology, which is beyond the scope of this paper. Thus in this section I shall confine myself to giving an outline of what I take to be the most plausible reading. As I want to suggest in the next section, however, this reading has at least one exegetical merit – it allows us to solve the exegetical puzzle raised by Kant’s division. To begin with, in the passage Kant describes knowledge as “apodeictic judging”, but he also points out that this does not rule out knowledge of empirical truths. In other words, my assent that there is a chair in the room might count as knowledge and hence as apodictic judging. This claim is puzzling, given that Kant tends to associate apodictic judgment with a priori knowledge. However, if we recall that Kant also describes knowledge in terms of an underlying ground, the following explanation suggests itself. When I know that p, my assent is based on a ground that makes it rationally necessary for me to assent that p. In other words, grounds of knowledge have what one might call strong normative force – they don’t merely license assent; they rationally require it.¹⁶ Let me briefly pause to say something about the terminology I will be using. In the sense that I am using the term, to say that x is rationally required to φ is to say that x ought to φ. However, in what follows I shall speak of rational requirement simply because this allows me to treat ought as a normative relation that holds between a ground, an agent and an assent that p. Moreover, to have a broader notion of that relation, I shall speak of a ground that makes it rationally appropriate for an agent to assent that p. Thus to say that a ground rationally requires an agent to assent that p is to specify the way in which the ground in question makes it rationally appropriate for the agent to assent that p. As we will see, however, certain grounds – grounds of opinion, that is – make it rationally appropriate for the agent to form an assent, but they don’t rationally require the agent to form an assent.¹⁷

 For a similar reading of Kant’s description of knowledge in terms of an underlying obligation, see Mattey :  ff.,  ff. Mattey also offers an interesting proposal as to how we might relate this description to Kant’s account of the laws of the understanding (Mattey :  f.).  Strictly speaking, to say that something – a fact or state – is a ground for an agent to assent that p is to say that this fact or state stands in a relation to the agent and the assent, namely, the relation of ‘being a ground for’ (for a similar view see Scanlon ). Now one might ask whether, according to Kant, this relation just is the rational appropriateness relation. At least on the face of it, Kant seems to allow for grounds that don’t make an assent rationally appropriate, such as, for example, ‘grounds of persuasion’ (see KrV A/B). In what follows, I shall thus attempt to leave the question open by using ‘ground’ simply as a placeholder for the relevant fact or state.

Kant on Opinion, Belief, and Knowledge

209

Returning to the passage before us, it seems that the description of knowledge in terms of an underlying requirement does not yet provide a full explanation as to why Kant thinks that knowledge involves apodictic judging. As Kant explains in the passage, “what I know […] I hold to be apodeictically certain, i.e., to be universally and objectively necessary (holding for all)” (9:66). In other words, a ground of knowledge does not simply make it necessary for the agent to assent to some judgment; it makes it universally and objectively necessary. The notion of universal and objective necessity that Kant is invoking here is reminiscent of his famous claim in the Prolegomena that objective validity and necessary universality for everyone are interchangeable concepts (Prol 4:298; see also Mattey 1986: 437). However, even without entering into a lengthy discussion of this latter claim, I think the upshot of Kant’s description is clear enough from the passage currently before us. Consider how Kant summarizes his description in parentheses. As Kant puts it, what I know holds “for all” rational agents. In other words, knowledge that p will entitle me to assert p to others, with the contention that they should assent to p as well. Now in light of what we have seen so far, it is tempting to explain this feature of knowledge in terms of an underlying ground. Thus we might say something along the following lines: what ultimately entitles me to assert p to others is the fact that my assent that p is based on a ground that imposes something like a universal requirement on me – the ground in question requires not only me but all rational agents to assent that p. However, as it stands, this can hardly be correct. For example, suppose I know that there is a chair in my room by virtue of seeing that there is a chair in my room (where seeing that p entails p). Now if Kant is correct, there will be something about this perceptual state, or about the fact with which it presents me, that requires me to assent that there is a chair in my room. It is hard to see, however, how this state, or the corresponding fact, could also require you to form the assent in question. You are simply not in a state that qualifies as seeing that there is a chair in Thomas’s room; nor are you presented with the corresponding fact that there is a chair in my room. In other words, Kant’s description must entail an important proviso. My ground of knowledge will require you to assent that there is a chair in the room provided that it is available to you. If you were in my room and were to see that there is a chair, it would – at least under ordinary circumstances – be just as necessary for you to form the relevant assent as it is for me.¹⁸  Note that statements like ‘My ground of knowledge that p is available to you’ allow for two readings. On a de dicto reading, what is available to you is the fact that some state or fact counts as a ground of knowledge with respect to my assent that p. By contrast, on a de re reading, what is available to you is something – a fact, or a certain kind of state – that also happens to count as

210

Thomas Höwing

The example I have just given is meant to suggest two things. First, it is meant to suggest quite generally that a ground – a certain fact or state – must be available to the agent if it is to make a corresponding assent rationally appropriate or even rationally necessary for her. In other words, the availability of a ground is a necessary condition for its having a normative grip on the agent. Second, the example makes explicit Kant’s claim that what I know holds “for all”. Suppose that a fact or state counts as a ground of knowledge with respect to an agent’s assent that p. Kant’s claim is not that this state or fact requires every agent to assent that p. Rather, his claim is that this state or fact requires every agent to whom it is available to assent that p. In other words, in the case of grounds of knowledge, availability is not only a necessary but also a sufficient condition for their having a normative grip on the agent. Of course, much more needs to be said about these claims. In particular, one would have to give a Kantian account of the kind of state or fact that might actually count as a ground of knowledge, and of what it is for that state or fact to be available to the agent. I am far from being able to provide such an account. But I suspect that any such account would have to involve the notion of a state that provides the agent with infallible access to truth, such as, in the aforementioned example, my state of seeing that there is a chair in the room. Suppose I merely seemingly see that there is a chair in my room, where seemingly seeing that p is compatible with both p and not-p. In such a case, we might very well conceive of another rational agent who is in the very same kind of state but who is not, or at least not by the state in question, required to assent that there is a chair in my room. The agent in question might simply know that her being in the relevant state is compatible with there being no chair in the room, or she might know that there is no chair in the room at all.¹⁹ Now be this as it may, it seems that Kant’s description of knowledge is ultimately based on a description of the grounds required for knowledge. This latter description might be summarized as follows. If a state or fact, g, counts as a ground of knowledge with respect to S’s assent that p, then g will satisfy the following two conditions:

a ground of knowledge with respect to my assent that p. Kant’s claim requires the de re reading. If you were to see that there is a chair in the room, what would be available to you would be a state or fact that also happens to count as a ground of knowledge with regard to my assent that there is a chair in the room. Note that in what follows I shall quite generally stick to the de re use of similar statements, such as ‘x’s ground of Belief that p is available to y’.  The view I am tentatively attributing to Kant here is similar to the disjunctivist view that John McDowell has defended with regard to perceptual experience; see, for example, McDowell , , as well as Matthiessen .

Kant on Opinion, Belief, and Knowledge

211

(i) g rationally requires S to assent that p, and (ii) necessarily, for any rational agent x, if g is available to x then g rationally requires x to assent that p. In other words, there is no possible scenario in which g is available to a rational agent x and g does not require x to assent that p. As for Belief, Kant claims that the latter qualifies as “assertoric judging”. As Kant puts it, “what I believe I hold to be assertoric, but not as objectively necessary, only as subjectively so (holding only for me)” (9:66). By associating Belief with assertoric judgment, Kant does not want to suggest that the latter is merely an ‘actual’ assent that is not formed on the basis of some ground.²⁰ As Kant indicates in the passage, in forming a Belief the agent is still responsive to a necessity, albeit to a merely subjective necessity. And once again Kant summarizes the upshot of this description in parentheses. An assent that qualifies as Belief will hold “only for me”. In other words, if I Believe that p, I am not entitled to assert p to others with the intention that they should assent to p as well. Taking the description of knowledge as our guide, we may likewise explain this feature of Belief in terms of an underlying ground. If I Believe that p, my assent will be based on a ground that rationally requires me to assent that p. However, this requirement, or this necessity, will be merely subjective. The ground in question will require me to assent that p, but it will not require everyone to whom it is available to assent that p. Now to see why this is so, it can be helpful to consider two examples that Kant gives for the attitude of Believing. In the first Critique, Kant makes the following remark about moral Belief: since my moral Belief “depends on subjective grounds (of moral disposition) I must not even say ‘It is morally certain that there is a God,’ […] but rather ‘I am morally certain’” (KrV A829/B857). Suppose I were to say to you that it is certain that God exists. In saying this, I would imply that I am in possession of a ground that potentially requires both of us – me and you – to assent that God exists. However, as Kant explains in the passage just quoted, I am not entitled to make such a claim, simply because my moral Belief depends on “subjective grounds (of moral disposition)”. In other words, the requirement that the relevant ground imposes on me is conditioned by my moral disposition; i.e. it is conditioned by my decision to act in accordance with the moral law. And  For a reading on these lines see Mattey . As Mattey suggests, an agent’s judgment that p is an assertoric judgment iff the agent actually affirms the truth of p (Mattey : –, ). Moreover, Mattey argues that a Belief is a mode of assent where one is subjectively caused to accept the truth of some judgment but at the same time realizes that one lacks sufficient theoretical justification (Mattey : ).

212

Thomas Höwing

since you might not have made this decision, I should say that I am morally certain that God exists.²¹ Now one might wonder whether this implies that the state or fact that constitutes my ground of Belief may not be available to you. Once again, a full answer to this question will depend on what kind of state or fact constitutes a ground of Belief. In light of what we have just seen, an obvious candidate would be an agent’s decision to act in a certain way. On this account, my ground for holding that God exists is simply my decision to act in accordance with the moral law. On the other hand, this account does not fit very well with the idea of a conditional requirement that Kant is invoking in the passages before us. The idea is not that there is an available ground – my decision to act in a certain way – that requires me to assent to some judgment. Rather, the idea is that there is an available ground – a state or fact – that requires me to assent to some judgment, conditional on my decision to act in a certain way.²² I think what Kant has in mind emerges in a second example from the first Critique. There, Kant presents the case of a doctor who is confronted with a patient in critical condition. As Kant puts it, the doctor “looks to the symptoms, and judges, because he does not know of anything better, that it is consumption” (KrV A824/B852). And Kant goes on to suggest that, given the doctor’s subsequent decision to treat the patient for consumption, he will be required to form the (pragmatic) Belief that the patient suffers from consumption. The passage is rather obscure, and Kant does not say what the doctor’s ground of Belief is. However, in light of what Kant does say, it is tempting to suppose that the doctor’s ground of Belief is simply constituted by those facts about the patient’s disease that are available to him. As Kant points out, these facts include facts about the patient’s symptoms, but presumably they also include

 Recent commentators have argued that moral Belief is valid for everyone (Wood : – ; Chignell a: ; Pasternack : ). Wood cites a passage in which Kant claims that moral Belief is subjectively sufficient “absolutely and for everyone” (KrV A/B; Wood : ). However, the context of the passage suggests that Kant’s claim may be read in a different way. According to Kant, the content of our moral Beliefs can be determined on a priori grounds. Even before I ask whether I should hold a moral Belief, I may know that this Belief will concern the existence of God, the immortality of the soul, and so on. As Kant puts it, “I […] know with complete certainty that no one else knows of any other conditions that lead to this same unity of ends under the moral law” (KrV A/B).  Note that, according to Kant, not just any decision will do. Rather, the decision has to qualify as a rational decision, and perhaps it also has to be responsive to a practical requirement, or to what Kant calls an imperative (the latter is clearly suggested by Kant’s account of moral Belief). However, since nothing hinges on the matter here, in what follows I shall simply speak of a ‘decision to act in a certain way’, leaving open how exactly the decision in question should be described.

Kant on Opinion, Belief, and Knowledge

213

more general facts about the kinds of diseases that are likely to generate such symptoms. Moreover, Kant suggests that, considered by themselves, these facts provide only limited epistemic support for the doctor’s estimation. In other words, taken by themselves, these facts would at best make a corresponding assent rationally appropriate for the doctor, but they would certainly not require such an assent. However, given the doctor’s decision to treat the patient for consumption, these facts will do exactly this – they will rationally require the doctor to assent that the patient has consumption. Admittedly, much more would be needed to spell out the reading of Kant’s account of Belief that I have just sketched.²³ However, the reading does seem to fit well with the idea of a conditional requirement that Kant associates with grounds of Belief. Moreover, the reading allows us to think of various scenarios in which the doctor’s ground of Belief is available to some other agent but fails to require the latter to form a corresponding assent. For instance, we might think of a second physician, who finds herself in the very same epistemic situation as the doctor but who does not decide to treat the patient for consumption. Or we might think of a third physician, who later investigates the case and who knows that the doctor’s patient did not suffer from consumption. Both physicians might have access to the very same facts that required the doctor to form a particular assent, but these facts won’t require them to form the assent in question. In sum, we might summarize Kant’s description of grounds of Belief as follows. If g counts as a ground of Belief with respect to an agent S’s assent that p, then g will satisfy the following conditions: (i) g rationally requires S to assent that p, and (ii) it is not the case that necessarily, for any rational agent x, if g is available to x then g rationally requires x to assent that p. As should be clear by now, Kant’s description of grounds of Belief reflects his overall account of non-epistemic justification. Putting things very roughly, ac-

 I discuss Kant’s conception of pragmatic Belief in more detail in Höwing, unpublished. One might also ask what kind of facts will constitute a moral agent’s ground for Believing that God exists, that the soul is immortal, and so on. I have to confess that I lack a full-blown answer to this question. As Kant argues, due to the limits of cognition, there is simply nothing available to us that would support the truth of these propositions. However, he also argues that, due to the limits of human cognition, there is nothing available to us that would support the falsity of these propositions. In other words, we have at least access to facts that completely rule out the possibility of counterevidence. And perhaps it is these facts that constitute the grounds of moral Belief (see Logik :.– for a passage that points in this direction).

214

Thomas Höwing

cording to Kant, non-epistemic justification is to be explained in terms of a requirement that arises from a combination of a lack of sufficient evidence and the decision to act in a certain way. On the one hand, the agent has access to certain facts that provide only insufficient evidence with regard to the truth of some proposition. On the other hand, she is required by these facts to assent to the proposition in question given that she has decided to act in a certain way. And this opens up a space for various scenarios in which these facts are available to some other agent but fail to impose a rational requirement on them. Finally, as for opinion, Kant explains: “[W]hat I merely opine I hold in judging, with consciousness, only to be problematic” (9:66). What Kant has in mind here, I think, emerges from the account of problematic judgment in the first Critique. There, Kant explains that “[t]he problematic proposition is […] that which expresses merely logical possibility […], i.e., a free choice to count such a proposition as valid” (KrV A75/B101). In light of this remark, it is tempting to read Kant’s description of opinion in a similar fashion. According to Kant, an agent’s opinion that p is based on a ground that leaves the agent a certain ‘freedom of choice’ – the ground in question does not rationally require the agent to assent that p.²⁴ To be sure, this is not to say that the ground in question will fail to make an assent rationally appropriate for the agent. In fact, in this latter regard an opinion seems to involve something more than a merely problematic judgment. As Kant points out in the first Critique, an opinion must be based on facts that are available to the agent in the mode of knowledge and that provide at least weak to moderate epistemic support for the proposition in question.²⁵ Thus, to say that an opinion counts as problematic judging is merely to say that the underlying ground fails to impose a rational requirement on the opining agent. To take a simple example, I may assent that it is going to rain in the next few hours on the ground that my somewhat unreliable weather app says so. In such a case, it seems that the ground in question makes it rationally appropriate for me to

 According to Mattey, judging problematically that p is neither accepting p as true nor accepting not-p as true. As Mattey suggests, this is because in making the judgment in question the agent acknowledges that she is under no rational constraint to take a stance (Mattey : ). Accordingly, Mattey defends the view that opinion is merely a “preliminary judging in which one does not accept the truth of the judgment” (Mattey : ). Mattey seems to be right in suggesting that an opinion is formed in the absence of an underlying rational requirement. However, as we have seen both in the first Critique and in the Logik, Kant makes clear that, just like the other two kinds of assent, opinion involves a “holding to be true”, i. e. accepting some proposition as true (see Logik :–; KrV A/B).  As Kant puts it, “I must never undertake to have an opinion without at least knowing something by means of which the in itself merely problematic judgment acquires a connection with truth” (KrV A/B).

Kant on Opinion, Belief, and Knowledge

215

assent that it is going to rain. But it certainly does not rationally require me to assent that it is going to rain. Interestingly, in the passage Kant remains silent as to whether a judgment that is held in the mode of opining is intersubjectively valid. To take up Kant’s wording, will what I opine hold for all, or will it hold only for me? In light of what we have seen so far, it is tempting to assume that the judgment in question will hold neither for me nor for all. This is because, ultimately, to say that a judgment is ‘holding’ (geltend) for an agent is to say that it is not left to the agent’s discretion to take the judgment on board. In other words, it is to say that there is something that rationally requires the agent to assent to the judgment in question. And as we have just seen, my opinion that p is based on a ground that does not rationally require me to assent that p – the ground in question leaves it to my discretion to take p on board. And therefore, in such a case p will not hold for me. Now it follows straightforwardly that p won’t be ‘holding for all’. However, while this is correct, it is important to note what it means. The notion of holding for all, as Kant uses it in the passage from the Logik, is a relativized notion. Suppose you know that it is going to rain in the next few hours. In such a case, it seems that, from your perspective, the judgment in question holds for all. That is, you are in possession of a ground that requires not only you but everyone to whom it is available to assent to the judgment in question. By contrast, suppose I opine that it is going to rain in the next few hours. In such a case, from my perspective, the judgment in question holds neither for me nor for all. That is, I am in possession of a ground that does not require me to assent to the judgment in question. And therefore, my ground doesn’t require everyone to whom it is available to assent to the judgment in question. In sum, we might say that, if g counts as a ground of opinion with respect to an agent S’s assent that p, then g will satisfy the following conditions: (i) g does not rationally require S to assent that p, and (ii) it is not the case that necessarily, for any rational agent x, if g is available to x then g rationally requires x to assent that p. Note that this description does not rule out a scenario in which g is available to some other rational agent T and in which g rationally requires T to assent that p. In such a case, g would obviously count as a ground of Belief with respect to T’s assent that p. In other words, g would require T to assent that p, but g would not require everyone to whom it is available (such as, for example, S) to assent that p. On the face of it, this might sound puzzling. But on closer inspection, the scenario in question fits very well with Kant’s conception of grounds of Belief. Consider once again Kant’s example of the doctor. As we have seen, the doctor’s

216

Thomas Höwing

Belief is based on facts that, taken by themselves, would at best make it rationally appropriate for him to assent that his patient has consumption. However, these facts would certainly not require him to form the assent in question. In other words, considered by themselves, these facts would at best license an opinion that the patient has consumption. However, we have also seen that, given his decision to treat the patient for consumption, the very same facts will rationally require the doctor to form the assent in question. In other words, in certain cases, a ground of opinion may convert into a ground of Belief – namely, where the agent has made a corresponding decision to act in a certain way.²⁶ Moreover, and perhaps more importantly, one might wonder why Kant’s description of grounds of opinion does not bring out an essential feature of these grounds, namely, that they make an assent rationally appropriate for the agent. Although I lack a full answer to this question, I think something like an explanation emerges from what we have seen in the first section. Kant’s description of grounds of opinion is simply not meant as a self-standing definition of these grounds. Rather, it is part of a more complex classification of different grounds. And presumably the ‘genus’ of this classification just is the notion of a ground that makes an assent rationally appropriate for the agent. In other words, the two conditions outlined above allow us to distinguish grounds of opinion from both grounds of knowledge and grounds of Belief. However, what we should take for granted in making this distinction is that all three grounds have one thing in common – they all make an assent rationally appropriate for the agent. This suggests that in the passage from the Logik, Kant is ultimately concerned with a classification of grounds of assent. In fact, the passage from the Logik yields a classification of grounds that is surprisingly similar to that drawn by Kant in the first Critique. The classification that emerges from the passage in the Logik proceeds

 As a matter of fact, Kant suggests that there is some such connection when, in the Logik, he presents the example of a businessman who asks himself whether he should strike a deal that has been offered to him. As Kant puts it, “the businessman, […] to strike a deal, needs not just to opine that there will be something to be gained thereby, but to believe it, i.e., to have his opinion be sufficient for an undertaking into the uncertain” (Logik : f. note). Kant’s remark is somewhat obscure, but it may perhaps be explained as follows. Suppose that the businessman has access to certain facts that make it likely that the deal will be a success. Taken by themselves, these facts at best justify an opinion that something will be gained from the deal. Yet in addition, given that the businessman has an interest in making a profit, these facts might also make it rational for him to actually strike the deal. In such a case, we may say, the businessman’s opinion will be ‘sufficient for an undertaking into the uncertain’. Accordingly, we may assume that the businessman goes for what is the rational thing to do and decides to make the deal. Now as Kant seems to suggest, in such a case the facts in question convert into a ground of Belief – they make it rationally necessary for the businessman to hold that something will be gained from the deal.

Kant on Opinion, Belief, and Knowledge

217

as follows. We begin by presuming that an agent’s ground of knowledge has two features: (i) it requires the agent to form a particular assent and (ii) it requires everyone to whom it is available to form the assent in question. We then ask which of these features also pertain to grounds of Belief and grounds of opinion. As it turns out, grounds of Belief have the first feature but lack the second, whereas grounds of opinion lack both features. Now recall the classification that Kant draws in terms of the SOIS-distinction. Here, we likewise begin by presuming that grounds of knowledge have two features: (i) subjective sufficiency and (ii) objective sufficiency. And we then ask which of these two features also pertain to grounds of Belief and grounds of opinion. As it turns out, grounds of Belief have the first feature but lack the second, and grounds of opinion lack both features.

3 A Solution to the Puzzle This suggests that we are on the right track. Thus, in the final section, I shall argue that the classification of grounds that emerges from the Logik can be fully mapped onto the classification that Kant draws in terms of the SOIS-distinction. Moreover, I shall suggest that spelling out the SOIS-distinction in this way allows us to solve the exegetical puzzle I presented in the first section of the paper. As we have just noted, Kant’s classification concerns the different ways in which a ground makes an assent rationally appropriate for an agent. Accordingly, to spell out Kant’s classification of grounds, we should begin by making the following assumption. There is some state or fact, g, and some agent, S, such that g makes an assent that p rationally appropriate for S. This assumption is meant to make explicit the domain that Kant’s classification is concerned with, namely, the domain of grounds that make an assent rationally appropriate for an agent. Now with this assumption in place, we may introduce a first distinction by saying that (R) is either true or false: (R) g rationally requires S to assent that p. In other words, either it is true that g rationally requires S to assent that p or it is not true. In the first case, g counts as subjectively sufficient with respect to S’s assent that p. In the second case, g counts as subjectively insufficient with respect to S’s assent that p. A second distinction concerns the question as to whether g requires every rational agent to whom it is available to assent that p. Accordingly, we may say that (UR) is either true or false:

218

Thomas Höwing

(UR) Necessarily, for any rational agent x, if g is available to x then g rationally requires x to assent that p. In other words, either it is true that g requires every rational agent to whom it is available to assent that p or it is not true. In the first case, g counts as objectively sufficient with regard to an assent that p. And in the second case, g counts as objectively insufficient with regard to an assent that p. Finally, let us consider the possible ways in which these two distinctions may be combined: 1. In the first case, g qualifies as subjectively and objectively sufficient with respect to S’s assent that p. In other words, g rationally requires S to assent that p, and g requires every rational agent to whom it is available to assent that p. As we have seen, in such a case, g counts as a ground of knowledge with respect to S’s assent that p. 2. In the second case, g qualifies as subjectively sufficient and objectively insufficient with respect to S’s assent that p. In other words, g rationally requires S to assent that p, but g does not require every rational agent to whom it is available to assent that p. As we have seen, in this case, g counts as a ground of Belief with respect to S’s assent that p. 3. In the third case, g qualifies as subjectively and objectively insufficient with respect to S’s assent that p. In other words, g does not rationally require S to assent that p, and g does not require every rational agent to whom it is available to assent that p. Recall, however, that we are assuming that g makes it rationally appropriate for S to assent that p. If this is taken into account, we may say that, in the third case, g counts as a ground of opinion with respect to S’s assent that p. Note that, within the domain we are considering, there cannot be a fourth case. In such a case, g would qualify as subjectively insufficient and objectively sufficient with regard to S’s assent that p. To see why this is conceptually impossible, consider the following line of thought. To begin with, recall our initial assumption that g makes it rationally appropriate for S to assent that p. Furthermore, in the previous section we saw that the availability of a ground is a necessary condition for its making an assent rationally appropriate for the agent. In other words, if g makes an assent that p rationally appropriate for S, then g will be available to S. Along with our initial assumption, it follows that g is available to S. Now consider the alleged fourth case. In this case, g counts as objectively sufficient with regard to an assent that p. In other words, g requires just any agent to whom it is available to assent that p. Since g is available to S, it follows that g requires S to assent that p. And this is just to say that g is subjectively sufficient with regard to S’s assent that p.

Kant on Opinion, Belief, and Knowledge

219

This, I take it, shows that the classification from the Logik may be consistently mapped onto the classification that Kant draws in terms of the SOIS-distinction. Now before we turn to the puzzle, we might define the two main concepts of Kant’s SOIS-distinction. Consider the following definitions: (OS) A state or fact g counts as an objectively sufficient ground with regard to an assent that p iff necessarily, for any rational agent x, if g is available to x then g rationally requires x to assent that p. (SS) A state or fact g counts as a subjectively sufficient ground with regard to a particular agent S’s assent that p iff g rationally requires S to assent that p. It might seem that the notions of subjective and objective insufficiency may be defined merely by negating the respective definiens in (OS) and (SS). However, while this is true with regard to objective insufficiency, there are two ways to define subjective insufficiency: (SIS1) A state or fact g counts as a subjectively insufficient ground with regard to S’s assent that p iff g does not rationally require S to assent that p. (SIS2) A state or fact g counts as a subjectively insufficient ground with regard to S’s assent that p iff (i) g makes it rationally appropriate for S to assent that p and (ii) g does not rationally require S to assent that p. The definiens that appears in (SIS1) simply denies the definiens that appears in (SS). By contrast, the definiens in (SIS2) adds the condition that g makes it rationally appropriate for S to assent that p. In presenting the classification, I have implicitly worked with (SIS1) since this fits better with Kant’s negative description of opining in terms of a merely problematic judgment. On the other hand, (SIS1) has crucial philosophical disadvantages. For one thing, (SIS1) allows us to speak of a subjectively insufficient ground in cases where the corresponding facts don’t lend any rational support to an assent that p, or, what seems to be worse, in cases where these facts rationally support an assent that not-p. What is more, (SIS1) allows for various combinations of the SOIS-distinction outside the domain of Kant’s classification. For instance, it makes room for a scenario in which g counts as subjectively insufficient1 and objectively insufficient with regard to S’s assent, but in which g fails to make the assent rationally appropriate for S. In other words, in such a scenario g would not qualify as a ground of opinion for S. In light of these considerations, I have a certain preference for attributing (SIS2) to Kant. But I simply don’t see clear textual evidence to support such a reading. Be this as it may, all this does not seem to be much of a problem

220

Thomas Höwing

when it comes to our understanding of Kant’s classification itself. As we have seen, this classification is concerned with the domain of grounds that make an assent rationally appropriate anyway. In other words, nothing hinges on the matter here, and thus I shall simply leave it to the reader to choose between (SIS1) and (SIS2).²⁷ With this reading of Kant’s classification now in place, let us turn to the second question of whether the reading solves the puzzle presented in the first section of the paper. In short, the puzzle was this. Kant describes grounds of Belief in terms of two features, namely, a positive feature shared with grounds of knowledge (subjective sufficiency), and a negative feature shared with grounds of opinion (objective insufficiency). And this seems to run counter to Kant’s claim that the grounds of Belief have at least one unique feature – they count as non-epistemic grounds. It should be clear by now that, if my reading is correct, there really is no such puzzle. This is because, according to the reading, Kant’s description of grounds of Belief does seem to reflect his account of non-epistemic justification. To begin with, we have seen that an agent’s ground of Belief has two features: (i) it requires the agent to form a particular assent (subjective sufficiency), and (ii) it does not require every agent to whom it is available to form the assent in question (objective insufficiency). Moreover, we have also seen that the first feature would also pertain to a corresponding ground of knowledge, whereas the second feature would also pertain to a corresponding ground of opinion. However, there doesn’t seem to be anything puzzling about this description. This is because it is precisely the combination of the two features that reflects Kant’s account of non-epistemic justification. As I have suggested, Kant holds the view that non-epistemic justification is to be explained in terms of a rational requirement that arises from a lack of strong epistemic support and the agent’s decision to act in a certain way. On the one hand, the agent has access to certain facts that, considered by themselves, exert only weak to moderate normative force on the agent to form a corresponding assent. On the other hand, once the agent decides to act in a certain way, the normative force of these facts increases dramatically – the facts in question rationally require the agent to form the assent in question. This explains why Kant describes an agent’s ground of Belief in terms of features that would also pertain to a corresponding ground of knowledge and a corresponding ground of opinion. Just like a corresponding ground of knowledge, a

 To be sure, I am not claiming that Kant commits an ambiguity by using two different notions of subjective insufficiency. Rather, I simply want to leave open the question which of the two notions Kant uses.

Kant on Opinion, Belief, and Knowledge

221

ground of Belief requires the agent to form a particular assent. However, just like a corresponding ground of opinion, a ground of Belief does not require everyone to whom it is available to form the assent in question. For if we take away the agent’s decision to act in a certain way, what is left are facts that provide only insufficient epistemic support to the assent in question. And this opens up space for possible scenarios in which these facts are available to some other agent but fail to impose a rational requirement on them.²⁸

References Baumgarten, Alexander G. 1973, Acroasis logica in Christianum L. B. de Wolff (1725), in: Christian Wolff, Gesammelte Werke, Abt. 3, Bd. 5, Hildesheim/New York: Olms Chignell, Andrew 2007a, “Kant’s Concepts of Justification”, Nous 41, 33–63 Chignell, Andrew 2007b, “Belief in Kant”, Philosophical Review 116, 323–60 Höwing, Thomas (unpublished), “Kant on Justification, Rational Choice, and Pragmatic Belief” Mattey, G.J. 1986, “Kant’s Theory of Propositional Attitudes”, Kant-Studien 77, 423–40 Matthiessen, Hannes Ole 2014, “Disjunktivismus. Die Auflösung des Dualismus von Anschauung und Welt”, in: Christian Barth/David Lauer (eds.) 2014, Die Philosophie John McDowells. Ein Handbuch, Münster: mentis, 51–66 McDowell, John 1995, “Knowledge and the Internal”, Philosophy and Phenomenological Research 55, 877–93 McDowell, John 1998, “Criteria, Defeasibility, and Knowledge”, in: John McDowell 1998, Meaning, Knowledge, and Reality, Cambridge/London: Harvard University Press, 369–94 Meier, Georg Friedrich 1997, Vernunftlehre, ed. Günter Schenk, 3 vols., Halle/Saale: Hallescher Verlag Motta, Giuseppe 2012, Die Postulate des empirischen Denkens überhaupt. KrV A 218–235/B 265–287. Ein kritischer Kommentar, Berlin/Boston: de Gruyter Pasternack, Lawrence 2011a, “Kant’s Doctrinal Belief in God”, in: Oliver Thorndike (ed.) 2011, Rethinking Kant, Vol. III, Newcastle: Cambridge Scholars Publishing, 200–18 Pasternack, Lawrence 2011b, “The Development and Scope of Kantian Belief: The Highest Good, The Practical Postulates and The Fact of Reason”, Kant-Studien 102, 290–315 Pasternack, Lawrence 2014, “Kant on Opinion: Assent, Hypothesis, and the Norms of General Applied Logic”, Kant-Studien 105, 41–82 Scanlon, T.M. 2014, Being Realistic About Reasons, Oxford: Oxford University Press Stevenson, Leslie 2003, “Opinion, Belief or Faith, and Knowledge”, Kantian Review 7, 72–101 Theis, Robert 2010, “Du Savoir, de la foi et de l’opinion de Wolff à Kant”, Archives de Philosophie 73, 211–28

 Work on this paper was supported by the Deutsche Forschungsgemeinschaft (DFG). I thank Kimberly Brewer, John Callanan, Andrew Chignell, Gabriele Gava, Johannes Höwing, Anselm Spindler and Eric Watkins for many helpful comments on earlier drafts of the paper. Marcus Willaschek provided both oral and written comments that helped me to improve the paper at various stages.

222

Thomas Höwing

Wolff, Christian 1978, Vernünfftige Gedanken von den Kräfften des menschlichen Verstandes und ihrem richtigen Gebrauche in Erkenntnis der Wahrheit [German Logic], Hildesheim: Olms Wolff, Christian 1983, Philosophia Rationalis sive Logica [Latin Logic], Hildesheim: Olms Wood, Allen W. 1970, Kant’s Moral Religion, New York: Cornell University Press

Marcus Willaschek

Must We Believe in the Realizability of Our Ends? On a Premise of Kant’s Argument for the Postulates of Pure Practical Reason Can it be rational to pursue an end the realization of which one takes to be impossible? It seems not. If I am convinced that I won’t be able to reach the station in time for my train, it would be irrational for me to even try. It seems to be a condition of rational agency that one believes, and is rationally justified in believing, that the end one pursues can be realized and, more specifically, that it can be realized by one’s own actions. Call this the realizablility principle, or RP. The question whether there are belief conditions on rational agency, and what exactly they consist in, concerns an important aspect of Kant’s doctrine of the “primacy of pure practical reason”, according to which one can be rationally warranted in holding a belief whose truth is not backed by sufficient evidence. According to Kant, beliefs of this kind, which he calls “postulates of pure practical reason” (cf. KpV 5:122), must satisfy two conditions, which we can call “theoretical undecidability” and “practical necessity”.¹ A belief is theoretically undecidable if even under epistemically ideal conditions neither its truth nor its falsity can be established in a way that would be sufficient for knowledge or “cognition”.² For Kant, since we can have cognition only of the empirical realm, all beliefs about non-empirical objects fall into this category. The practical necessity of a belief, by contrast, consists in its relation to a necessary demand of practical reason. Precisely what relation is relevant here is a complicated question, but let us say for the moment that it is one of rational consequence or co-commitment: a belief B is practically necessary for subject S if (i) there is some unconditionally binding norm N that holds for S and if (ii) someone who accepts N as unconditionally binding is rationally committed to hold belief B. According to Kant, this is the relation between the moral law and the belief in the truth of the postulates of pure practical reason, i.e. belief in God, freedom and immortality. Kant’s thesis of the primacy of pure practical reason can be stated as follows: we are rationally warranted to hold beliefs that are morally necessary,  Cf. Willaschek .  This does not exclude the possibility of empirical evidence or a priori arguments for or against the belief in question. What exactly theoretical undecidability amounts to depends on what is required for knowledge or cognition. On Kant’s concept of “cognition” (Erkenntnis) and its relation to the concept of knowledge, cf. Watkins/Willaschek (unpublished).

224

Marcus Willaschek

even if they are theoretically undecidable. While Kant’s argument for the primacy of pure practical reason is highly complex,³ the main idea on which it is built is quite simple: assume that there is a belief that is both practically necessary and theoretically undecidable. That it is practically necessary means that we are rationally required to hold it.⁴ But then, the fact that the belief is theoretically undecidable cannot mean that we are rationally required not to hold it. Otherwise, we would be rationally required to believe something (on moral grounds) that we were rationally required not to believe (on epistemic grounds), which would violate reason’s constitutive norm of consistency. Hence, we are rationally warranted, and indeed even rationally required, to accept as true all beliefs that are practically necessary, even if these beliefs are theoretically undecidable and hence epistemically unjustified. Of course, whether there are any beliefs that are practically necessary (and, if so, what these might be) is a further question. Kant’s arguments for the specific postulates of God and immortality do not seem convincing, or even promising.⁵ But Kant’s argument for the primacy of pure practical reason is interesting and important because, if successful, it is not restricted to Kant’s postulates, since it shows for any belief that is both practically necessary and theoretically undecidable that we are rationally required to accept it. Since it seems indisputable that there are many potential beliefs that are theoretically undecidable, the question is whether some of them are practically necessary. This is far from obvious. If we think of practical propositions as those aiming at what “ought to be the case”, and of beliefs as theoretical attitudes aiming at “what is the case”, it may be difficult to see how accepting a practical proposition as normatively binding could commit one to holding a belief. The way in which Kant establishes the practical necessity of certain beliefs presupposes some principle, such as the realizability principle mentioned above, which

 For reconstructions of the overall argument, cf. Gardner ; Willaschek .  Note that Kant holds that we are rationally required, but not morally obligated, to accept the postulates (cf. section  below).  Kant’s arguments for the postulates of God and immortality rest on two assumptions that seem highly problematic: () that moral perfection (Heiligkeit, holiness) is impossible for human beings (so that we can only approximate it in an infinite afterlife) and () that the highest good requires the distribution of happiness in proportion to people’s different degrees of virtue (which only God can accomplish). But it is unclear () why moral perfection for human beings should consist in holiness (instead of humanly possible virtue) and () why the highest good should not, as Kant himself sometimes seems to think, consist in general and perfect virtue in combination with general and perfect happiness (cf. Willaschek ).

Must We Believe in the Realizability of Our Ends?

225

links the rational pursuit of an end to the belief in its realizability.⁶ Because the moral law requires us to make the highest good our end, and because we cannot make it our end without believing (in a general, not specifically Kantian sense of “believe”) in its realizability, there is at least one belief to which we are rationally committed simply by being bound by the moral law – namely, the belief in the realizability of the highest good (which belief is thus practically necessary). If it could be shown that this belief rationally presupposes other beliefs (such as belief in God and immortality), the rational necessity of our commitment to the moral law would be transmitted to these other beliefs in virtue of their being rational presuppositions of believing in the realizability of the highest good.⁷ Since making the highest good our end is not optional, belief in the necessary conditions of its realizability is not optional either.⁸ Thus, Kant’s argument for the postulates hinges on (some version of) the realizability principle.⁹ The aim of this paper is to discuss the validity of this principle in the context of Kant’s argument.¹⁰ In the next section (2), I will suggest a reconstruction of Kant’s argument for the postulates in the Critique of Practical Reason, which will allow us to see in more detail how that argument depends on some version of the realizability principle. I will then raise three objections to the realizability principle and the use Kant makes of it in his argument for the postulates. Each of these objections challenges the idea that we really have to believe our ends to be realizable in order to be rational in pursuing them. Instead, the objections suggest that three less demanding alternatives may be sufficient: first, using the end, even though it cannot be fully realized, as a regulative idea (3); second, acting as if the end could be realized (4); and third, not believing that the end cannot be re-

 There may be other relations between our practical and our theoretical commitments that could make certain beliefs practically necessary (the postulate of freedom seems to be a relevant example), but I will not pursue this possibility here.  Besides Kant’s postulates, there are other prima facie plausible examples of practically necessary beliefs, ranging from belief in other minds and the external word to belief in the possibility of world peace and of overcoming poverty. (As we will see, world peace is an example Kant himself discusses.)  Paul Guyer’s interpretation of Kant’s doctrine of the postulates rests on the idea that it “is stated within the limits of human psychology” (Guyer : ). According to Guyer, the “point of the postulates is to give us psychological assistance in the performance of morally requisite action” (Guyer : ). By contrast, I think that the point of the postulates is to make explicit a requirement of rational agency that results from the realizability principle.  That Kant’s argument rests on some version of the realizability principle has been noted by Allen Wood (cf. Wood :  ff.; Wood :  ff.).  By doing so, I hope to resolve some of the problems I have raised, but not solved in a satisfactory way, in an earlier paper (Willaschek : –).

226

Marcus Willaschek

alized (5). While answering the final objection will require a modification and weakening of the realizability principle, I will argue that there is a version of this principle that remains unaffected by the objections and is still sufficient as a basis for Kant’s argument for the postulates of practical reason (6).

1 Kant’s Argument for the Postulates in the Second Critique and the Realizability Principle If we step back a little from Kant’s own presentation of the argument for the postulates of God and immortality in the second Critique, we can present it in the following form: (1) The highest good (“of a possible world”) consists of happiness in proportion to virtue (for each individual in that world) (cf. KpV 5:110–11). (2) We are morally obligated (and thus rationally required) to make the highest good our end (cf. KpV 5:113, 114). (3) We can be rational in making the highest good our end only if we rationally believe that the highest good is practically possible (i. e. that we can realize the highest good through our own actions). (4) The highest good is practically possible only if either virtue necessarily causes proportionate happiness or happiness necessarily causes proportionate virtue (cf. KpV 5:113). (5) Compelling arguments seem to show that virtue does not necessarily cause proportionate happiness and that happiness (and its pursuit) does not necessarily cause proportionate virtue (cf. KpV 5:113). (I take this to be the heart of the “antinomy of practical reason”.)¹¹ (6) In order for it to be rational for us to believe that we can realize the highest good, we require an account of how, in spite of (5) (i.e. in spite of the antinomy), the highest good is practically possible (cf. KpV 5:112, 115, 145). (7) We can give an account of how virtue can be the (indirect) cause of proportionate happiness, and thus how it is possible for us to realize the highest good, only if we believe in God (cf. KpV 5:115, 124–5) (and immortality; cf. KpV 5:122).¹² (8) We are rationally required to believe in God (and immortality).  For a reconstruction of the antinomy of practical reason, cf. Watkins .  One may wonder whether it is not sufficient to believe that God and immortality are possible (although perhaps not actual) in order to secure the possibility of the highest good. I will address this issue, which is important in itself but not central to the concern of this paper, in the Appendix.

Must We Believe in the Realizability of Our Ends?

227

According to this reconstruction, Kant’s argument relies on a version of the realizability principle, which appears here in premise (3).¹³ However, Kant never seems to make this premise fully explicit.¹⁴ What he does say is that if it is morally necessary to make the highest good our end, then the highest good must be practically possible (that is, realizable for us): “If […] the highest good is impossible according to practical rules, then the moral law which commands that it be furthered must be fantastic, directed to empty imaginary ends, and consequently inherently false” (KpV 5:114; cf. 5:115). This does not necessarily mean that we have to believe the highest good to be practically possible in order to make it our end. It only means that we cannot be morally required to make it our end unless it is practically possible. Hence, one may wonder whether there is an alternative reconstruction that is closer to Kant’s text and that replaces premise (3) with (3*) If we are morally obligated to make the highest good our end, it must be possible for us to realize the highest good.¹⁵ The problem with this intuitively plausible principle (an application of ought implies can) is that (given Kant’s overall argumentative strategy) it doesn’t take us to the postulates. According to Kant, God and immortality are not objectively necessary conditions for the practical possibility of the highest good (cf. KpV 5:145). Rather, they are only “subjective conditions of reason” (KpV 5:145), that is, conditions without which human reason cannot conceive of the practical possibility of the highest good.¹⁶ “Objectively speaking”, it may be possible for the highest  Instead of what I have called the realizability principle, Paul Guyer sees two different principles at work here: “First, a canon of rational willing according to which it is only rational to adopt a maxim that defines a certain end or goal and to act according to that maxim if we have reason to believe that the realization of that end is possible; second, an assumption about possibility itself, namely that the real possibility of any object requires not just that the concept of it be free of internal contradiction but also that the possible object have some sort of ground in something that actually exists” (Guyer : –). While Kant’s distinction between “real possibility” and merely “logical possibility” may stand in the background of Kant’s argument for the postulates, what matters for Kant, at least in the version in the second Critique, is not the “real possibility” but the “practical possibility” of the highest good. Kant discusses “real possibility” only with respect to the content of the postulates, that is, to the ideas of God, freedom, and immortality.  Kant comes closest to accepting () in a footnote in his Logik (:n.).  Paul Guyer seems to me to go back and forth between () and (*) without marking the difference between the two principles (cf. Guyer : –).  While Kant makes this point explicit only later in the Dialectic of the second Critique (in a section entitled “Of Taking Something to be True from a Need of Pure Reason”, KpV : ff.), his wording in the antinomy section carefully prepares for it. There, he says that a necessary connection between virtue and happiness is impossible insofar as it “is not to be expected”, to which formulation he

228

Marcus Willaschek

good to be realized without divine assistance and an immortal soul.¹⁷ Thus, Kant’s argument cannot take the following form: we are morally obligated to make the highest good our end; the highest good is practically possible only if God exists; therefore God exists – a form in which the argument would rely on (3*). Rather, Kant must argue as follows: we are morally obligated to make the highest good our end; we can rationally believe the highest good to be practically possible only if we believe that God exists; therefore, it is rationally necessary for us to believe that God exists. For this argument to be valid, what is needed is a principle that links the moral necessity of making the highest good one’s end to the belief that the highest good is practically possible. (3) is such a principle, whereas (3*) is not. Hence, it seems that Kant’s argument depends on the principle that we cannot make something our end unless we believe that it is “possible”, or, as Kant repeatedly says, “practically possible” – that is, realizable through our own actions. In what follows, I will focus on the question whether this realizability principle is valid and, if so, how exactly it must be formulated if it is to be (a) plausible and (b) sufficiently strong as a premise in Kant’s argument for the postulates. Let’s start with the following principle: RP1 An agent A is rational in pursuing an end E by doing D only if A (rationally) believes that E is practically possible (i. e. that it is possible for A to realize E by doing D). To be sure, if something like RP1 is a valid principle of rational agency at all, it needs further refinement in order to be valid without exception. For instance, there may be cases where E cannot be realized by A alone – such that it is in some sense not possible for A to realize E by doing D – but where E can be realized by doing D provided that other people contribute to realizing E. Kant seems to acknowledge this point. While he does say that we are morally obligat-

refers back at KpV : (“Above I had said […] not to be expected […]”), now adding that such a connection is not objectively impossible, but only subjectively inconceivable for beings like us. One reason for this move seems to be that Kant wants to avoid the conclusion that belief in God and immortality is morally obligatory (KpV :; cf. Willaschek : –).  “In fact, the impossibility referred to is merely subjective, that is, our reason finds it impossible for it to conceive, in the mere course of nature, a connection so exactly proportioned and so thoroughly purposive between events occurring in the world in accordance with such different laws, although, as with everything else in nature that is purposive, it nevertheless cannot prove – that is, set forth sufficiently on objective grounds – the impossibility of it in accordance with universal laws of nature” (KpV :).

Must We Believe in the Realizability of Our Ends?

229

ed to realize the highest good, he also says that it is our duty to promote (befördern) the highest good (e. g. KpV 5:114), which latter phrase does not imply that any one human being on their own could realize the highest good. A related point concerns what it means, exactly, to say that someone realizes some end E by doing D. Obviously, for this to be the case, doing D must in some sense causally contribute to the realization of E. But in precisely which sense? Consider three ways of realizing the end of cooking spaghetti: first, there is the normal way of putting a pot on the stove, bringing water to boil, and then cooking the spaghetti. Second, there is the lucky way of putting a pot with water and spaghetti out in the yard in clear weather where, out of the blue, lightning happens to strike the pot and cooks the spaghetti. Finally, there is the divine way of simply turning uncooked spaghetti into cooked spaghetti – or into spaghetti phenomenally indistinguishable from cooked spaghetti – by an act of one’s will. Clearly, the divine way of doing things is not open to human beings, which means that our realizing some end always depends on causal background conditions and on the working of causal connections outside our control. (In what follows, I will ignore the special class of cases where the end is identical with the action by which it is realized.) This means that our actions can only be contributing causes of the realization of our ends and are not in themselves sufficient to bring about these ends. On the other hand, if our success in bringing about the end depends on causal factors that are highly unlikely, such as lightning striking a pot with spaghetti, then even where these factors happen to obtain, this would not count as realizing the end in question by what one does. Rather, there must be some reliable causal relation between D and E, which means that the obtaining of the additional causal conditions besides D must not be too unlikely. (How likely they must be may depend on the availability of alternative ways of realizing E, on the importance of realizing E, and on other factors.) Note that the additional causal conditions can include other people’s actions, so that we can allow that some agent A realizes some end by doing D, even though the realizing of the end requires the collaboration of other people.¹⁸ These brief reflections suggest the following gloss on what it means for an end to be practically possible in the sense required for RP1:

 In “On the Old Saw”, Kant speaks of the “highest good that is possible through our collaboration” (“höchstes auch durch unsere Mitwirkung mögliches Gut in der Welt”) (Gemeinspruch :n.).

230

Marcus Willaschek

PP It is practically possible for an agent A to realize some end E by doing D only if (i) A’s doing D causally contributes to realizing E and (ii) the obtaining of a set of causal conditions that are jointly sufficient to realize E (of which A’s doing D is a part) is not unlikely.¹⁹ Inserting PP into RP1, we get: RP2 An agent is rational in pursuing an end E by doing D only if A (rationally) believes that (i) A’s doing D causally contributes to realizing E and (ii) the obtaining of a set of causal conditions that are jointly sufficient to realize E (of which A’s doing D is a part) is not unlikely. But even against RP2, various objections can be raised. I will consider three objections, all of which can be based on things Kant himself explicitly acknowledges. First, one might object that it is not necessary to believe that E is practically possible if we consider E an ideal, or a regulative idea. Second, we may not be required to have any doxastic attitude towards the practical possibility of E as long as we can act as if E were possible. And third, it may suffice not to believe that E is practically impossible (instead of believing that it is practically possible).

2 The Highest Good as a “Regulative” Idea In various places in his published and unpublished writings, Kant calls our concept of the highest good an “idea”, a “practical idea” or an “idea of practical reason”.²⁰ An “idea”, according to Kant’s official definition, “is a concept from notions [viz. pure concepts of the understanding] that transcends the possibility of experience” (KrV A320/B377). Alternatively, Kant says that an idea “is a necessary concept of reason to which no congruent object can be given in the senses” (KrV A327/B383). The reason why there cannot be an empirical object corresponding to an idea is that ideas contain the synthesis of an absolute totality, which in turn is something “unconditioned”. Experience, however, presents us only with conditioned objects and not with absolute totalities. It seems that the concept of the highest good is a clear case of an idea, since (a) it is the concept of a world (which is the concept of a totality of objects, in this case the totality of all moral  This formulation is meant to leave open the difficult questions connected with causal overdetermination by not requiring that A’s doing D be causally necessary.  E. g. KrV A/B, A/B, A/B; KpV :; Theodizee :; Gemeinspruch :n.; Logik :n.; Fortschritte :; :.

Must We Believe in the Realizability of Our Ends?

231

agents) and (b) because both individual virtue and individual happiness as the defining features of this world are totalities, namely the totality of a person’s actions (all of which are done from the motive of duty) and the totality of a person’s desires or volitions (all of which are satisfied). Kant himself repeatedly emphasizes the element of completeness in both concepts and calls both happiness and virtue “ideas”.²¹ For all these reasons, the highest good cannot be an empirical object. But this means that the highest good cannot be realized in the empirical world, which seems to be the only world in which we can act in order to realize the former. And this seems to imply that we simply cannot realize the highest good by anything we do, which would mean that the highest good is not practically possible for us. However – and here comes the objection to the realizability principle – this does not show that we cannot make the highest good our end. Kant’s account of the “regulative use of ideas” can serve as our model here. According to what Kant says in the “Appendix to the Transcendental Dialectic”, even though the transcendental ideas of pure (speculative) reason cannot be of “constitutive use” (that is, they cannot be used to cognize objects), they can be of regulative use, namely “to direct the understanding toward a certain goal” (KrV A644/ B672). In particular, the ideas of homogeneity (each pair of species falls under a common genus), specification (there are no lowest species), and continuity (no two species are closest) (cf. KrV A657 ff./B685 ff.) can serve to guide empirical research toward the goal of systematic unity of empirical knowledge, even though these ideas must not be used constitutively (that is, as principles that govern empirical reality), and even if the goal of complete systematicity, which itself is only an idea of reason (cf. KrV A645/B673), can never be fully realized. The reason why it still makes sense to search for higher genera and lower and closer species is that, in doing so, we asymptotically approach a completely systematic picture of nature, even though it is not possible for us ever to reach that end completely. We can, however, at least approximate it. We can now transfer the distinction between the regulative and the constitutive use of ideas from the transcendental ideas of reason to the idea of the highest good. Just as the three ideas of homogeneity, specification, and continuity cannot be used to cognize given objects, but only to direct our research toward the goal of systematic knowledge, the idea of a necessary connection between virtue and happiness is not used to cognize empirical objects, but rather to direct us toward the ultimate end of practical reason, that is, the highest good. Al-

 For “virtue”, cf. e. g. KrV A/B; KpV :; RGV :n.; for “happiness”, cf. GMS : and KU :.

232

Marcus Willaschek

though, as empirical beings, we can never fully realize the highest good, this does not mean that we cannot rationally pursue its realization as an end. More generally, the fact that some state of affairs (such as virtue, happiness, systematic unity of cognition) can never be fully realized does not mean that we cannot rationally make it our end. It only means that we must be aware that all empirical steps toward this end will at best be part of an asymptotic approximation that will never reach its endpoint.²² If such an attitude is not generally irrational, as Kant himself seems to allow, something must be wrong with the realizability principle and with Kant’s argument for the postulates. In order to evaluate this objection, we will have to distinguish between how it affects the realizability principle (a) and how it affects Kant’s argument for the postulates (b). (a) The possibility of a regulative use of ideas as such does not undermine the realizability principle. On a general level, the realizability principle can be defended by insisting that, in the case of the asymptotic approximation of some state of affairs that can never be fully realized, our end is not the state of affairs we approximate, but rather to approximate this state as far as possible. If I know that I cannot produce an absolutely plane surface, it would be irrational to try to produce one. This is just what the realizability principle implies. But it would not be irrational to try to make a surface that is as plane as possible, even though one is aware that a perfectly plane surface is physically impossible. The idea of a perfectly plane surface serves here not as an end, but rather as an ideal (in the ordinary, non-Kantian sense of that term) that gives content to the otherwise empty conception of a plane physical surface: a physical surface is plane to the degree that it approximates the ideal of a perfectly plane surface. It is the physically possible approximation that can serve as an end of human activities, not the ideal that determines its content. Something similar can be said about the idea of a complete system of empirical knowledge. If it should be the case that this idea cannot be empirically realized, this means that the end of human reason in its theoretical employment, the final end of theoretical inquiry, is not the complete system of knowledge but its best possible approximation. But then, the general strategy Kant employs with respect to the regulative use of reason in order to give the ideas of reason some positive function is compatible with the realizability principle.²³

 Cf. Kant’s argument for the postulate of immortality.  This is not an argument against the possibility of rationally pursuing an idea as one’s end even if the idea cannot be realized. Rather, I have presupposed with RP that this is impossible. My point is only that putative cases of pursuing ideas as ends can be plausibly re-described as

Must We Believe in the Realizability of Our Ends?

233

(b) This response, however, does not help with Kant’s argument for the postulates. We cannot argue that even though the highest good cannot be realized, this is compatible with the realizability principle, because what we make our end is not the realization of the highest good, but only its best possible approximation. Obviously, this would undermine Kant’s argument for the postulates, since we don’t need God and immortality in order to approximate the highest good, but only to realize it fully.²⁴ Hence, Kant is committed both to the claim that we must be able to realize the highest good fully and to the claim that the concept of the highest good is an idea that cannot be fully realized empirically. But nor is the highest good something that can be fully realized in an intelligible world alone, since human beings depend for their happiness on many empirical conditions. Granted that the regulative use of ideas is compatible with the realizability principle as a premise of his argument for the postulates, Kant’s insistence that our concept of the highest good is an idea seems to conflict with a different premise of that argument (namely, that the highest good must not only be approximated, but fully realized). This apparent problem disappears, however, if we think of the empirical/intelligible distinction not as a distinction between ontological realms, but as a distinction between aspects under which we consider human agency. The highest good as the totality of the object of pure practical reason is not an empirical object; moreover, it contains an element, namely the moral quality of people’s character, that is not empirically accessible but requires that we regard people as rational and autonomous beings acting from principles. But since these rational beings are the same beings whose happiness depends on a multitude of empirical factors, the highest good is an object that combines both intelligible and empirical aspects.²⁵ That the concept of the highest good, like any idea, cannot be fully realized empirically does not mean that it can be fully realized only in an intelligible word (e.g. an

cases in which the end consists in the best possible approximation of the idea. Hence, they are not counterexamples to the realizability principle.  I assume here that Kant, in the context of his argument for immortality, identifies the infinitesimal approximation of virtue in an infinite series of morally good acts with the full realization of virtue.  That this is how Kant thinks of the highest good is confirmed, for instance, by the following passage from the resolution of the antinomy of practical reason: “But since I am not only warranted in thinking of my existence also as a noumenon in a world of the understanding but even have in the moral law a purely intellectual determining ground of my causality (in the sensible world), it is not possible that morality of disposition should have a connection […] as cause with happiness as effect in the sensible world” (KpV :–; my emphasis). While the morality of my character depends on an intelligible determining ground (the moral law) and hence places me in an intelligible world, its effect, happiness, belongs in the sensible world (also cf. RGV :n.).

234

Marcus Willaschek

“afterlife”), but rather that its realization engages both the empirical and the intelligible sides of our existence. None of this implies that the highest good cannot, in principle, be fully realized. So the fact that Kant repeatedly calls the highest good an idea does not require him to deny its full realizability, or to claim that it can be realized only in an afterlife. It only means that its full realization is not an empirical object (because it includes our moral disposition, which belongs to the nonempirical aspects of our existence as both sensible and rational beings). Therefore, claiming that we have to make this idea our end does not bring Kant into conflict with the realizability principle. In sum, while the possibility of approximating an idea that can never be fully realized is compatible with the realizability principle (a), Kant’s claim that our concept of the highest good is an idea commits him to saying not that it cannot be fully realized, but only that its full realization is not a possible object of empirical cognition (b).

3 “Acting as If” Let us next turn to a seemingly more radical objection to the realizability principle. Its source, again, is Kant himself. In an intriguing but neglected passage in his “Conclusion” to the Doctrine of Right (MS 6:354–5), Kant is concerned with what he there calls “the highest political good”, namely perpetual peace. Kant starts by pointing out that there are cases in which we can demonstrate neither that some thing “is” nor that it “is not”, but where we are interested in assuming that the thing “is” in order to realize a morally necessary end, that is, an end “such that the maxim of adopting it is itself a duty”. So far, Kant’s characterization of the case he is talking about is in keeping with the two defining characteristics of the postulates: theoretical undecidability and practical necessity. And Kant continues: “Now it is evident that what would be made our duty in this case is not the assumption (suppositio) that this end can be realized […] for there can be no obligation to do this (to believe something). What is incumbent upon us as a duty is rather to act in conformity with the idea of that end, even if there is not the slightest theoretical likelihood that it can be realized, as long as its impossibility cannot be demonstrated either. […] So the question is no longer whether perpetual peace is a thing or a non-thing [Ding oder Unding] and whether we are not deceiving ourselves in our theoretical judgment when we assume the former. Rather, we must act as if it were a thing [so handeln, als ob das Ding sei], though perhaps it is not” (MS 6:354; trans. modified; my emphasis).

Must We Believe in the Realizability of Our Ends?

235

The distinction between thing and non-thing (Ding und Unding) is the distinction between things that are possible and things that are not (for instance because their concept involves a contradiction; cf. MS 6:377; Ref1 6338a 18:659) and parallels Kant’s distinction between “something” and “nothing” (etwas and nichts) in the Critique of Pure Reason (KrV A290/B346–A292/B349; cf. 23:20). Kant’s point here seems to be that we don’t have to bother with the question whether perpetual peace is “a thing” (that is, something possible) or “a non-thing” (something impossible), as long as it cannot be demonstrated that it is impossible. Nor do we have to assume that it is possible (even though we may take an interest in that hypothesis). All we have to do in order to make perpetual peace our end is to act as if it were possible (which presupposes that its impossibility cannot be demonstrated), even though there may not be the least likelihood that it can be realized.²⁶ As before, we must distinguish the question whether this is a valid objection to the realizability principle (a) from the question how the objection affects Kant’s argument for the postulates (b). (a) What does it mean to “act as if” the highest end could be realized, even though one believes that it most probably cannot? With respect to perpetual peace, Kant says that “acting as if” means that we “work toward establishing perpetual peace”, for instance by trying to establish the rule of law (cf. MS 6:355) and “a republicanism of all states” (MS 6:354). Even if, as Kant says, “the complete realization of this objective [viz. perpetual peace] always remains a pious wish, still we are certainly not deceiving ourselves in adopting the maxim of working incessantly towards it. For this is our duty” (MS 6:355). Now the rhetoric of “working towards” some end that may never be fully realizable is reminiscent of the language with which Kant describes how we can approximate the ideal of systematic knowledge. And indeed, Kant closes by saying: “The attempt to realize this idea [viz. the rule of law] should not be made by way of revolution […]. But if it is attempted and carried out by gradual reform in accordance with firm principles, it can lead to continual approximation to the highest political good, perpetual peace” (MS 6:355). Hence it seems that the attitude to the idea of perpetual peace that Kant recommends in the Doctrine of Right does not differ from the attitude towards ideas in general that we considered in the previous section: even though they cannot be fully realized (at least not empirically), it is possible to approximate their re-

 Paul Guyer briefly mentions Kant’s treatment of the highest political good in the Doctrine of Right (Guyer : ), but does not discuss the apparent conflict with the realizability principle (or the “canon of rational willing” that Guyer attributes to Kant instead, cf. note  above).

236

Marcus Willaschek

alization. Acting “as if” the idea could be realized just means taking the appropriate steps to approximate it as far as possible. As we have seen, this can be understood as making the approximation of the idea one’s end (instead of its full realization). So it seems that what Kant says about the highest political good does not, after all, undermine the realizability principle, since “acting as if” perpetual peace is possible can be re-described as making its best possible approximation one’s end (which end is guaranteed to be practically possible).²⁷ (b) As before, if this “as if” approach were applicable to the case of the highest good, it would immediately undermine Kant’s argument for the postulates in the second Critique. In order to make the highest good our end we would not have to believe that it can be realized, i. e. that it is practically possible, but only that we can approximate it. Obviously, we wouldn’t need the postulates of God and immortality for that. However, as we have seen, where this “as if” approach is applicable to some end, Kant insists that “its impossibility cannot be demonstrated” (MS 6:354). Even though we don’t need any positive evidence that our end can be realized in order to act as if it were practically possible, the latter attitude is incompatible with a demonstration that its realization is impossible. So here we see a difference between the practical attitude towards ideas in general (as discussed in the previous section) and the “as if” attitude Kant recommends with respect to perpetual peace. Kant’s point seems to be that only if the realization of an end is not demonstrably impossible is it rational to regard our approximating that end “as if” it were its gradual realization. In the case of ideas that, in principle, can never be realized (such as the complete systematic unity of human cognition), such an attitude would be irrational, whereas it is appropriate with respect to an end such as perpetual peace which, however unlikely, cannot be demonstrated to be impossible. Now without the postulates, the condition of not being demonstrably impossible is not satisfied in the case of the highest good, since the antinomy of practical reason (premise 5 above) seems to provide a demonstration that the highest good is practically impos A problem with this solution is that (on the assumption that people cannot be as happy as they morally deserve unless they live in peace) perpetual peace seems to be a necessary condition for the realization of the highest good. So how can the one be realizable, while it is an open question whether the other can be realized? (Thanks to Christina Engelmann for raising this problem.) A possible answer might be that with respect to perpetual peace, Kant is concerned with its political possibility, while the highest good is supposed to be practically possible, where the political possibility of an end requires more than mere practical possibility (e. g. compatibility with the anthropological conditions of political action). Thus, perpetual peace would be practically possible (i. e. something that could at least in principle be realized by human agents). But we cannot be sure that it is also politically possible (i. e. something we can realistically expect to be realized).

Must We Believe in the Realizability of Our Ends?

237

sible. As long as that demonstration is not refuted, i.e. as long as we don’t have an account of how the highest good is possible that resolves the antinomy, an “as if” attitude towards the highest good on the lines adopted by Kant towards perpetual peace is not rationally available. In sum, the possibility of pursing an end that cannot be fully realized in an “as if” mode is compatible with the realizability principle (a), since it is possible to re-describe the end in terms of approximating it. Moreover, it does not undermine Kant’s argument for the postulates in the second Critique (b), since adopting an “as if” attitude toward pursuing the highest good presupposes the absence of a proof of its impossibility, which condition, because of the antinomy, is not satisfied independently of the postulates.

4 “Not Believing” Instead of “Believing That Not” This takes us to a third objection against the realizability principle in its application to the concept of the highest good. That principle requires us to believe in the realizability or practical possibility of our ends. But wouldn’t it be enough to require for the rational pursuit of an end that we don’t believe that it cannot be realized? This suggests the following principle: RP3 An agent A is rational in pursuing some end E by doing D only if A (rationally) does not believe that E is practically impossible (i. e. that it is impossible for A to realize E by doing D). RP3 is considerably weaker than RP2 in that it does not require a belief in the possibility of one’s end, but only the absence of a belief in its impossibility. But precisely because it is weaker, the truth of RP3 in itself would not be a problem for a defender of the stronger principle RP2, since RP3 can plausibly be seen as a logical consequence of RP2. The stronger principle would be in danger only if there are possible cases that satisfy RP3 but violate RP2. Here is a candidate case: Imagine I have to take a test and score at least 50 out of 100 points in order to pass. By any standards, I’m insufficiently prepared for the test. Hence, I don’t believe I will succeed. I am even uncertain as to whether it is possible, under these conditions, for me to succeed. Being uncertain, I don’t believe it to be possible. But neither do I believe it to be impossible. Thus, I’ll do my best and try to pass the test. Is this a plausible case of someone’s pursuing an end that he does not believe he can realize, but for whom it is nonetheless not irrational to pursue the end, given that he does not believe the realization of that end to be impossible? If so, this would be a case of rational action that satisfies RP3 but violates RP2.

238

Marcus Willaschek

One may perhaps worry here that even though not believing that p obviously does not imply believing that non-p, in this particular case there is no psychological space left between not believing that I can succeed and believing that I cannot succeed. If I don’t believe that I can succeed, how can I avoid believing that I cannot? Well, here is how. I might ask myself: can I do it? Is it really possible for me to pass the test or is it out of the question? The honest answer may be that I don’t know. In this case, it seems both rationally and psychologically plausible that I don’t take a stand at all: neither do I believe that I will be able to pass the test nor do I believe that I will not be able to do so. So the question is whether in such a case it is rational for me to try. Here, a word about trying is in order. In general, there are two ways in which we can pursue an end. In many cases, I know that I will be able to realize my end. In these cases, I do not try to realize my end; I just do it. If my end is to get up from my chair and stretch my legs, and I am not paralyzed or tied to my chair, I normally just get up and stretch my legs. Of course, for some odd reason I may fail and then say that I tried in vain to get up. But if I do not fail, it would be misleading to say that I tried; rather, I just did it.²⁸ In other cases, such as passing a test, I don’t know whether I will succeed. In these cases, it makes sense to say from the outset that I try to realize the end in order to emphasize the fact that I am not sure that I will succeed. Now I cannot “just do” something if I don’t believe I will succeed, which trivially implies that I also believe that it will be possible for me to succeed. But when it comes to genuine cases of trying, it seems sufficient that I don’t believe that it is impossible for me to succeed. It does not seem to be irrational to try to pass a test even if I don’t believe I can succeed, as long as I do not believe that I must fail. So we must distinguish between “just doing” and “trying”.²⁹ Whereas RP2 or some similar principle may hold for cases where someone does something without having to try, this principle is too strong for cases of trying: it is not generally the case that an agent A is irrational in trying to realize some end E by doing D if A does not believe that E is practically possible. Rather, cases of trying fall under the weaker principle RP3.

 Cf. Dancy .  It is unclear to me whether Kant himself draws this distinction in any systematic way. Since the question we are currently investigating is whether there is a version of the realizability principle that is both plausible and appropriate to Kant’s argumentative purposes, and since the point of distinguishing between doing and trying to do is to raise a problem for RP, it does not matter much whether this distinction is one Kant himself draws explicitly. The distinction is commonsensical enough anyway.

Must We Believe in the Realizability of Our Ends?

239

But even RP3 may still be overly demanding, which becomes evident when, instead of using the abbreviated expression “practical possibility”, we round things out by using the more complex formulation of PP (cf. section 2): RP4 An agent A is rational in trying to realize some end E by doing D only if A (rationally) does not believe that it is not the case that (i) A’s doing D causally contributes to realizing E and that (ii) the obtaining of a set of causal conditions that are jointly sufficient to realize E (of which A’s doing D is a part) is not unlikely. RP4 is too strong since it may well be rational to try to realize an end “against all odds”, that is, even if condition (ii) is not satisfied. Whether or not it is rational to try to do something even though success is highly unlikely will mainly depend (a) on the importance of the end pursued and (b) on the costs involved in trying. As long as success is not absolutely impossible, it seems that trying can always be rational if the stakes are high enough. Moreover, it may be rational to try to realize some end even if condition (i) is not satisfied (that is, even if A believes that doing D does not causally contribute to realizing E) as long as A does not believe it to be impossible that doing D should causally contribute to realizing E. For instance, I may believe that homeopathic medicine will not causally contribute to curing a serious illness, but since I am not absolutely certain about this (I do not believe this to be impossible), I try it anyway. This does not seem to be irrational – particularly in cases where very much is at stake. If we take these two things into account, we arrive at the following principle: RP5 An agent A is rational in trying to realize some end E by doing D only if A (rationally) does not believe it to be impossible that (i) A’s doing D should causally contribute to realizing E and that (ii) a set of causal conditions that are jointly sufficient to realize E (of which A’s doing D is a part) obtains.³⁰

 It is an interesting question, which I will not be able to pursue here, in which sense of “possibility” A must not believe (i) and (ii) to be impossible. It can’t be practical possibility (which is captured by (i) and (ii)). Both logical and metaphysical possibility seem clearly too weak. Perhaps “physical possibility” will do. Alternatively, RP might be reformulated in terms of epistemic possibility (“compatible with all the evidence available to A”), which would fit with the internalist account of rationality in terms of the internal consistency of a subject’s commitments that seems to be at work in Kant’s argument.

240

Marcus Willaschek

So we’ve come to the conclusion that some revision of RP, at least for cases of trying, is indeed called for. And this revision directly concerns Kant’s argument for the postulates, since realizing the highest good is also not something we “just do”, but rather something we at best can try to do. So it seems that we are not rationally required to believe that the highest good is practically possible, but rather required not to believe that it is practically impossible.³¹ What does this mean for Kant’s argument for the postulates? With respect to the reconstruction offered above (in section 2), it means that premise (3) – according to which we can be rational in making the highest good our end only if we rationally believe it to be practically possible – is not valid in this context, since all we can require of one rationally trying to realize the highest good is that one does not believe it to be practically impossible. Hence, we have to replace (3) with (3**) We can rationally make the highest good our end only if we do not believe that it is impossible for us to realize the highest good, and (6) with (6*) In order for it to be rational for us to believe that we can realize the highest good, we require an account of how, in spite of (5) (i. e. in spite of the antinomy), it is not impossible for us to realize the highest good. Even though (6*) may not be as intuitively compelling as (6), it seems highly plausible. The antinomy of practical reason appears to show that the highest good is practically impossible, and thus that we are rationally required to believe that it is practically impossible. As obedient subjects of the moral law, however, we nevertheless make the highest good our end, which, according to (3**), rationally requires us not to believe that the highest good is practically impossible. Hence, in order to avoid irrationality in the form of incompatible doxastic attitudes, we need some explanation of how, even in light of the antinomy, it is not the case that the highest good is practically impossible. The resolution to the antinomy provides such an explanation by showing that we can conceive

 Kant seems to employ an analogous principle in his argument against Mendelssohn in “On an Old Saw” (Gemeinspruch :–): “Empirical arguments against the success of these resolutions, which are taken on hope, accomplish nothing here. For, that what has not succeeded up to now will therefore never succeed does not even justify abandoning a pragmatic or technical purpose […], still less a moral purpose that, if only it is not demonstratively impossible to effect it, becomes a duty”.

Must We Believe in the Realizability of Our Ends?

241

of the practical possibility of the highest good if (and only if) there is an intelligible author of the world. Hence, we can move on to (7*) We can give an account of how virtue can be the (indirect) cause of proportionate happiness, and thus why it is not impossible for us to realize the highest good, only if we believe in God (and immortality), and from there to (8) We are rationally required to believe in God and immortality. Thus, Kant’s general argument remains intact even if we replace RP with RP5 and (3) with (3**).

5 Conclusion Our first result is that we have to distinguish different versions of the realizability principle for cases of doing things, on the one hand, and cases of trying to do something, on the other. In particular, rationally trying to realize some end does not in general presuppose that one believes that one’s end can be realized. Hence it does not presuppose a belief that the necessary conditions for realizing that end are given. But rationally trying to realize some end requires that one does not believe it to be impossible to realize one’s end. If there is reason to believe that this is impossible, as is the case with trying to realize the highest good by acting virtuously, we therefore need some positive account as to how realizing the end is not impossible after all. Our second result is that the weaker version of the realizability principle that holds for cases of trying (RP5) is still strong enough to function as a premise in Kant’s argument for the postulates. The role of God in Kant’s argument is to forge a causal connection between virtuous action and happiness and thereby to show how realizing the highest good is not practically impossible. Thus, Kant’s argument relies not on a strong version of the realizability principle such as RP3, but only on the weaker version, RP5. Obviously, this is not a defence of Kant’s complete argument. In particular, it is not a defence of Kant’s specific arguments for the postulates of God and immortality (which I have not discussed here at all). But even if one finds fault with Kant’s overall argument or some of its steps, its rejection cannot be based on objections to the realizability principle, since there is a version of that principle that seems to be both valid and sufficiently strong for Kant’s purposes.

242

Marcus Willaschek

Appendix Kant argues that we must believe that God exists in order for it to be rational for us to believe that the highest good is possible. But why is it not sufficient for that purpose to believe that it is possible that God exists? If it is possible that God exists, it is possible that God does whatever is necessary to make the highest good possible. And isn’t that enough of a reason not to believe that it is impossible? The answer is no, because there is an illicit shift in the modalities involved in this line of thought. Consider an analogy: I want to run (and finish) a marathon today, even though I haven’t done any training. A friend points out to me that that is (practically and even physically) impossible, since running a marathon requires considerable training. According to RP5, this means that I rationally must abandon my goal of finishing a marathon today. Now I realize that it is metaphysically possible that there is a pill that immediately induces the effects of a year’s worth of marathon training.³² Does this make it rational for me to set myself the end of finishing a marathon today? Obviously not, since as far as I know there is no such pill. What would be required to make it rational to try to finish the marathon in light of my friend’s objection is the belief that there really is such a pill (and that it is possible for me to take it). Mere belief in its metaphysical possibility is insufficient. The same holds for Kant’s argument for the postulates: the antinomy is supposed to show that the highest good is practically impossible (where practical possibility is the possibility of bringing about an effect by one’s own actions). Now if one believes that it is possible that God exists, one obviously does not believe that God is practically possible, but rather that he is metaphysically possible (and, perhaps, epistemically possible).³³ But from (a) God’s being metaphysically possible and (b) the highest good’s being practically possible if God exists, it does not follow that the highest good is practically possible.³⁴ If it is metaphysically possible that God exists, there is at least one possible world in which God exists. Let us assume that if God exists in some world, he forges a causal link between virtue and happiness such that the highest good is practically possible in that world. But if God does not actually exist, the possible world in which virtue causes happiness is not our world (which, after all, is the actual word). The only way in which God can forge a causal link between virtue and happiness in our world is by existing in our  Something is metaphysically possible if it exists/is true in at least one possible world.  Something is epistemically possible if its existence/its being true is compatible with all the available evidence.  Eric Watkins makes what is substantially the same point in Watkins : .

Must We Believe in the Realizability of Our Ends?

243

(the actual) world, that is, by being actual. Hence, the mere metaphysical possibility of God’s existence is not sufficient to counter the antinomy argument against the practical possibility of the highest good and thus to make it rational not to believe that the highest good is impossible.³⁵

References Dancy, Jonathan 1995, “Arguments from Illusion”, Philosophical Quarterly 45, 421−38 Gardner, Sebastian 2006, “The Primacy of Practical Reason”, in: Graham Bird (ed.) 2006, A Companion to Kant, Oxford: Blackwell, 259–73 Guyer, Paul 2000, “From a Practical Point of View: Kant’s Conception of a Postulate of Pure Practical Reason”, in: Paul Guyer 2000, Kant on Freedom, Law, and Happiness, Cambridge: Cambridge University Press, 333–71 Watkins, Eric 2010, “The Antinomy of Practical Reason”, in: Andrews Reath/Jens Timmermann (eds.) 2010, Kant’s Critique of Practical Reason. A Critical Guide, Cambridge: Cambridge University Press, 145–67 Watkins, Eric 2013, “Kant on the Natural, Moral, Human, and Divine Orders”, in: Eric Watkins (ed.) 2013, The Divine Order, the Human Order, and the Order of Nature. Historical Perspectives, Oxford: Oxford University Press, 219–36 Watkins, Eric/Willaschek, Marcus: “Kant on Cognition and Knowledge” (unpublished) Willaschek, Marcus 2009, “Kants Begründung für das Primat der praktischen Vernunft”, in: Heiner F. Klemme (ed.) 2009, Kant und die Zukunft der europäischen Aufklärung, Berlin: de Gruyter, 251–68 Willaschek, Marcus 2010, “The Primacy of Practical Reason and the Idea of a Practical Postulate”, in: Andrews Reath/Jens Timmermann (eds.) 2010, Kant’s Critique of Practical Reason. A Critical Guide, Cambridge: Cambridge University Press, 168–96 Wood, Allen W. 1970, Kant’s Moral Religion, New York: Cornell University Press Wood, Allen W. 1994, “Rational Theology, Moral Faith, and Religion”, in: Paul Guyer (ed.) 1994, The Cambridge Companion to Kant, Cambridge: Cambridge University Press, 394– 416

 Thanks to Thomas Höwing, Florian Marwede and Eric Watkins, as well as audiences in Antwerp, San Diego and Frankfurt for many helpful comments on earlier versions of this paper. Thanks also to Hannes Matthiessen, Andreas Maier and Daniel Karanovic for discussions about the modal aspects of Kant’s argument. Finally, I’d like to thank Carolyn Benson for revising my English.

Andrea Marlen Esser

Applying the Concept of the Good: The Final End and the Highest Good in Kant’s Third Critique ¹ In the Critique of Judgement, towards the end of §87, Kant specifically mentions Spinoza as an example of a “righteous man” who “takes himself to be firmly convinced that there is no God and […] there is also no future life” (KU 5:452), but who nonetheless obeys the “moral law which he actively honors”. Shortly before this passage, Kant clearly reaffirms that the moral law is “the formal rational condition of the use of our freedom” and, as such, “obligates us by itself alone, without depending on any sort of end as a material condition” (KU 5:450). In view of this assurance, it seems quite clear that even individuals who are not convinced of the existence of God must still recognize that they are bound by moral laws, for these laws “are formal and command unconditionally, without regard to ends” (KU 5:451). Yet although Kant emphatically praises Spinoza’s moral attitude, immediately after this assessment he proceeds to criticize him, claiming that “his [Spinoza’s] effort is limited” (KU 5:452). And the conclusion that Kant finally draws with regard to Spinoza at the end of §87 is that even this “well-intentioned person” is ultimately faced with a choice: either he must abandon as impossible the end, which he “had and should have had before his eyes in his conformity to the moral law”, namely the “one ideal final end”, or, if he does not want to “weaken” this final end and thus do injury to his moral disposition, he must “assume the existence of a moral author of the world, i.e., of God” (KU 5:452).² The notion that our practical reason provides us with a certain “intrinsic purposiveness” and furnishes us with a “final end”, which we should also try to realize as the “highest good” that is possible in the world, is one that Kant had

 I am indebted to Andreas Eckl for discussing the Critique of Judgement with me over many years. A lot of his ideas and his helpful suggestions are reflected in this paper. I also want to thank Jens Timmermann for his invitation to the Paton Colloquium, which gave me the opportunity to present a version of this paper at St Andrews and discuss it with Alix Cohen, Kyla Ebels and Allen Wood.  In the second Critique, the “virtuous Epicurus” (KpV :), who according to Kant regarded the existence of the Gods as “evident” even though they exert no influence upon us or upon the world, is spared the kind of strong criticism to which the “righteous” Spinoza is subjected. Kant spares him even though his error concerns the principle of morality itself and even though Kant accuses him of circular reasoning when it comes to defining the highest good.

246

Andrea Marlen Esser

already discussed in earlier writings. Thus in the Critique of Practical Reason, in the section entitled “The Antinomy of Practical Reason”, we read that if “the highest good is impossible in accordance with practical rules, then the moral law, which commands us to promote it, must be fantastic and directed to empty imaginary ends and must therefore in itself be false” (KpV 5:114). But even if the doctrine of the moral final end as the highest good undoubtedly reveals certain conceptual affinities in Kant’s various writings (and especially in the second and third Critiques), this doctrine nonetheless acquires a further and indeed new systematic significance in the Critique of Judgement insofar as it is here specifically integrated with Kant’s developed conception of purposiveness and his doctrine of reflective judgement. In what follows, I shall attempt to clarify this thesis in relation to Kant’s criticisms of Spinoza (section I). From there, I shall proceed in section II to bring out three distinctive features of the doctrine of the highest good as it is presented in the Critique of Judgement: a) the explicit focus on the “conditions for realizing” the practical moral law in the world and the connection between the doctrine of the highest good and the notion of the purposiveness of nature (which is important for Kant’s aim of unifying his philosophy into a “systematic whole”); b) the newly defined logical relation between this doctrine and the doctrine of reflective judgement; and c) the even more emphatic critical implications for religion, which are revealed in Kant’s argument with Spinoza and his profound reservations regarding the physico-theology which was enthusiastically defended by many thinkers at the time.

I Kant’s Argument with Spinoza and the “Virtuous Atheist” The choice and presentation of the particular example used by Kant in the Critique of Judgement to show why a doctrine of the “final end” or of a “highest good” must be developed in the first place is already clearly marked by the systematic concerns of the third Critique: the issue of mediating “the nomothetic of freedom on the one hand and that of nature on the other” (KU 5:448) and the issue of overcoming a supposed “two-world” dualism (cf. KU 5:176).³ The “righteous man”, who is firmly convinced that there is no God but nonetheless acknowledges all the demands of the moral law, and who unselfishly follows these demands to the utmost of his power, finds himself within a world, within a system of nature, which at the very least is indifferent to the realization of his moral aims and ends

 Kant makes it quite clear that this is merely a metaphorical mode of expression and specifically points out that it is wrong to speak of “two worlds” in any literal sense here (cf. KU :).

Applying the Concept of the Good

247

(when it does not actively resist or impede their realization).⁴ In this connection there is a wonderful passage in §87 of the third Critique which I would like to quote here: “Deceit, violence, and envy will always surround him, even though he is himself honest, peaceable, and benevolent; and the righteous ones besides himself that he will still encounter will, in spite of all their worthiness to be happy, nevertheless be subject by nature, which pays no attention to that, to all the evils of poverty, illness, and untimely death, just like all the other animals on earth, and will always remain thus until one wide grave engulfs them all together (whether honest or dishonest, it makes no difference here) and flings them, who were capable of having believed themselves to be the final end of creation, back into the abyss of the purposeless chaos of matter from which they were drawn” (KU 5:452).⁵ But such a righteous individual cannot expect from nature any assistance or “agreement […] with the ends to act in behalf of which he still feels himself bound and impelled” (KU 5:452). But even if the “righteous man” were not misled or deterred by this and expected no reward or support from the world for duly following the moral law, his “endeavour” would still be limited or “circumscribed” in a quite specific sense, as Kant emphasizes. This is not because such a virtuous atheist has a sufficient psychological motive for moral conduct only if he assumes that a final end and a highest good are possible (as Frederick Beiser maintains),⁶ such that, unable to realize his intended ends in the world, he must inevitably yield to resignation. Rather, his endeavour is circumscribed because, given the particular way in which he sees the world and himself, he is incapable of appropriately articulating “his own inner purposive determination by the moral law” (KU 5:452). Thus he is

 Terrence Malick’s film The Tree of Life () can be interpreted as an artistic engagement with this question. We hear the voice of the mother telling us of her religious education: “The nuns taught us there were two ways through life – the way of nature and the way of grace. You have to choose which one you’ll follow […]. Grace doesn’t try to please itself. Accepts being slighted, forgotten, disliked. Accepts insults and injuries […]. Nature only wants to please itself. Get others to please it too. Likes to lord it over them. To have its own way. It finds reasons to be unhappy when all the world is shining around it. And love is smiling through all things […]. The nuns taught us that no one who loves the way of grace ever comes to a bad end”. Malick pictures nature as sublime and entirely indifferent to human aims, action and harm. But the beauty of nature as well as the end of the narration might be signs of a “practical” (in a Kantian sense) connection between the seemingly two ways of leading a life.  To suppose, for example, that man is also the final end of nature, a position defended by many representatives of physico-theology, would amount to a substantively misleading and ungrounded conception of the final end and of one’s own status as an end in itself. I shall return to this question in II..  Cf. Beiser :  ff.

248

Andrea Marlen Esser

also incapable of understanding himself as the final end of the creation of a moral author of the world, and, finally, of promoting the realization of the highest good within this creation, which is entrusted to us through this conception of the final end.⁷ For the practical moral law requires its complete realization, that is to say, its systematic and not merely partial realization in terms of specific acts, in this world;⁸ it requires us in this world, together, to bring about what is “best for the

 The precise relationship between man as the final end of creation and the final end which practical reason calls upon human beings to realize may seem rather complex at various points in the third Critique. I think we can reconstruct the argument along the following lines: as a rational being, man must be regarded as the final end of creation, for by analogy we may ascribe a ground or end to our conception of the moral author of the world, as an end which the latter pursues with his creation. And this ground or end can only be a rational or moral one, such as the way in which the creation maintains certain beings who respond to this moral end of creation, or at least attempt to do so. But the only beings (known to us) who are capable of reason and morality at all are human beings. We can thus regard ourselves as the final end of creation, but only insofar as we bring reason and morality into the world through our own action, and thus insofar as we strive to realize rational and moral structures and relations within animate and inanimate nature and within the rational and the non-rational worlds. The moral author of the world (God) has unquestionably accomplished his final end and purpose with the creation of a world that contains beings capable of realizing these things. For us, on the other hand, the insight that we must regard ourselves as this final end of creation means that we must “qualify” ourselves for this in the first place precisely through our moral action: “But morality, or a causality according to ends that is subordinate to morality, […] in the order of ends is the sole principle possible which is absolutely unconditioned in respect of nature, and it is what alone qualifies [my emphasis] the subject of such causality to be the final end [Kant’s emphasis] of creation, and the one to which the whole of nature is subordinated” (KU :n.; trans. Meredith): The final end which we ought to try to realize as “creatures” is at the same time the highest good that is possible in the world, which we can therefore also posit as the sole and a priori object of the pure will. Reductive and therefore false conceptions of nature, like those which have been disseminated by the doctrines of Spinoza (and of course also those of Epicurus), may deter or distract us from realizing this final end and must be criticized precisely for this reason.  Therefore it should be clear that the doctrine of the highest good is a necessary part of Kant’s moral theory and cannot be isolated from it. In my opinion, Reath is perfectly right in elaborating the complexity of Kant’s moral theory and emphasizing “that the Highest Good may not be viewed as a theological notion” (Reath : ). Reath identifies two distinct doctrines of the highest good in the third Critique – one “theological”, the other “secular” – and states that only the latter can be seen “as the best expression of Kant’s view” (Reath : ). In contrast to this reading, I will suggest that it is the specific outcome of the Critique of Judgement to show that these two dimensions only work together if we describe the world in terms of teleology. It is one of the maxims of enlightenment (“the consistent way of thinking”; cf. KU : f.) – and not of theology – that obliges the adoption of a critically proven, and only subsequently theological, understanding of the world. It is therefore the “secular” conception which forces us for pragmatic reasons to adopt the “theological” hypothesis of an “author of the world (i. e. a God)” (cf. KU :) and the idea of an afterlife.

Applying the Concept of the Good

249

world”, in the sense of a sure and certain relationship between the greatest possible satisfaction for rational beings and the highest condition of the good. Spinoza cannot even begin to set before himself an end such as this. For to set this final end before ourselves and actively promote the realization of the highest good is only possible under two conditions, both of which Spinoza – if he actually wishes to comply with the moral law – would have to acknowledge. The objective condition under which human beings can set such a final end has already been clearly formulated: this is the unqualified acknowledgement of the moral law. But the subjective condition under which human beings as finite and sensuous beings can alone undertake to realize the required final end is that of happiness. And the highest good – the sole end of practical reason – also corresponds to both of these conditions since it is defined precisely as “happiness – under the objective condition of the concordance of humans with the law of morality” (KU 5:450). Against the background of Kant’s general theory of action, this notion can perhaps be clarified as follows: we can only set this final end for ourselves if we fulfil both the objective condition (i. e. acknowledge the practical moral law and strive to realize it) and the subjective condition, insofar as we anticipate the subjective feeling of happiness which arises through the realization of the moral task. Thus I think that what Kant means is this: we can actually only set a final end for ourselves if we can entertain the legitimate hope that our moral ends can be realized and that, should this actually prove successful, we will also be happy in correspondence with the realized moral demands. But it is important that the subjective condition also be fulfilled with the setting of the final end, at least in the anticipation of happiness (a happiness in correspondence with moral demands). This is important because the highest good is an end that we must assume specifically as human beings, namely as beings who are not only rational but also finite and sensuous in character – as beings who strive for happiness by nature. This is why it is not possible simply to cast this subjective condition aside, as Spinoza implies.⁹ It is not only that Spinoza is unable to grasp his own “inner purposive determination” in terms of the moral law (because he must assume the connection of ends in the world as unintentional [cf. KU 5:391] and “deprive this idea of all reality” [KU 5:393]). He is also unable to set any final end for himself because from the start he only seeks to realize the objective condition of the highest good (name-

 What Spinoza fails to recognize in his admittedly rather heroic attitude is that the latter conception cannot be adopted by any and every human being. For heroism is not something which can or ought to be morally demanded (cf. also Kant’s criticisms of “fantastical” virtue and the ethical outlook attributed to the heroes of novels at the end of the Critique of Practical Reason and in the Metaphysics of Morals [KpV :; MS : and ]).

250

Andrea Marlen Esser

ly to comply with the moral law).¹⁰ If he were really “to think in a morally consistent manner” (KU 5:451n.), as Kant requires, then he would be able to see that he must actually abandon the final end, which he is purportedly striving to realize, or indeed assume the existence of God in the sense of a moral author of the world. In this discussion of Spinoza it might seem, prima facie, that Kant is simply returning to the problem of motivation and moral incentives, which he had not yet resolved to his own complete satisfaction.¹¹ But in fact Kant’s reference to Spinoza here also leads to a fundamental systematic argument in which the morally destructive effects of certain forms of atheism and of certain notions of God are exposed and criticized, and in which the morally beneficial contribution of a specific use of teleological judgement – namely the (critical) use of the reflective power of judgement – is also supposed to be shown. With this reference to the “righteous” Spinoza, Kant also deliberately alludes to a familiar topic of discussion in the Enlightenment which was recognized as particularly relevant to the first issue here, namely the morally destructive effects of atheism. Kant’s source for the theme of the “virtuous atheist” was the article entitled “Spinoza” in Pierre Bayle’s Dictionnaire. ¹² Bayle here defended the view that the commitment to morality and the commitment to atheism do not necessarily contradict one another. According to Bayle, both convictions are entirely compatible with one another because it can be shown that morality is independent of faith and religion and is therefore autonomous, such that the claims and laws of morality can be legitimated without reference to faith, religious convictions, or religious arguments of any kind. In order to demonstrate the autonomy of morality, Bayle specifically refers to Spinoza, who seems to offer an excellent example for his argument on account of the virtuous life that the latter was widely recognized to have led. Yet later on in the same article, Bayle also emphatically warns his readers of the morally destructive effects of a systematic doctrine of atheism. Bayle actually goes on to mention Spinoza again, but this time as an example of a proponent of just such a systematic doctrine of atheism. This obvious ten With the notion of the final end one also inevitably accepts the assumption of a moral author of the world and thereby the immortality of the soul, insofar as we must assume that God, understood in this sense, has established nature in such a way that we (or any beings that respond to the moral law) can become happy in the context of this nature.  Cf. also the following passage from Kant’s essay On the common saying: That may be correct in theory but it is of no use in practice: “The need to assume a […] highest good in the world that is also possible through our collaborative activity as the final end of all things is a need that arises not from some defect in our moral incentives, but rather from external relations in which, solely in accordance with these incentives, an object can be presented as an end in itself (as a moral final cause)” (Gemeinspruch :n.; trans. modified).  Cf. Bayle : –.

Applying the Concept of the Good

251

sion within his article has been remarked upon by many readers.¹³ Some have attempted to resolve the difficulty simply by strictly separating the two perspectives at work in the article: on this reading, Bayle praises the virtuous conduct of Spinoza’s life from a specifically moral perspective, while from a second, specifically theoretical perspective, he criticizes Spinoza’s ethical and metaphysical position and even rejects it as “repellent”.¹⁴ This kind of reading is not wholly implausible, for there were certainly some characteristic systematic features of Spinoza’s philosophy, such as the pantheism defended in the Ethics and the claim that the force or power belonging to any and every thing in the world is ultimately only an expression of God’s power, which led Bayle to his critical objection that Spinoza’s God must appear as a God who seems to hate himself, who begs himself for mercy which he then denies to himself:¹⁵ in short, as a God to whom we must also correspondingly ascribe all of the mutually conflicting and mutually contradictory attributes which we see at work in the world.¹⁶ But there is also another way of interpreting the obvious tension in Bayle’s article:¹⁷ this reading insists that Bayle was concerned with far more than diagnosing the contradictory aspects of Spinoza’s Ethics and apportioning praise and censure here and there as required. According to this very convincing reading, as it seems to me, Bayle’s article was intended above all to warn his readers very clearly that Spinoza’s atheism would have effects which are deleterious to morality, tending to weaken or destroy our moral attitudes or even morality itself. As a particular concrete individual, Spinoza may well have been an unquestionably virtuous person, as Bayle expressly concedes; yet his image of the world and his metaphysical theory also stand, from a systematic perspective, in fundamental contradiction with the way in which he led his own life. This tension is one which cannot be resolved and is therefore not resolved by Bayle. And the whole point of his article was to make this unambiguously clear. The harsh judgement which Bayle passes on Spinoza at the end of the article – if we accept this reading of the text – is intended above all to show that certain kinds of atheism, namely those that construct their systematic position on the basis of determinism and defend a pantheistic conception of the world, do not merely partially undermine the

 Cf. Schroeder : .  Cf. Schroeder : –.  Cf. Bayle : ; Bayle : : “And therefore all the phrases made use of to express what men do one against another have no further true sense than this: God hates himself; he asks favours of himself; and he refuses them to himself: he persecutes himself, kills himself, eats himself, calumniates himself, executes himself”.  Cf. Bayle : .  This is the position Schroeder is arguing for in his essay (cf. Schroeder : ).

252

Andrea Marlen Esser

possibility of an autonomous form of morality, but rather destroy it in principle. Now considerations that point in a very similar direction to Bayle’s article can also be found in Kant’s own references to Spinoza in §72 and §73 of the Critique of Judgement. In these particular passages, Kant engages with what he calls the “system of fatality” and identifies Spinoza as the most prominent defender of such a system.¹⁸ Kant’s basic criticism of this position in §72 and §73 is that Spinozism is unable to explain the supposed explanatory ground, and thus the required unity, of the “connection of ends (which it does not deny)” (KU 5:393).¹⁹ Given the theoretical structure of his system, Spinoza is compelled to explain the relationships and interconnections of things in the world in terms of some ultimate cause which cannot be further elucidated in any way that we are able to comprehend. Kant objects that such an approach involves both an unintelligible concept of God (cf. KU 5:391) and a fundamental “fatality” in the order of the world. As a result – and here Kant agrees entirely with Bayle – this approach no longer leaves any room for an autonomous grounding of morality and exerts a morally destructive effect on our action in the world. Now of course Kant’s concern in the Critique of Judgement is not primarily to offer a critique of Spinoza and his Ethics. His discussion of the problems involved in this “system of fatality” is rather meant to show both that it is possible to overcome these problems and how it is possible to develop a view of the world which is compatible with morality and may perhaps even promote morality – a view that both reveals that the moral ends we set ourselves are “realizable” and makes the purposive relations that we notice in the world intelligible. For the systematic question raised by the Critique of Judgement as a whole is this: how can we mediate our practical understanding of ourselves as free beings with our experience of nature and our image of the world in which we live in such a way that the systematically disturbing contradictions between the fundamental laws of these two “worlds” can be overcome? This is nothing less than the task of overcoming the great “gulf fixed between the domain of the concept of nature […] and the domain of the concept of freedom” (KU 5:175 f.) which was already identified in the Introduction to the third Critique – an issue that not only presents a problem for the systematic articulation of philosophy, but above all threatens to undermine the reality of morality. For morality itself requires that its claims and demands should be realized in the sensible world (cf. KU 5:176). But this would be impossible if the central concept of moral philosophy, the idea of freedom,  He also recognizes what might be called an “ancient” form of Spinozism, however, for this conception “is to all appearances much older” (KU :).  In order to maintain this systematically required unity, one would specifically have to identify or furnish an end, rather than a cause, which is what Spinoza provides.

Applying the Concept of the Good

253

were simply a “phantom of the mind”,²⁰ indeed thought on the basis of this “gulf” but quite incapable of producing any effects in the world since the conditions of realizability required for freedom (and the demands of morality) are not given, or at least cannot be reliably assessed.²¹ These considerations lead us to the first distinctive feature of Kant’s doctrine of the highest good.

II Distinctive Features of the Doctrine of the Highest Good in the Critique of Judgement 1 The Focus on the “Conditions of Realizability” of the Practical Moral Law in the World If the concepts of practical reason are not to be empty or mere “phantoms of the mind”, it is necessary for us to demonstrate their “objective reality”.²² In the context of practical philosophy, “objective reality” signifies practical reality in the sense of the realizability of whatever is demanded of us (cf. KU 5:457).²³ Now the practical moral law requires, as we have already indicated, the complete realization of its object (the highest good) in this world. For as Kant already emphasizes in the Introduction to the third Critique (in section IX; cf. KU 5:195), the consequences or effects of the concept of freedom should be encountered in the world.²⁴ But in that case, it is also necessary to bear in mind and reflect upon the conditions of realizability of this demand. If in this connection we were to reach the conclusion that it is impossible to realize the sole final end and object

 As Kant puts it in the Groundwork for the Metaphysics of Morals (GMS :).  See KU : and also KU :, where Kant emphasizes that the moral law gives us a reason to assume, “in order to apply our powers to realize it, its possibility, its realizability”, and hence also a nature that can be judged as purposive insofar as it accedes to our struggle.  For further discussion of this cf. Esser : , where I try to show that transcendental explanations have to fulfil a generative task. They show the conditions under which the concepts in question can possibly have corresponding objects. Practical concepts get their objects through human actions that are performed in compliance with the claims of the moral law.  See also Förster : : “If the concepts of practical reason are not empty, their real possibility must be demonstrated. Here, however, real possibility or objective reality amounts to ‘practical reality’, and consists in their achievability or ‘realizability’ [Ausführbarkeit] (:)”.  Cf. KU :: “The effect in accordance with the concept of freedom is the final end which (or the appearance of which in the sensible world) should exist, and this presupposes the condition of the possibility of that end in nature (i. e. in the nature of the subject as a being of the sensible world, namely as a human being)”.

254

Andrea Marlen Esser

of morality within our world (an end entrusted to us), then the moral law would also be “fantastic” and ultimately even “false”, as Kant says in the second Critique (KpV 5:114). Kant goes on to pursue this thought in §87 of the third Critique and presents it concretely in the context of what he calls “moral teleology”. It is here that Kant uses the expression “moral teleology” for the first time. This moral teleology is not concerned with the question of what specific ends we should set for ourselves, but addresses the relationship between the moral ends we set ourselves, along with the associated “causality of freedom”, and the world in which we are supposed to realize these ends. It is true that this moral teleology has its ground “in ourselves”, namely in our practical reason (KU 5:447), but it nonetheless concerns us not simply as rational beings, but also as beings involved in the world. Ultimately, as the rational “beings in the world” that we are (KU 5:447, my emphasis), that is to say, as beings in this world who are also involved with other things and with other beings like us in the world, we are called upon to realize the highest good.²⁵ The question raised by this moral teleology is, in the first place, whether we are compelled, if we consider this final end in the light of “rational judgement” (KU 5:448), “to go beyond the world and seek an intelligent supreme principle for that relation of nature to what is moral in us, in order to represent nature as purposive even in relation to the morally internal legislation and its possible execution” (KU 5:448). Kant has already provided an answer to this question in an earlier section of the text (namely §84; cf. also KU 5:382), and of course he argues that it is certainly not possible to clarify and determine the relationship in question unless we “go beyond the world” and seek an answer in an order of ends. Kant has already found his answer, and he believes that it also corresponds with common human understanding. This is the claim that only human beings standing under moral laws can be the final end of creation – and the emphasis here must be placed specifically on the expression “under moral laws” rather than on the human beings per se, in terms of their purely particular ends and interests. All of these earlier steps in the argument are simply recalled in order to develop the notion of moral teleology. For moral teleology then requires – and there is no question that it thereby goes “beyond the world” – that nature as a whole (the organisms, all the things and beings in the world) be considered from the perspective of the realizability of our calling to pursue the final end. We are therefore called – on practical grounds and also only  The importance of the interpersonal dimension of the doctrine of the highest good is elaborated in an essay by Engstrom (Engstrom : ). In addition, Engstrom convincingly shows that the doctrine of the highest good connects the theory of morality with the concept of happiness more closely than many interpreters assume and in this way claims to satisfy even our natural self-interest.

Applying the Concept of the Good

255

from the practical point of view – to subject nature in its entirety to the thought of the moral end, and not only to judge nature in its entirety as an instrument for the realization of our own ultimate end but also (if we pursue the argument to its ultimate conclusion) to treat nature correspondingly from the same practical perspective. For as Kant explains, appealing to an analogy with the political sphere, just as a complete description of the state involves not only the provision of a constitution but also the provision of an executive, since both of these necessarily belong to the (rational) concept of a state, the nomothetic of freedom also necessarily involves the nomothetic of nature in a moral teleology. We should therefore consider nature and the world as a whole from the perspective of the conditions required for realizing the final end, which is our moral calling. And this also implies that we should make it quite clear to ourselves that we are part of this world, that we cannot therefore simply isolate ourselves from it or from the other beings that inhabit it (as Spinoza believes), and also that we therefore cannot hope to realize the highest good independently of the conditions and relations of this world. And this implies – if we genuinely and consistently wish to live up to the demands of the practical moral law – that we must think of nature and the world as if a moral author of creation has disposed it in such a way that we are able to realize the final end in the world (although we cannot therefore necessarily infer that this is actually the case, as Kant emphasizes in §90 – cf. KU 5:464). For our setting the final end for ourselves in the first place depends on our assuming this view of the world (the subjective condition of the final end can only be fulfilled on this assumption). And we can also undertake to modify or transform the concrete “conditions of applicability” of the practical moral law precisely because (in contrast to Spinoza) we may entertain the hope that we can also realize the final end and the highest good in this world (and if not in this world then in a future life). In this specific form – namely through the connection between the moral demand to bring about what is “best for the world” and our teleological judgement of the world in terms of moral teleology – the doctrine of the highest good fulfils the programme formulated by Kant in the Introduction to the third Critique. If we proceed in accordance with this approach, then it is possible that the concept of freedom, admittedly the only “idea of pure reason” to do so, “proves its objective reality (by means of the causality that is thought in it) in nature, through its effect, which is possible in the latter” (KU 5:474). And so it is possible that on this practical path “the supersensible […] (freedom), by means of a determinate law of causality arising in it, […] demonstrates the fact of its reality in actions” (KU

256

Andrea Marlen Esser

5:474).²⁶ Nothing of this is possible in Spinoza’s “system of fatality”, where it would be excluded for systematic reasons and consequently dismissed as an illusion. From a theoretical point of view, of course, Kant insists that we cannot “reason out” anything more from these considerations. We are certainly not compelled to regard the existence of God as something that has thereby been theoretically proved, and we should not even imagine that we could legitimately infer it on these grounds. This brings us directly to the second point.

2 The New Logical Relation Between the Doctrine of the Highest Good and the Theory of Reflective Judgement In referring the doctrine of the highest good to the faculty of judgement, Kant clearly indicates, above all, the limits of the validity of the moral proof of the existence of God and thus also the limits of its legitimate application (cf. §88; KU 5:457). In §88 and the following sections, Kant once again clarifies the procedure and the utility of analogy as a method. This is the procedure that the power of judgement employs in order to elucidate certain issues and relations – and to determine their relevant sense at least in practical terms – where no theoretical knowledge of them is possible. And here I should like to indicate in outline what I take to be significant and important for the status of the doctrine of the highest good as far as these sections of the text are concerned. To begin with, it is important to note that Kant introduces what he calls a “practical power of reflecting judgement” (KU 5:456) in this connection (another expression new to his thought). For the theoretically reflective power of judgement could certainly develop a physical teleology in connection with the purposiveness of nature (which is indeed the true form of physico-theology) and venture the thought that an intelligent author of the world should be assumed in relation to the organization of nature. But this approach is also compelled to recognize that such an author of the world cannot be defined or determined any further in the context of physical teleology. For there is no path that leads from nature, from natural teleology, and therefore also from physico-theology, to a determinate concept of God. This is only possible if we adopt the moral or practical path of thought, and such an inference can only be drawn by a practical reflective power of judgement which employs the practical concept of the final end on  Cf. also Kant’s observation: “and that is the idea of freedom, the reality of which, as a particular kind of causality (although it would be extravagant of us to try and form a concept of it in a theoretical sense) can be shown through practical laws of pure reason and in accordance with these in real actions, and thus in experience” (KU :; my emphasis).

Applying the Concept of the Good

257

the basis of the moral teleology I have already mentioned. The inference drawn by this practical reflective power of judgement (reflective namely with regard to practical-moral concepts) is framed “only by analogy”. Thus it can only make the final end of the moral author of the world and his attributes “comprehensible” to us (i. e. we can only “think” it in analogy with our own intrinsic moral calling), and it certainly cannot furnish any knowledge in this regard. In short, the entire doctrine of the highest good is based on a concept by means of which the power of judgement creates “objects” for its own use: objects which ultimately possess no objective or even objectifiable character but are the results of “heautonomy” or “self-legislation”. On this basis, we also furnish ourselves with the notion of a highest morally-legislating author of the world precisely in order to be able to think the realization of our final end as something that is possible, and in this way to strive after manifest practical results (cf. KU 5:456). In projecting the notion of a moral author of the world, we can think of a supreme cause of the world for the purpose “of determining our own selves and our will in accordance” (KU 5:457) without thereby assuming the existence of such a cause or perhaps even attempting to discover what “it” could want of us. In the section where Kant attempts to clarify the specific status of this whole approach by referring to the use of analogy (§88), we also find an illuminating example of this analogical procedure. The example in question serves to remind us how often and how easily we employ explanations based on analogical inference. In this case, the example makes it clear that it is the realm of the practical on which the power of judgement draws for its teleological judgements of nature, and from which it also borrows the causa finalis, the second form of causality we recognize alongside the causa efficiens. And it makes clear – which is of capital importance – that the entire realm of the practical itself is only disclosed in an analogical manner. The issue is rather complicated, but I would like at least to sketch the essentials of the argument here. In §88 (KU 5:457), Kant writes that we ascribe a vis locomotiva (and thus a force) to the soul, and we do so because we observe that bodily movements are effected or initiated by our thoughts or representations (which thereby function as causes).²⁷ But we do not therefore ascribe this vis locomotiva to the soul itself in a factual (i. e. theoretical) sense, and we would never think that the soul is a “cause” that produces its effects in exactly the same way as the causes we recognize in the realm from which we have drawn this analogy (thus in terms  Cf. KU :: “We may name a cause after the concept which we have of its effect (though only in respect of the relation in which it stands to this effect). And we may do this without on that account seeking to define intrinsically the inherent nature of that cause by the only properties known to us of causes of that kind, which properties must be given to us by experience”.

258

Andrea Marlen Esser

of impact and pressure, etc.). For the soul is not a physical entity in any sense; it has no extension and therefore cannot satisfy the conditions under which we may also ascribe to it the properties of things that stand under the laws of motion. Nonetheless, although we cannot further determine this cause – which “we name […] after the concept that we have of its effect (though only with regard to its relation to the latter)” (KU 5:457), and which is only another expression for “end” – we still insist that it does in fact produce effects in the sensible world. What this passage once again makes clear, in my view, is that Kant is not engaged in promoting the “illusory construction” (MacIntyre 1981) of a further, intellectually unjustifiable and hence “metaphysical” (Williams 1972) form of causality. On the contrary, the entire realm of the practical and its specific form of (final) causality is itself disclosed through an analogy with the causality of “mechanism”, of the causa efficiens. In this way we try to make “comprehensible” to ourselves something that we recognize in our actual life-world, namely the fact that we accomplish actions in accordance with ends that we represent to ourselves and are thus capable of realizing these ends. But since we have recourse to analogy to explain this phenomenon, the situation in question remains “incomprehensible” to us from a purely theoretical or scientific perspective, as Kant already emphasized in the Groundwork for the Metaphysics of Morals (GMS 4:456) and in the Critique of Practical Reason (KpV 5:48). Extending this analogical approach to the realm of the practical, the third Critique assumes a moral author of the world, who “contains the ground of the possibility and practical reality, i. e., the realizability, of a necessary moral final end” (KU 5:457). This moral author of the world is then the very one who has created the world in which we discover causal relations (with regard to organisms and their relations to one another), which we can only explain by recourse to analogy and by drawing upon a certain conformity to law with which we are familiar in the realm of the practical. But from this practical perspective we may then interpret the purposive relations of organisms and their relations to one another as an indication that we inhabit a world that “accedes” to (or assists in) the realization of our final moral end. The status of the doctrine of the highest good as an analogically formulated exercise in moral teleology and the characterization of the moral author of the world which derives from it brings us to our third point, which I shall now outline.

3 The Critical Religious Implications of Kant’s Position First, it seems unmistakably clear in this context that all physico-theological attempts to determine the concept of God are doomed to fail since the predicates

Applying the Concept of the Good

259

that can be drawn from nature in order to characterize an “author of the world” are selected without recourse to legitimate criteria and must therefore remain arbitrary. The error of physico-theology lies in the fact that it observes nature from a privileged human perspective that is already presupposed, and then identifies natural features and properties of human beings through which man as such, namely as a part of nature, is already qualified as the final end of creation. When Kant says in his discussion of Spinoza that they “who were capable of having believed themselves to be the final end of creation” are hurled back “into the abyss of the purposiveless chaos of matter from which they were drawn” (KU 5:452), this almost cynical-sounding remark is surely directed specifically at the defenders of physico-theology. Second, it is clear that the concepts of God proposed by Spinoza and Epicurus – neither of whom can explain purposive relations in the world – must be criticized as essentially reductive. For both theories simply degrade these purposive relations, along with the notion of a causality of freedom, to the level of a mere semblance, without being able to explain how this semblance might arise in the first place. Finally, the doctrine of the highest good decisively subordinates religion to the sphere of morality. Thus the doctrine neither endangers the autonomous grounding of morality nor opens the door to the return of religious dogma. On the contrary, this subordination also signals an emphatic rejection of theologies that are really only “demonologies” (cf. KU 5:444). This is how Kant describes religions that endow their Gods with capricious human qualities or ascribe rage or favour to them (probably on the basis of special interests or power-political reasons of one kind or another). But the “honour of God”, as Kant says in the third Critique (KU 5:449n.), can only concern the moral law, for this is the only way, as the critical philosophy attempts to show, that the concept of God can be developed in a truly consistent fashion. Thus the concept of the supreme author of the world can only be further determined by reference to morality and its criteria, and cannot meaningfully be framed from the perspective of other standpoints (like those springing from self-interested or power-political considerations). In the context of the ethico-theological approach, all that can legitimately be defended is a critical religion of reason conceived under the limiting conditions of morality. And within this critical religion of reason, the concept of a supreme cause of the world can in turn only be understood as a moral author of the world, where this latter concept can in turn only enjoy the status of a “functional” rather than a “substantial” concept, to use the terminology of Ernst Cassirer (cf. Cassirer 2000: 20). Thus the concept of the supreme cause of the world can and may only be further developed in order to present the world as organized in such a way that the realization of the unconditioned claim of the moral law is possible. In this context, the substantive

260

Andrea Marlen Esser

character of any religion is then no longer simply a matter of arbitrary decision and can no longer be justified simply by reference to some given culture or tradition. Religious commandments and prohibitions can therefore no longer evade critical evaluation and investigation, for they too, like all other forms of human activity, are subject to moral judgement. Their claims must be examined and where necessary qualified or challenged in the light of moral, and not simply theological, criteria. As Kant himself implies, the conception of God which is compatible with morality cannot be that of a despotic deity, who insists on slavish obedience and prohibits all further “rational” enquiry, and nor can “articles of faith” be elevated to the status of supposedly factual claims or uncritically accepted commands. For, as the Critique of Judgement shows us, the entire realm of religious thought rests on the ground of analogy. When we participate within a religious “language game” we must also remember that claims and statements in the communicative context of religion serve to give metaphorical expression to fundamental experiences of our life-world. As Kant puts it in the Metaphysics of Morals, such claims and statements present our duties as “divine commandments” and are thus really aesthetic means of supporting our practical orientation and helping us to realize it in practice, even though this orientation is grounded solely in the moral law.²⁸ As Kant trenchantly expresses this thought in the Opus Postumum: “God is not a being outside of me, but merely a thought in me. God is morally practical self-legislating reason” (Op. Post. 21:145). God cannot (and therefore need not) be understood as an intervening power, who helps us immediately as individuals to attain the happiness we deserve. Rather, God is to be thought of as the moral author of a purposive world of nature which “assists” our vocation as a final end. The distinctive feature of the doctrine of the highest good in the Critique of Judgement, it seems to me, is that this idea of the moral author of the world is not derived solely from our moral vocation as a final end. Rather, Kant develops the idea by bringing a specific experience of nature and the principle of purposiveness, with which we can explain it, into relation with our moral vocation, one step at a time. In terms of a complete explanation of nature as a whole, this principle certainly leads us “beyond the world”; but on its basis we can explain empirically given objects, successfully make judgements regarding them, take a specific aesthetic delight in them, and finally even make productive use of this kind of explanation for framing hypotheses in the context of natural scientific inves-

 Cf. MS :.

Applying the Concept of the Good

261

tigation and stimulating progress in the explanation of causes.²⁹ We can regard all these forms of experience as indications that we too fit or “belong in the world” and that our moral vocation as a final end can be realized there. Of course, there is no immediate moral obligation for us to adopt this view of the world; but we are called upon to think in a “morally consistent” fashion. This can only mean that we do not merely will the end, i. e. the good, but also recognize that in addition we are commanded to will the means that are required to realize the adopted end. For otherwise, as Kant’s theory of hypothetical imperatives shows, we must relinquish the end. The doctrine of the highest good and the specific attitude to the world it opens up is required in order for us to improve the conditions for applying the good and thereby also to strengthen ourselves in all our moral efforts. Considered from this perspective, the doctrine of the highest good reveals an aspect of Kant’s later thought which might almost be described as “pragmatic”, an aspect which surely deserves our full attention.

References Bayle, Pierre (ed.) 1826, A Historical and Critical Dictionary, Vol. 3, London: Hunt and Clarke Bayle, Pierre 2003, “Spinoza”, in: Pierre Bayle 2003, Historisches und kritisches Wörterbuch. Eine Auswahl, ed. and trans. by Günter Gawlick and Lothar Kreimendahl, Darmstadt: Wissenschaftliche Buchgesellschaft, 367–439 Beck, L.W. 1960, A Commentary on Kant’s Critique of Practical Reason, Chicago/London: Chicago University Press Beiser, Frederick C. 2006, “Moral Faith and the Highest Good”, in: Paul Guyer (ed.) 2006, The Cambridge Companion to Kant and Modern Philosophy, Cambridge: Cambridge University Press, 588–629 Cassirer, Ernst 2000, Substanzbegriff und Funktionsbegriff, in: Ernst Cassirer, Gesammelte Werke, Vol. 6, Hamburg: Felix Meiner Verlag Cohen, Alix 2009, Kant and the Human Sciences: Biology, Anthropology and History, London: Palgrave Macmillan Engstrom, Stephen 1992, “The Concept of the Highest Good in Kant’s Moral Theory”, Philosophy and Phenomenological Research 52, 747–80

 Förster concludes that “[t]ranscendental philosophy must remain undecided as to whether the highest good is possible or not, or whether there is a final end of creation” (Förster : ). This position from the third Critique, as Förster emphasizes, shares the philosophical agnosticism of the Critique of Pure Reason and remains, even in Kant’s late works, only an ideal of self-constituting practical reason. Although I do fully agree with this position, I want to complete it with the hint Kant gives in § of the Critique of Judgement. Here, the antinomy of judgement is solved by demonstrating how the two “maxims” of judgement work hand in hand, such that it is possible to prove the effects of our self-constituting practical reason as events in the world.

262

Andrea Marlen Esser

Esser, Andrea-Marlen 2004, Kants Tugendlehre in der Gegenwart, Stuttgart-Bad Cannstatt: Frommann-Holzboog Verlag Förster, Eckart 2010, “What is the ‘Highest Point’ of Transcendental Philosophy?”, in: Pablo Muchnik (ed.) 2010, Rethinking Kant, Cambridge: Cambridge Scholars Publishing, 257– 74 Guyer, Paul 2003, “Beauty, Systematicity, and the Highest Good: Eckart Förster’s Kant’s Final Synthesis”, Inquiry 46, 195–214 MacIntyre, Alasdair 1981, After Virtue, Notre Dame: University of Notre Dame Press Pasternack, Lawrence 2011, “The Development and Scope of Kantian Belief: The Highest Good, The Practical Postulates and The Fact of Reason”, Kant-Studien 102, 290–315 Reath, Andrews 1988, “Two Conceptions of the Highest Good in Kant”, Journal of the History of Philosophy 26, 593–619 Sala, Giovanni 1990, Kant und die Frage nach Gott: Gottesbeweise und Gottesbeweiskritik in den Schriften Kants, Berlin/New York: Walter de Gruyter Schroeder, Winfried 2004, “Zwei ‘tugendhafte Atheisten’. Zum Verhältnis von Moral und Religion bei Pierre Bayle” in: Karl Eibl/Norbert Hinske/Lothar Kreimendahl/Monika Neugebaur-Wölk (eds.) 2004, Aufklärung, Hamburg: Meiner, 9–20 Williams, Bernard 1972, Morality: An Introduction to Ethics, Cambridge: Cambridge University Press Wood, Allen W. 1992, “Rational Theology, Moral Faith, and Religion”, in: Paul Guyer (ed.) 1992, The Cambridge Companion to Kant, Cambridge: Cambridge University Press, 394– 416 Zammitto, John 1992, The Genesis of Kant’s Critique of Judgement, Chicago: Chicago University Press

Günter Zöller

“The supersensible … in us, above us and after us”: The Critical Conception of the Highest Good in Kant’s Practico-Dogmatic Metaphysics “der alte Lampe muß einen Gott haben” (Heine 1976, vol. 5, 604)

This contribution addresses Kant’s ethico-teleological conception of the “highest good, insofar as it is possible in the world” (Fortschritte 20:294) in his late fragmentary work, Prize Essay on the Advances of Metaphysics. ¹ The focus is on the practically based validation of theoretical concepts of reason and on the teleological nature of the unity of theoretical and practical reason in Kant. Special consideration is given to the fabricated character (“made”, “supposed”) of the rational concepts of freedom, God and soul in Kant’s post-critical metaphysics.² Section 1 details the change in teleological thinking and in thinking about teleology between Aristotle and Kant. Section 2 explores the extent of the analogy drawn by Kant between theoretical and practical reason. Section 3 presents Kant’s teleological conception of the unity of theoretical and practical reason. Section 4 details Kant’s integration of the highest good into a critically warranted practico-dogmatic metaphysics.

1 On Purpose During most of its history, Western philosophy has been shaped by teleological thinking. Under the formative influence of Aristotle’s work in physics, metaphysics, natural history, ethics and politics, philosophy in the ancient, medieval and early modern periods has conceived of entities of all kinds and events of all sorts as structured by and functioning in accordance with purposive principles that regulate their being and becoming. In its original, Aristotelian form the teleology of all things natural and supranatural had operated without explicit recourse to

 Here and elsewhere in this contribution all translations from Kant, including the very title of the work by Kant under consideration, are by the author.  On the relation between critique and metaphysics in Kant, see Zöller  and .

264

Günter Zöller

divine design. The purposive properties were considered to reside in the nature of things themselves. With the advent of Christian theology in Western philosophy, purpose took on the additional trait of intention reflecting supreme intelligence and superior wisdom. In technical terms, Aristotelian purposiveness was a form of causality, reflecting a fourth kind of cause (aitia), in addition to the causality of form, of matter and of motion (causa formalis, causa materialis, causa efficiens, causa finalis). But the finality involved in Aristotelian teleology was not based on the setting, adopting or pursuing of ends. Rather it consisted in ends being embedded and operative in the workings of the world and the beings in it. Even when, as in human action by deliberation and choice, ends could be said to be pursued intentionally, or “on purpose”, the finality of such action was governed, normatively, by a teleological order reflecting the nature of things, including the nature of rational animals of the human kind. Objective teleology prior to personal purposes and individual ends also governed Aristotelian ethics and its extension into politics. Formally considered, the inherent end (telos) of all human action – that “for the sake of which” (hou heneka) an action was undertaken – was some “good” (agathon), the latter being the generic term of approval for an end naturally given or humanly elected. On Aristotle’s account, the multiple ends of human action along with the types and tokens of “the good” they involved formed a hierarchy of ends serving as means to other, higher ends that culminated in what was good in and of itself or the highest (human) good. In contentual terms, Aristotle identified the highest good with eudaimonia, variously rendered as “living well”, “doing well” or “happiness” and understood as virtue-guided activity in one’s social comportment. By contrast, Aristotle rejected alternative forms of ethics – and of politics – that equated the highest good with pleasure (hedone) or honor (time), and he pointedly ignored the pursuit and accumulation of wealth (chremata), which was deemed to turn a mere means into an altogether unworthy end in itself. For Aristotle the highest good under the guise of happiness (eudaimonia) also represented a good that was both complete (teleios) and self-sufficient (autarkes). A serious challenge arose for Aristotelian teleology with the development of modern experimental science and its allied mechanistic metaphysics of nature, especially in the latter’s materialist variant. By reducing the causality of nature to relations of cause and effect (of the Aristotelian kind of efficient causality), modern physics eliminated ends or purposes from the range of causal factors to be found and measured in nature. The absence of intention and the denial of design behind the order of nature were particularly poignant in the materialist account of the world, along with the human beings in it, to be found in the ma-

“The supersensible … in us, above us and after us”

265

terialism and mechanism of Thomas Hobbes and in the reputedly atheist and fatalist naturalism of Baruch de Spinoza. The suspension of teleological thinking even affected modern political philosophy, to the extent that the functioning of social entities was seen as distinct from the intentions and interest of the individuals involved. In Bernard Mandeville’s fabulous account of modern commercial society (The Fable of the Bees) personal preferences that flagrantly forego common mores for selfish interests (“private vices”) unintentionally result in an overall social order marked by wealth and well-being (“public benefits”). On Adam Smith’s less cynical account, the socio-civic whole exhibits properties that exceed the reach and grasp of individual planning and pursuit and are due to a supervenient systemic surplus (“invisible hand”). The philosophy of Immanuel Kant partakes in the modern critique of teleological thinking. In theoretical philosophy, effectively reduced to a non-empirical theory of the grounds of nature and the principles of natural science, Kant maintains blind mechanism in the governance of things and events. In natural history, formerly the chief province of Aristotelian teleology, the critical Kant reduces purposiveness from an objective property to a subjective principle (“maxim”) of reflecting on biological structures and functions. In philosophical theology, Kant’s critique of reason does away with the argument from design, along with all the other theoretical proof pretensions of the tradition. Kant’s departure from Aristotelian teleology is most remarkable, though, in his mature moral philosophy. Kant demotes the pursuit of happiness from its former status as the very principle of ethical conduct, replacing it with the principle of rational obligation that imposes the absolute condition of universal form on an individual’s self-sought rules of conduct (“maxims”).³ Moreover, Kant redefines the ultimate norm of ethical pursuit (“unconditional good”) in strictly non-objective terms by locating it not in some entity or state to be obtained but in the moral qualification of the subjective principle of human action (“the good will”) (GMS 4:393). With the moral law and the good will fulfilling the former function of Aristotelian happiness, Kant moves on to redefine the highest good in formal as well as contentual terms. Formally speaking, the highest good now assumes the role of the ideal sum-total of human endeavors, a remote goal ever to be approached and never to be reached. Formerly a form of life, pursued for its own sake and cultivated for its own good, the highest good has been turned into the prepara-

 On the seismic shift from a teleological to a deontological conception of ethics in Kant, see Zöller a.

266

Günter Zöller

tion and premonition of an afterlife. In material terms, the highest good looses its previous complete identification with happiness by having the latter placed under the necessary condition of morality (Sittlichkeit), which is outright defined as the normative qualification (“worthiness”) for happiness (KrV A806/B834). In a further move, Kant merges the ethico-cosmological sense of the highest good as the ultimate end of all pursuits “in the world” with its ontico-theological significance as the perfect entity residing “outside the world”. The complex composition of the highest good in Kant thus links metaphysical and moral motives. From a systematic perspective, Kant’s doctrine of the highest good serves to fill the void left by the pointed destruction of traditional metaphysics and to replace the obsolete object of speculation with the global goal of striving.

2 Theoretical and Practical Reason in Comparison The twofold character of the highest good in Kant – as involving ethical ends and as including metaphysical objects – reflects its chief systematic task of assuring the unity of reason amidst the latter’s specific differentiation into a theoretical and a practical use or employment. Aristotle and an entire tradition following him had distinguished theory (theoria) and practice (praxis) as generically distinct life forms (bioi) with characteristically different epistemic status and ethical standing. In particular, Aristotle had limited the scientific mode of cognition (episteme), as the most elevated rational life form, to the serene contemplation of objects, including the divine being, and assigned the acquaintance with ethical and political matters to practical insight (phronesis) subject to lesser epistemic requirements. For Kant, by contrast, both theory and practice are subject to the standard of “one and the same” reason (KpV 5:121). In particular, theoretical and practical cognition, when guided by the rules of reason, admit both of apodictic certainty in ascertaining what is the case (theory) or what ought to be the case (practice), respectively. Still Kant acknowledges the irreducible difference between the theoretical and the practical employment mode of reason, with the former involving the determination of objects for theoretical purposes (Gegenstandsbestimmung)⁴ and the latter the determination of the will for practical purposes (Willensbestimmung). Unlike both his rationalist predecessors and his idealist successors, who sought to retrieve the root of reason this side of its divide into theoretical and practical reason, Kant links the epistemic elevation of practical reason, which brings the latter

 For a systematic reconstruction of Kant’s account of (theoretical) objective reference, see Zöller .

“The supersensible … in us, above us and after us”

267

on a par with theoretical reason, with its systematic separation from theoretical reason. The graphic equivalent of the relation between theoretical and practical reason in Kant therefore is neither the botanical branching nor the maritime merging of the two employment modes of reason but their perpetual parallelism, a close proximity that remains constantly so, combining equality with equidistance. Logically speaking, the relation between the parallel conceptual universes of theoretical and practical reason in Kant introduces analogy as the chief tool for assessing the similarities and the difference between reason’s chief modes. On Kant’s account, analogy in the philosophically relevant sense consists not in the “imperfect similarity of two things” but in the “perfect similarity of two relations between entirely dissimilar things” (Prol 4:357). An analogy thus involves the close comparison of relations in one domain, such as that of theoretical reason, with relevantly similar relations from another domain, such as that of practical reason. As part of his distinction and demarcation of theoretical and practical reason, Kant provides detailed considerations about the extent as well as the limits of the analogies to be drawn between theoretical and practical reason. A particularly instructive instance of Kant’s self-interpretive stand on the relation between theoretical and practical reason is the concluding section of the Analytic of the Critique of Practical Reason, entitled “Critical Elucidation of the Analytic of Pure Practical Reason” (KpV 5:89–106).⁵ In the Critical Elucidation Kant notes the role of reversal in the architectonic move from the Analytic of the first Critique, with its ascent from sensible intuitions (space and time) through concepts (categories) to judgments (principles of the pure understanding), to the second Critique’s systematic sequence from principle (moral law) through concepts (good and evil) to sensibility (moral feeling). Moreover, Kant contrasts the recourse to the fact of an established a priori science, which serves as evidence for the validity of pure theoretical reason, with the pre- and extra-scientific “fact” (Faktum) of morality and its basis in common practical reason (KpV 5:31). A further feature singled out by Kant for the analogical presentation of theoretical and practical reason is the common conception of autonomy or self-legislation that links the two employment modes of reason as so many basic ways of imposing lawful order onto some domain. To be sure, the term “autonomy” first is introduced by Kant for the specific purpose of identifying the peculiar order imposed among the principles of human volition (maxims) by the formal condition of universal legislation, as expressed in the moral meta-law (Sittengesetz) (GMS 4:436). Moreover, the term “autonomy” derives from the sphere of pol-

 On the systematic function of the Critical Elucidation in the second Critique, see Zöller .

268

Günter Zöller

itics, where it designates the condition of independence and self-rule that obtains within a state or between states. But in addition to having the term “autonomy” designate the legislation of practical reason for a possible “system of freedom” (KrV A815/B843). Kant also draws on the notion of autonomy to convey the legislation of theoretical reason – under the guise of the understanding (Verstand) – for a possible “realm of nature” (KrV A812/B840). The negative feature of independence from foreign rule and the positive feature of self-rule inherent in Kant’s politico-practical concept of autonomy thus is carried over to the rule of reason in the imposition of the objective principles (laws) that turn the undetermined objects of sensible intuition (appearances) into logically determined objects of nature (phenomena). The shared feature of the legislation of nature and the legislation of freedom is the crypto-political notion of there being, in each case, a mass (manifold) in need of being ruled and regulated into some shape, form or order and, moreover, as amenable, even open and ready for such treatment. The extended analogy between theoretical and practical autonomy notwithstanding, Kant ultimately resists a more far reaching assimilation of theoretical reason to the standards of practical reason and pointedly avoids merging the two employment modes of reason into one. In particular, he seems to shy away from applying the term closely connected with the practical conception of autonomy, viz., freedom, to the autonomy of theoretical reason (understanding). Instead Kant attributes the independent and self-active character of the theoretical legislation of nature to the mind’s “spontaneity”, as opposed to its passive, factual feature of “receptivity”. Moreover, the specifics of Kant’s account of the workings of (transcendental) spontaneity in the establishing of the order of nature, including the foremost formal character of spontaneous thinking and the latter’s constitutive need for material provisions to be supplied by sensory intuition, differ sharply from the self-sufficiency of (pure) practical reason and offer little to support an autonomist reading of reason tout court in Kant. In addition to rendering the relation between theoretical and practical reason by means of parallels drawn and analogies established, Kant seeks to connect the two main modes of reason’s employment by detailing their complementary relation to each other. In particular, he cites the fit between the restriction of theoretical reason to objects in space and time (appearances) and the reach of practical reason beyond the confines of experience. The conceptual space left open by the self-limitation of theoretical reason (things in themselves) matches exactly the self-assertion of (pure) practical reason in its occupation of the intelligible world as the world of (moral) freedom. Throughout Kant stresses the balance observed or realized in the relation between the two main modes of reason, with one providing what the other is lack-

“The supersensible … in us, above us and after us”

269

ing, and vice versa. The resulting composite effect of the reciprocal relation between theoretical and practical reason is one of equilibrium and stability. In such a comparative and conjunctive perspective, it is not the similarity between theoretical and practical reason, much less their secret, radical identity that assures the regimen of reason but the compensatory character that they possess for each other. Theoretical and practical reason emerge as the two sides of a set of scales that strike a balance in which no side can stay in place without the opposed support of the other one.

3 The Teleological Unity of Theoretical and Practical Reason The libral linkage of theoretical and practical reason iconically indicates Kant’s overall strategy for addressing and articulating the unity of the twin modes of reason. Rather than exploring and exposing their elusive joint origin, Kant focuses on their cooperative constellation. The unity of theoretical and practical reason that emerges in Kant’s account is not the unity of a pre-disjunctive origin but that of a common function and a joint purpose. In Kant theoretical and practical reason are united in teleological rather than archaeological terms. Moreover, the teleological unity of reason in Kant does not simply consist in one of reason’s major modes being the means to the other one’s end. The teleology involved is not modeled on the practical purposiveness of probate means for chosen ends. Instead of being based on the model of purposive action, the teleology attributed to the relation between theoretical and practical reason in Kant is informed by the paradigm of natural teleological processes, chiefly those to be found – or rather supposed – in organic beings and their peculiar life form. The tertium comparationis of reason and organism in Kant’s philosophical imaginarium is the notion of system, which combines the orderedness and stability of an architectonic structure with the animateness and functionality of an organic whole. On Kant’s account, reason in its entirety is structured and organized in two chief spheres that jointly make up reason in its integral functionality. In considering reason as a teleological system, Kant manages to preserve the wholeness of reason as a basic, unitary and comprehensive intellectual capacity (Vernunftvermögen) over and against its division into parts and pieces without turning reason into some essentialist entity that precedes its specific modes and manifestations as their self-same origin. As in the organic being to which it is likened, reason not so much precedes and produces its theoretical and practical employment modes, as that it consists in their distinct but joint articulation as the twin manifestations of a common capacity that has no realization apart

270

Günter Zöller

from its original differentiation. In Kant’s systematico-teleological perspective, reason is turned from substance into function. The teleological conception of reason’s identity and difference in Kant is not limited to its organic unity amidst the functional difference between theoretical and practical reason. Kant also conceives of each of the employment modes of reason as itself teleologically structured and purposively operative. In particular, he has each form of reason be subject to a specific guiding rule of engagement that orients as well as motivates the deployment of reason in each of its major modes. Kant’s technical term for the directedness and drivenness of reason-inuse is “interest” (KrV A804 f./B832 f.; KpV 5:119 f.), a term that lends a proto-practical purposiveness to what otherwise might seem a self-sufficient sum-total of rational rules removed from application and advancement. When Kant declares accordingly that all interest on the part of reason is “ultimately practical” (KpV 5:121), this is not to suggest that there is but one basic type of interest, the practical kind. Rather the practical nature of all interest conveys the generic character and function of interest to assure the actual application of reason and to turn the mere “faculty” (Vermögen) of reason into the actual activity of reason. Moreover, the priority attributed to practical reason reflects Kant’s general view of the systematic subservience of theory to practice and of the natural to the moral – an order of rational ranking that does not exclude and even requires the independent and “autonomous” operation of reason in each of its specific employment modes. The specific interest attributed by Kant to theoretical reason and propelling the practice of theoretical reason is the determination of objects, which aims at ascertaining their forms and features by means of reason’s cognitive capacity (Erkenntnisvermögen). By contrast, the specific interest of practical reason, so Kant, consists in reason’s determination of the will in accordance with reason’s conative capacity (Begehrungsvermögen) in order to identify and impose the moral qualification of willing. Kant epitomizes the contrast between the purposive orientation of theoretical and practical reason by correlating each of them with a specific critical question that addresses the intended end of the respective exercise of reason. Reason’s theoretical interest culminates and coalesces in the question “What can I know?” (KrV A805/B833), which places the cognition of objects under the critical concern with the maximal extent and thereby with the grounds and bounds of theoretical reason. By contrast, the practical interest of reason is conveyed by the question “What shall I do?” (KrV A805/B833), which places the formation of the will under the critical concern with the required universal form of morally qualified willing. The answer to the former question is given by the first Critique’s principal restriction of theoretical cognition (“knowledge”, Wissen) to possible experience. The first Critique’s answer to the latter question is the nor-

“The supersensible … in us, above us and after us”

271

mative restriction of actions to those that morally qualify the agent to the measured award of the agent’s naturally given final purpose, viz., the agent’s own happiness (Glückseligkeit). To be sure, Kant was to replace the answer to the question involving practical reason given in the first Critique, which was cast in eudaemonistic terms, even though it did not involve a eudaemonistic principle of morals (KrV A806/B834), with a deontologically dimensioned answer under the guise of the categorical imperative in the second Critique. But Kant retained the inclusion of happiness into the answer to the third question concerning reason’s interest, first asked and answered in the Critique of Pure Reason and subsequently developed in greater detail in the Postulates of Pure Practical Reason of the Critique of Practical Reason. The third question concerning the interest of reason guiding the latter’s employment that Kant poses in the first Critique does not represent a further, third mode of reason-in-use though. Rather the third question combines the different and separate interests of theoretical and practical reason, as expressed in the first and second question, into the complex, theoretically as well as practically natured question “if I do what I ought to do, what then may I hope for?” (KrV A805/B833).⁶ Kant’s third question is remarkable for its conditional form, which places the possibly permitted object of hope under the presupposition of acting morally. Moreover, in its basic set-up the composite third question combines the practico-moral requirement, considered under the condition of its sufficient satisfaction, with the theoretico-speculative perspective to its hoped-for consequence. On Kant’s conception, the desired object of rationally justified hope is human happiness (Glückseligkeit), more precisely an extent and degree of such happiness that is warranted by the individual agent’s prior moral conduct. The antecedent of the third question’s conditional operates in reason’s practical register, which involves the prescription of moral conduct. By contrast, the conditional question’s consequent belongs to the theoretical domain, which is concerned with the determination of an object. The logical sequence of the third question thus links practical and theoretical determination by moving from practical rules, under the condition of their compliance, to a reasonably to be hoped for object, viz., a future state of happiness. The hoped-for object as such is not morally required and hence not practically natured, as are the moral rules that condition it. Rather the happiness reasonably to be hoped for is a future state of affairs and as such the theoretically determined object of a cognition of sorts.

 On the composite character of the “third question”, see Zöller .

272

Günter Zöller

To be sure, the cognition of some future morally warranted happiness as the object of reasonable hope does not concern an object of experience, not even of possible experience. After all the happiness in question is not given or possibly given but only to be hoped for. Still the object, or rather the state, of such practically warranted hope is not itself practical like the rules of conduct, but theoretical as an object of a cognition envisioned and embraced as an expected commensurate consequence of prior moral conduct. On Kant’s account of the third question’s complexion of theoretical and practical modalities, the practical serves as the “guiding thread” (KrV A805/B833) for the theoretical – with the fulfillment of a specifically practical requirement (morality) in effect warranting the extension of specifically theoretical cognition to otherwise unknown, even unknowable objects or states (future happiness). Moreover, on Kant’s account, the extraordinary object of morally warranted hope is not exhausted by some (future) morally warranted state of happiness. In order to lend credibility to the actual obtainment of such happiness as the object of moral hope further conditions have to be introduced and – for that matter – included as non-empirical (quasi‐)objects of morally justified hope. In particular, Kant introduces – already in the Critique of Pure Reason and again in the Critique of Practical Reason and also in the Critique of the Power of Judgment – the post-existence of human spiritual being (“immortality of the soul”) and the existence of a perfectly capable judge of moral conduct (“God”) as the two necessary conditions for the possible realization of morally warranted happiness in beings who do not have at their own disposal the means for turning merit into award and desert into receipt. As in the case of merited happiness, human post-mortem existence and the existence of a most perfect divine being are not part and parcel of the practical norms of morality as such. Rather they are to be regarded, on Kant’s account, as theoretical implications of morality’s norms – and as exceptional extensions of theoretical cognition into a domain otherwise inaccessible to humanly possible cognition, viz., the supersensible, at that. With the explorations of the requirements for possible morally deserved happiness completed, the objects of morally warranted hope include, in addition to the hoped-for morally warranted future happiness itself, that of a future life and of a most perfect being. Taking up traditional terminology but putting it to a characteristically different use, Kant calls the object of morally warranted hope the “highest good”. More specifically, he regards morally merited happiness as the “dependent” highest good and the divine being in its quality as the arbiter of enacted morality and the guarantor of morally deserved happiness as the “independent” highest good (KrV A810 f./B838 f.). Generally speaking, the designation “highest good” recognizes and conveys the status of morally conditioned happiness, along

“The supersensible … in us, above us and after us”

273

with its (quasi‐)theoretical presuppositions, as the ultimate objective involved in human practical pursuits and by extension in human existence altogether. To be sure, in Kant the title “the highest good” does not denote the principle or the ground of human moral conduct, as it did in the Aristotelian tradition, but only the implied complete object of that principle. The principle itself, in Kant, resides in the formal or meta-law of human rational practice (moral law).

4 Metaphysics in a Practico-Dogmatic Regard After its initial presentation in the Canon of Pure Reason of the Transcendental Doctrine of Method in the Critique of Pure Reason, which introduced the immortality of the soul and the existence of God as subjectively necessary presuppositions of successful moral activity, Kant refined the doctrine of the highest good in the Dialectic of Pure Practical Reason of the Critique of Practical Reason. The second Critique distinguished even more clearly between the originally independent and self-sufficiently effective principle of pure practical reason (moral law), which possesses necessary validity and possible efficacy on its own, on the one hand, and the ideally complete object of human willing, which essentially includes happiness to the extent of being morally merited and involves a freely adopted, rationally grounded belief (freies Fürwahrhalten) or rational faith (Vernunftglaube) in metaphysical objects, on the other hand. The Doctrine of Method of the Critique of the Power of Teleological Judgment in the Critique of the Power of Judgment further detailed the epistemology of rational faith in the highest good by placing it in the context of critical considerations on the ethico-theological argument for God’s existence (moral theology). Kant was to take up the doctrine of the highest good yet another time in his later fragmentary contribution to the Berlin Academy’s prize question about the recent advances in metaphysics from 1793.⁷ Kant produced three incomplete and overlapping drafts of his answer to the prize question, which were edited into a quasi-complete treatment of the topic by Rink (1804), who used the first draft for the first half, the second draft for the second half and the third draft for a supplement to constitute Kant’s answer. Within the Academy Edition the work figures among Kant’s literary remains (20:253–311 and 313–32).⁸ The edition of Kant’s works by Wilhelm Weischedel includes the late Prize Essay under the  On the prize competition and Kant’s abandonned participation in it, see Kant , vol. , –. See also the introduction by the editor-translator in Kant : –.  The quotations of terms and phrases from the late Prize provided in the remainder of this section disregard emphases and declensions to be found in Kant’s text.

274

Günter Zöller

writings in metaphysics and logic (Kant 1968, vol. 6, 585–676). In the revised version of the Academy Edition currently under planning and in preparation the work is scheduled to be included among Kant’s published works (vol. 9). The late Prize Essay, while lacking a final authoritative form, constitutes an important document of Kant’s philosophical self-interpretation. In particular, it situates Kant’s critical philosophy in relation to the metaphysical tradition past and present. Moreover, it combines a focus on theoretical philosophy in general and the “metaphysics of nature” (Fortschritte 20:293) in particular with the final inclusion of an ethico-theological perspective that links natural and moral teleology under the conception of the highest good. The work’s key concept for Kant’s assessment of the success of metaphysics past and present, “Fortschritte”, is only inadequately rendered by the customary English translation as “progress”. The wording of the prize question as well as the terminology favored in Kant’s attempts at an answer employ a plural formulation (Fortschritte) that allows to distinguish distinct advancing steps undertaken over the extended historical course of metaphysics, especially in recent times, for which Kant lists three such steps (der erste Schritt, der zweite Schritt, der dritte Schritt; Fortschritte 20:266, 261, 262). Moreover, on Kant’s account, historically the presence of progression in metaphysics has been challenged by the occurrence of regression in the field, such that the “forward steps” (Fortschritte) possibly achieved in metaphysics are obstructed by the incidence of “recession” (Rückgang; Fortschritte 20:263), with the additional intervention of a complete lack of movement on the part of metaphysics resulting in “standstill” (Stillstand; Fortschritte 20:281). On the whole, Kant distinguishes in the late Prize Essay three phases in the development of metaphysics, for the designation of which he employs the clinical term “stadium” (Stadium), indicating a course of development that involves pathology as much as cure. In particular, Kant records the sequence of dogmatic pseudo-advancement, skeptical stalemate and critical completion (Vollendung; Fortschritte 20:281) in the overall progression of metaphysics. The concept of metaphysics underlying Kant’s analysis is doubly defined: in terms of its “final purpose” (Endzweck) as the “science of progressing from the cognition of the sensible to that of the supersensible by means of reason” (Fortschritte 20:260) and in terms of its epistemic status as the “system of all principles of pure theoretical cognition of reason by means of concepts”, in short, the “system of pure theoretical philosophy” (Fortschritte 20:261). The first, finalist definition of metaphysics suggests the ultimate orientation of metaphysics as a science (Wissenschaft) toward metaphysics as a “doctrine of wisdom” (Weisheitslehre; Fortschritte 20:261) and hence a close connection to the necessary practical, i.e., moral use of reason. By contrast, the second, epistemological

“The supersensible … in us, above us and after us”

275

definition confines metaphysics to the sum-total of the principles of theoretically employed reason, at the explicit exclusion of “practical doctrines of pure reason” (Fortschritte 20:261). Still Kant pointedly includes those theoretical principles within the purview of his narrow epistemological definition of metaphysics that are presupposed for very possibility of pure practical or moral philosophy. Kant’s exclusion of pure practical philosophy (“metaphysics of morals”; Fortschritte 20:293) from the orbit of his answer to the prize question has a strategic reason; it allows him to focus exclusively on the theoretical means available for achieving, or lacking, success in past and present metaphysical investigations. Still the possible progress of metaphysics, as diagnozed by Kant, while not taking place within the confines of a metaphysics of morals, involves a “practical intent” (praktische Absicht; Fortschritte 20:292) which alone proves capable of procuring a measure of validation to concepts of the supersensible (“ideas”) that remain elusive from a strictly theoretical, “speculative” point of view. In particular, Kant draws on the antinomy of pure reason, as presented in the Critique of Pure Reason, in order to show that the “skeptical stalemate” (Stillstand der reinen Vernunft) issues in “dogmatic advances” (dogmatische Fortschritte) that afford a cognition of sorts regarding a supersensible entity removed from possible theoretical cognition (Noumenon, Sache an sich) (Fortschritte 20:292). More specifically, Kant cites the freedom of the faculty of choice (Freiheit der Willkür), insofar as it stands under moral laws, as that supersensible which allows a quasi-theoretical, para-cognitive determination, and a twofold such determination at that: as “actually given in the subject” (wirklich im Subject gegeben) and as determinative in view of an object, albeit only “in a practical regard” (in praktischer Rücksicht) (Fortschritte 20:292). The reality of the freedom of choice that Kant locates in the subject refers to the latter’s generic consciousness of moral obligation and to the ability for moral conduct entailed thereby. By contrast, the object of freedom under moral laws, while not to be cognized in a specifically theoretical regard, is said to constitute an object of practically informed dogmatic cognition, of “practico-dogmatic cognition”, as opposed to a cognition of the “theoretico-dogmatic” kind (Fortschritte 20:281). According to Kant, this hybrid form of cognition – dogmatically geared and practically based – constitutes the “proper final purpose of metaphysics” (eigentliche Zweck der Metaphysik) (Fortschritte 20:292). More specifically, Kant identifies the two dynamical ideas and the associated third and fourth antinomial conflicts from the first Critique, with their featured progressive synthesis of heterogeneous elements (cause and effect, the contingent and the necessary), as suitable for the transition required of metaphysics from the sensible to the supersensible, and as the only suitable such transition at that.

276

Günter Zöller

Accordingly, the ingredients of the object constituting the final purpose of metaphysics are “the supersensible in the world (the spiritual nature of the soul) and that outside the world (God), hence immortality and theology” (Fortschritte 20:292). In including among theoretical metaphysics the quasi-theoretical implications of practical freedom and its moral laws, both with regard to the subject and to the object of morally ruled freedom, Kant effectively has opened up metaphysics for the inclusion of the highest good and its two key components, psychic immortality and divine existence. The latter two constituents of the complete object of pure practical reason emerge themselves as objects of a kind of cognition that is different from the theoretical determination of objects through the pure concepts of the understanding (categories), the employment of which is restricted to objects of possible experience. As instances of a practically informed cognition concerning things supersensible, the cognition of the highest good, for Kant, constitutes a case of critically warranted dogmatic cognition on the part of metaphysics. To be sure, the object of such cognition is not determined in strictly theoretical terms. Rather the soul in its spirituality and God in his actuality are objects of a cognition that is based on and conditional upon the reality of freedom and its (moral) laws. Still the cognition so obtained is (practically) certain and its object (practically) real. Based on the general suitability of the dynamical ideas of pure reason for effectuating the cognitive transition from the sensible to the supersensible required of metaphysics, Kant’s offers a specific argument for the projected “practico-dogmatic step-over (Überschritt) of the metaphysics of nature” (Fortschritte 20:293) that turns on the “teleology of nature” (Teleologie der Natur; Fortschritte 20:294). On Kant’s account in the late Prize Essay, which in turn is informed by the Critique of the Teleological Power of Judgment in the third Critique, the complete cognition of nature includes a peculiar concept that is imported into nature in order to render intelligible particular natural properties, which otherwise would remain unintelligible. The concept in question originates in reason and is not properly part of the object to which it is attributed. It is the concept of the “purposiveness of nature” (Zweckmäßigkeit der Natur; Fortschritte 20:293), which Kant considers exhibited in objects of experience (natural purposes), although not as, strictly speaking, inhering in nature as such but as being brought to bear onto nature by human reason in an effort to render intelligible animate nature and by extension the overall purposive structure (“organization”) of nature. Kant traces the origin of the conception of purposiveness to the human “faculty of purposive connection” (Vermögen der Verknüpfung nach Zwecken; Fortschritte 20:294) inherent in human beings as rational agents able to conceive of and enact ends.

“The supersensible … in us, above us and after us”

277

Kant goes on to transpose the mode of imposing purposiveness from the order of nature, where it never amounts to a theoretico-dogmatic cognition of objects, to the moral order, where it results in the articulation of the ultimate object of practicodogmatic cognition, viz., the “final purpose of pure practical reason”, which is none other than “the highest good, insofar as it is possible in the world” (Fortschritte 20:294). Materially speaking, the latter includes happiness as individually qualified by ethically lawful conduct. Formally speaking, the highest good in the world, along with its psychological and theological conditions, is not something to be found among objects but an idea to be imported into objects (hineinlegen; Fortschritte 20:294). On Kant’s account, the concepts that enter into the practico-dogmatic cognition of the highest good are fabricated or “made concepts” (gemachte Begriffe; Fortschritte 20:295). In particular, Kant distinguishes three such made concepts involved in the highest good: freedom as the “supersensible in us”, God as the “supersensible above us” and immortality as the “supersensible after us” (das Übersinnliche … in uns, über uns und nach uns; Fortschritte 20:295). Kant’s late vindication of metaphysics is remarkable for its inclusion of freedom (along with God and immortality) among the ingredients of the highest good, and hence as the object of an epistemic attitude, “faith” (Glaube), that is different both from theoretical cognition or knowledge (Wissen) concerning the natural order and from practical cognition or knowledge of the moral order. But the freedom so considered in the late Prize Essay is not identical with the freedom presupposed and hence, at least indirectly, known in Kant’s second Critique under the guise of autonomy. Rather the freedom included by Kant among the ingredients of the highest good encompasses, in addition to moral self-legislation or “autonomy”, ethical self-mastery or “autocracy” (Autokratie; Fortschritte 20:295), which adds the “faculty to achieve morality […] already here in earthly life” (im Erdenleben), hence the qualification for the actual enactment of morality against all contravening circumstances to be encountered by a moral agent. Since the transition from principal autonomy to actual autocracy is practically contingent and cannot be taken for granted with the establishment of the principle of autonomy as such, freedom qua enacted freedom remains the object of a “faith in virtue” (Glaube an die Tugend; Fortschritte 20:295). Kant consider all three “made concepts” (effective freedom, immortal life and divine being) the object of a deep seated “concern” (Anliegen; Fortschritte 20:295) of human reason that is not in need of any “proof”, just as metaphysics in its perpetual preoccupation with this concern is beyond the need of “justification”. Yet given the essentially fabricated, albeit reason-based nature of the concepts and objects involved, Kant insists that the epistemic attitude toward the highest good and its ingredients can only be “faith” (Glaube; Fortschritte 20:297) and, moreover, a faith that cannot be commanded in the manner of an imperative to believe (kein crede;

278

Günter Zöller

Fortschritte 20:298) but can only the result of a “free assumption” (freies Annehmen) issuing in a profession of faith (credo) (Fortschritte 20:298). On Kant’s late considered view, God, morally successful freedom and human self-survival are selfmade objects of human reason to which an ontological status of sorts (“objective reality”) is granted “freely” (freiwillig). More importantly, the former chief entities of metaphysics are accorded such status with an eye toward the possible “return effect” (Zurückwirkung) of their assumption “on the subjective principles of morality and their strengthening” (Fortschritte 20:299). The highest good, while salvaging metaphysics from remaining void and empty, populates Kant’s post-critical metaphysics with prudent products of morally interested reason. As Heine suggests in the text from which the motto of the contribution was taken, the “resurrection” undertaken by Kant may not so much have been owed, on a anecdotal level, to his man servant Lampe but, on a civico-political level, to a concern for his fellow human beings – in an effort to assure peace and order by means of an act of ethical policing, for the optimal operation of which Heine actually uses the German term, “Polizei”, in its older, ordo-political meaning (die Polizei wegen unternommen; Heine 1976, vol. 5, 605). The religious outgrowth of Kantian ethics in the doctrine of the highest good thus appears as a carefully crafted form of moral psycho-politics that lends Kant’s late practico-dogmatic metaphysics the character of a Rousseauian civil religion.⁹

References Heine, Heinrich 1976, Zur Geschichte der Religion und Philosophie in Deutschland, in: Heinrich Heine, Sämtliche Schriften in zwölf Bänden, ed. Klaus Briegler, vol. 5, Munich and Vienna: Hanser Kant, Immanuel 1968, Werke in zwölf Bänden, ed. Wilhelm Weischedel, Frankfurt/Main: Suhrkamp Kant, Immanuel 2004, Theoretische Philosophie, 3 vols., ed. Georg Mohr, Frankfurt/Main: Suhrkamp Kant, Immanuel 2013, Les progès de la métaphysique, ed. Antoine Grandjean, Paris: Garnier Flammarion Zöller, Günter 1984, Theoretische Gegenstandsbeziehung bei Kant. Zur systematischen Bedeutung der Termini “objektive Realität” und “objektive Gültigkeit” in der “Kritik der reinen Vernunft”, Berlin/New York: Walter de Gruyter Zöller, Günter 2004, “Metaphysik nach der Metaphysik. Die limitative Konzeption der Ersten Philosophie bei Kant”, in: Karin Gloy (ed.) 2004, Unser Zeitalter – ein postmetaphysisches? Würzburg: Königshausen und Neumann, 231–43

 For a comparison of Rousseau and Kant on civil religion, see Zöller .

“The supersensible … in us, above us and after us”

279

Zöller, Günter 2010, “Critique: Knowledge, Metaphysics”, in: Will Dudley/Kristina Engelhard (eds.) 2010, Kant: Key Concepts, London: Acumen, 13–27 Zöller, Günter 2013, “Höffen-Dürfen. Kants kritische Begründung des moralischen Glaubens”, in: Dietmar H. Heidemann/Raoul Weicker (eds.) 2013, Glaube und Vernunft in der Philosophie der Neuzeit/Foi et raison dans la philosophie moderne, Hildesheim/New York: Olms, 245–57 Zöller, Günter 2014, “The Virtuous Republic. Rousseau and Kant on the Relation Between Civil and Moral Religion”, in: Steve Hoeltzel/Halla Kim (eds.) 2014, Kant, Fichte and the Legacy of German Idealism, New York: Lexington Books/Rowan&Littlefield, 31–51 Zöller, Günter 2015, “L’intelligible en nous. Liberté transcendantale et chose en elle-même dans l’Élucidation critique de l’analytique de la raison pratique pure de Kant”, in: Sophie Grapotte/Margit Ruffing/Ricardo Terra (eds.) 2015, Kant. La raison pratique. Concepts et héritages, Paris: Vrin, 53–70 Zöller, Günter 2015a, “‘Without Hope and Fear’. Kants Naturrecht Feyerabend on the Ground of Obligation to an Action”, in: Robert Clewis (ed.) 2015, Reading Kant’s Lectures, Berlin/Boston: Walter de Gruyter, 346–62

Index of Names Albrecht, Michael 29–30 Allison, Henry 56, 181, 185, 187–8 Ameriks, Karl 126 Aristotle 7, 76–8, 81, 87, 89–90, 92, 97, 104, 108, 196, 263–6, 273 Audi, Robert 112, 123, 127 Basaglia, Federica 2, 17–8, 20, 25, 73 Baumgarten, Alexander Gottlieb 144, 201 Baxley, Anne M. 196 Bayle, Pierre 250–2 Beck, Lewis W. 18, 29, 33, 44, 52, 143, 181, 185, 187 Beiser, Frederick 39, 126, 247 Bentham, Jeremy 1 Bojanowski, Jochen 5, 49, 181 Brandt, Reinhard 24–5, 115, 123 Broadie, Sarah 92 Cassirer, Ernst 259 Chignell, Andrew 122, 126, 192, 201–2, 204–7, 212, 221 Cohen, Alix 245 Cranz, August Friedrich 174 Cunico, Gerardo 118, 120 Dancy, Jonathan 238 Denis, Lara 52, 68 DiCenso, James J. 168 Dörflinger, Bernd 112 Düsing, Klaus 29–30 Engstrom, Stephen 3–4, 34, 44, 52, 54, 56, 61–2, 64, 89–90, 92, 95, 254 Epicurus 3, 76, 79, 83, 89, 103, 116, 245, 248, 259 Esser, Andrea M. 6, 72, 245, 253 Euclid 144, 193 Förster, Eckart 30, 112, 253, 261 Fugate, Courtney 44 Gardner, Sebastian 224 Georges, Karl E. 152

Gerhardt, Volker 81, 86 Gesang, Bernward 18 Guyer, Paul 4–5, 9, 68, 143, 157, 178, 225, 227, 235 Hegel, Georg Wilhelm Friedrich 79–80 Heine, Heinrich 263, 278 Herman, Barbara 17–18, 34, 92 Hill, Thomas E., Jr. 60, 92 Hobbes, Thomas 265 Höffe, Otfried 29, 115 Horn, Christoph 55 Höwing, Thomas 6, 18, 49, 64, 68, 108, 128, 201, 213, 221, 243 Infield, Louis 92 Insole, Christopher

124

Kleingeld, Pauline 2–3, 25, 33–4, 38, 48, 68, 181, 185, 187 Kluxen, Wolfgang 151 Lambert, Johann Heinrich 157 Lavater, Johann Caspar 174 Leibniz, Gottfried Wilhelm 9, 103 MacIntyre, John 258 Maly, Sebastian 146 Mandeville, Bernard 265 Mariña, Jacqueline 52 Marwede, Florian 3, 51, 243 Mattey, George J. 207–9, 211, 214 Matthiessen, Hannes O. 210, 243 McDowell, John 196, 210 Meier, Georg Friedrich 201 Mendelssohn, Moses 4–5, 157–62, 165–6, 169, 172, 174, 176–8, 240 Menzer, Paul 92 Mill, John Stuart 1 Moore, Adrian 117, 127–8 Motta, Giuseppe 207 Nagel, Thomas

181

282

Index of Names

O’Connell, Eoin 111, 116 Oggionni, Eva 27, 30, 32 O’Neill, Onora 55, 62 Palmquist, Stephen 174 Pasternack, Lawrence 111, 124, 202, 205–7, 212 Pieper, Annemarie 148 Pinkard, Terry 111 Pistorius, Hermann Andreas 17–20, 22, 25– 30, 34–6 Plato 92, 96, 104, 159, 161 Reath, Andrews 34, 39, 44, 56, 64, 67–8, 248 Ricken, Friedo 119–18, 143 Rink, Friedrich Theodor 273 Ross, William David 96 Sala, Giovanni 18, 27 Scanlon, T.M. 208, 221 Schopenhauer, Arthur 85, 172 Schroeder, Winfried 251 Seneca 76–8, 81 Setton, Dirk 75 Smith, Adam 265 Socrates 159, 161–2, 165 Specht, Ernst K. 151

Spinoza 6–7, 123–4, 245–6, 248–52, 255– 6, 259, 265 Spree, Axel 145 Stark, Werner 10, 92 Sulzer, Johann Georg 157 Surprenant, Chris W. 182, 184, 187–8 Tafani, Daniela 112 Theis, Robert 201 Timmons, Mark 55 Torralba, José M. 25 Tugendhat, Ernst 77 Vossenkuhl, Wilhelm

117, 128

Watkins, Eric 27, 30–1, 38, 111, 116, 221, 223, 226, 242–3 Weischedel, Wilhelm 273 Wentscher, Max 53 Willaschek, Marcus 6, 34, 68, 120, 123, 131, 192, 221, 223–5, 228 Williams, Bernard A. O. 181, 258 Winter, Aloysius 143 Wolff, Christian 9, 144, 162, 201 Wood, Allen W. 9, 52, 65, 135, 212, 225, 245 Zimmermann, Stephan 4, 131, 145 Zobrist, Marc 29–8 Zöller, Günter 7, 263, 265–7, 271, 278

Subject Index antinomy of pure practical reason 68, 74–5, 78, 83–5, 94, 150, 226, 233, 236–7, 240–2, 246, 261 antinomy of pure reason 83, 275 argument from design 4, 113, 265 assent (Fürwahrhalten) 112–13, 120, 122, 125–8, 143, 194, 201–21, 273 atheism 121–5, 246–53, 265 autonomy 21–2, 74, 93, 233, 250, 252, 257, 259, 267–8, 270, 277 beatitude (Seligkeit) 52–3, 71, 151 belief (Glauben) 4–6, 86–7, 111–13, 125–8, 137, 143, 182, 184, 190–5, 201–7, 210–13, 215–16, 218, 220–1, 223–5, 228, 237, 241– 2, 273, see also faith – that God exists 1, 4, 111–13, 119–28, 201, 211–13, 228, 242–3 – that the soul is immortal 1, 4–5, 131, 139, 157–79, 181, 190–5, 201, 213, 247 – and acceptance 85, 112–3, 127–8, 135, 201 – doctrinal 206 – pragmatic 193, 206, 212–13, categorical imperative 3, 24, 33, 38, 42–7, 49, 51, 54, 58–9, 62, 67–8, 73, 133–5, 138, 140–1, 150, 154, 163, 271, see also moral law creation 116, 118, 161, 247–8, 254–5, 258– 9 desire 20, 22, 64, 101–4, 107, 231 – faculty of 20–4, 26, 63, 72, 79, 82, 101– 2, 104, 163 dualism 3–4, 71–2, 83–7, 90–1, 93, 100, 107–8, 246 duty (moral obligation) 3, 5, 29, 38–45, 47, 49, 51–5, 57–67, 81–2, 85, 87, 89–90, 94, 111, 134, 138, 181–3, 185–8, 190, 226–8, 234–5 – to promote the highest good 2–3, 33–4, 38–45, 47–8, 51–4, 60, 63, 67–8, 120, 182–3, 185, 229

– to promote the happiness of others 40–1, 51–2, 57–68, 163–5

3, 33,

empirical realism 74 end 6, 20–7, 34–37, 39–44, 46–8, 73, 79– 80, 114–16, 118–20, 158, 163–6, 169, 177, 190–2, 223, 225–42, 245–55, 257– 8, 263–6, 269–70 – final 6, 42–3, 45–7, 113–27, 245–50, 253– 61, 271, 274–7 – in itself 61, 247, 250, 264 enlightenment 1, 92, 131, 147, 154, 174, 181, 248, 250 faith

4, 86, 111–13, 124–8, 131–5, 137–44, 147, 150–3, 158, 168–9, 173–5, 201, 204, 250, 260, 277–8 – rational (Vernunftglaube) 170, 190, 273 – fiducial 112, 127–8 – attitudinal 112 God

4–5, 28, 39, 71, 84, 94, 111–14, 117– 28, 147–54, 157, 161–2, 164–9, 171–3, 175–6, 184, 196, 223–8, 241–3, 245–6, 248, 250–2, 256, 258–60, 272–3, 276–8 – as creator 83–4, 118, 148, 258 – as supreme intelligence 4, 71, 113–14, 117–19, 127, 145, 185, 256, 264 – as omniscient being 94, 139 good (notion of) 2–3, 6, 17–20, 24–31, 33– 8, 44, 89–90, 103–4, 107, 245 – in itself 17–18, 25, 27, 31, 35–6, 82, 107, 184 – will 17–20, 34–5, 91, 96–7, 99, 106, 175, 178 – supreme 72–3, 78–9, 92–3, 98, 106, 108, 111, 184–5, 187, 190, 196 – complete 72, 74, 78, 90, 92–3, 95, 97–8, 108, 111, 184 happiness 1–5, 17, 30–1, 33–4, 37–42, 48, 51–3, 55–8, 60–6, 68, 71–87, 90–5, 98– 103, 105–7, 111, 116–18, 119–20, 126,

284

Subject Index

157, 161–7, 169–70, 175, 181–4, 231–3, 236, 264–6, 271–3, 277 – universal 3, 51–2, 58–9, 61–8, 106, 176 – individual 33, 51–2, 58–9, 72–3, 164–5, 231 – of others 3, 33, 40–2, 51–3, 56–8, 60–3, 68, 73, 81, 105, 163–4 – as eudaimonia 90, 264 heteronomy 21–2, 28, 46, 89, 91 holding-to-be-true (Fürwahrhalten), see assent holiness 5, 52–3, 85, 166–7, 170–1, 182–9, 195–6, 224 idealism, transcendental 74, 85, 157, 167, 171 immortality 4–5, 74, 85–6, 112, 157–62, 164–70, 172–9, 181–2, 184–5, 187–96 – of the species 173 – as endless progress toward moral perfection 53, 84–5, 166, 174, 186–9 justification 2, 6, 126, 128, 132–3, 192–4, 201–20 – non-epistemic 6, 125–6, 128, 192, 201–6, 213–4, 220–1 kingdom of ends 84, 93 knowledge (Wissen) 4, 6, 89–90, 94, 96– 107, 122, 173, 201–2, 207–11, 214, 270, 277 moral argument 2, 4–6, 111–13, 119–22, 124–8, 131–5, 144–6, 149, 162–8, 177, 226–30, 242–3 moral law 4, 19–22, 24–9, 33, 36–44, 51–2, 54–7, 60, 65–8, 91, 94–5, 100, 103–5, 112, 116, 120, 122–4, 126–7, 134, 138– 40, 163, 166–7, 169–72, 178, 184–6, 190, 211–12, 223, 225, 245–50, 253–5, 267, 273 moral world 34, 38–43, 46, 48, 53, 163 opinion (Meinung) 4, 214–21

6, 23, 140, 194–5, 201–

perfection 5, 28, 38–40, 73, 97, 161–2, 165–72, 174, 178, 181, 183–6, 188–9, 195–6, 224 – moral 5, 26, 40, 42, 169, 172–3, 181–2, 183–6, 188–9, 195–6, 224 physico-theological proof 4, 84, 113–4, 258–9 pleasure 21, 26, 28, 33–4, 36, 38, 41, 64, 76, 79, 89, 101–2, 104 postulate (notion of) 4, 120, 131–44, 147, 150–1, 153, 181–2, 191–4, 223–5 – of pure practical reason 2, 4–6, 54, 84–6, 94, 109, 120, 131–44, 154, 157–8, 162, 164–8, 174–5, 177–8, 181–2, 184–5, 187–95, 223–8, 232–7, 240–3, 271 practical reason 7, 17–18, 20–31, 34–8, 48–9, 89–90, 94, 98, 100–5, 128, 135– 41, 192–6, 223–6, 230–3, 253–4, 263, 266–73 – object of pure 2, 17, 20, 20–31, 37–40, 47, 90, 93–4, 104–8, 233, 276, see also good (notion of) proportionality 31, 33–4, 52, 68, 72–4, 78, 84–5, 87, 90, 93, 95, 103–8, 163–4 self-contentment (Selbstzufriedenheit) 51, 53, 75, 151–2 self-love 57, 75, 105, 167, 172 synthesis 38, 42, 45–7, 72, 74, 77, 79–80, 117, 119, 137, 157, 160, 169, 230, 275 teleology 7, 27, 92, 112–16, 154, 162, 248, 250, 255–7, 263–5, 269–73, 276 – moral 113–17, 121, 254–8, 274 – physical 6–7, 113–17, 246–7, 256, 258–9 theology 85, 113–18, 121, 123, 131, 146, 151, 248, 259–60, 264–5, 276 – moral 114–17, 273 – physico-theology 7, 114, 246–7, 256, 258–61 theoretical reason 4, 7, 48, 74, 132, 135, 137–9, 141–3, 149, 153–4, 192, 195, 267–8, 270 transcendental idealism 74, 85, 157, 167, 171

Subject Index

virtue 53, 72–80, 83–5, 87, 90–5, 97–8, 157–8, 162–3, 165–70, 174–5, 178, 182– 3, 185–8, 192, 195–6, 224, 231–3, 241– 2, 245–7, 249–51, 264, 277

285

wisdom 98–100, 108, 274 worthiness to be happy 71–4, 78–9, 83–4, 90–1, 117, 119–20, 139, 163, 165–7, 169– 70, 266