Freedom and Anthropology in Kant's Moral Philosophy

Freedom and Anthropology in Kant’s Moral Philosophy This book is the first comprehensive account of Kant’s theory of freedom and his moral anthropology. The point of departure is the apparent conflict between three claims to which Kant is committed: that human beings are transcendentally free, that moral anthropology studies the empirical influences on human beings, and that anthropology is morally relevant. Frierson shows why this conflict is only apparent. He draws on Kant’s transcendental idealism and his theory of the will and describes how empirical influences can affect the empirical expression of one’s will in a way that is morally significant but still consistent with Kant’s concept of freedom. As the first work on Kant to integrate his anthropology with his philosophy as a whole, this book will be an unusually important source of study for all Kant scholars and advanced students of Kant. Patrick R. Frierson is an assistant professor of philosophy at Whitman College, Washington.

Freedom and Anthropology in Kant’s Moral Philosophy

PATRICK R. FRIERSON Whitman College

cambridge university press Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo Cambridge University Press The Edinburgh Building, Cambridge cb2 2ru, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521824002 © Patrick R. Frierson 2003 This publication is in copyright. Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published in print format 2003 isbn-13 isbn-10

978-0-511-17009-6 eBook (MyiLibrary) 0-511-17009-2 eBook (MyiLibrary)

isbn-13 isbn-10

978-0-521-82400-2 hardback 0-521-82400-1 hardback

Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

for J.C.

Contents

page ix

Preface Introduction: A Problem with Kant’s Moral Anthropology

1

i. the problem 1 2 3

The Asymmetry in Kant’s Conception of Freedom Anthropology as an Empirical Science The Moral Importance of Kant’s “Pragmatic” Anthropology

13 31 48

4

Moral Anthropology in Contemporary Neokantian Ethics

68

5 6

ii. the solution Transcendental Idealism, Radical Evil, and Moral Anthropology Moral Influence on Others Epilogue: Incorporating Moral Anthropology and Defending Kantian Moral Philosophy

95 136 163

Notes References

167 200

Index of Kant’s Works Name Index Subject Index

205 207 209

vii

Preface

From my earliest exposure to Kant’s moral theory, I was drawn to his emphasis on the centrality of freedom but bothered by the apparent abstractness of the moral law. Thus when I first approached the Anthropology many years ago, I did so with excitement. I hoped that Kant would incorporate all the rich details of human life that I found lacking in his Grounding, and that he would show how these details fit with the distinctive focus on freedom that attracted me to his moral theory. At first, the Anthropology seemed more amusing than philosophically satisfying. However, as I came to appreciate the details of Kant’s Anthropology and as more neokantians incorporated anthropological insights into moral theory, I saw that Kantians could provide as rich and concrete a moral theory as anyone. During graduate school, especially as a result of interaction with neoAristotelian ethical theories and more recent neokantian accounts, I became interested in Kant’s treatment of the cultivation of moral character. In parts of Kant’s anthropology he seemed to suggest that there could be empirical influences on moral development, and these showed that Kant could provide the sort of nuanced theory of human nature that often made Aristotle attractive. But I was still puzzled about the compatibility of these new (for me) aspects of Kant’s account with the treatment of freedom that initially attracted me to Kant. This puzzle led me to write my dissertation on Schleiermacher’s critique of Kant’s Anthropology, in which Schleiermacher argues that transcendental freedom is incompatible with a robust anthropology. And Schleiermacher’s critique finally led to this book, which is my attempt to show how Kant’s moral theory can incorporate the anthropology that I ix

x

Preface

have grown to appreciate into the theory of freedom that first drew me to Kant.

Acknowledgments Without the generous and perceptive criticism of Karl Ameriks, this book would not exist. As my advisor at the University of Notre Dame, he pushed me to refine and expand ideas that eventually made their way into this book. Since that time, he has continued to offer suggestions for which I am extremely grateful. G. Felicitas Munzel, Philip Quinn, and Robert Solomon also read various drafts and provided extensive and invaluable comments. Robert Louden was extremely generous, not only to give me an early draft of his book, but to discuss my comments on his work in detail at an earlier stage of this project. Natalie Brender also gave very helpful comments on an early draft of this book. I also thank Eric Newman and my anonymous reviewers at Cambridge University Press. Travis Exstrom at Whitman College provided help with the index of Kant’s works and the name index. Samuel Fleischacker deserves a special thanks for introducing me first to philosophy, and then to Kant, while I was an undergraduate at Williams College. Finally, this work is built on the support and sacrifice of family. I am grateful to my parents for their encouragement and patience. And I thank my wife, Katheryn, whose calls for clarity and relevance in my work certainly improved this book and made me both a better philosopher and a better person.

Introduction A Problem with Kant’s Moral Anthropology

1. Kant’s Anthropology and Schleiermacher’s Objection In 1798, Immanuel Kant published his Anthropology from a Pragmatic Point of View. In this work, he discusses “what man makes, can, or should make of himself” (7:119). The book offers detailed, even if incomplete, accounts of human capacities and character, and these accounts help flesh out Kant’s Critical philosophy with empirical information about human beings. This 1798 Anthropology was not Kant’s first foray into anthropology. Starting in 1772, Kant offered yearly lectures on anthropology that parallel the published work. In addition, anthropological insights are scattered throughout Kant’s other publications. The essays on history (primarily from 1784 to 1786), the third Critique (1790), Religion within the Boundaries of Mere Reason (1793), and the Metaphysics of Morals (1797) all present various aspects of Kant’s anthropology. But the Anthropology of 1798 is the most detailed, systematic, and public treatment of anthropological issues in Kant’s corpus. Within a year and a half of its publication, at least eleven reviews of Kant’s Anthropology appeared.1 Among these was an important review by a young Friedrich Schleiermacher, published in the Romantic journal Athaeneaum. In his review, Schleiermacher sarcastically suggests that Kant’s Anthropology must have been intended as a “negation of all anthropology” (Schleiermacher 1984: v.1, p. 366, cf. Schleiermacher 1998) because it blatantly conflicts with the rest of Kant’s philosophy. The review criticizes the Anthropology for disorganization and triviality and accuses Kant of failing to combine systematicity and popularity in it.2 But the most philosophically important objection comes in the form of a 1

2

Introduction

challenge to Kant to choose between his anthropology and his theory of freedom, insisting that Kant cannot have both. The conflict is between two claims, that “nature is choice” and that “choice is nature” (Schleiermacher 1984: v.1, p. 366). Kant’s anthropology, according to Schleiermacher, must affirm that “choice is nature.” That is, human beings and all their choices must be considered objects in nature if they are to be studied by anthropology. But Kant must also affirm that “nature is choice” – that is, that an individual’s human nature is due to that individual’s choice. The “nature” that anthropology studies cannot be merely the result of natural causes. Schleiermacher gives two reasons for insisting that Kant must reconcile his anthropology with this strong claim about freedom. First, he points out that the conception of freedom developed in Kant’s Critical philosophy commits Kant to the view that freedom grounds human choices, and thereby human “nature.” Second, he suggests that any anthropology must have some account of choice underlying nature to make sense of the epistemic norms implicit in scientific inquiry.3 Kant could have offered an anthropology that would not conflict with his Critical philosophy. The problem is the particular sort of anthropology he presents, one that is both empirical and morally relevant. The dilemma can be stated in terms of a conflict among three claims to which Kant seems committed: r Human beings are transcendentally free, in the sense that empirical influences can have no effect on the moral status of a human being and in the sense that choice is fundamentally prior to natural determination. r Moral anthropology is an empirical science that studies empirical influences on human beings. r Moral anthropology is morally relevant, in that it describes influences on moral development.

Schleiermacher suggests that Kant is committed to all three of these claims and that the claims are inconsistent with one another. Any two of them could be held consistently, but all three cannot. Unless Kant is willing to sacrifice the conception of freedom on which his moral philosophy depends, his Anthropology can be nothing more than a “negation of all anthropology.”4

A Problem with Kant’s Moral Anthropology

3

2. The Practical Problem of Moral Anthropology At this point, it is important to distinguish Schleiermacher’s objection to Kant’s anthropology from a familiar objection to Kant’s metaphysics. From Kant’s earliest critics, such as Rehberg, Fichte, and Hegel,5 to more recent commentators, Kant’s account of freedom has been criticized as an incoherent form of compatibilism. This criticism takes different forms, but the basic point is that one cannot claim both that one’s actions are causally determined in a series of natural events and that one is free in the sense that one’s actions are ultimately caused by some more fundamental freedom. There is simply no room for both natural and free causes, when freedom is understood in Kant’s anti- determinist sense. Responses to this objection have been almost as varied as the formulations of the objection itself.6 The problem that Schleiermacher raises is not primarily this metaphysical one. Schleiermacher is not claiming merely that studying human beings as natural objects is impossible because they are metaphysically free. His objection also, and more fundamentally, involves a problem from the standpoint of practical reason. This problem arises in the context of moral anthropology.7 The practical problem is how to account for moral judgments that make use of anthropological insights regarding helps and hindrances for moral development. Anthropological insight into empirical influences on moral choice is not merely scientific knowledge of human beings. For one thing, it is knowledge that is specifically articulated for practical use. In his anthropology, for example, Kant argues that politeness promotes virtues.8 This observation leads to a duty to promote politeness. The anthropological perspective enters into the practical one. Of course, scientific perspectives enter into practical deliberation all the time. When I serve tea to guests, I make use of my knowledge that tea will contribute to their happiness. A murderer who decides to pull a trigger makes use of his or her knowledge that the gun will fire and kill the victim. But moral anthropology enters into deliberation in a more problematic way. Specifically, a human agent must be considered at once as both empirically influenced and morally responsible. One must consider people as capable of influence by polite society, or one does not have any responsibility to promote polite society. But one must also think of this influence as bearing on the moral status of those people, because that is the particular sort of influence that makes politeness so important.9 But then people must be considered from a practical perspective and thus as free from any empirical

4

Introduction

influence. The conflict between freedom and moral anthropology arises as a practical problem even if metaphysical issues surrounding freedom can be resolved.

3. Kant’s Susceptibility to the Problem Ultimately, this book defends Kant’s account of freedom and anthropology. But the first half of the book shows just how closely Schleiermacher’s objection makes contact with Kant. One easy way to defend Kant would be to deny one of the three claims that constitute Schleiermacher’s dilemma. If Kant does not affirm transcendental freedom, or does not hold that anthropology is empirical, or restricts anthropology to nonmoral contexts, then he is easily saved. But Kant cannot be saved that easily. He does affirm all three of the claims that form the dilemma. The project of saving Kant thus involves showing that there is a way that they can all be held consistently. In Chapter 1, I show that Kant has a strong, noncombatibilist conception of human freedom. I focus on one crucial feature of Kant’s account that makes his anthropological work difficult. That feature is an asymmetry in the causal relation between the noumenal free self and its phenomenal appearance in the world, an asymmetry that arises whether one holds a two-object, two-aspect, or two-perspective account of Kant’s metaphysics. Kant resolves the third antinomy of the first Critique by suggesting that although there cannot be a free cause in nature, there can be a free ground of effects that are in nature (see A537f./B565f.). This ground has a relation to its effects in the world similar to that of a natural cause, though it is not spatiotemporal. In the first Critique, Kant does not show that there is a free ground of empirical effects, only that for all we know there can be such a ground. What is important, however, is that Kant specifies the metaphysical place that such a ground would occupy. A free ground of effects in the world would have to lie outside of nature in the sense that it would not be susceptible to being an effect of natural causes. This is precisely what it means for such a cause to be free. Theoretical reason provides a basis for saying that if there are free causes, they must not be influenced by other causes in the empirical world. In that sense, the relationship between freedom and nature is asymmetrical. In the second Critique, Kant argues that human beings actually are free agents. Human beings fill the spot left open but empty by the first

A Problem with Kant’s Moral Anthropology

5

Critique. And in the second Critique, Kant again emphasizes the asymmetry between the free cause – the agent – and the empirical world in which one’s agency is effective. Not only for theoretical reasons (the nature of causation) but for moral ones (the conditions of possibility of moral responsibility), the free agent must affect but not be affected by the world. In Chapter 2, I show that Kant’s anthropology is empirical. This is the most consistent claim that Kant makes about his work in anthropology, persisting through all his lectures on it and enduring in his published work from his earliest works to his published Anthropology. In the Groundwork, he distinguishes pure morality, which is the rational part of ethics, from “practical anthropology,” which is the “empirical part” (4:388). In the Metaphysics of Morals, he explains that in anthropology “we shall often have to take as our object the particular nature of human beings, which is cognized only by experience” (6:216–17). In the Anthropology itself Kant makes clear that anthropological investigation is a matter of empirical observation, not a priori theorizing. Although he dismisses observations that are not put to use as “speculative theorizing” that “is a sheer waste of time,” proper anthropological investigation also consists in “observations,” but only when one “distinguishes between those observations which have been found to hinder and those which have been found to promote” the faculty under investigation (7:119). Both the fruitless and the proper sorts of anthropology are empirical. The difference is that proper anthropology puts empirical observation to use. In Chapter 2, I articulate what it means for anthropology to make claims that are at once universal, related to the free self, and empirical. In Chapter 3, I describe how Kant’s empirical anthropology is a moral anthropology. That is, anthropology takes moral choice as one of its objects. Based on Kant’s distinction between moral, pragmatic, and technical considerations in the Groundwork, one might think that the title of Kant’s published work, Anthropology from a Pragmatic Point of View (emphasis added), speaks against its inclusion of moral considerations. However, the term pragmatic is used in several different ways in Kant’s work, and I show that the primary sense of pragmatic in the Anthropology contrasts not with moral but with merely theoretical or physiological concerns. Kant’s anthropology is meant to be useful, rather than mere knowledge of human beings. And one of the uses of anthropology is to cultivate and encourage good moral choices.

6

Introduction

Thus one finds in the Anthropology examples of empirical helps and hindrances to having a good will. Kant’s discussions of politeness, of the passions, and of character all present these as helps or hindrances to morality itself, so I briefly discuss each of these examples. The chapter concludes by pointing out the systematic place of moral anthropology in Kant’s anthropology as a whole. The first three chapters thus show that Kant’s moral anthropology makes him susceptible – at least at first sight – to Schleiermacher’s criticism.

4. Freedom and Anthropology in Contemporary Moral Theories In recent years, neokantian moral theorists have begun to pay more attention to Kant’s moral anthropology. Onora O’Neill, in a chapter of Constructions of Reason called “Action, Anthropology, and Autonomy,” accounts for a “gap between Kant’s practical philosophy and contemporary would-be Kantian writing on ethics” in part by pointing out that “modern protagonists of ‘Kantian’ ethics are mainly interested in rights, which for Kant are one element in a broader picture” (O’Neill 1989: 66). Anthropology helps to flesh out this broader picture. Allen Wood explicitly articulates his conception of Kant’s ethical thought in contrast to approaches that are open to “common charges that Kantian ethics is unconcerned with the empirical realities of psychology, society, and history, that it sees no value in the affective side of our nature, and that it is individualistic” (Wood 2000: xiv). This new approach to Kant involves an extensive treatment of Kant’s anthropology (see pp. 193–320). Robert Louden’s Kant’s Impure Ethics is devoted entirely to drawing attention to “the second part of Kant’s ethics, a part that . . . unfortunately remains a well-kept secret [and that] Kant referred to . . . as ‘moral anthropology’ ” (Louden 2000: vii). And G. Felicitas Munzel’s Kant’s Conception of Moral Character (1999) seeks to integrate Kant’s anthropology into his moral philosophy through a study of the notion of “character.” Even neokantians who do not discuss Kant’s anthropology as a whole often turn to specific aspects of moral anthropology to flesh out their Kantian moral theories. Kant’s remark that people are “not to shun sickrooms or debtors’ prisons” (6:457) has become a popular text to point out Kant’s awareness of the importance of cultivating sympathy for the moral life.10 Barbara Herman has drawn attention to the role that community and education can play in promoting moral behavior (Herman 1993, esp. pp. 82–3), and she has drawn attention to the importance of “character” for Kant (Herman 1996). Nancy Sherman, in a series of

A Problem with Kant’s Moral Anthropology

7

books and articles, has shown that Kant’s account of emotions can contribute to a richer Kantian ethical theory (Sherman 1990, 1995, 1997, and 1998). All of these developments spring from Kant’s anthropological observations. The attention to Kant’s anthropology in contemporary ethics is not merely an attempt to be historically accurate. Neokantian moral theorists find in Kant’s anthropology a richness of detail and attention to human particularity that should be an important part of any moral theory. The recent rise of neo-Aristotelian, Humean, and anti-theoretical approaches to moral theory has presented serious challenges for Kantian moral theories.11 These apparent alternatives to Kant tend to focus on character rather than action, virtues rather than rights or duties, and take into account a wide range of features of human psychology that Kantians have sometimes ignored. They thus present sensitive accounts of moral development and the role of emotions in moral motivation, and they can seem to provide a very nuanced account of ethical life. The focus on formulaic applications of the categorical imperative, and a general emphasis on the Groundwork in Kantian moral theory, has made some Kantians particularly susceptible to challenges from these alternative accounts of ethics. Kant’s anthropology provides effective responses to many of these objections. His moral anthropology includes extensive discussions of the importance of community and education for moral development. He discusses and differentiates different sorts of emotions and various roles that these can play in moral life. His moral anthropology focuses on cultivating a virtuous character, rather than on merely doing good deeds. And throughout his anthropological writings, Kant discusses character, disposition, and virtue.12 Moreover, his anthropology provides detailed, even if scattered, accounts of the particulars of human life. He analyzes the psychology that underlies sexual temptation, gives a sophisticated treatment of the role of politeness in modern life, and even provides advice on conducting an excellent dinner party. Even when these descriptions of human life fall short of what one might hope for, they go far beyond the abstraction of the categorical imperative. Thus neokantians have been right to look to Kant’s anthropology for an ethical theory that can hold its own against recent virtue-based and anti-theoretical approaches to ethics. But there has been insufficient attention to the problems that Kant’s anthropology presents for his overall moral theory. In Chapter 4, I take up three of the more prominent

8

Introduction

current neokantians who draw extensively from Kant’s anthropology – Nancy Sherman, Robert Louden, and G. Felicitas Munzel. All three explicitly articulate their accounts as theories that can meet some of the challenges recently raised against Kant, especially by those sympathetic to Aristotle. Because of their use of Kant’s anthropology, these contemporary neokantian moral theories are susceptible, in varying degrees, to Schleiermacher’s objections to Kant. Unfortunately, no one has yet offered a sufficient integration of Kant’s moral anthropology with his conception of freedom. Thus contemporary accounts often fall short of seeing the full significance of Kant’s moral anthropology. In some cases, they simply fail to recognize all the ways in which Kant’s anthropology affects his ethics. Nancy Sherman, for example, allows for important anthropological influences but ultimately does not give anthropology the range of moral significance that Kant allows. In other cases, neokantians fail to save Kant’s theory of freedom. The result is a moral theory that is so tied to anthropology that it loses its distinctive Kantian emphasis on freedom. At times, Louden and Munzel go in this direction. Given the increasing emphasis on moral anthropology as an important part of a contemporary Kantian ethics, there is a need to articulate an answer to Schleiermacher’s challenge that can justify the integration of anthropological insights into a genuinely Kantian moral theory.

5. Solving Schleiermacher’s Dilemma The second half of this book, especially Chapters 5 and 6, offers the needed solution to Schleiermacher’s dilemma. In Chapter 5, I show that Kant has the resources to distinguish between the empirical will, which can be affected by empirical influences, and the free will, which cannot. The connection between these is such that the empirical will is morally relevant as the expression of the moral status of the free will. In the simplest case, an action in the world such as making a false promise for personal gain expresses an evil will. But the situation is complicated by the presence in human beings of what Kant calls radical evil. Radical evil involves both choosing badly and making choices that reinforce one’s tendency to choose badly. This evil forces Kant to reconceive of the nature of the human good will and its expression in the world. According to Kant’s account in Religion within the Boundaries of Mere Reason, the best that humans can hope for is a “revolution” against evil, the expression of which is a constant struggle against evil in one’s nature. Although the revolution

A Problem with Kant’s Moral Anthropology

9

itself is not something temporal, the struggle that expresses a revolution in the will is temporally extended. To combat radical evil, one must not only choose rightly but also act to counteract one’s tendency to choose poorly in the future. In cool hours of moral self-control, one who is in revolution will act to effect a good empirical will not only in the present but in the future as well. In this context, moral anthropology is crucial. Moral anthropology explains the means for effectively correcting and improving one’s empirical will. And the effort to correct and improve one’s empirical will is part of the struggle against evil that expresses the will in revolution. Promoting a good character through methods explained in moral anthropology is an expression of one’s free will, so the asymmetry between nature and freedom in Kant’s philosophy is preserved. But because it is an expression of one’s moral status, anthropology has moral significance. This solution to Schleiermacher’s dilemma is worked out in detail in Chapter 5. In Chapter 6, I take up an important remaining problem with the account offered in Chapter 5. If considerations from moral anthropology are relevant because they enable one to express a revolution against evil, it is not clear how interpersonal moral influence can be morally significant. I argue that Kant is not as committed to the possibility of interpersonal moral influence as some have suggested. He does not think that one should seek to effect moral revolutions in others. Nonetheless, acting to improve the character of others is morally significant for several reasons, which I explore in Chapter 6. Most important, because one’s own empirical will is connected to the wills of others, acting to improve the wills of others expresses one’s own struggle against evil. I even argue that there is some room for Kant to allow that the actions of one agent can genuinely affect the moral status of another. Still, one can never know how this occurs and should not consider it a reason for acting to promote moral development in others. In the brief Epilogue, I reflect on where the debate between Kant and Schleiermacher, and Kant’s many other critics, stands given the account of Kant’s moral theory offered in this book. Although this book does not show that Kant’s moral theory is the only reasonable option, it does show that one of the most important objections to that moral theory fails. Kant can integrate moral anthropology into his ethics without sacrificing the account of freedom that lies at the core of his philosophy.

part i THE PROBLEM

1 The Asymmetry in Kant’s Conception of Freedom

In the Introduction, I raised a problem with Kant’s moral anthropology. For this problem to make any contact with Kant’s philosophy, Kant’s theory of freedom must preclude the influence of empirical causes on moral choice. In this chapter, I argue that Kant does articulate just such a conception of human agency. In section 1, I briefly sketch Kant’s mature argument for freedom and draw attention to the role of moral responsibility in this argument. In section 2, I argue that whatever one believes about the relationship between freedom and nature, or noumena and phenomena, or practical and theoretical perspectives, it is essential to recognize that for Kant, freedom is the ground of nature and not vice versa. Although Kant cannot explain how freedom grounds nature, he can say something about what this relation means, and why one must assume it. To avoid entering into recent debates about whether Kant offers primarily a two-object or a two-perspective account of the self, I defend the asymmetry of the relation between nature and freedom on both the two-object and two-perspective interpretations of the relation. In section 3, in order to clarify this asymmetry, I point out how the priority of the practical precludes any traditional determinist account of freedom. Finally, in the last section I touch on the nature of our knowledge or lack thereof of the free self. The nature of knowledge of the free self is important to clarify the sort of resources available to Kant to explain the relation between freedom and anthropology.

13

14

The Problem

1. Kant’s Argument for Freedom Kant’s argument that persons are free has two stages. In the Critique of Pure Reason, he argues that the possibility of freedom is not precluded by the nature of our experience of the world, the requirements of scientific knowledge, or any justifiable metaphysical theories. This argument depends on Kant’s transcendental idealism. Kant claims that experience is structured according to two forms of intuition – space and time – and twelve categories, one of which is cause and effect. Because these intuitions and categories provide the structure of our experience, one knows a priori that any experience of ours will be spatiotemporal and will fit into a series of natural causes that are objects of possible experience. However, because these intuitions and categories structure our experience, one cannot apply them in order to know “things in themselves.”1 In the context of arguing for this transcendental idealism Kant presents several antinomies, the third of which highlights the conflict between the claims that there is freedom and that there is no freedom.2 The antinomy arises because while understanding demands that every object of experience be explained in terms of a (temporally)3 prior cause, reason demands that this cause be sufficient, such that it leaves nothing to be explained. Only a cause that is itself not in the world can serve this purpose, because the world is structured by spatiotemporal causation such that every cause has a prior cause. If a cause c1 of an effect e1 has another prior cause c2 , then c1 does not fully explain e1 , because even after granting c1 , there is something left to explain (namely, what caused c1 ). To satisfy the demands of reason, one would have to find some cause that does not require (or does not allow) any further explanation. Ultimately, Kant argues that the resolution of the antinomy depends on the fact that one can distinguish things as they are in themselves from things as they appear.4 Because experience is structured according to the laws of causality, nothing can be experienced except as determined by prior causes. Thus nothing can be experienced as free. But it is thinkable that free things in themselves provide a ground for the series of appearances that is ordered according to natural laws, and these free things neither need nor allow further explanation in terms of further causes. Because these things in themselves are not structured according to the schematized categories of human understanding, they cannot be thought of as possible objects of our experience. The way in which they “ground” appearances is at best only analogous to the way appearances ground one another.5 But Kant

Asymmetry in Kant’s Conception of Freedom

15

at least shows that nothing that one knows about natural causation rules out some sort of freedom. As he explains, It should be noted that we have not been trying to establish the reality of freedom. . ., [and] we have not even tried to prove the possibility of freedom . . . . [To show] that . . . nature at least does not conflict with causality through freedom – that was the sole thing we could accomplish, and it alone was our sole concern. (A558/B586)

In the Critique of Pure Reason, Kant’s goal is simply to open a space for freedom. In the second stage of the argument, Kant argues for the reality of freedom.6 In the second Critique, Kant claims it is a “fact of reason” that people have moral obligations. Kant uses this fact “of which we become immediately conscious” to support the reality of freedom (5:29, cf. 5:6, 31–2, 42, 47, 55, 91, and 104–5; and Rawls 2000: 253–72; Beck 1960: 166–75; Allison 1990: 230–49; and Ameriks 1981). Moreover, this moral obligation does not depend upon any particular details about oneself. Rather, the good concerns something that anyone ought to do in the same situation. The moral law that underlies this obligation binds universally. And this universality is not merely contingent. Obligation is not due to some feature that agents happen to have in common such that if particular persons came to lack this shared characteristic they could be excused from morality. Moral obligation is both universal and necessary. From these general features of moral obligation, Kant argues in the Groundwork that the moral law that underlies obligation must have nothing to do with any particular or contingent content or context of choices but with the form of morally acceptable choice.7 Because whether or not a choice is permitted does not depend on the particular empirical influences in the context of which the choice is made, the way in which one is obligated to act is not determined by natural causes in the world. And yet, to be obligated, it must be possible to act in the way required. But then one must be free in the sense that one can act in a way that is not causally determined by empirical influences alone. Because the moral law requires action on a basis distinct from natural causes, and because transcendental idealism allows freedom beyond natural causes, it must be possible to act on such a basis whenever the moral law is applicable. When morality is at stake, then, one is free in the sense that one’s actions need not have their ultimate ground in natural causes but can be chosen freely in accordance with a universal moral law.8

16

The Problem

2. The Priority of Freedom Although every person is free from determination by natural causes, a person’s actions, as appearances in the world, still are explicable in terms of those causes. In this section I briefly consider the issue of how a free and morally responsible person is related to that person’s actions as objects of experience. In “Kant’s Compatibilism” (Wood 1984), Allen Wood has dealt with several of the metaphysical issues that arise in this context. My aim here is to show merely that freedom is in an important sense prior to actions that can be empirically described. Exactly how this priority is described and defended depends on whether one takes Kant’s distinction between things in themselves and appearances to describe merely two different ways of thinking about things (two perspectives) or two different sorts of things that might interact (two objects). In this section, I do not arbitrate between these two interpretations of Kant. Instead, I explain how the priority of freedom can be explained and justified on either interpretation. A. Asymmetry on a Two-Object Account of Kant’s Transcendental Idealism Richard Aquila offers the best defense of a two-object account of Kant’s transcendental idealism. Aquila claims to “elaborate a defense of the twoobject approach” to Kant’s transcendental idealism (Aquila 1979: 295) and explicitly applies this approach to the nature of the free self. Aquila focuses on the apparent problem that in the case of the person, “a number of passages concerning these phenomena do seem to require the conception of a single entity which may be regarded both phenomenally and as a thing in itself ” (Aquila 1979: 305). Aquila deals with these passages by articulating a two-object intepretation of the notion of a “phenomenal person.” He explains, It is compatible with the two-object interpretation to maintain that a “phenomenal person” is to be identified not with a particular phenomenal object, but rather with a noumenal entity regarded as “ground” of at least some of the behavior exhibited by some phenomenal object. . . . [A] phenomenal person is a certain sort of composite : a composite of a noumenal subject and a phenomenal object. . . . All propositions about phenomenal persons (qua persons) are analyzable into propositions about both a noumenal subject and a phenomenal object. (306)

This account is a two-object account because there is a real distinction between the noumenal subject and the phenomenal object of which that subject is the ground.

Asymmetry in Kant’s Conception of Freedom

17

The “grounding” relationship between the subject and the phenomenal object is akin to a causal relationship, and this relationship is asymmetrical. The free subject is always the ground of the phenomenal object, while the phenomenal object never exerts causal influence on the free subject. Both the metaphysical reflections of the first Critique and the practical arguments of the second Critique defend this asymmetry. In the first Critique, Kant explains how it is possible to talk about relations between the phenomenal and the noumenal. He admits that one cannot make sense of the relation between the free subject and the phenomenal appearance of that subject using schematized categories of the understanding. But one can think about this relation in terms of unschematized concepts. Kant is not totally consistent about whether these concepts are unschematized categories or merely logical forms of judgment. He often says that the free self is the ground (Grund) of its appearance (see, e.g., A545–6/B573–4), making use of the ground-consequent relation from the table of judgments. But at other times he claims that the free self is the cause (Ursache or Kausalit¨at) of its effects in appearance (see, e.g., A539/B567), making use of the category of cause and effect. Kant is best understood as holding that categories, but not schematized categories, can be used to think of things in themselves. When used to think about a free self, “cause” must be taken in a sense that does not involve intuitions of space and time. Kant’s use of the term Grund can then be seen as a way to draw attention to the fact that things in themselves do not cause appearances in the way that we ordinarily understand causation – that is, spatiotemporally. On any interpretation of this relation, it is crucial that things in themselves ground appearances, and not vice versa. In the personal case, the free self in itself is the ground of the appearance of that self, hence of one’s actions and thereby of changes in the world. The relationship is asymmetrical in that the morally responsible agent determines the way the world is, whereas changes in the world do not determine the status of a moral agent. And that means that the ultimate ground of one’s actions is oneself, not prior empirical causes.9 The theoretical need for this asymmetry is found in Kant’s resolution of the third antinomy. This resolution involves positing the possibility10 of freedom as a ground of appearances that is not within the series of appearances themselves. As Kant explains, While for every effect in the [field of ] appearance a connection with its cause in accordance with the laws of empirical causality is indeed required, this empirical causality, without the least violation of its connection with natural causes, is itself an effect of a causality that is not empirical but intelligible. (A544/B572)

18

The Problem

Whereas there can be no freedom in the series of natural causes, freedom can characterize a purely intelligible cause of the empirical series. This need not require that there in fact be such a free intelligible cause. But it establishes that if there is such a cause, it is prior to the appearances that it causes, prior just by virtue of the fact that the intelligible causes the empirical and not vice versa. To this theoretical basis for asymmetry, Kant adds a practical one. Practical reason establishes not just a space for freedom but for the reality of freedom. But the practical argument for freedom also requires an asymmetry between freedom and nature, because Kant’s conception of moral responsibility precludes the freedom of a “marionette or an automaton,” in which one is “free” but at a deeper level determined by natural causes (5:101, cf. 28:267). For practical reasons, the noumenal self must be truly free. It must not be itself affected by the series of appearances that it grounds. B. Asymmetry on a Two-Perspective Account of Kant’s Transcendental Idealism When one interprets Kant according to a two-perspective theory, the picture gets more complicated. On this account, there is neither a real distinction nor causal interaction between things in themselves and their appearances. Instead, the free self and the self as appearance merely refer to two different perspectives one can take on the self. Christine Korsgaard explains the distinction in terms of two different “standpoints”: The deliberating agent, employing reason practically, views the world as it were from a noumenal standpoint. . . . The theorizing spectator, on the other hand, views the world as phenomena, mechanistic and fully determined. The interests of morality demand a different conceptual organization of the world than those of theoretical explanation (MM 217; 221; 225). Both interests are rational and legitimate. (Korsgaard 1996a: 173)

People have different reasons to give accounts of actions. Depending on the “interests” that motivate the account, one assumes either a practical standpoint according to which one is the ultimate free cause of the action or a theoretical standpoint within which one can trace the natural causes that give rise to the action. Korsgaard is right to point out that both are rational and legitimate. But she then goes on to say that “it is important that neither standpoint is privileged over the other – each has its own territory” (173). Korsgaard quickly catches herself, recognizing that this symmetry is contrary to Kant’s own account, and adds, “Or, if either is privileged, it is the practical, because, according to Kant,

Asymmetry in Kant’s Conception of Freedom

19

‘every interest is ultimately practical’ (C2 121)” (173). Korsgaard’s concession to Kant is important here, but she does not sufficiently work out a Kantian justification of the asymmetry.11 As a result, it is all too easy to think that it really does not matter whether one perspective is privileged over another. They just have separate spheres. And, it might seem, there is no third “metaphysical” perspective that gets at what we “really” are. For Kant, however, it is crucial to establish the priority of the practical.12 Kant does not establish this priority through any third perspective that goes beyond the theoretical or the practical. Instead, he argues that from within each of these perspectives, the status of the other perspective is raised. In the third antinomy of the first Critique, for example, the demands of reason within the theoretical perspective raise the problem of the need for an unconditioned ground of the series of appearances. In the course of discussing the relationship between the theoretical and practical perspectives, John Rawls explains the interests of the theoretical perspective: Theoretical reason has two legitimate interests: one is the positive interest in regulating the understanding and unifying into the highest possible systematic unity the low-level empirical knowledge it provides; the other is the negative interest in restricting speculative folly. (Rawls 2000: 324)

Rawls goes so far as to say that “the unity of reason is established . . . by the full satisfaction of the legitimate claims of theoretical and practical reason as articulated in the form and structure of the two points of view” (324). In his description of this unity of reason, Rawls focuses on the fact that the theoretical and practical perspectives do not make conflicting claims. Theoretical reason “abdicates” a “space” within which practical reason can make practically justified claims (324). The interest of theoretical reason in the systematic unity of knowledge leads to a need on the part of theoretical reason for some fundamental ground. In the context of a two-perspective account of Kant’s idealism, this cannot be some fundamental metaphysical “cause” of appearances but must be a more fundamental perspective on the same world that theoretical reason investigates. Now Kant insists that one cannot give any theoretical argument to show that there is such an ultimate perspective. The demand of theoretical reason for completeness within empirical investigation goes beyond the conditions of empirical knowledge. But reason’s theoretical demands are real and legitimate, and the Critical solution is to hold open the possibility, and even the hope, of a new perspective that is not itself restricted by the forms of intuition. By positing

20

The Problem

the priority of this new perspective to its own, theoretical reason can hold open the possibility of satisfying its own demands. Thus Kant suggests in the first Critique that “the architectonic interest of reason” provides a speculative recommendation for the thesis of freedom, because without it, “a complete edifice of [empirical] knowledge is . . . altogether impossible” (A474–5/B502–3). And in the second Critique, Kant reminds his readers that the transcendental freedom for which he argues on the basis of the fact of the moral law is the unconditioned ground that “speculative reason needed” (5:3, see also 5:7, 48–9, 107, 120). This implies that the structure of the two points of view involves an asymmetry between points of view. In “What Is Orientation in Thinking” – published just after the Groundwork and before the second edition of the first Critique – Kant offers an extended account of this sort of “need which reason imposes on itself ” (8:136). Kant suggests that the “need of reason can be regarded as twofold in character: firstly, it has a theoretical use, and secondly, a practical use” (8:139). In this essay, Kant is particularly interested in the need of reason to posit God, but the need of theoretical reason to posit God has a parallel in reason’s need for a unconditioned free cause in the first Critique. Kant suggests that the need of reason to posit something unconditioned as a means of “orientating ourselves in thought” can give a “subjective ground for presupposing and accepting something which [theoretical] reason cannot presume to know on objective grounds . . . purely by means of the need of reason itself ” (8:137). Kant is careful, later in the essay, to point out that this subjective ground is merely conditional and thus does not provide the kind of justified belief that is provided by the postulates of practical reason in the second Critique. That is, whereas theoretical reason requires positing God or freedom only “if we wish to pass judgment on the first causes of all contingent things,” practical reason compels us “to assume that God exists . . . because we must pass judgment ” in practical matters (8:139). As a result, insofar as freedom is established as a condition of moral responsibility, it justifies a “conviction of truth [that] is not inferior in any degree to [empirical] knowledge . . . even if it is totally different from it in kind” (8:141). Because of its “need” for completeness in empirical investigation, theoretical reason itself points to a perspective that it cannot justify or characterize. Moreover, for this other perspective to satisfy the need of theoretical reason, it must provide an explanation that conditions but is not conditioned by what is described empirically. In the second Critique, Kant nicely summarizes both the range and limits of this demand on the part of theoretical reason. He there explains

Asymmetry in Kant’s Conception of Freedom

21

what could be said on the basis of theoretical reason alone, in the first Critique, about the relationship between the theoretical and practical perspectives: We could defend the thought of a freely acting cause. . . . On the one side, in the explanation of events in the world and so too of the actions of a rational being, I grant the mechanism of natural necessity . . ., while on the other side I keep open for speculative reason the place which for it is vacant, namely the intelligible. . . . (5:49)

Theoretical reason has a need for and holds open a vacant space for freedom. And this vacant space, as described by theoretical reason, must be a perspective that is “causally” prior to the theoretical space the need of which it satisfies. Whatever perspective fills that vacant space must be considered, even by theoretical reason, to provide the ultimate explanation that empirical investigation, because the forms of intuition condition it, can never provide. Theoretical reason holds open the possibility13 that the ultimate explanation of actions is beyond the conditions of possibility of experience because only on this sort of foundation can it hope for that “complete edifice of knowledge” which is the regulative ideal of empirical investigation. What theoretical reason can not do, however, is “realize this thought” – that is, show that there actually is another legitimate perspective on the actions of a rational being. “Pure practical reason now fills this vacant place” (5:49).14 At the abstract level of theoretical reason’s need, the perspective that theoretical reason privileges is any perspective from which freedom can be justified, not particularly the practical one. There is a space left open, from within the perspective of theoretical reason, for a perspective that satisfies theoretical reason’s interest in systematic completeness. But this new perspective can play this role only by being more fundamental than the theoretical perspective itself. Once the practical perspective provides a basis for freedom, it satisfies the demand raised within the theoretical perspective and is considered, from that theoretical perspective, to have priority.15 From the practical standpoint, one also considers the status of the empirical. An agent is interested in evaluating actions or volitions that occur in the realm of experience. This is true for both moral evaluation and moral deliberation. When one evaluates oneself or another, one evaluates an action or a volition that is experienced. Only insofar as a volition enters the realm of experience16 can it be an object of evaluation, but if a volition is considered as experienced, it must be considered according

22

The Problem

to the perspective of empirical investigation. This does not mean that the empirical perspective is the proper perspective for moral evaluation; it clearly is not. But when we evaluate something, the perspective of empirical investigation is at hand. We can privilege the moral perspective for the purpose of evaluation only because we consider the moral perspective more fundamental than the empirical one. The self considered as free is the ultimate ground of what is observable, and not vice versa. In the case of moral deliberation, the relationship between the empirical and practical perspectives also arises. When one deliberates, one deliberates about what to do. The decision that I make is a decision that is a possible object of empirical investigation. What is more, and this is crucial, I care about the decision insofar as it is an object of possible empirical investigation. My choice has weight for me because it will have effects in the world. This does not mean that all I care about is results. It is crucial for Kant that morality not be based on consequences. But what I care about is an actual decision, a decision that enters the series of causes in experience. As Kant explains in the Critique of Practical Reason, “this [moral] law is to furnish the sensible world . . . with the form of . . . a supersensible nature “(5:43, emphasis added). What I deliberate about from the practical perspective is what I will choose to do, where this choice and the action that proceeds from it are possible objects of empirical investigation. Thus it makes all the difference in the world, from the practical perspective, that this choice can be thought of from an empirical perspective. But it also makes all the difference in the world that the practical perspective, according to which I am free, is more fundamental than this empirical perspective. In deliberating I take my choice understood from a practical perspective to be the real ground of that choice understood from an empirical perspective, and not vice versa. Thus from both the theoretical and the practical perspectives, the practical is considered more fundamental than the theoretical. Even if one does not accept in general that the theoretical standpoint must be prior to the practical, the practical perspective has priority over the theoretical in the arena of moral anthropology. As we will see, moral anthropology involves empirical insights about human beings that are put to practical use in cultivating moral character. The person putting the insights to use must be considered from a practical perspective because he or she is acting, and the one in whom moral character is cultivated must be considered from a practical perspective because what is cultivated is moral character. And within the practical perspective itself, the practical

Asymmetry in Kant’s Conception of Freedom

23

certainly has priority over the empirical. Thus even if the priority of the practical is not maintained in general, it must be affirmed in precisely the context in which it poses problems for Kant’s anthropology. Whether one holds a two-object or a two-perspective view of the relation between the free self and the naturally determined self, these two objects or perspectives are not symmetrical. The free self (or the self considered as free) is the ground of the empirical (or empirically considered) one and not vice versa. Even if the empirical self is also phenomenally determined according to natural causation, nothing empirical determines the fundamental nature – in particular, the moral status – of the free self. Kant rightly describes the relation between the free self and its appearance, or between two perspectives on the self, as one of ground and consequence, and even of cause and effect.17 And so he rightly suggests that when one empirically investigates oneself or others, one merely observes what people look like, not what they really are.

3. Incompatibilist Freedom The belief in the priority of freedom precludes any straightforward compatibilism. Like compatibilists, Kant holds that human actions are both causally predetermined and free. But unlike most compatibilists – so-called “soft determinists” – Kant insists that the free cause of one’s actions cannot be predetermined. Whereas soft determinists interpret “freedom” in such a way that a free cause can be determined by prior natural causes, Kant affirms that people are free causes of their actions, and that those actions are determined by an infinite series of natural causes, but also that a free cause cannot itself be causally determined. In this sense, Allen Wood’s description of Kant’s position as a “compatibilism between compatibilism and incompatibilism” is apt (Wood 1984). Kant is adamant about the insufficiency of standard compatibilist accounts of freedom as grounds for moral imputation.18 He is able to avoid these standard accounts and the problems with them because of his belief in the priority of freedom over natural causation. In the Critique of Practical Reason, Kant clearly expresses his dissatisfaction with “soft determinist” accounts of freedom. If such an account were true, Kant says, A human being would be a marionette or an automaton . . . built and wound up by the supreme artist; . . . the consciousness of his own spontaneity . . . would be mere delusion. (5:101)

24

The Problem

Freedom that has its ultimate cause in anything else, whether this is a natural world or divine Creator, is not real freedom but “mere delusion.”19 And this soft determinist freedom is simply not enough to provide a condition of the possibility of moral responsibility, as Kant makes clear elsewhere in the second Critique: It is a wretched subterfuge to . . . say . . . that the kind of determining grounds of [man’s] causality in accordance with natural law agrees with a comparative concept of freedom . . ., e.g., that which a projectile accomplishes when it is in free motion, in which case one uses the word “freedom” because while it is in flight it is not impelled from without. (5:96)

A few sentences later, Kant explains why he believes that this compatibilist solution is unacceptable: They therefore leave no transcendental freedom, which must be thought as independence from everything empirical and so from nature generally . . .; without this freedom . . ., which alone is practical a priori, no moral law is possible and no imputation in accordance with it [is possible]. (5:97)

Kant clearly believes that a compatibilist account of freedom does not leave room for the freedom necessary for moral responsibility. If one is to be held morally responsible for an action, it must be possible to be free of empirical determination, because this sort of autonomy is morally required. If forces of nature ultimately cause all action, the best that anyone can hope for is action in conformity with the demands of the moral law. One could never act from the moral law itself, because the moral law is not a force of nature. And acting from the moral law is precisely what the moral law demands. But at the same time that Kant holds as a condition of possibility of moral responsibility a sort of freedom that is not compatible with soft determinism, he is committed to the view that human actions can be described in terms of natural causes. Whereas most reconciliations of freedom and natural determination insist that the ultimate explanation of an action is according to natural causation – such that one’s “freedom” is just a proximate but predetermined cause of an action – Kant insists that the fundamental level of explanation is freedom. Natural causes just express that freedom. The difference between Kant’s “compatibilism” and the accounts that he attacks is “the difference between the laws of a nature to which the will is subject and of a nature which is subject to a will ” (5:44). That is, the primacy of freedom over nature distinguishes Kant from his compatibilist predecessors. Freedom is not a means by which natural causes affect the world. Nature with its causes is instead

Asymmetry in Kant’s Conception of Freedom

25

the means by which freedom effects change in the world. There are, of course, numerous problems with this view.20 What is crucial here, though, is just that this description allows Kant to hold to a peculiar kind of compatibilism without a soft determinist conception of freedom. Freedom and natural causation are compatible only because free choices, totally undetermined by anything prior to them, are themselves the grounds of a series of actions in the natural world. There is one problem with Kant’s account that I take up directly. Kant’s remarks about the cultivation of character and the role that empirical influences can have in shaping and enabling freedom might lead one to think that his conception of freedom cannot be as absolute as this account makes it out to be. This problem is the subject of the rest of this book. Before turning there, however, I must make one more provisional comment on Kant’s account of freedom.

4. The Inscrutability of Freedom Crucial to Kant’s account of freedom is the notion that one can distinguish between a free self (or practical perspective on the self) and a causally determined self (or empirical perspective on the self). As we have seen in sections 2 and 3, at least some detail with respect to the relationship between these selves or ways of thinking of the self is important for differentiating Kant’s theory from soft determinist versions of compatibilism. But Kant is also wary of giving too much detail about the nature of freedom and its relation to the empirical self. When he introduces the distinction between intelligible and empirical character in the first Critique, for example, he insists, This intelligible character can never, indeed, be immediately known, for nothing can be perceived except in so far as it appears. It would have to be thought in accordance with the empirical character. (A539–40/B567–8)

This theme, that the intelligible character can be known only through its appearance, or empirical character, is reiterated throughout Kant’s moral philosophy. Still, Kant does insist that one can intelligibly say at least some things about freedom. The most important of these is that freedom is real. In our judgments in regard to the causality of free actions, we can get as far as the intelligible cause, but not beyond it. We can know that it is free. (A557/B585)

26

The Problem

In making the claim that there actually is an intelligible cause that is free, Kant goes beyond21 the strict speculative limits of the first Critique and acknowledges as much. His argument for the reality of freedom is based, even in a work that is supposed to limit itself to theoretical reason, on the fact that we impute responsibility to human agents, the fact of practical reason. Kant’s fuller articulation of the way in which practical reason establishes the reality of freedom comes in the Critique of Practical Reason: We could [in the first Critique] defend the thought of a freely acting cause . . . by showing that it is not self-contradictory to regard all its actions as physically conditioned insofar as they are appearances and yet also to regard their causality as physically unconditioned insofar as the acting being is a being of the understanding; . . . I do not cognize at all the object to which such causality [i.e. freedom] is attributed . . . but I nevertheless remove the obstacle inasmuch as on the one side, . . . I grant the mechanism of natural necessity . . ., while on the other side I keep open for speculative reason the place which for it is vacant, namely the intelligible. . . . Pure practical reason now fills this vacant place with a determinate law of causality in an intelligible world (with freedom), namely the moral law. (5:49)

While speculative reason can establish a “vacant space” for freedom, practical reason establishes its reality. Although both of these passages suggest room for some knowledge of the free self, Kant is careful, in both contexts, to limit such knowledge to the mere fact that one is free without any understanding of how one is free. Hence he says in the first Critique that we can get as far as the intelligible cause – that is, its reality – but not beyond it to understanding its nature. He goes on, But to explain why in the given circumstances the intelligible character should give just these appearances and this empirical character transcends all the powers of our reason, indeed all its rights of questioning. (A557/B585)

Likewise in the second Critique, Kant insists that there is an important sense in which one cannot even extend the category of causation to the free self: Even the concept of causality, which has application and so too significance strictly speaking only in reference to appearances, in order to connect them into experiences (as the Critique of Pure Reason proves) is not enlarged in such a way as to extend its use beyond the boundaries mentioned. For, if reason sought to do this it would have to try to show how the logical relation of ground and consequence could be used synthetically with a kind of intuition different from the sensible, that is, how a causa noumenon is possible; this it cannot do. (5:49)

Asymmetry in Kant’s Conception of Freedom

27

This inscrutability of freedom is reiterated later in the Critique of Practical Reason (see 5:72), throughout Religion within the Boundaries of Mere Reason (6:22, 43–4, 49fn, 50, 59, 62), and in Kant’s lectures in both metaphysics and ethics (27:505, 29:1020–2). Despite this inscrutability, however, Kant claims, “Freedom is regarded as a kind of causality . . . with respect to actions possible through it as appearances in the sensible world” (5:67). This application of “a kind of causality” to freedom seems to contradict Kant’s insistence that the categories cannot apply to the free self. The apparent contradiction can be resolved if we get clearer on exactly why and in what sense Kant prohibits the application of the categories beyond experience. Kant is concerned with three related problems that could arise if one applies the categories to the free self. First, his restriction of the categories to empirical objects is part of a general limitation of theoretical knowledge to objects of possible experience. Because the free self is not an object of possible experience, one cannot have theoretical knowledge of it. The categories as concepts through which one theoretically knows objects do not apply to the free self. Hence, after establishing the reality of the free self, Kant warns, I do not now claim to know theoretically by this concept the constitution of a being insofar as it has a pure will; it is enough for me to . . . consider myself authorized to make no other use of it than with regard to the moral law which determines its reality, that is, only a practical use. (5:55–6)

Insofar as the categories apply to the free self, they apply only for practical use. One cannot have any theoretically warranted knowledge about the causality of a free being, nor can one put the knowledge that one is free to use in any theoretical justifications of further claims. Second, one of the important uses of the categories is to establish the real possibility of an object. To show that something is a real possibility, one must show that it is a possible object of experience, and this is done by showing that the categories and forms of intuition apply to it. In the first Critique, Kant considers the possibility of something that cannot be an object of possible experience for human beings because it does not conform to the form of human intuition. About such an “object,” Kant says, If we suppose an object of a non-sensible intuition to be given, we can indeed represent it through all the predicates which are implied in the presupposition that it has none of the characteristics proper to sensible intuition; that it is not extended or in space, that its duration is not in time, that no change (succession of determinations in time) is to be met with in it, etc. But there is no proper

28

The Problem

knowledge if I thus merely indicate what the intuition of an object is not, without being able to say what it is that is contained in the intuition. For I have not then shown that the object which I am thinking through my pure concept is even so much as possible. . . . To such a something not a single one of all the categories could be applied. (B149)

Here Kant considers precisely the situation in which he finds himself in his resolution to the third antinomy, where he supposes something that is free and hence not sensible. In this passage, he suggests first that one is entitled to make various negative claims about such a free subject. That entitlement is important for his future accounts of human freedom because it allows him to deny to the free will any properties that depend on space and time. Here Kant also denies that the categories can apply to such an object. What is important to recognize, however, is that the reason Kant gives for this restriction is that “save in so far as empirical intuition provides the instance to which I apply [the category], I do not know whether there can be anything that corresponds to such a form of thought” (B149). The problem with applying the categories is that their application depends upon the reality of the object to which they are applied. In the case of freedom, one has no theoretical knowledge of its reality, and hence one cannot put the categories to any theoretical use with respect to the free self. However, once one establishes the reality of freedom on practical grounds, one applies the categories – the only way in which one can think of a thing – in one’s practical use of the concept of freedom. In this sense, one thinks of freedom as the ground or cause of one’s actions. Third, Kant is suspicious of the application of the categories to freedom because knowledge of objects is always knowledge of objects under schematized categories, and these cannot be applied in any sense to a free subject. We have seen that concepts are altogether impossible, and can have no meaning, if no object is given for them, or at least for the elements of which they are composed. . . . We have already proved that the only manner in which objects can be given to us is by modification of our sensibility. . . . The conditions of sensibility [thus] constitute the universal condition under which alone the category can be applied to any object. This formal and pure condition of sensibility to which the employment of the concept of understanding is restricted, we shall entitle the schema of the concept. (A139–40/ B178–9)

In the first Critique, Kant emphasizes that the categories can be applied in a way that is intelligible for theoretical reason only when they are schematized according to the intuitions of time and, for external objects,

Asymmetry in Kant’s Conception of Freedom

29

space. Whenever one seeks to picture what it means for something to be the cause of some other thing, one can imagine this category only under the conditions of time. But time does not apply to the free agent, which is not a part of experience. Hence the categories, insofar as they can be imagined by human beings, cannot apply to free agents.22 Nonetheless, one can apply the categories from the standpoint of practical reason, where they take on purely practical significance. Thus one says that the free agent “causes” some appearance, not in the sense that a free act precedes and is necessarily succeeded by that appearance but in the sense that moral responsibility for the appearance is imputed to the agent. Likewise, a free agent is a “substance” not in the sense that it persists through time but in the sense that the appearances of the past, present, and future are imputed to a single agent. In this context, it is appropriate to think of the categories as applying to free agents in a way that is analogous to the way schematized categories apply to objects in the world. One cannot strictly explain the nature of that agent, but one can understand the relationship between it and its appearance by analogy with relationships in experience. Kant explains this in the Critique of Judgment: Analogy . . . is the identity of the relation between bases and consequences (causes and effects) insofar as it is present despite what difference in kind there is between the things themselves. . . . Though I can conceive of the causality of the supreme world cause when I compare its purposive products in the world with the works of art of man, by analogy with an understanding, I cannot by analogy infer that it has these [same] properties; for in this case the principle that authorizes such an inference is just what is lacking, i.e., we do not have paritas rationis for including the supreme being in one and the same general kind as man (as regards their respective causalities). The causality of world beings (which includes the causality through understanding) is always conditioned by the sensible, [and so] cannot be transferred to a being that has no generic concept in common with them except that of a thing as such. (5:464fn)

Here Kant explicitly allows that the relation of causality can be applied even to the relationship between something noumenal (the supreme world cause) and the empirical world. However, this relation is applied without giving knowledge of what is noumenal. It merely forms the basis of an analogy. Similarly, Kant can allow, once the reality of the free self has been established on practical grounds, the use of the relation of causality to make analogical statements about freedom, but these must be understood as merely analogical and put to purely practical use. Kant’s restriction of the scope of knowledge of freedom has two important implications for his account of the relationship between freedom and

30

The Problem

anthropology. First, that account need not take the form of an explanation of freedom. Kant has theoretical reasons for claiming that it is impossible for human beings to understand the nature of freedom and how a free self underlies an empirical one. His account must show that moral anthropology is consistent with freedom, but it need not explain how freedom itself is possible. Second, his response cannot make use of claims about freedom in theoretical arguments, because these claims have no theoretical warrant. Kant cannot apply the categories in their schematized form to free agents. For that matter, Kant cannot apply to free agents any concepts that make use of the intuitions of space and time. If he does make statements that seem to apply the schematized categories, his argument must work when these are interpreted analogically.23 Moreover, he cannot make use of the categories for any theoretical knowledge of the self. Kant is perfectly free to discuss freedom in terms of the categories because we cannot think of it in any other way, but such discussion must be purely practical. It must be limited to pointing out the conditions of possibility of moral obligation.

5. Conclusion In this chapter, I presented Kant’s arguments for human freedom, arguments based on the fact of moral responsibility. I have shown that the freedom established on the basis of these arguments must be considered prior to one’s appearances in the world, whether one understands this relationship to hold between two objects or between two perspectives. I have further shown why Kant takes this asymmetry to both imply and allow for a stronger conception of freedom than is available from soft determinist accounts. This conception of freedom raises several problems, but the one on which I focus is that it seems to preclude any robust moral anthropology. If the asymmetry between nature and freedom is absolute, such that freedom can influence nature but never vice versa, as I have indicated here, then there seems to be no room for observations about empirical influences on the free will. The relation between freedom and nature allows influence in only one direction. If moral anthropology requires mutual influence, there seems to be no way out of the dilemma. In the next two chapters, I take up the question of just what sorts of influences anthropology does investigate. I show that the influences are both empirical and morally relevant, and thus that there is at least a prima facie problem with Kant’s mature moral theory. Chapters 5 and 6 will then show how Kant can solve that problem.

2 Anthropology as an Empirical Science

In the previous chapter I showed that Kant’s theory of freedom implies an asymmetry according to which freedom is prior to nature. On one account, this means that the free self can influence but cannot be influenced by the empirical world. To interpret in another way, one might say that the empirical perspective on the self cannot play a fundamental role within the practical perspective, whereas the practical perspective is necessary in order to complete the empirical one. Whichever interpretation one prefers, one would not expect to find Kant outlining empirical helps and hindrances to moral progress. Such empirical influences on morality represent a reversal of the required asymmetry. However, throughout Kant’s mature moral philosophy, one finds references to apparent influences on morality that seem to be empirical. Among the helps and hindrances to which Kant refers are beliefs such as the practical postulates (5:122–35, 452–3, 472), belief in grace (6:44–78), and even belief that virtue is rewarded in one’s life (6:216, 474); feelings such as sympathy (6:457), respect (5:71–89, 4:400, 6:39), and love (6:399, 8:337–8); social institutions such as polite society (6:473–4, 7: 151f., 25:502–5, 1455), a peaceful republic (8:375), and churches (6:93–202); the beautiful or sublime (5:268–9, 299, 354–6); and moral education (5:151f., 6:474f., 9:480– 99). A great deal could be said about each of these influences. Some are more clearly empirical than others are. Some are more clearly influences on morality than others are. All of them potentially raise problems for Kant’s theory of freedom. The problem outlined in the Introduction, and my eventual solution, applies to most of the helps and hindrances discussed in Kant’s moral philosophy. But the explicit focus of my attention is the way in which Kant’s 31

32

The Problem

anthropology presents empirical helps and hindrances to having a good will. In this chapter and the next, I show the extent to which Kant’s anthropology satisfies the description of anthropology that Schleiermacher criticizes. That is, Kant’s anthropology is at once empirical and morally relevant. In this chapter, I show that anthropology is empirical. In the next, I show that it is morally relevant. One of the difficulties in offering an analysis of Kant’s anthropology is that there remains disagreement about where that anthropology is to be found. There can be no doubt that the published Anthropology from a Pragmatic Point of View and the lectures on anthropology should be considered anthropology for Kant, although even in these cases there is disagreement about the relation of this anthropology to the moral anthropology referred to in the Groundwork and Metaphysics of Morals.1 Many also include as “anthropological” works such as Religion within the Boundaries of Mere Reason, parts of the Critique of Judgment, and many of Kant’s shorter essays, especially those dealing with history (cf. Anderson-Gold 2001, Munzel 1999, and Louden 2000). My account focuses on the published Anthropology and related lectures because it is this subject matter more than any other that bothered Schleiermacher, and this is the most uncontroversially anthropological of Kant’s works. This focus should not be interpreted as a denial of the anthropological nature of Kant’s other works, however, and in the course of my discussion I occasionally point out connections between his explicitly anthropological works and his other writings. There are two important dimensions to the question of whether anthropology is empirical, which I take up in turn. First, one can ask whether anthropology is empirical in its method. When one undertakes the study of anthropology, does one investigate empirically? Second, one can ask whether the subject matter of anthropology is empirical. Is the anthropos that is studied by anthropology the human as a free, noumenal agent, or humans as observable in the world? And insofar as anthropology finds influences on human beings, are these purely empirical influences? The status of both the method and the subject matter is important in the context of Kant‘s account of freedom, and the two issues are intimately related. Insofar as moral anthropology is empirical in method, it suggests that there is a perspective within which human beings must be considered objects of empirical study but at the same time as morally responsible. Moreover, if it can be shown that anthropology is empirical

Anthropology as an Empirical Science

33

in its method, this proof will provide an important piece of evidence that the subject matter is empirical. Whether or not the subject matter is empirical is important both to highlight the problem of moral anthropology for Kant and to move toward a solution to this problem. The problem becomes clearer because if empirical anthropology finds empirical helps and hindrances, and if these are also moral helps and hindrances (which will be discussed in the next chapter), then there is a good prima facie case that this sort of anthropology is incompatible with Kant’s account of freedom. At the same time, the fact that Kant’s anthropology is primarily related to the self as observed, rather than to the free noumenal self, may provide a way to escape this apparent incompatibility. There is no problem with empirical influences’ affecting the observed self. To show how this limitation helps solve the dilemma, it will be necessary to explain how the observed and free selves (or the self considered as free and as observed) relate to each other. I will take up this task in Chapter 5. At this point it is enough to mention briefly that by drawing attention to the observed self, anthropology may begin to point toward Kant’s overall account of anthropology and freedom.

1. Kant’s Conception of His Empirical Anthropology That Kant’s anthropology is empirical is the most consistent claim that Kant makes about his work in anthropology, persisting through all his lectures on it and enduring in his published work from his earliest works to his published Anthropology. As far back as 1755, in his Allgemeine Naturgeschichte, Kant says, “consciousness and the senses should teach us what man is” (1:366). In the Groundwork, he distinguishes pure morality, which is the rational part of ethics, from “practical anthropology,” which is the “empirical part” (4:388). In the Metaphysics of Morals, Kant explains that in anthropology, to which a metaphysics of morals can be applied, “we shall often have to take as our object the particular nature of human beings, which is cognized only by experience” (6:216–17, emphasis added). And in the Anthropology itself Kant makes clear that anthropological investigation is a matter of empirical observation, not a priori theorizing. Observations about the causes of memory that are not put to any use are “speculative theorizing” that “is a sheer waste of time.” However, proper anthropological investigation also consists in “observations,” but only when one “distinguishes between those observations which have been found to hinder and those which have been found to promote” the faculty under investigation (7:119). Both the fruitless and the proper sorts of

34

The Problem

anthropology are empirical. The difference is that proper anthropology puts empirical observation to use. The empirical nature of anthropology is also evident in the difficulties that Kant raises for the prospect of anthropology: (1) one “who notices that he is being observed . . . will not be able to behave as he really is”; (2) “if he wants to observe himself . . . he does not observe himself when his impulses are in action, and when he observes himself, the impulses are at peace,” so he cannot truly observe his own impulsive behavior; and (3) “conditions of time and place, when lasting, result in habits which, it is said, constitute second nature, which makes man’s judgment of himself more difficult” (7:121, cf. 4:471, 15:660 [Reflexion 1482], and 25:1212). All three difficulties pose a problem for anthropology only because anthropology is a science that depends upon accurate observations.2 Anthropology is dependent upon good observation precisely because it is an empirical science. Showing in exactly what sense anthropology is empirical, however, requires a closer look at Kant’s explanation of the methods of anthropology. Before turning to this closer look, however, we must confront one seeming difficulty with the view that all of Kant’s anthropology is empirical.

2. A Problem: How Can Anthropology Be Empirical and Universal? Kant’s anthropology is universal. Kant says that anthropology must begin with “universal knowledge” and that the subject of anthropology is human beings as such. In fact, one of the difficulties for anthropology is precisely that one might mistakenly include contingent details reflecting particular times and places as part of the scientific core of anthropology. This universality does not refer to all anthropological knowledge. Some of the most well-known and most embarrassing parts of Kant’s anthropology are his reflections on women and various races.3 These are meant only to refer to the gender or race under discussion. Kant also distinguishes different characteristics of people with different ages or temperaments, or at different stages in the cultural development of the species. But most of his anthropological observations are meant as universal claims about all human beings. This poses an apparent problem because of Kant’s general account of knowledge. There, Kant connects universality with necessity and argues that the only possible basis for universal claims is a priori reasoning. In the Critique of Pure Reason, he says, “Experience tells us, indeed, what is, but not that it must necessarily be so, and not otherwise. It therefore

Anthropology as an Empirical Science

35

gives us no true universality” (A1–2, cf. B3–4, A91/B124, A196/B241). Whereas in the Anthropology Kant seems to want to extend his claims to all people without exception, his general claims about universality would make this extension seem “merely fictitious” (A196/B241). One seems forced to choose between anthropology’s being universal and its being empirical, when Kant clearly states that it is both. This tension has not been discussed with reference to the Anthropology in particular,4 but it has provoked discussion in the context of Kant’s claim in the Religion that humans are radically evil. In his discussion of radical evil in Kant’s Theory of Freedom, Henry Allison insists that the proper Kantian grounds for believing in radical evil in human nature must be a priori synthetic, although Kant himself does not give the requisite argument.5 Although this is not specifically an argument about anthropology per se, the discussion of radical evil is often taken rightly as an example of Kantian anthropology, and Kant himself claims that it “can only be proved [by] anthropological research” (6:25). Moreover, Allison’s primary reason for insisting that radical evil requires a priori justification also applies to many of the claims of Kant’s Anthropology. Specifically, Allison points out that Kant regards radical evil as “ ‘rooted in humanity itself’ and, therefore, universal” and asks “what grounds . . . does Kant offer for this claim?” (Allison 1990: 154). He then rightly points out that Kant simply “seems to treat it as an unproblematic empirical generalization” (see 6:32–4) but objects that “the most that this evidence can show is that evil is widespread, not that there is a universal propensity to it” (Allison 1990: 154). This is exactly the sort of objection that one seems able to raise against the Anthropology. Moreover, if there is any case in which the universality of an anthropological claim is crucial, it is the case of radical evil.6 Radical evil thus provides a useful test case for the relationship between being empirically based and holding universally. There are two stages to Allison’s argument that Kant’s ascription to humans of a universal propensity to radical evil must be a priori. First, he argues against Kant’s empirical defense of universal radical evil. Second, he offers his own a priori argument. I take these up in turn. Allison first objects to Kant’s empirical argument on two grounds: [1] the most that this [empirical] evidence can show is that evil is widespread, not that there is a universal propensity to it. [2] Moreover, since Kant insists that this propensity concerns only the ultimate subjective ground of one’s maxims and is perfectly compatible with a virtuous empirical character, it is difficult to see what could conceivably falsify his claim. Consequently, it is difficult to take seriously the suggestion that it is an empirical generalization. (Allison 1990: 154–5)

36

The Problem

The second problem – that because Kant could not give empirical conditions that would falsify his claim, he cannot mean it seriously – is not relevant to Kant. It is true that there is no empirical means by which one can falsify Kant’s claim, but this does not mean that it cannot be empirically established. The claim cannot be falsified because a good empirical will is consistent with either a good or an evil intelligible one. One might look good to all appearances – even to oneself – and yet have hidden motives that make one evil. But while a good empirical will is consistent with really being either good or evil, an evil empirical will is consistent only with genuine evil.7 If one can show the presence of evil in a person’s empirical will, one can be sure that the person is evil. And showing evil in the empirical will is something that can be accomplished empirically. Kant goes so far as to say that “we call a man evil . . . because [his] actions are of such a nature that we may infer from them the presence in him of evil maxims” (5:20).8 Although one cannot falsify the claim that one is evil by showing that one is good, one could undermine support for that claim by showing that one’s will is empirically good. Unless Kant can show that one has an evil empirical will, he cannot confidently assert anything about one’s intelligible will. In this sense he has an empirical negative test for his claim, though failing that test does not prove that his claim is false, only that it might be false. Still, even if one can demonstrate an evil intelligible will from an evil empirical one, Allison seems right that one could never show on the basis of empirical criteria that evil or the propensity thereto is a fact of human nature, universally shared by all. Here, as we have already pointed out from the first Critique, Kant must agree with Allison at least in a sense. His claims that human beings are radically evil or have a propensity thereto cannot have strict universality if they are empirical. But Kant also might add that there is another sense of “universal” according to which his claims can hold universally. Moreover, it is crucially important that these claims not have strict universality. In the first Critique, Kant consistently connects “strict universality” with “necessity,” going so far as to insist that these are “inseparable from each other” (B4). For a predicate to hold of human nature according to strict universality, it must be not only true of all people but necessarily true of all. But for Kant it is crucial that the propensity to evil not be universal in that sense. It must not be strictly universal that one have a propensity to evil precisely because this propensity is the result of one’s choice. Kant is adamant on this point: “this propensity must itself be considered as morally evil . . . [and hence] as something

Anthropology as an Empirical Science

37

that can be imputed to man” (5:32). Allen Wood points out, in contrast to Allison, Since this propensity is one for which we ourselves are culpable, it could not have derived its existence merely from the finitude of human nature. The moral imperfections of our volition are limitations we impose upon ourselves, limitations for which we ourselves are responsible, and which must be inferred from the use of our freedom, and not merely from the concept of finite rational volition in general. (Wood 1970: 223–4)

Allison is surely correct that no empirical investigation can yield a judgment about radical evil that would have strict universality. But Kant’s judgments about radical evil impute that evil to free human beings as their own fault, and hence as the result of free choice. Because necessity accompanies strict universality and undermines moral accountability, to argue for strict universality would undermine the moral imputability of the propensity to radical evil. Before moving on to discuss the sense in which Kant’s claim is universal, it is important to address briefly the second stage of Allison’s argument, in which Allison claims to offer an a priori argument showing the propensity for radical evil. Kant would not want an a priori argument for the propensity of radical evil, because such an argument would imply that radical evil is a necessary propensity in human beings rather than a moral fault imputable to free agents. But Allison does not just claim that such a proof is needed; he claims to give such a proof.9 His deduction rests on the claim that unless it is possible for human beings to have propensities to good, it is necessary that they have propensities to evil. Although some have denied this claim (see Wood 1970: 222–3), I agree with Allison’s general thrust here, and it is not necessary to contest this point to show where he goes wrong. Instead, I argue that for Kant it is possible that human beings have what Allison calls a propensity to good, and thus not necessary that they have a propensity to evil. Allison identifies a propensity to good with “a spontaneous preference for the impersonal requirements of morality over one’s own needs as a rational animal with a built-in desire for happiness” (Allison 1990: 155). He then claims that if someone had such a propensity, “the moral incentive would, as a matter of course, always outweigh the incentive of self-love” (155). This account of one with a propensity to good is just a description of one who always does what morality requires. To take one’s moral interest to outweigh all interests of self-love is what Kant requires for one to be morally good. The moral law precisely demands that one choose it over self-love, and it is

38

The Problem

through choice that the moral law becomes an incentive that outweighs self-love.10 Allison claims, however, that if this were the case, For an agent blessed with such a propensity, there would be no temptation to adopt maxims that run counter to the law and, therefore, no thought of the law as constraining. Within the Kantian framework, this means that the law would not take the form of an imperative and moral requirements would not be viewed as duties. (Allison 1990: 155)

Here Allison goes too far. The fact that one takes the demands of the moral law to outweigh incentives of self-love does not in any way remove the incentives of self-love. And as long as self-love is present, it is a temptation. The fact that one resists that temptation, that one chooses what is right over what is in one’s interest, does not mean that there is no temptation to face. And this would not be any different if one always chose the moral law over self-love. Allison’s a priori argument fails, as Kant would hope. Still, Allison’s contention that empirical arguments cannot establish universality is strong. If there is no a priori argument for radical evil, it seems that Kant should not have claimed that it is universal.

3. Anthropological Universality as Empirical Universality In fact, however, although Kant must certainly deny that his ascription of the propensity to evil to the human race is strictly universal, strict universality is not the only kind of universality there is. In the first Critique, Kant repeatedly refers to a kind of universality that experience can provide. Although “experience never confers on its judgments true or strict” universality, it can confer on them “assumed and comparative universality, through induction” (B3). Later, he says that “empirical rules . . . can acquire through induction only comparative universality, that is, extensive applicability” (B124/A91). According to this kind of universality, We can properly only say, therefore, that, so far as we have hitherto observed, there is no exception to this or that rule. . . . Empirical universality is only an arbitrary extension of a validity holding in most cases to one which holds in all, for instance, in the proposition “all bodies are heavy.” (B 3–4)

The first Critique downplays the role of this kind of empirical universal judgment in its search for a priori conditions of experience. Still, Kant’s insistence that this universality is merely “extensive applicability” is only a contrast to that strict universality according to which “no exception is allowed as possible” (B4, emphasis added). As his example of heavy

Anthropology as an Empirical Science

39

bodies makes clear, these judgments need not be taken as merely claims that something is widespread. They hold without exception, but with the possibility of exception.11 In the Critique of Judgment, Kant even explicitly distinguishes these two kinds of universality: “Here it is understood that the universality [Allgemeinheit] is only comparative, so that the rules are only general [generale] (as all empirical rules are), not universal [universale]” (5:213). Allgemein is the term generally translated as “universal” in Kant’s writings and in particular is the term used to describe the “universality” of radical evil in the Religion (6:29). In the Anthropology as well, Kant is careful to use the term Generalkenntnis to describe the “universal” knowledge involved in anthropology (7:120). In both cases, the knowledge at issue is universal only in the comparative sense. As Kant repeatedly insists, “there is no cause for exempting anyone from it” (6:25), but this universality is merely contingent (6:29).

4. The Primary Source of Anthropology: Inner Sense In support of universal empirical claims, in both the Religion and the Anthropology, Kant offers several supporting lines of investigation.12 Through arguing in all of these ways, Kant hopes to establish claims that go legitimately beyond empirical claims about merely widespread human phenomena and that are therefore generalizations that apply to all human beings.13 As we will see, this generality need not imply that Kant takes these claims to be infallible, but if they are correct then they apply to everyone. At the beginning of his Anthropology, Kant mentions three different sorts of observation that can play a role in anthropological investigation. He says, [1] Travel is among the means of enlarging the scope of anthropology even if such knowledge is only acquired by reading books of travel. One must, however, have gained his knowledge of man through [2] interaction with one’s fellowmen at home if one wishes to know what to look for abroad in order to increase one’s range of knowledge of man. Without such a program (which presupposes [3] the knowledge of man), the anthropology of the citizen of the world will remain very limited. (7:120)

Kant here distinguishes three levels of observation. Travel or travel books, through which one is exposed to a wide range of peoples, are the final stage of anthropology. Prior to that, Kant claims that one must study one’s fellow human beings locally. Even prior to this investigation of others,

40

The Problem

one must acquire a certain “knowledge of man,” a general knowledge that investigation of others refines. Kant makes the source of this general knowledge clearer later in the Anthropology, insisting on the importance of “knowledge of man through inner experience, according to which he judges others” (7:143, emphasis added). This emphasis on inner experience as a basis for anthropology fits well with Kant’s account of the difficulties one encounters in attempting to do anthropological research. Of three difficulties that he mentions, two arise when one is “observing oneself” or seeking to issue a “judgment about oneself” (7:121).14 Later, Kant points out that “on the whole, knowledge of man through inner experience, according to which he judges others, is of great importance” (7:143). In an early draft of the Anthropology, Kant goes so far as to say that “anthropology is supplied by the inner sense with content” (7:398). In the lectures on anthropology, this emphasis on inner sense is even clearer.15 There Kant describes anthropology as “the knowledge of people as objects of inner sense” (25:473). In one lecture, Kant begins with a passage very similar to his treatment of travel in the published Anthropology: One who is well-traveled, and has come to know many people . . . cannot say that he knows the person, for he has only known the condition, which is very changeable. If, on the other hand, I know humanity, it must fit all kinds of people. Anthropology is thus not a description of people, but of the nature of people. (25:471)

Kant immediately follows this up with an explanation of the source of the knowledge of humanity that provides the basis for further anthropological investigation: The knowledge of humanity is at the same time my knowledge. . . . Hence we must study ourselves and because we want to apply this to others, we must study humanity. (25:471)

Although anthropological knowledge is knowledge of all people, hence of others as well as oneself, it properly begins with introspection. Only with respect to oneself does one have an inner experience, by means of which motives, inclinations, and desires can be observed. Hence one is equipped to apply anthropology to others only once one has selfunderstanding. Appealing to inner sense is important because it provides a means by which judgments can be made that are both empirical and universal. Claims based on inner sense are empirical, for Kant, because observations

Anthropology as an Empirical Science

41

of oneself are empirical observations. This insistence on the empirical nature of self-knowledge is one of the most important and controversial claims in his Critique of Pure Reason. In the second edition in particular, Kant goes out of his way to insist that “we know our own subject only as appearance, not as it is in itself” (B156, cf. B69). In the preface to that edition, he explains that “through inner experience I am conscious of my existence in time” (Bxl, fn). But existence in time is necessarily the existence of an appearance only, because things in themselves are not dependent upon the intuitions of space or time. In the “Refutation of Idealism” and the B edition of the “Paralogisms,” Kant further reiterates that selfknowledge through inner experience is merely empirical knowledge of oneself (see B275–9, B422, and B429f.). Although judgments based on inner sense are empirical, however, they can assume a certain kind of universality. Inner sense is generically human, and when one observes the structure of one’s cognition, feelings, and desires, one observes a structure that is likely to be shared by others. This is not to say that the mere form of inner sense implies the details of one’s inner experience. If one’s reflections were determined in that way, they would be necessary, synthetic a priori judgments, whereas they are in fact contingent, empirical, and still “universal.” But unless one has a reason to doubt that others have in common a structure in the context of which they experience themselves, one can explore that structure within oneself as a means of discerning general truths about humanity. In particular, the investigation of inner sense gives one the means of understanding the motives, desires, and other aspects of inner experience that underlie the actions of others. The problem, of course, is that one does not observe a generic human inner sense but only one’s own. Thus there is a danger of mistaking contingent facts about one’s particular inner experience for universal facts about humans in general. Kant is well aware of this difficulty. In the Anthropology, two of his three basic warnings about anthropological investigation relate directly to the problem of extending particular facts about oneself to all people. The first of these points out limits to the depth of self-observation. The second points out that “conditions of time and place, when lasting, result in habits which, it is said, constitute second nature, which makes man’s judgment of himself more difficult” (7:121). This is a “difficulty inherent in human nature itself,” and it seems to threaten the possibility of any universal anthropology. Before looking at how Kant mitigates against these difficulties, it is important to remember that for anthropology to be universal, it need not

42

The Problem

establish anthropological claims on grounds that cannot be doubted, but only on grounds that one reasonably claims to hold for all people. Good judgment is needed to distinguish those aspects of one’s inner experience that can rightly be applied to others and those that are peculiar to oneself. One can always make mistakes in one’s judgments about what is truly universal. Luckily, there are some aids to correcting these sorts of mistakes.

5. Secondary Sources of Anthropology: Corrections to Inner Sense First, there is one’s experience of others. Self-observation is only the first stage in anthropology. The next sphere of investigation is the circle of one’s acquaintances. Although Kant suggests that self-observation is the primary method of anthropology, he is clear that anthropology also involves interaction with and observation of others. Self-centered observation provides a basis for further observations, but it is not in itself sufficient because of the problem of identifying particular characteristics of one’s second nature with universal human nature. Thus Kant writes that “one must have gained his knowledge of man through interaction with one’s fellow man at home” (7:120). And in his lectures, after asking, “Where will we then get to know the world, without traveling around it?” he answers, 1) The observation of people that are around us, and a sharp reflection, can replace for us extensive experience and widely surpass whatever an unthinking traveler gets. . . . 2) Civil association. What is essential here is the attention to the human dispositions that often appear in many forms. (25:734)

It is essential to recognize that when Kant points out the value of observing others for anthropology, he insists that a “sharp reflection” accompany this observation. Anthropological knowledge from self-observation comes first, especially insofar as it leads to understanding the way in which inner dispositions are related to external forms in which they appear. But it should be supplemented with observations of one’s fellow citizens. Even the observation of one’s fellows, however, may not sufficiently broaden one’s horizons. At least two potential sources of error remain. First, the observation of one’s fellows at home may be insufficient to distinguish human nature from a second nature that is common to a group. The scope of investigation calls for expansion to ensure that men and women around the world share the characteristics one identifies in one’s

Anthropology as an Empirical Science

43

fellow men at home. Second, one’s observations of others ultimately depend on interpretations of their actions and expressions, interpretations that are rooted in one’s observations of oneself. Thus one is in danger of imposing on others what properly belongs only to oneself. One might, for instance, interpret expressions of extreme feeling as a sign of loss of selfcontrol even when they are not, simply because if one were to express such feeling oneself it would involve such a loss. Or one might na¨ıvely suppose that another’s smiles and affection are genuine acts of kindness, simply because one lacks the experience of certain emotions or of being duplicitous. To further refine the judgment necessary for anthropology, one must look beyond one’s fellow citizens to observations that are both broader and deeper. One can broaden one’s horizons through travel or reading books of travel. Throughout his lectures Kant insists that travel is neither a necessary nor a sufficient condition of good anthropology, but he does recognize it as a useful means for expanding and refining one’s knowledge. Thus he allows that “travel is among the means of enlarging the scope of anthropology, even if such travel is only acquired by reading books of travel” (7:120). In addition, one can both broaden and deepen one’s observations through literature. In answering the question of the sources of anthropology in his lectures, Kant adds to observations of others at home and abroad a third source of anthropological insight: “3) Plays, novels, histories, and various biographies” (25:734). This source is in fact Kant’s favorite and most consistent external source of anthropology. In part this is because he lacked the resources to travel, but Kant also values literature because a well-written book incorporates both the author’s observations of others and his or her inner reflection on those observations, such that one gets an account rich in anthropological insight. Thus Kant says in a lecture in 1775, From where does anthropology spring? Through collections of many observations of people [in] various authors that have had a sharp knowledge of man, e.g. Shakespeare’s theatrical works, the English Spectator, and Montaigne’s essays together with his life. (25:472)

In his published Anthropology, Kant defends the use of literature for anthropology, saying, Such characters as are created, for instance, by Richardson and Moli`ere, still had to be derived in their basic traits from observance of the actual doings of man. Exaggerated as these traits may be in degree, they must still conform to human nature. (7:121)

44

The Problem

Books and plays present the world interpreted according to the inner experiences of their authors. Thus one is exposed not only to a broader extent of experience (in Kant’s case, being able to explore parts of the world he would never be able to visit) but also to an enriching intensive perspective on human affairs. When one reads Richardson, Shakespeare, or Montaigne, one comes to see more clearly what aspects of one’s own inner and outer experience are truly universal. This refinement is especially important for those whose judgment is immature, who are prone to mistake particular details of themselves for universal phenomena. By being exposed to good literature, one comes to see both the variety of human experience and the universal aspects of human nature. Even for someone (Kant, perhaps?) with a relatively cultivated anthropological judgment, reading Jane Austen, Charlotte and Emily Bront¨e, and George Eliot can help refine that judgment. One may find that what seemed universal is more limited than one imagined. If one goes on to read the Mahabharata and the Tale of Genji and encounters literature from people all over the world, one may again rethink one’s judgments. This broadening of one’s horizons need not lead one to renounce all universal anthropological claims, but it may require adjusting one’s views about which claims are truly universal.16 What, then, are the sources of anthropology? The primary source is one’s own inner experience, on the basis of which one can make reflective judgments about the actions of others. Aids to the cultivation of this judgment include not only direct engagement with and observations of others but also the rich anthropological data provided by books and plays. All of these aids to anthropological judgment, it is worth pointing out, are empirical. One does not give a priori arguments based on the concept of a human being or the necessary structure of experience. One’s arguments are based on experience, enriched by a reflective judgment that is itself shaped by experience. So anthropology is deeply empirical. But then just how universal is anthropology? It is properly universal, although not in the strict sense, and not infallibly. When Kant says, for instance, “what the affect of anger does not accomplish quickly will not be accomplished at all” (7:252), he takes this to describe a universal fact about the feeling that he characterizes as the affect of anger. Human beings, he supposes, are not capable of sustaining this particular sort of feeling for long periods of time. If he is correct, then he is making a true and universal, but not a necessary, claim. Is it possible that he is wrong? Of course. In fact, Kant changes several of his anthropological

Anthropology as an Empirical Science

45

views over the course of his lectures.17 And he emphatically denounces various politicians of his time for holding false anthropological beliefs. In Perpetual Peace, he criticizes “moralizing politicians,” who try to cover up political principles which are contrary to right, under the pretext that human nature is incapable of attaining the good which reason prescribes as an idea. . . . They may boast to know men (which is certainly to be expected, since they have to do with so many of them), although they certainly do not know man and what can be made of him, for this requires a higher anthropological vantage-point. (8:373)

Although Kant does not detail the higher anthropological vantage point to which these politicians should turn to refine their false anthropological claims,18 he does make clear that interaction with and observations of many people do not guarantee that one’s anthropological judgments will be true. People can always make mistakes. And given that certain errors, especially errors about human capacities, can be useful for justifying wickedness, one might even expect mistakes to turn up in anthropology. But in addition to pointing out that these anthropological facts give no excuse for error, which Kant does in Perpetual Peace and throughout his moral philosophy, it is important to fight bad anthropology with good anthropology. And that is what Kant does.19 The fact that a claim holds for all people does not mean that one affirms that claim with absolute certainty, nor that it holds necessarily. The danger with empirical investigations of any sort, regardless of how universal they are, is that one might be wrong. Even to claim that some particular person has some characteristic at some particular time might be wrong. So it surely might be wrong that every person has a characteristic at every time. But one still can make reasoned judgments of these sorts. One may conclude, on the basis of a refined judgment about what sorts of things are due to one’s nature, that such-and-such a characteristic is universal. One’s judgment can be refined by more experience, and as it is one may change some of one’s anthropological claims. None of this means that one is incapable of ever being right. And until one has a reason to doubt one’s judgment, one does best to trust it. This is especially true given the scope of the application of anthropology. One is most likely to use anthropology in the context of those with whom one is most familiar, and hence those who have primarily formed the basis of one’s anthropological claims. Errors are less likely to have serious practical implications than if one’s anthropology were applied more broadly. This assumes that one gives close attention in

46

The Problem

one’s anthropological investigations to all of those with whom one has contact. If one systematically excludes from anthropological investigation a class of people with whom one regularly deals, one is not only in danger of being in error, but these errors are likely to have serious practical consequences.20

6. The Empirical Objects of Anthropological Study Thus far I have shown only that the discipline of anthropology is empirical in method, not that it studies empirical helps and hindrances. It is necessary to point out that anthropology is not only an empirical science but also a science with an empirical subject matter. One might think that there is no reason an empirical science would have to highlight empirical helps and hindrances. Perhaps it could be an anthropological fact about human beings that they must believe the postulates of practical reason to sustain commitment to obeying the moral law, but the postulates could be considered an influence of reason itself rather than a purely empirical influence.21 Or one might argue that it is an empirical fact about human beings that one’s happiness is promoted by acting on the basis of the principle that to achieve an end one must pursue the means to that end. But this principle could be a purely rational influence on happiness. For Kant, however, the fact that a science is empirical implies that its subject matter is also empirical. One of the most important implications of the first Critique is the general point that experience is always only of appearance and never of things in themselves, and Kant specifically applies this to knowledge of the self through inner sense (see B155). Thus all empirical knowledge can take only objects of experience as its objects. The dialectic in the first Critique shows that even the use of pure theoretical reason, which is based on the a priori conditions of the possibility of experience, errs in seeking to discern the existence or properties of objects that transcend experience. To attempt any such extension of knowledge on empirical grounds would be an even greater error.22

7. Conclusion Anthropology, for Kant, is an empirical study of human beings. It makes claims that are contingent but universal, in the sense that they apply to all human beings without exception. Moreover, anthropological claims study relations among empirical objects, because this is the proper subject matter of an empirical discipline. Thus insofar as anthropology studies

Anthropology as an Empirical Science

47

influences on human beings and human faculties, it studies only empirical influences. In that context, and given the asymmetry between nature and freedom discussed in Chapter 1, one might expect Kant to restrict the scope of anthropology. In particular, because there can be no empirical influences on one’s free and morally responsible self, one might think that anthropology would be limited to spheres that do not involve specifically moral development. But as we will see in the next chapter, Kant does not limit his anthropology in this way.

3 The Moral Importance of Kant’s “Pragmatic” Anthropology

In the preceding chapter, we saw that anthropology for Kant is an empirical science with empirical subject matter. Not only is its method empirical, but the helps and hindrances that anthropology finds are empirical helps and hindrances. This anthropology will seem to raise problems for Kant’s account of freedom if it is put to moral use, such that empirical helps and hindrances aid moral development. If anthropology were merely an empirical description of human beings or merely pointed out empirical influences on faculties like memory, it would not even appear to conflict with the asymmetrical account of the relation between nature and freedom outlined in Chapter 1. If Schleiermacher’s criticism is to make contact with Kant’s anthropology, that anthropology must be not only empirical but also morally relevant. Thus this chapter takes up the issue of the extent to which Kant’s “pragmatic” anthropology is a moral anthropology. Before showing that Kant’s anthropology is moral, it is important to clear up a potential misconception about what moral anthropology involves. In particular, moral anthropology in Kant’s mature moral theory is not the kind of anthropology needed to apply the categorical imperative to concrete human situations.1 That is, moral anthropology does not have the kind of role indicated by Kant’s claim in the Groundwork that moral “laws require a power of judgment sharpened by experience . . . in order to distinguish in what cases they are applicable” (4:389). If this were the only role that anthropology played in acting well, anthropology would not raise any prima facie problems for Kant’s account of freedom, because such an anthropology need not consider moral agency as itself susceptible to empirical influence in order to apply the moral law to human beings. 48

Kant’s “Pragmatic” Anthropology

49

To apply the general moral requirement to promote the happiness of others, one might need to know something specific about what makes people happy. And to know that false promises are immoral, one must know that human beings are capable of basing trust on promises, incapable of reading one another’s minds, and so on. But none of these anthropological claims need have any bearing on the transcendental freedom of human beings. Kant ascribes a different role to moral anthropology in his mature moral theory. Even in the Groundwork, he describes moral anthropology as important to “gain for [the moral law] access to the human will” (4:389).2 In the Metaphysics of Morals, this role is more clearly defined: “moral anthropology . . . would deal only with the subjective conditions in human nature that help or hinder in fulfilling the laws of a metaphysics of morals. It would deal with the development, spreading, and strengthening of moral principles” (6:217). And in his lectures on ethics, Kant explains the distinction between practical philosophy and anthropology by pointing out that “consideration of rules [in practical philosophy] is useless if one cannot make men ready to follow them [which depends on anthropological insight], so these two sciences are connected” (27:244). In Kant’s mature moral philosophy, applying moral principles is part of the metaphysics of morals. In his Metaphysics of Morals, Kant even goes so far as to offer various “casuistical questions,” including, for example, “Can one at least justify . . . a use of wine bordering on intoxication, since it enlivens the company’s conversation and in doing so makes them speak more freely?” (6:428). Anthropology is very important for these sorts of applications of the moral law, but anthropology that merely serves to apply the moral law is not “moral anthropology.” Moral anthropology deals specifically with subjective conditions that spread and strengthen, rather than merely specify, moral principles. That is, moral anthropology studies the influences that help or hinder one in adopting and acting on moral principles. In a certain sense, of course, this anthropology still involves the application of the moral law to particular human situations, but its scope is greatly restricted. The application with which moral anthropology is concerned is the working out of the imperfect duty to develop one’s moral perfection. This narrower task of laying out helps and hindrances to moral choice is particularly problematic from the standpoint of Kant’s theory of freedom. And this narrower role is prominent in Kant’s Anthropology from a Pragmatic

50

The Problem

Point of View. Unfortunately, whereas Kant is very explicit about the fact that his Anthropology is empirical (as we saw in the previous chapter), he does not as clearly indicate that it is moral. In fact, there are suggestions that might lead one to think that Kant deliberately avoids calling his anthropology moral. Reinhard Brandt points out that whenever Kant discusses anthropology in his purely moral works (whether lectures or published work), he designates it “moral” or “practical” anthropology (Brandt 1997). Yet in the published Anthropology and throughout his lectures on anthropology, he never calls the subject “moral” and only rarely refers to it as “practical.”3 His use of the term “pragmatic” (pragmatisch) to describe anthropology seems to distance Kant’s specifically anthropological work from the anthropology to which he refers in his moral philosophy. The title of Kant’s published work, Anthropology from a Pragmatic Point of View (emphasis added), seems to reinforce this.4 In this context, one would expect Kant’s anthropology to focus on that sphere of practical reasoning which he elsewhere calls “pragmatic.” And one might expect this sphere to be distinct from the realm of the moral. But Kant’s use of the term pragmatic varies throughout his work. This term has at least three different meanings in Kant’s moral philosophy, and aspects of the Anthropology reflect each of these different senses. “Pragmatic” reasoning can involve (1) one’s happiness, (2) the whole sphere of the practical, and/or (3) the use of others to achieve one’s ends.5 Of these, only the first excludes moral elements from its consideration, while the second necessarily includes them. The third involves morality, though more obliquely. In the rest of this chapter I first show that Kant uses the term pragmatic in all of these senses, and hence that the term need not be taken to exclude morality. Then I point out three examples of moral anthropology in Anthropology from a Pragmatic Point of View and the related lectures. Finally, I conclude with a brief remark on the systematic place of moral anthropology in Kant’s anthropology as a whole.

1. Three Conceptions of Pragmatic in Kant’s Philosophy A. Pragmatic as Relating to Happiness The term pragmatic can refer to the sphere of practical reasoning related to the promotion of happiness. In this context, pragmatic reasoning is distinguished from technical and moral types of reasoning. In the

Kant’s “Pragmatic” Anthropology

51

Groundwork, Kant distinguishes three types of imperatives according to corresponding spheres of practical consideration: The first kind of imperatives might also be called technical (belonging to art), the second kind pragmatic (belonging to welfare), the third kind moral (belonging to free conduct as such, i.e., to morals). (4:416–17)6

In this context, Kant contrasts pragmatic and moral imperatives. Whereas moral imperatives are concerned with the necessary laws governing an autonomous being, pragmatic ones are merely “counsels of prudence” that help one promote happiness. This is reiterated in the Metaphysics of Morals, where Kant distinguishes “a human being’s duty himself to develop and increase his natural perfection, that is for a pragmatic purpose” with “a human being’s duty to himself to increase his moral perfection, that is, for a moral purpose only” (6:444, 446). In fact, this distinction between technical, pragmatic, and moral spheres is a persistent theme in Kant’s practical philosophy (cf. 5:12n, 27:255–6). If the Anthropology were presented from a “pragmatic point of view” in this sense, it would primarily offer investigations of human beings that are conducive to achieving happiness. One would expect to find investigations of what sorts of things make people happy, what the most reliable means of achieving those things are, and how to improve one’s future capacities for happiness. One does find these sorts of concerns throughout the Anthropology. To name just one example, Kant’s discussion of the “highest physical good” offers suggestions about the importance of meaningful work to a happy life: “The greatest sensuous pleasure . . . is found under healthy conditions in resting after work” (7:276).7 If this were the only or primary consideration of anthropology, it would not raise any problems for Kant’s theory of freedom. There might still seem to be a problem, in that human beings would be considered objects of empirical investigation and at the same time free in the sense that they can act on counsels of prudence. But this weak sort of prudential freedom is not, on Kant’s understanding, incompatible with ultimate determination by empirical influences. In particular, while it is crucial for Kant’s moral theory that one always be absolutely free to do what is morally required, it is not essential that one always be capable of doing what is in one’s best interest. The freedom involved in prudence could be a soft determinist freedom that would eliminate the problems raised in the Introduction.8

52

The Problem

B. Pragmatic as Practical A second use of pragmatic contrasts the term not with technical and moral but with physiological or theoretical reasoning. In this context, pragmatic considerations include not only those related to happiness but moral and technical ones as well. Anything that is practical is pragmatic. In their introduction to Kant’s lectures on anthropology, Brandt and Stark explain this sense of pragmatic: The overarching concept “pragmatic” ends up being for Kant an approximate frame for the triad of technical skill (for any ends), pragmatic prudence (for our actual end in association with others) and wisdom (for moral ends). . . . The anthropological doctrine of prudence . . . includes the whole territory of possible praxis. Pragmatic is on the one hand a general concept, whose opposite concept is “speculative”; on the other, it is one of three specific concepts, between “technical” and “moral.” (25:xvi–xvii.)9

Brandt and Stark defend this interpretation of pragmatic by drawing on Kant’s letter to Marcus Herz, in which Kant contrasts his anthropology with that of Ernst Platner: This winter I am giving, for the second time, a lecture course on anthropology, a subject that I now intend to make into a proper academic discipline. But my plan is quite unique. I intend to disclose the bases of all sciences, the principles of morality, of skill, of human intercourse, of the method of molding and governing men, and thus of everything that pertains to the practical. (X:146, emphasis added)10

Kant wrote this letter in 1773, when he first began teaching his anthropology lectures. Later, in a Reflexion written sometime between 1776 and 1789, Kant reiterates this general notion of pragmatic: “The historical kind of teaching is pragmatic, when it . . . is not merely for the school, but also for the world or ethics” (Refl. 3376, 16:804, emphasis added, cf. 25:xv). The broad focus on all that is practical characterizes Kant’s lectures in anthropology. Thus in a set of lecture notes taken by Kant’s student Pillau in 1777–8, he outlines the “Uses of Anthropology,” of which one is that “it gives us contact with the subjective principles of all the sciences. And these subjective principles have a great influence . . . in morals” (25:734–5).11 In the published Anthropology, Kant is not as explicit about including moral purposes in his general description of anthropology, but his claim that “pragmatic knowledge of man aims at what man makes, can, or should make of himself as a freely acting

Kant’s “Pragmatic” Anthropology

53

being” (7:119) suggests a broad notion of pragmatic as practical. Aiming at what man “should make” of himself in particular suggests a moral purpose. Further, Kant’s use of the term pragmatic in the published Anthropology (1798) is specifically designed to contrast his anthropology with “physiological” studies of human beings. Kant explains, “A systematic doctrine containing our knowledge of man (anthropology) can either be given from a physiological or a pragmatic point of view, [where] physiological knowledge of man aims at the investigation of what Nature makes of man” (7:119). This contrast between a purely physiological anthropology and his own is exactly the contrast that Kant sought to make, twenty years earlier, between his own pragmatic anthropology and Platner’s. This similarity suggests another reason to interpret the “pragmatic” nature of Kant’s Anthropology in a broad sense, meaning “everything that pertains to the practical.” C. Pragmatic as Pertaining to the Use of Others A third use of pragmatic, related to but different from the previous two, considers pragmatic reasoning to involve the use of others for one’s ends. This is related to the connection between prudence and pragmatic reasoning in the Groundwork and to Kant’s suggestion that pragmatic anthropology teaches prudence, or Klugheit (cf. Sturm 1997 and Munzel 1999:162). In the lectures, for instance, Kant says that “we must study mankind, but not psychologically or speculatively, but rather pragmatically, for all pragmatic instruction is instruction in prudence [Klugheits Lehren]” (25:471), and that lectures that promote “prudence [are] pragmatic” (25:1436).12 Kant explains in an important footnote in the Groundwork what he means by prudence : “The word ‘prudence’ . . . can mean worldly wisdom, . . . the skill of someone in influencing others so as to use them for his own purposes” (4:416n.).13 This sense of prudence as worldly wisdom is the primary sense of prudence in Kant’s discussions of his pragmatic anthropology. In his lectures, for instance, immediately after explaining that anthropology is connected with prudence, Kant says that “the reason that morals and preaching that are full of admonitions . . . have little effect is the lack of knowledge of man” (25:471–2, cf. 27:358). Kant’s lectures on anthropology are meant to supply the knowledge of man necessary so that his students, who will go on to be doctors, lawyers, teachers, or priests, can affect their patients, judges, students, and parishioners. In his later lectures, Kant again emphasizes the

54

The Problem

topic of influence on others in connection with the pragmatic nature of anthropology: We must trouble ourselves to form the way of thinking and the capacities of those people with whom we have to do, so that we are not too hard nor too offensive. So we are taught anthropology, which shows us how we can use people to our ends. (25:1436)

In the published Anthropology the same theme emerges. Kant says in his discussion of the different characters of various national groups: We are interested only in [what] would permit judgment about what each has to know about the other, and how each could use the other to its own advantage. (7:312)

The emphasis in these passages is on the ability to use others for one’s own ends. Although this use of others is occasionally connected with promoting what is “to one’s own advantage,” prudence need not be limited to ends of self-interest. It is involved in any context in which one must learn how to influence others to promote one’s ends, whether those be particular ends, one’s happiness, or moral ends. Kant even specifically says, precisely in the context of pointing out that anthropology helps one manipulate others, that “morals must be bound together with knowledge of man” (25:472, emphasis added). Thus Kant explicitly discusses the uses to which anthropology can be put, and he specifically mentions moral cultivation. Kant says, Anthropology is . . . indispensable and manages great uses. 1. In pedagogy. 2. With respect to the influence that we can have on others. Especially commanders, who with a proper knowledge of man can get total opposites to work [together], which otherwise can be set right only through violence. 3. With respect to the influences on morals and religion, that through this knowledge one can give these duties the power of inclinations. (25: 1437)

Each of these three uses of anthropology can concern moral development. Education, which Kant mentions first, is emphasized throughout his anthropology. Insofar as one learns how to direct others, not only in action but also in thought and choice, one can effectively educate those under one’s care. In several sets of lectures Kant discusses education in

Kant’s “Pragmatic” Anthropology

55

depth and often ends his course on anthropology with applications of his insights to the education of others (see, e.g., 25:722–8). In all of these discussions, one of the most important kinds of education for Kant is moral. In the Anthropology, Kant’s discussion of education includes man’s need to be “moralized,” to be “educated to the good” (7:324, 325). In his lectures, he claims that “respect for the worth of humanity in one’s own person is the final level of education” (25:727). And in his lectures on education, Kant discusses at length “moral culture [moralische Cultur]” and a “practical education [praktischen Erziehung]” that includes “morality” (9:480, 486). The second important use of anthropology relates to a passage we looked at earlier, where Kant insisted that learning to direct others to one’s ends can help one avoid being “too hard or too offensive.” Here Kant examines the most extreme case, insisting that the lessons of anthropology are an important alternative to violence in getting others to do one’s will. Kant does not suggest means by which one can or should force or deceive another into promoting one’s ends. Rather, by learning about others, one can affect them without force, deceit, or even offense, by playing off of their own natural inclinations and, in some cases, weaknesses. This is perfectly consistent with the categorical imperative, whereby one can never use another merely as a means. The commander who uses anthropological prudence not only acts in an acceptable way; she does positive good insofar as she promotes social cooperation by getting opposites to work together. This social cooperation is a crucial part of a moral community, for it establishes the necessary condition of stability under which such a community can flourish. In the Religion and his writings on history, Kant points out the need for political stability as a precondition of moral stability – that is, for a political commonwealth prior to a moral commonwealth (cf. Wood 1999, Louden 2000, and Anderson-Gold 2001). Peaceful social cooperation is thus an important part of promoting moral community. While the first two uses can apply to moral and nonmoral concerns, Kant’s third use for pragmatic anthropology connects it directly with his explanations of moral anthropology in the Metaphysics of Morals. By teaching how to get others to adopt one’s ends, one gains specific insight into how someone can be influenced to adopt obedience to the moral law as an end. The unconditional value of the moral law may be an end that one seeks to influence another to adopt in the case of moral education and may even be an end that one wants to influence oneself to adopt, as we will see more clearly in Chapter 5. Not only does this “pragmatic” nature

56

The Problem

of anthropology not exclude consideration of how to promote morality, but Kant’s remarks on the uses of anthropology suggest that increasing the force of duty is one of the most important roles of anthropological prudence. This brief discussion of the various senses in which Kant uses the term pragmatic shows that his description of the Anthropology as being given from a “pragmatic point of view” does not preclude important moral dimensions to that work. Moreover, Kant’s own claims about his anthropology, how it differs from Platner’s as well as the uses that he mentions to his students, suggest that Kant meant to include moral anthropology in his anthropology from a pragmatic point of view. Of course, neither the possibility that anthropology includes moral anthropology nor Kant’s intention to include moral anthropology shows that there actually is anything moral about Kant’s specifically anthropological work. And if Schleiermacher is right about the conflict between Kant’s Critical philosophy and his anthropology, one might even think that Kant cannot, or should not, succeed in actually carrying out a moral anthropology. Moreover, much of Kant’s anthropology is clearly not moral anthropology in the narrow sense. Almost all of his anthropological observations are relevant to applying the moral law, but only some relate directly to the task of showing what influences strengthen or weaken the force of the moral law on the will. This is not surprising, because there is no reading of pragmatic that identifies it with moral concerns. But Kant’s pragmatic anthropology does include moral anthropology, and the most effective means of illustrating that is to offer examples. To that task we now turn.

2. Three Examples of Kant’s Moral Anthropology In order to show that Kant’s anthropology includes empirical influences that promote a good will, I examine three particular cases in which anthropological insights are morally relevant in the sense of giving subjective conditions that hinder or help people in fulfilling the laws of a metaphysics of morals. Exploring the full richness of these examples is not necessary to establish the presence of moral anthropology in his work, so I focus in each case on showing that the example is moral, in the sense that it reflects a help or a hindrance to moral maturity. When relevant, I will show how the Anthropology enriches accounts in Kant’s more specifically moral works in just the ways that one would expect from moral anthropology. I also briefly show that each example is practical, in the

Kant’s “Pragmatic” Anthropology

57

sense that it is to be put to use rather than merely a matter of theoretical knowledge, and empirical, in the sense discussed in the previous chapter. In the concluding section of this chapter, I briefly point out how the structure of the Anthropology and Kant’s corresponding lectures shows a deep concern on Kant’s part with the moral dimension of his work that comes out in the examples. A. Politeness Kant’s discussion of politeness occurs primarily in a section of the Anthropology called “on the admissible moral perception” (7:151–3) and in Kant’s discussion of “sociability” in the Metaphysics of Morals (6:473–4).14 In Kant’s lectures on anthropology, politeness is included within a more general discussion of the distinction between illusion and deceit (see, e.g., 25:502–5, 1455). Kant’s account of politeness is not the most prominent example of moral anthropology, but it is one of the most clear. Kant insists that because of certain facts about human beings, we are susceptible to influence through politeness, and this influence should be used for the strengthening of moral principles in oneself and others. Politeness is an empirical aid to the development of a good will, and it is explained in detail only in Kant’s anthropology. Thus it highlights Kant’s attention to the influence of empirical helps and hindrances on subjective conditions of morality, as well as his recognition that anthropology is the proper place for such attention. Kant’s account of politeness is clearly moral anthropology, in the sense that it discusses an aid to moral cultivation. One should be polite because politeness fosters virtue in oneself and others. In the Metaphysics of Morals, Kant explains that although polite graces “are, indeed, only . . . a beautiful illusion resembling virtue . . ., yet they promote the feeling for virtue itself” (6:473). The Metaphysics of Morals focuses on the claim that “it is a duty . . . to associate the graces with virtue” because these graces promote virtue itself (6:473). The Metaphysics of Morals does not discuss, however, how politeness promotes virtue. And this is proper, because such a discussion of the empirical aids to virtue is the proper subject of moral anthropology. In the Anthropology, Kant reiterates the moral purpose of politeness, saying that “when men play [virtuous] roles, virtues are gradually established, whose appearance had up until now only been affected” (7:151, cf. 7:152–3). But he also goes further, pointing out how such politeness works to cultivate morality. Politeness cultivates morality by “deceiving the deceiver in ourselves” (7:151), which occurs in two main ways. First, one who is polite

58

The Problem

must often refrain from directly satisfying illegitimate desires, and this self-restraint both reveals and cultivates a capacity for self-control that may be ignored or undeveloped otherwise. As Kant explains, “This betrays a self-mastery and is the beginning of conquering oneself. It is a step towards virtue or at least towards a capacity thereto” (25:930). Second, politeness presents a “beautiful illusion resembling virtue” (6:473), such that “one who loves the illusion of the good eventually is won over to actually loving the good” (25:931).15 Kant’s account of politeness in his Anthropology is fundamentally moral anthropology, because it studies an aid to the cultivation of virtue, and it provides needed anthropological background to flesh out Kant’s reference to sociability in the Metaphysics of Morals. Kant’s account of politeness is also not merely speculative theorizing; it is meant to be put to use. In the Anthropology itself, this is clearest in the language that Kant uses to describe its effects. In pointing out that through politeness “the virtues are gradually established,” Kant adds that this is “a good thing . . . in this world” (7:151). Throughout he speaks of being polite in concrete terms, directing his readers’ attention to what they can actually do in response to his anthropological insights. And in case one is unmotivated by the good effects of politeness, the discussion in the Metaphysics of Morals points out that one has a duty to put into practice the insights of the Anthropology: “It is a duty to oneself as well as to others not to isolate oneself but to use one’s moral perfections in social intercourse . . . and so to associate the graces with virtue” (6:473). Unlike Platner’s physiological anthropology, Kant’s discussion of politeness aims at what one can and should do; it is practical. Finally, Kant’s discussion of politeness in the Anthropology makes clear that politeness is an empirical influence. In the published Anthropology, Kant repeatedly emphasizes the fact that politeness is merely an appearance (Schein). Virtue involves an “appearance of attachment” (7:151) to others, an “appearance of affability” (7:152), and “good and honorable formal behavior [which] is an external appearance” (7:152). It is precisely “the appearance of good in others [which] must have some value for us, because in the long run something serious can come of [it]” (7:153). Politeness fosters virtue precisely as an illusion, an external appearance that can affect behavior even when one knows that it is mere appearance. Hence politeness is a paradigm case of an empirical influence on morality, and it is discussed in detail only in Kant’s anthropology. Thus it highlights not only Kant’s attention to empirical aids but also his recognition that anthropology is the proper science to which such attention belongs.

Kant’s “Pragmatic” Anthropology

59

B. Affects and Passions Politeness is an important empirical help for developing a good will, but Kant’s Anthropology also discusses two important empirical hindrances: passions and affects.16 Passions and affects are not significant for morality alone. In both his published Anthropology and his lectures, Kant points out that passions and affects hinder one’s pursuit of happiness (25:592). But they also hinder virtue, and thus they are properly a part of moral anthropology. The moral importance of affects and passions is clearest in Kant’s lectures in ethics, where Kant insists that “in duty to ourselves, and for the dignity of mankind, the demand upon a man is that we have no passions or affects at all” (27:368). In the Metaphysics of Morals, Kant reiterates this theme, saying that affects involve a “lack of virtue” and passions are “properly evil” (6:408). The key to their hindering effect, Kant points out, is the way in which they inhibit self-mastery. Kant’s Anthropology reiterates this, saying that “to be subject to affects and passions is . . . an illness of mind because both . . . exclude the sovereignty of reason” (7:251). But Kant’s anthropology goes further than simply pointing out that affects and passions are morally dangerous.17 Anthropology explains the differences between passions and affects and explains how each hinders morality. The primary purposes of the descriptions of affects and passions in the Anthropology and related lectures are, first, to make clear in what way each is an “illness of mind,” and, second, to classify various different affects and passions that hinder morality (and the pursuit of happiness) in different ways. In this context, Kant also clarifies the claim in the Metaphysics of Morals that affects are merely a “lack of virtue,” whereas passions are “properly evil,” and he presents at least some suggestions regarding the “distinct . . . method[s] of prevention as well as . . . cure that the physician of souls would have to employ” to treat affects and passions (7:251). With regard to the first purpose, that of clarifying what affects and passions are, Kant explains, The inclination that can hardly, or not at all, be controlled by reason is passion. On the other hand, affect is the feeling of a pleasure or displeasure at a particular moment, which does not give rise to reflection (namely the process of reason whether one should submit to it or reject it). (7:251)

Here Kant makes clear that his condemnation of affects and passions is not a condemnation of all emotions, nor even of all intense emotions. What distinguishes affects and passions from ordinary emotions is the “want of reflection” (7:254). This want of reflection, and “not the

60

The Problem

intensity of a certain feeling,” qualifies a feeling as an affect (7:254). In this passage, Kant also distinguishes affects and passions based on the different faculties of soul that they affect. Passions are uncontrolled inclinations, related to the faculty of desire, while affects are uncontrolled feelings, related to the distinct faculty of pleasure and displeasure. The Anthropology clearly distinguishes these faculties, and this difference plays an important role in understanding how passions can be evil while affects are merely a lack of virtue. In particular, the faculty of desire is the seat of choices (see 7:251 and 6:212–13). One who is under the influence of a passion makes one inclination the overriding inclination in life. This choice of one inclination over all others does not involve the comparison of that inclination with others, nor the consideration of the demands of reason. But it does not consider those other interests because it deliberately rejects them. Thus Kant claims that “passion is an illness that abhors all medication” (7:266). Because it is deliberate, it is “properly evil” (6:408). By contrast, affects, which merely infect one’s faculty of pleasure and displeasure, bypass normal processes of choice. Kant offers as an example one who is paralyzed by seeing a child fall into the water, who is “so shocked that one thereby cannot do anything” (25:592). Because affect bypasses rather than corrupts choice, it “produces a momentary loss of . . . self-control” (7:267) and reflects merely a “lack of virtue” (6:408). As we will see later, however, even this “lack of virtue” is a hindrance to the cultivation of a good character, because it involves a rest in a struggle against evil that should be incessant. These general points about affects and passions help classify and characterize two important hindrances to the good will. Kant also gives details about specific affects, such as anguish, rage, and shock; and specific passions, such as ambition, thirst for revenge, and lust for power. The details of these accounts are unnecessary here. The primary purpose of these discussions is to draw attention to the symptoms of particular cases so that one can be more aware in oneself and others of the presence of these hindrances to virtue. In addition to classifying and characterizing passions and affects, Kant offers at least some advice on how to treat and prevent these hindrances. Thus this discussion is properly practical. With respect to passions, Kant points out that “most of them are incurable because the sick person does not want to be cured” (7:266). Thus the most important measures with respect to passions are preventative, and Kant gives specific suggestions for preventing various particular passions. For instance, he explains that

Kant’s “Pragmatic” Anthropology

61

“in nature there is a propensity for frugality in elders. But reason requires that this inclination be resisted and that one saves only for a purpose” (25:520). By being aware of the passions that are most dangerous for one, one can take preventative measures. In addition, there are at least some things that one can do to lead another person out of passion despite himself. In this context, for example, Kant reiterates the importance of politeness as an illusion that can free one from passions: “The passion of love is much moderated through [politeness], when one . . . conceals the red-hot inclination that otherwise would be difficult to suppress” (25:930). With respect to affects, Kant suggests some preventative measures, such as avoiding “romances and maudlin plays” (5:273), and suggests that refining feeling through culture can help prevent and treat affects (25:622–3). Because affects are only “precipitate” (7:252; 6:407) and “what the affect . . . does not accomplish quickly will not be accomplished at all” (7:252), affects provide much more room for self-treatment. One can and probably will seek out refining influences of culture and avoid situations that prompt affects. Finally, affects and passions are empirical influences. Kant describes affects as “surprise through sensation” (7:252) and offers an empirical description of desire that culminates in his claim that “a subject’s sensuous desire that has become customary . . . is called inclination . . . [and] inclination that hinders the use of reason . . . is passion” (7:265). In both cases, Kant describes empirical processes that characterize passions and affects. And this is proper, not only because they are included in his Anthropology, which is empirical in general, but because each is a disorder of a different faculty of soul, and these faculties are themselves outlined in Kant’s empirical description of the human being. Whereas politeness is an empirical aid to virtue, affects and passions are empirical hindrances to virtue. And Kant describes and characterizes them in his Anthropology, providing a needed anthropological background to the brief treatment in the Metaphysics of Morals. Moreover, he suggests practical measures to take in the light of his account, confirming that his anthropology is not only moral, but also properly practical. C. Character as a Subjective Condition of Moral Agency The most important aid to morality discussed in the Anthropology is character. In the Anthropology Kant discusses several different senses of character, only some of which serve to characterize human beings in general. The

62

The Problem

discussion is divided into sections on the character of the person, of the sexes, of nations, of races, and of the species. Of these, the character of the person is the most morally significant. The character of the person is explicitly identified with “moral character” (7:285) and described as “above all price” (7:292), linking this discussion with specifically normative concerns. In the other sections, the term character often merely refers to characteristics.18 Thus my discussion of character focuses on the character of the person. Character is clearly relevant to morality. The repeated reference to character as “moral” (moralische) in Kant’s introductory section on character and the claim that “character has an inner value and is above all price” (7:292) might even suggest that character is identical to a good will. But Kant makes clear that character and virtue, though related, are not identical. Kant’s only example of character in his discussion is Sulla, “a person of evil [b¨osem] character . . . [who] is, nevertheless, also an object of admiration, because we generally admire strength of soul” (7:293). The strength of soul that is possessed even by the evil Sulla is not identical to a good will, but it is an invaluable aid to a good will. In his lectures on anthropology, Kant defines “character” as “the use of our choice [Willk¨uhr] to act according to rules and principles” (25:630, cf. 25:1384). In the Anthropology, he expands on this: Simply to have a character relates to that property of the will by which the subject has tied himself to certain practical principles. . . . Although these principles may sometimes indeed be false or defective, nevertheless the formal element of will as such, which is determined to act according to firm principles (not shifting hither and yon like a swarm of gnats), has something precious and admirable to it, which is also something rare. (7:292)

When one remains firmly committed to defective principles, as in the case of Sulla, one is not morally good. But in order to be morally good, one must remain firmly committed to good principles. Thus having a character, a commitment to principled action, is a necessary but not a sufficient condition for complete virtue. As Kant explains, “before a good or evil character is built for a human being, a character altogether must be built, with which he first has a character in general, i.e., he first must get into the habit of acting from principles” (25:630–1). Hence Kant says later that “such people [without character] are like soft wax19 ; every instant they take up another rule. Towards evil they are quite steerable but not to good, for to become good already requires principles” (25:631). People without character are steerable to evil but not

Kant’s “Pragmatic” Anthropology

63

good because good action is necessarily action according to principle. To be good, for Kant, just is to act according to maxims that conform to the categorical imperative because they conform to the categorical imperative. But this implies an overall commitment to a principle (the categorical imperative) as a guide to action. Thus, for Kant, “character rises to the level of morality insofar as it is determined as either good or evil” (25:652), such that “good character would be the good will” (25:648). The close connection between character and morality might lead one to forget that the study of character is part of an empirical anthropology. In that context, it is important to note that Kant’s account of character is not an account of what constitutes a good or evil will. That subject is proper to moral philosophy as such. Character is a subjective condition of a will choosing to be either good or evil. Hence Kant’s discussion focuses exclusively on character as such, not on how character becomes good or evil. Kant’s account of the origin and development of character further emphasizes its empirical basis. Although Kant claims that “character comes not from nature, but rather must be acquired” (25:1172), he adds that the “seed” of character comes from “nature” (25:1172). This is fleshed out in terms of an innate and natural propensity for character (25:823, 1176). Kant also suggests that some human beings are better prepared for character than others by virtue of another natural characteristic – temperament. Kant says, “not every temperament is inclined to adopt a character, e.g. the melancholy one adopts a character first, the sanguine one not so easily” (25:1388), and of one with a phlegmatic temperament, Kant claims, “without being brilliant, he will still proceed from principles and not from instinct” (7:290). Kant’s discussion of temperaments is “psychological” but still empirical, and temperaments can even “be influenced . . . by the physical condition of a person” (7:286). Thus insofar as they play a role in the formation of character, they show the extent to which character is empirical. Kant discusses further aids to the cultivation of character that are not natural endowments but are nonetheless empirical influences. Among these, education is the most important.20 “The acquisition of good character with people happens through education” (25:1172). Throughout his writings Kant’s accounts of character are accompanied by specific pedagogical recommendations. He suggests that because “imitation . . . greatly hinders character, in education one must never refer one’s children to the neighbor’s children, . . . but rather build their character directly,

The Problem

64

[using] principles of good and bad to inspire righteousness and nobility” (25:635, cf. 7:325; 5:154; 25:599, 722ff., 1386). Politeness also plays an important role in the cultivation of character by combating passions and promoting self-control. In his writings on history and politics, Kant gives further explanations for how natural inclinations can give rise to various social institutions – including stable and just political regimes, peace, and even progress in the arts and sciences – that may have beneficial effects on character.21 Thus “experience and history” provide reasons that “we should not despair about our species’ progress towards the better” (7:329). The increasing presence of stable political structures and advancing culture, like the presence of stable norms of polite society, can help the cultivation of constancy in principled action. These external influences, like the more direct influence of education, support the establishment of character. None of these external supports for character absolve people from responsibility for developing character themselves, however. Thus in addition to the practical suggestions for promoting character in others, Kant’s discussion of character also provides a practical program for developing character in oneself. He suggests five basic principles to aid the cultivation of character. a) b) c) d) e)

Not to speak an untruth intentionally . . . not to dissemble . . . not to break one’s legitimate promise . . . not to join the company of evil-minded people . . . not to pay attention to slander. . . . (7:294, cf. 25:1387–8, 1392)

These are all practical principles that support and constitute the development of character as such. The pursuit of these methods for developing character depends on one’s already having at least some level of character. Unless one can act on the basis of principles, one will be unable even to follow the principles for developing character. But keeping these principles even sporadically can have some beneficial effect. The more one avoids duplicity, bad company, and slander, the easier it will be for one to stick to principles in the future. Insofar as one has some minimal level of constancy, these principles can reinforce one’s character. They are important aids to self-improvement, even if they are not sufficient. And their inclusion here is enough to show that character as an empirical aid to virtue is part of a practical anthropology, rather than mere speculative theorizing.

Kant’s “Pragmatic” Anthropology

65

3. Conclusion: The Systematic Place of Moral Concerns in Kant’s Anthropology These three examples of the moral anthropology within Kant’s “pragmatic” anthropology show that his anthropology is not concerned merely with techniques for achieving happiness or worldly success. And this, of course, is what one would expect, given that Kant considered his course in anthropology a pre- or at least co-requisite of his course in ethics.22 But although there are several areas in which Kant’s anthropology is morally relevant, one still might think that these are only occasional and peripheral.23 And Kant’s anthropology certainly contains much that is not particularly moral, including a long section on cognition that has little to do with subjective conditions of moral choice.24 Nonetheless, there is a clear structure to Kant’s anthropology that highlights the importance of moral anthropology. Throughout his lectures and in his published anthropology, Kant begins his account with a description of the cognitive faculty, then discusses pleasure and displeasure, and ends the first part of his work with an account of the faculty of desire, the faculty most directly related to morality. Consistently, Kant ends this first part of his anthropology with discussions of passions and affects.25 This discussion is especially prominent in the published Anthropology, in which Kant’s account of the faculty of desire is taken up almost entirely with the passions and affects. Not only does this prominence reflect an extremely narrow range for a discussion that is supposed to be about the entire faculty of desire, but it does not fit Kant’s systematic division of faculties into cognition, feeling, and desire, because affects are matters of feeling. Kant seems more concerned with placing a discussion of passions and affects prior to his account of character than with holding rigorously to an account divided by faculties of the soul. In all of his lectures in anthropology as well, the discussion of passions and affects precedes an account of character that makes up the second major part of his anthropology. The importance of this structure should not be missed. Kant’s accounts of passions and affects and character are the two aspects of his anthropology that are most directly related to moral anthropology, and they mark the climax of nearly all his lectures as well as his published Anthropology. Moreover, they are closely related, such that Kant uses his account of passions and affects to introduce and provide a transition into his account of character. In his lectures, this is especially dramatic. Kant ends the first part of his anthropology in his 1775–6 lectures with

66

The Problem

“general remarks on the passions and affects” (25:613). He concludes this section: All these [passions and affects] are set against the steadfastness, wherein one does not deviate from his principle and persists firmly in his decision. It is important to have firmness in one’s decisions and not to depart from one’s principle, and better to endure disadvantage than to let go one’s principle. When then a human being knows something for sure, he carries it out. But whoever is not steadfast therein often seizes a principle that he is certain that nothing will come of, because he knows that already he has frequently broken principles. Thereby the human being is a cream puff in his own eyes; he believes himself capable of nothing more, from which springs hopelessness. That is a comfortless condition, when one always postpones hope. So it is with late conversions. So it is with other things, that one wants to break the habit of, e.g., oversleeping, for it always says, just one more time but then no more, and so one philosophizes oneself further away from carrying out [one’s resolution]. In such a condition one never has hope to become better; this is an important point in morals. One must therefore seek to hold himself ever so promptly to this word as to others. From that springs a firm confidence in ourselves. (25:624)

This passage is a perfect conclusion to a long discussion of passions and affects, in that it highlights the importance of holding firm to principles and the hopelessness of a life lacking in self-mastery, in which one cannot follow through on one’s own resolution. But it is also a perfect introduction to the second part of the anthropology, dealing with character. After a long discussion of human capacities, Kant always discusses the danger posed by affects and passions, a danger whereby all of one’s capacities can prove fruitless through one’s loss of self-mastery. He then turns to a discussion of character that stresses the importance of constancy. Kant’s anthropology, which begins with discussions of cognitive capabilities, ends with the importance of developing character in oneself and in the species as a whole. And although this character is not necessarily morally good, it is a form of self-mastery that is a subjective condition of morality. It is especially significant that in these sections on character (and only in these), Kant describes his account of the character of the person in language that makes clear its deep connection to pragmatic anthropology as a whole. Recall that in his preface Kant distinguished pragmatic from mere physiological anthropology by saying that whereas physiological anthropology investigates “what Nature makes of man,” pragmatic anthropology deals with “what man makes . . . of himself” (7:119). It is only in his discussion of character that Kant points out explicitly that that

Kant’s “Pragmatic” Anthropology

67

purpose of pragmatic anthropology is fully realized: “Here it does not depend on what Nature makes of man, but what man makes of himself” (7:292). Kant’s account of character finally ends in a discussion of the human species in which he explains how the race can “advance constantly from the evil to the good” (7:333). As he summarizes it, The sum total of findings generated by pragmatic anthropology as to the classification of man and the characterization of his development is as follows: man is destined by his reason to live in a society of other people and in this society he has to cultivate himself, civilize himself, and moralize himself by the arts and sciences. . . . (7:324, emphasis added)

Pragmatic anthropology deals with cultivating all human capacities, to be sure, but it reaches its conclusion only in the civilization, and eventually moralization, of humanity itself. This means that although it is empirical, anthropology is deeply concerned with helps and hindrances to moral development. Politeness and character mark two important helps, and Kant’s discussion of character points to a wide variety of other helps, such as education, republican government, world peace, and the development of the arts and sciences. Passions and affects are two significant hindrances, for which Kant proposes at least some treatments. And all of these influences are empirical influences, the proper objects of an empirical science of anthropology. Schleiermacher thus seems to have accurately characterized Kant’s anthropology, and he has drawn attention to a real tension between this anthropology and Kant’s conception of transcendental freedom. In the next chapter, I draw from important work in contemporary neokantian ethics to articulate several ways in which Kant might seem to escape Schleiermacher’s dilemma. While all are promising for explaining certain sorts of empirical helps and hindrances, all fall short in one of two ways, either by failing to sufficiently account for an empirical moral anthropology in the robust sense outlined in this chapter, or by sacrificing Kant’s theory of freedom. In Chapter 5, I offer my own resolution of the tension, drawn from Kant’s Religion and later writings in moral philosophy.

4 Moral Anthropology in Contemporary Neokantian Ethics

Neokantian moral theorists have recently begun to pay more attention to Kant’s moral anthropology. Robert Louden devotes the longest chapter of Kant’s Impure Ethics (Louden 2000, cf. Louden 1986) to the treatment of the role of anthropology in the “second part” of Kant’s moral theory. And G. Felicitas Munzel devotes considerable portions of her Kant’s Conception of Moral Character (Munzel 1999) to pointing out the insights that can be gleaned from the anthropology for Kant’s moral theory. Aside from these direct treatments of anthropology, many of the most influential contemporary appropriations of Kant’s moral theory have shown increasing attention to various little-known gems of anthropological reflection in Kant. Nancy Sherman and others have pointed out the important role of the emotions in Kant’s moral philosophy, drawing in large part from the Anthropology and the most anthropological sections of Kant’s other writings. Allen Wood and Sharon Anderson-Gold, in their attention to the social dimension of ethical life (Wood 1999; Anderson-Gold 2001), draw from anthropological considerations in Kant’s Religion and historical essays. Paul Guyer’s rich accounts of the experience of freedom also draw extensively on anthropological considerations (Guyer 1993). And this just scratches the surface of the explosion of attention to anthropology in contemporary Kantian moral philosophy. In part, this attention is a response to challenges from Kant’s critics, especially those sympathetic to contemporary virtue ethics, who often accuse Kantian moral theory of paying insufficient attention to character, virtue, and the richness of human life. In part, it is the result of the recent publication in Germany of Kant’s Lectures on Anthropology (vol. 25 of the Academy Edition of Kant’s works). And in part it is simply a 68

Moral Anthropology in Neokantian Ethics

69

long-overdue recognition of the importance that Kant himself gave to anthropology. These developments are welcome. They flesh out a moral theory that though provocative and appealing seemed to have lost touch with human life. But this enrichment of moral theory through anthropology has led some Kantians to neglect the central importance of freedom in Kant’s moral theory. Even those who continue to attend to Kant’s theory of freedom while seeking to incorporate anthropological insights fail to explain adequately how one can coherently integrate moral anthropology into Kant’s theory of freedom. As we have seen in the past three chapters, Kant’s theory of freedom does not fit well with his empirical and morally relevant anthropology. As neokantians incorporate anthropology into moral theory, they often fail to give moral anthropology the significance it deserves, or they sacrifice Kant’s commitment to freedom. In this chapter, I examine four roles for moral anthropology that seem to offer promising possibilities for integrating empirical influences with Kant’s theory of freedom. I point out the place of these possibilities in the work of some of the most innovative recent work in Kantian ethics today, especially that of Barbara Herman, Nancy Sherman, Robert Louden, and G. Felicitas Munzel.1 I argue that none of these possibilities offers an adequate account of how a genuine moral anthropology can be integrated with Kantian freedom. In section 1, I consider the possibility that morally beneficial empirical influences play an epistemic role in helping human beings determine the content, relevance, or proper application of the moral law in concrete situations. In section 2, I consider the possibility that these influences play a necessary or useful instrumental role in carrying out the demands of morality. That is, they provide resources that one can use to do more good deeds. In section 3, I consider the possibility that helpful “influences” are actually constitutive of the good will, such that to have character, or to be polite, or to be part of a good community, is part of what it is to be good. In sections 4 and 5, I consider the suggestion that empirical influences are a necessary or helpful “propaedeutic” to morality. The first possibility plays an important role in the work of both Nancy Sherman and Barbara Herman (Herman 1993 and 1998; Sherman 1990, 1995, 1997a, 1997b, and 1998). Nancy Sherman is also the primary source for the second and third options. The final role is discussed by Robert Louden, who raises it in the context of an explicit treatment of the relationship between anthropological influences and Kant’s theory of freedom (Louden 2000). This

70

The Problem

final option – of influences as propaedeutics – is also central to recent accounts offered by G. Felicitas Munzel (Munzel 1999) and Allen Wood (Wood 1999). Wood deals extensively even if obliquely with the issue of the relationship between history, political and social structures, and morality. He suggests that empirical influences can play roles as necessary preconditions of moral choice, and he seems to think that at least some empirical influences are necessary preparations for choosing rightly. But Wood does not offer as much detail as Louden, because his book is focused more generally on Kant’s ethical theory. Moreover, on the issue of helps and hindrances there does not seem to be any significant difference between their views.2 Thus although I focus on Louden, most of what I say here could apply to Wood as well. In section 5, I take up Munzel’s propaedeutic account. At times, Munzel suggests a conception of “propaedeutic” that leaves room for “influences” that are not necessarily “prior” to choice, but she does not sufficiently develop this conception.

1. Empirical Influences as Epistemic Aids The first role that one might ascribe to empirical influences is an epistemic one. According to this account, certain influences, such as polite society, a well-functioning political system, a good moral education, or well-cultivated emotions, aid in discerning morally relevant features of a situation. One who lives in a society with rigidly enforced gender or class roles, for instance, might not recognize the harm done to women or to members of a lower class as morally relevant in deliberation. These empirical influences would thus hinder one from being morally good. On the other hand, emotional sensitivity to the pain of others or a moral education that teaches one to recognize morally relevant features of a situation can be invaluable aids to choosing rightly. By revealing or concealing morally relevant features of one’s choices, empirical influences can – or so it seems – affect whether one turns out to be good or evil. But because these influences affect merely one’s cognitive faculties, one still freely chooses whether to act on good or evil maxims. Assigning this role to moral anthropology thus seems to reconcile empirical influence with transcendental freedom. Nancy Sherman assigns this sort of role to the emotions, the particular sphere of moral anthropology on which she focuses.3 Sherman not only suggests that emotions can be part of a structure that supports the good will, but she lays out, clearly and in several essays, “what this

Moral Anthropology in Neokantian Ethics

71

supportive structure could be like.”4 The first “way in which emotions can serve supportive roles” is “as modes of attention that help us track what is morally salient in our circumstances, and thus locate possible moments for morally permissible and required actions” (1997a:145; 1997b:273). In this respect emotions “serve instrumental (epistemic) roles indispensable for acting from moral interest” (1997a:158; 1997b:278). In this explanation of the role of emotions, Sherman draws heavily from Barbara Herman’s work on moral judgment. Herman points out the importance of “rules of moral salience” for moral evaluation and deliberation. These rules of moral salience “structure an agent’s perception of the world so that what he perceives is a world with moral features” (Herman 1993:77). Only in this context can an agent properly test maxims against the moral law, because such a test requires that one’s maxims pick out those features of a situation that are morally relevant. Like Sherman, Herman closely connects these rules of moral salience with empirical aids. She says that these rules are “acquired as elements in a moral education” (Herman 1993:77) and that “they are acquired in childhood as part of socialization” (78). External influences thus determine, at least in part, the structure of our moral sensitivity. And this structure affects not only evaluation of situations but also practical deliberation. As Herman explains, “issues of excuse and justification [that depend on rules of moral salience] do not enter a moral thought only after an action has been done; they are part of the way a normal moral agent explains the permissibility of his actions to himself” (Herman 1993:78). Like Sherman, Herman suggests the possibility that “affective capacities of response” may be necessary to properly perceive moral features of a situation, and if these are necessary, then “we will have found a Kantian argument for the development of the affective capacities” (Herman 1993:82). Because Herman does not show that affects are necessary, Sherman’s arguments that emotions actually are necessary provide a crucial supplement to Herman’s account. On the other hand, Herman makes clear that the epistemic instrumental role that Sherman assigns to the emotions is a role that can and probably must be filled by a wide variety of abilities that can be encouraged by empirical influences. Herman is also much clearer than Sherman about precisely what sort of help is provided by emotions or other influences that contribute to rules of moral salience. And it is clear in Herman’s discussion that these sorts of aids are not aids to actually having a good will. On Herman’s account, there are two critical ways in which rules of moral salience relate

72

The Problem

to the good will. First, an agent must have some rules of moral salience in order to be capable of choosing to be good or evil. Empirical helps are thus necessary for being a moral agent at all. As Herman explains, “the Kantian moral agent must have a characteristic way of seeing if he is to judge [morally] at all. To be a moral agent one must be trained to perceive situations in terms of their morally significant features (as described by the RMS [Rules of Moral Salience])” (Herman 1993:83). For Kant, the moral status of an action depends on the maxim that underlies that action. Beings that “act” merely from impulse, without any maxims at all, cannot be morally accountable at all. Likewise, those who act on maxims that incorporate no moral features at all cannot be morally accountable. One can imagine, for example, a “person” with a certain sort of cognitive failing such that he does not recognize basic rules of moral salience. Such a person might “kill” another but recognize in the action nothing more than the lifting of an arm and the moving of a finger. The person does not even perceive – or does not see as relevant – that the hand at the end of the arm is holding a gun and that the finger pulls the trigger and that the gun is pointed at another person. A person for whom every action is of this sort might act, say, for the pleasure of experiencing new bodily positions, without any awareness of the mayhem caused by these new positions. For such a person, moral issues simply do not arise. Awareness is so constrained that it never captures anything of any moral salience. Because of this peculiar cognitive failure, such a person would be incapable of choosing either good or evil. That person would not be a moral agent at all. A second role of rules of moral salience is to allow a moral agent – and Herman discusses only the case of morally good agents – to apply moral principles more easily to particular situations. In this case, better rules of moral salience allow one who is committed to acting only on good maxims to choose maxims that are not only consistent with the moral law but that also “more accurately reflect moral facts that can be obscured from plain sight” (Herman 1993:88). A person without good rules of moral salience would be someone who “is attentive to duty, but not very perceptive, and so [who] does not see that his circumstances fall under a principle of duty. (He fails to recognize some situation as one calling for help.)” (Herman 1993:81). In this case, as Herman is careful to point out, the moral status of the agent is not at stake. More or better rules of moral salience do not affect whether or not the agent has a good will. Because on Kant’s account “there is no way to judge actions apart from the way they are willed,” “morally defective RMS [rules of moral salience] may not yield

Moral Anthropology in Neokantian Ethics

73

morally defective actions” (Herman 1993:91). “It can be permissible for agents with mistaken RMS to act in ways that would be judged impermissible if their RMS were correct” (Herman 1993:89). For example, one may be morally good, act in accordance with the moral law, and refer to women as “girls,” or to African-Americans as “niggers,” because one fails to recognize that referring to them in these ways is demeaning. Failing to recognize this, one may deeply harm women or African-Americans while acting on a maxim that is consistent with or even based on respect for the dignity and happiness of women and African-Americans. One just fails to consider important dimensions of their situation.5 Because Kantian moral evaluation of agents is based on the status of their maxims, better rules of moral salience do not affect the status of a person’s will. Instead, as Herman points out, “the scope of beneficent actions (how much good is done) will be greater for persons who can more readily perceive the distress of others” (Herman 1993:81). Insofar as one has better epistemic access to relevant features of a situation, one is able to do more good, given a good will. But epistemic access is not relevant to the status of the will itself.6 On Herman’s account, then, empirical influences7 play an important epistemic role for the good will, but this in no case affects whether a given moral agent has a good will or not. Empirical influences may be necessary for someone to become a moral agent at all, or they can enable someone who already chooses rightly to perform more and better actions. But they cannot help a morally responsible person to choose rightly. They cannot help or hinder having a good will. Nancy Sherman draws from Herman’s account but adds one other way in which empirical influences – in particular, emotions – can serve as epistemic helps toward moral improvement. She suggests that they can provide better epistemic access not only to morally salient features of a situation but also to self-knowledge: The notion that emotions are modes of attending to moral salience may also have application to the duty of self-knowledge. . . . Emotions can turn inward to alert us of our own inner states, to help us to read, as he says, our own heart. As such, they may be thought of as mediums by which we can become more aware of our thoughts and motives. (Sherman 1997a:146; this passage is not in 1997b)

This additional epistemic role of emotions is similar to Kant’s accounts of the importance of politeness,8 and Kant does argue that this kind of self-knowledge aids moral development. But it is important to notice

74

The Problem

that Sherman does not give us any account of how this aid would work, nor even how it could be possible, given Kant’s account of freedom. In particular, if one comes to think – based on insights into oneself provided by the emotions – that one is in fact committed to the moral law, there is no obvious reason why this awareness should affect one’s commitment. One might even think that such knowledge would lead to an unhealthy complacency in one’s moral status.9 Moreover, if one is committed to some evil maxim, there is no obvious reason why becoming aware of that commitment will lead to any kind of moral improvement. One might simply regard it with indifference as a moral fact about oneself, or one might be led to a moral despair that would undermine one’s commitment to pursue the good.10 Nothing is this section should be taken to deny the important epistemic role of empirical influences in general or of emotions in particular. Sherman and Herman enrich contemporary Kantian ethics by pointing out the crucial roles of empirical influences as preconditions of moral agency and as aids to the performance of beneficent actions. But neither of these roles explains the distinctive role of empirical influences in moral anthropology. Neither explains how a moral agent can become morally better through empirical influence.

2. Empirical Influences as Instruments Rather than ascribing a purely epistemic role to moral influences, one might think that they are necessary or useful means for acting well in the world. A stable political order, for instance, provides a framework within which one can make and honor promises, help others achieve their morally worthy ends, and practice the virtues of self-cultivation. A good education and certain basic material goods may be necessary or at least very useful to make it possible for one to act in a variety of morally praiseworthy ways. Certain emotions might be needed to minister to a friend in need, to honor one’s duties to one’s family, or to effectively serve one’s nation. On the other hand, empirical influences that incapacitate or weaken one physically, mentally, and emotionally can hinder one’s performance of one’s duties. One might argue that in none of these cases do empirical influences help or hinder in a way that violates the priority of transcendental freedom. They merely extend or limit one’s abilities in ways that affect the extent to which one can perform the deeds that are morally required. Thus this instrumental

Moral Anthropology in Neokantian Ethics

75

role for empirical influences seems to integrate moral anthropology and freedom. In her account of the ways in which emotions support virtue, Sherman details the way in which emotions in particular can affect moral life as instruments: Not only do we read circumstances through the emotions, but we respond through them. Manifest affect is a vehicle or mode for conveying moral interest. Through emotions, we communicate or signal moral interest to others in ways tailored to particular circumstances and needs. (1997a:147; 1997b: 273)11

Emotions are often means by which one can carry out duties to others. When friends suffer, one’s efforts to provide for their happiness and one’s obligations to them as friends require that one do more than merely say compassionate things to them, spend time with them, and so on. One ought to have and express certain emotions such as sympathy and compassion. Sherman is certainly correct to point out the important role that emotions can play in carrying out one’s duties, but there is an important danger in taking this role too far. Given that emotions are not totally within one’s control, Kant cannot allow them to have a necessary role in the fulfillment of duties. Sherman rightly points out that Kant never says that emotions are totally beyond one’s control.12 But for emotions to be necessary for carrying out one’s duty, it is not enough that they be partially within one’s control. They would have to be totally within one’s control. Otherwise, one would be required to do something – feel sympathy, for example – that in certain cases would be impossible for one. And this would violate Kant’s fundamental principle that ought implies can. Sherman at times seems to recognize that emotions themselves cannot be necessary helps to carrying out one’s duty. As she explains, We cannot ensure success in actually expressing the emotions that we think are appropriate. For the Kantian, the most we can demand is good willing, that is, sincere effort in certain relevant endeavors, constrained by the principles practical agency itself generates. If an agent tries to be sympathetic because she realizes it is important for morality, but simply cannot make it happen – if she makes the attempt part of her sincere maxim of what is relevant to helping another, but can’t bring it off – is she subject to moral self-criticism? Presumably, a Kantian would argue “no,” especially if the agent is one who, on the whole, takes seriously a project of developing sympathetic feelings, exhorts herself to do better when she fails, cultivates the right outward gestures, and monitors her behavior to see how she is doing. A moral defect in character, on this Kantian view, shows up not

76

The Problem

so much in the finished product as in whether or not there is adequate striving of a will toward the right ends. (1997a:184)13

Emotions may serve important roles in more effectively pursuing virtuous ends, and as a result the cultivation of emotions may be a requirement to satisfy one’s duties of self-perfection and beneficence, but the emotions themselves cannot be required. More generally, anything that is useful for carrying out one’s duty can be the necessary object of an imperfect duty. If some empirical influences help one more effectively promote the happiness of others, for instance, then one has a derivative obligation to seek those influences. Unfortunately, whether one emphasizes the importance of emotions themselves or merely the cultivation of emotions, this role does not capture the distinctive role of emotions in moral anthropology. Sherman certainly highlights an important role that emotions play in moral life, but this role is not relevant to whether or not one actually has a good will. Rather, emotions, like much in the empirical world, can serve a will once that will is already good. For carrying out its purposes, emotions are important, and one therefore has a duty to cultivate them, just as physical strength can be important and one therefore has a duty to develop physical strength. But one also has a duty to develop and control one’s emotions in a variety of ways because such control can actually help one acquire and sustain a will that can make proper use of them. Sherman fails to show how emotions – or other empirical influences – can serve this role.14 Empirical influences that serve as instruments to acting well are important, but they do not have the special moral importance of the helps and hindrances that are central to moral anthropology.

3. Empirical Influences as Constitutive One might argue, however, that some empirical influences are important not merely as means by which a will does what is good but also as constitutive parts of a good will. Some “instruments” might be so enmeshed with acting well, for instance, that to act well necessarily includes acting with those instruments. Being a part of polite society may play such an important role in acting well that part of what it is to have a good will is to act politely and be part of a polite society. A just government, or at least some minimal level of justice, may be so intrinsic to respecting the rights of others that part of what it is to have a good will is to be a citizen under such a government. Perhaps most plausibly, one might think that certain

Moral Anthropology in Neokantian Ethics

77

emotions are just so important for virtue that part of what it is to have a good will is to have those emotions. Nancy Sherman suggests this last view as an important way in which emotions fit into one’s moral status. Sherman explains that “in the light of this last claim [that emotions are “ways in which we act and convey moral interest”] especially, I would argue that emotions are constitutive of virtue in human embodiment” (1997a:158; 1997b:278). The notion that emotions can be constitutive of a good will draws on Kant’s account of imperfect duties in the Doctrine of Virtue as well as his description of virtue as a “hierarchically structured composite” in the Religion. In this section, I examine Sherman’s suggestions about the way in which empirical aids can be constitutive of the good will in the context of both the Doctrine of Virtue and the Religion. In the Doctrine of Virtue, Kant insists that one has at least two obligatory ends, one’s own perfection and the happiness of others (see 6:385–6). One’s own perfection includes anything that helps the good will either in an epistemic sense as described in section 1 of this chapter or in an instrumental sense as described in section 2. Insofar as emotions serve either of these roles, they are necessary ends of a good will, and having them may seem to be constitutive of a good will. Moreover, if emotions are required for the moral awareness needed to be capable of choosing either good or evil, then they are constitutive of a good will in that they are constitutive of any morally responsible will at all. As in the preceding two sections, I agree with Sherman that emotions as well as other empirical aids are constitutive of a good will at least in the sense that a will that fails to cultivate them fails to fulfill its imperfect duties. Moreover, I grant that there may be some minimum level of emotional sensitivity or other empirical assistance that is necessary if one is to be morally responsible at all. However, Sherman sometimes takes this insight too far. She suggests not only that some level of moral awareness is necessary to be morally responsible and that a deeper level is a required end for a good agent but also that some deep level of moral awareness is a necessary attribute of a good agent. But all that a Kantian – or anyone who takes seriously that ought implies can – should require is that one must pursue, to an extent that may be different for different people, emotional sensitivity. Moreover, Kant specifically distinguishes in the Doctrine of Virtue between instrumental aids for the good will – which would include not only certain emotions, moral attentiveness, and so on, but also such things

78

The Problem

as physical strength and wealth – and specifically moral helps and hindrances. A person’s duty to perfect himself includes “a duty to cultivate his natural powers (powers of spirit, mind, and body), as means to all sorts of possible ends” (6:444). Some of the natural powers that one cultivates, such as sensitivity to the pain of others, may serve primarily moral ends, but none are uniquely moral.15 In contrast to these “natural powers,” however, Kant points out that the duty to perfect oneself also requires that one “increase his moral perfection” (6:446). Moral anthropology – as opposed to anthropology that is primarily technical or pragmatic – points out empirical helps and hindrances to this sort of perfection. In this context, Sherman, Guyer (Guyer 1993, Chapters 9 and 10), and others are correct to point out that the pursuit of various empirical aids is constitutive of having a good will. And it is true that these empirical aids are “helps” to having a good will in the sense that they are objects of an imperfect duty to cultivate one’s natural powers. But these accounts fail to capture the distinctive sense in which specifically moral helps and hindrances affect the status of one’s will. It is constitutive of having a good will that one interact with polite society or develop a sense of good taste because these empirical influences can aid in the cultivation of a good will. But to claim that they aid in this cultivation only in the sense that their pursuit is constitutive of having a good will is to argue in a circle. We can say that the pursuit of emotional sensitivity is constitutive of having a good will16 because it is an important means for promoting the happiness of others, which is itself a necessary end of the good will. And a good will necessarily wills the necessary or helpful means to its necessary ends. But some empirical influences serve not only as means to general ends that a good will ought to have but as means by which one comes to have a good will itself. These influences are constitutive of the good will, and in that sense their pursuit is required, but this cannot be the end of the story. They are means to having a good will prior to being constitutive of such a will, and are constitutive only because they are a means. Saying that they are constitutive does not solve the problem of moral anthropology. It only pushes it to a new level.

4. Empirical Helps as Necessary Propaedeutics One final possible role for empirical influences is as necessary propaedeutics. This view is in some ways an expansion of the epistemic option discussed in section 1. There we saw that certain influences could be considered necessary means for formulating proper maxims. In that way,

Moral Anthropology in Neokantian Ethics

79

they seemed to provide important preparatory steps to being morally good. But the fact that these were merely epistemic aids limited the significance of their role. They could serve as basic requirements for someone to be morally responsible at all, such that to lack such aids would preclude choosing good or evil. Or these epistemic aids could improve the effectiveness of people already committed to choosing well. Thinking of empirical aids as necessary propaedeutics broadens their scope by allowing that even those who properly formulate maxims may need some help in resisting temptations to act wrongly. This need would be due to their human limitations and thus would not – or so some say – interfere with freedom. On this account, empirical influences prepare free and responsible agents to choose well. They are not sufficient conditions for morally good choice, so there is still an important role for freedom, but they are necessary propaedeutics. Although various versions of the “propaedeutic” role have been offered in recent years, this section focuses on Robert Louden’s use of the account.17 Louden repeatedly emphasizes the role of empirical helps as “necessary preparations” for morality. To offer just a few examples, he claims that “human morality does presuppose a sufficiently developed, interconnected web of cultural institutions as a necessary condition for its own presence” (Louden 2000:21), that “moralization . . . necessarily presupposes the preparatory steps of culture and civilization” (41), that “art serves as an important and necessary preparation for morality in the lives of human beings” (109) and “the aesthetic is indispensible in cultivating human moral virtue” (116), that “history . . . ‘sets the stage’ for morality; it presents us with ‘quasi-moral “veneers” that are not themselves wholly moral’ but are necessary preconditions for morality” (149), that “the state plays a necessary preparatory role in the development of morality” (169), and that “political and legal progress are both necessary presuppositions for this deeper moral progress, as are cultural and scientific advances, growth in foreign trade; indeed as we have seen, even war itself” (160).18 In some of these discussions, it is not entirely clear to what the term morality refers. It could be taken to mean simply that certain preconditions are necessary before one can be capable of choosing either good or evil. This claim would be similar to the important preparatory role that Herman at times sees for rules of moral salience. In these cases, Louden would be saying that some people cannot be held morally responsible at all because they lack those aids that enable moral choice. The term morality could also refer merely to the moral appearances. Certain empirical

80

The Problem

aids might be necessary for good wills to manifest themselves in certain ways, or for more actions that conform to morality to be performed in spite of people’s evil wills. “Moral” progress even might be taken to refer to something that is merely a matter of the human species as a whole and not also related to the moral status of individuals. Louden clearly has in mind, however, a notion of morality and moral progress that is individual, even if also present at the level of the species, and that already assumes that agents who need these preparations are morally responsible. Empirical influences enable already morally responsible wills to choose in conformity with the moral law. He makes clear that morality at least includes individual moral choice when he says, for example, that There is no guarantee that people who have been exposed to these preparatory steps will be morally good, but human beings who lack all contact with them cannot possibly be morally good. (Louden 2000:53)

Moralization is a matter of individuals being morally good. Moreover, this moral goodness is not merely a matter of appearance. Louden makes this clear in an extended discussion of how Kant can justifiably talk about true moral progress throughout history. Louden points out that moral progress involves a “transformation of the way of thinking” that takes place at the level of an unobservable “radical innerness” (Louden 2000:151). He goes on to suggest, We must try the best we can to decipher the inner from the outer. And since certain manifestations of the latter (e.g., law, art, culture, religious community) are themselves necessary preparatory steps for the former, it is correct to say that, from a human perspective, clear signs of outer progress also afford us clues regarding inner progress. (Louden 2000:152)

We will come back to this passage again when I discuss a central problem with Louden’s account, but for now it is important to note that various empirical influences such as law, art, and culture are necessary preparations for an inner, moral conversion. Louden ends his book with an assessment of Kant’s ethics after its “impure” part has been reintegrated. He ends this assessment with the question of the “internal coherence” of Kant’s ethics and specifically questions whether “Kant’s commitment to transcendental freedom” is consistent with his impure ethics (Louden 2000:180). Louden does not offer a detailed response there. Instead he says, “I have demonstrated the coherence of Kant’s concept of empirical ethics in this study”

Moral Anthropology in Neokantian Ethics

81

(Louden 2000:180). Has he? Does Louden’s account of empirical aids as necessary preparations for having a good will really cohere with Kant’s doctrine of transcendental freedom? After raising a basic objection to Louden’s view, I consider the two most explicit responses that Louden seems to offer in his book and show that neither of these resolves the basic incoherence between his account of necessary preparations and Kant’s account of transcendental freedom. I then show that the problem of incoherence arises not merely when empirical aids are necessary preparations, but on any account according to which they are ultimately prior in a practical sense to the good will.19 To put it simply, Kant cannot ascribe moral responsibility to any agent that lacks what is necessary to act in accord with the moral law. As a result, if Louden is correct that empirical aids – such as rational and cosmopolitan churches, a stable political order, and aesthetic sensitivity – are necessary for one to be virtuous, then unless these aids are present, one is not morally responsible. Freedom means – if it means anything for Kant – that one can choose to obey the moral law regardless of empirical conditions. There are no excuses. So empirical influences cannot be necessary preparations for morality unless they are necessary for moral responsibility itself, and neither Kant nor Louden claims that all empirical aids are necessary in that sense. There is a straightforward contradiction between the claim that an agent is morally responsible and thus transcendentally free and the claim that certain empirical aids are in addition necessary for that agent to be morally good. Throughout his book, Louden at times seems to tackle this contradiction, but in almost every case he actually addresses a narrower and less significant problem. For example, at the end of his chapter “Education,” Louden considers several objections to the claims that “culture and education . . . are necessary . . . for human moralization” (Louden 2000:53) and that there must therefore be progress in human morality insofar as earlier generations must work to establish the culture and education that are necessary preconditions of the morality of later generations. In this context, the contradiction between freedom and necessary conditions arises in a particularly poignant way. It looks as if either earlier generations – who establish the conditions of possibility of morality – are not transcendentally free, or the “conditions of possibility” are not really necessary conditions.20 Louden considers several objections that seem to reflect concern about this basic contradiction between transcendental freedom and the necessary social conditions of moralization. First, “Kant’s philosophy of education and history contradicts the

82

The Problem

individualist strand of his ethical theory which holds that each moral agent has absolute moral worth” (Louden 2000:55). Second, there is “a charge of moral unfairness” because past generations lacked even the possibility to be good (Louden 2000:55). Third, there is the charge that Louden explicitly calls “Education vs. Transcendental Freedom,” which raises the question “how exactly is the crossing-over from nature to freedom to be achieved, and how can two such qualitatively different realms interact with one another?” (Louden 2000:56), and finally there is the problem of moral luck, that “later generations are eventually forced into a moral condition for which they themselves deserve no credit” (Louden 2000:57).21 Unfortunately, Louden’s solutions to these problems avoid rather than solve the real issue of tension between freedom and empirical influence. In response to the first problem, Louden only makes the tension more severe. He describes this problem in such a way that the conflict between “progress” and “dignity” is a matter of a supposed conflict between valuing people based on what they accomplish and valuing them based on their intrinsic moral worth. Louden thus reconstrues the problem as a conflict between Kant’s “teleological view concerning the true worth of the species” and the view that a good will would have dignity “even if [that] morally good will were not able to accomplish anything” (Louden 2000:57). His solution to this tension is to argue that insofar as the cultural and political development of the species is itself a precondition of human moralization, this conflict is only apparent. The dignity of human beings presupposes the development of the species. Without the latter, the former does not exist (Louden 2000:58).22 What Louden appears to be saying is that humans at later stages of development are not valued merely on the basis of the increase in good consequences, but that they actually come for the first time at the end of history to have good wills. Thus the two forms of evaluation do not conflict. But this response just makes the tension between transcendental freedom and teleological history all the more problematic. It looks as if progress, which establishes the necessary preconditions of morality, really does conflict with holding past generations to have even the minimum dignity that allows us to hold them morally responsible for their actions. And Kant, who insists that all human beings are morally responsible, cannot think that.23 Moreover, given that many of the supposed necessary conditions for moralization – conditions like world peace and high-quality moral education – are still not present, not even people today can be held morally responsible.

Moral Anthropology in Neokantian Ethics

83

With respect to the problem of unfairness, Louden suggests two responses. First, “once we recognize that complete perfection is an unattainable ideal, it becomes clear that no human being or group participates fully in this destiny: we can only approximate it” (Louden 2000:58). According to this response, it looks as if there is no moral unfairness, because all people and groups fail to achieve perfection, and all can in some way approximate it. The problem, of course, is that insofar as this is a claim about what is possible, Louden solves one problem – unfairness – by introducing the unfortunate consequence that no one is ever even capable of being morally good. But given that one is morally responsible on Kant’s account only if one is capable of acting morally, no one is ever morally responsible. Louden’s second response is to treat the issue of unfairness as one of progressive happiness, where the misery of early generations makes possible the happiness of later ones. It is not clear that Kant holds such a view, but if he does, then Louden’s response is correct. Louden explains that this unhappiness “is a matter of theodicy [in that it] makes possible the progress of the species” (Louden 2000:58), where the progress in arts, sciences, and (for Louden) morality is the final end of the species. While this solution would be legitimate in the context of happiness, it does not shed any light on the conflict between freedom and empirical influence. Louden’s response to the third problem, the most directly relevant formulation of the tension, is the most troubling. Louden bites the bullet. Kant does believe that efficacious moral education is education that somehow cuts through the surface causal network in order to effect the grounding of character. How this process works is something human beings cannot fully understand; we cannot know intelligible character, nor can we ever know with certainty that our attempts to shape and influence it are effective. But we can assume that such efforts may succeed, and, indeed, this assumption is a necessary presupposition of any program of moral education. (Louden 2000:59)

This is an absolutely amazing claim, and if true it would mean that Kant rejects the asymmetrical relationship between the intelligible and the empirical character that we saw in Chapter 1. If it is true that empirical influences can “cut through the surface network” to actually influence intelligible character, then Kant must give up the asymmetry in his theoretical account of the relationship between appearances and things in themselves as well as the practical insistence that freedom

84

The Problem

precludes determination by anything tainted with empirical influence. Unless it is really a “necessary presupposition of any program of moral education” – and I argue in the next two chapters that it is not – this should be our last resort. And if it is a necessary presupposition, we should just admit that Kant’s transcendental freedom is sacrificed at the altar of impure ethics and not pretend that the conflict has been resolved. In Louden’s response to the last problem, moral luck, he returns to an emphasis on freedom and moral responsibility. He admits that “luck plays a role in many areas of human life” but goes on to claim that “the normative status of a person’s moral character is a function of how well this person ‘has tied himself to certain practical principles that he has unalterably prescribed for himself through his own reason’ [7:292]. Here the question is what we do to ourselves, not what nature and the environment have done to us” (Louden 2000:60). This is an accurate account of what is required for moral approbation on Kant’s account. But it does not fit well with Louden’s conception of aids as necessary preparations for morality nor with his extreme claims that education can somehow cut through the causal network to effect changes at the intelligible level. The latter claim suggests that one’s moral education can – at least in part – determine whether or not one is morally good. In this context, whether one is lucky enough to have good moral education has a lot to do with one’s moral status. It is not only a matter of what one does to oneself but is also a matter of “what nature and the environment have done to us.” And even to say that moral education is merely a necessary preparation means that those who are unlucky enough to be denied moral education or other empirical aids will not be morally good. Again, what nature or the environment does to us would have a lot to do with our moral status, a conclusion that would conflict with Kant’s conception of human freedom. One final proposal that Louden occasionally offers to reconcile transcendental freedom with empirical propaedeutics is somewhat more promising. He says, Culture (in particular the arts and sciences) and education are, along with law, politics, and religion, all necessary but not sufficient conditions for human moralization. There is no guarantee that people who have been exposed to all these preparatory steps will be morally good, but human beings who lack all contact with them cannot possibly be good. (Louden 2000:53, emphasis added, see too pp. 149–50, 159)

Moral Anthropology in Neokantian Ethics

85

Louden suggests that because empirical influences are not sufficient conditions for morality, they do not preclude transcendental freedom. Regardless of how favored persons are in terms of empirical aids to morality, they still will not be morally good unless they choose to be morally good. In fact, Louden points out, Kant thought that in the case of peace as an empirical aid, “peace can (and according to Kant will ) be attained without a fundamental change in moral character occurring” (Louden 2000:159).24 This reconciliation of necessary preconditions with transcendental freedom does successfully solve the contradiction, but only in those cases in which all the necessary preconditions are present. By insisting that no conditions are sufficient, Louden can preserve transcendental freedom and hence moral responsibility for all those people who may live at that wonderful time in which all necessary preconditions are present. But those living without those advantages are not transcendentally free and hence not morally responsible. And perpetual peace – to take just one empirical precondition – is not yet available to people today. This would imply the untenable conclusion that no one now or at any point in the past can be held morally responsible, because good willing is not even an option for us. So far I have argued that Louden is mistaken to take empirical aids to be necessary preconditions of morality. There may be some preconditions to moral responsibility, but these must be minimal enough to allow us to hold most human beings today and in the past morally responsible. Kant clearly thought that the basic cognitive and motivational requirements for being morally responsible at all are not very rare. He insists that The moral law commands compliance from everyone, and in fact the most exact compliance. Appraising what is to be done in accordance with it must, therefore, not be so difficult that the most common and unpracticed understanding should not know how to go about it, even without worldly prudence. (5:36)

By everyone in this passage, Kant presumably can exclude most nonhuman animals, some children too young to be held morally responsible, and perhaps even people with certain severe forms of insanity or disability. But he cannot exclude everyone who lacks all the advantages that Louden suggests are necessary. To make all empirical aids necessary is simply to go too far. But even the claim that empirical aids are propaedeutics of any kind fails to adequately address the problem of transcendental freedom. First, to say

86

The Problem

that empirical influences are preconditions that “somehow cut through the surface causal network” to exercise an effect on one’s intelligible character does not explain how this influence can be consistent with our freedom. Louden insists that “how this process works is something human beings cannot fully understand” (Louden 2000:59), an appropriate Kantian move given the limits on knowledge of the intelligible. But Louden must at least show that this influence does not contradict Kant’s insistence that the will is free. Kant himself, in resolving the antinomies, did not simply assert that certain things are beyond human understanding. He showed how two apparently contradictory theoretical claims could be made consistent. Louden seems to recognize the apparent contradiction between two practically grounded claims – the transcendental freedom of the will and the potential for that free will to be empirically influenced – but he just dodges the central issue of how they can be reconciled. The claim that empirical aids are “propaedeutics” does not resolve the prima facie contradiction in Kant’s ethics. The suggestion that there is an important practical sense in which empirical aids, as “propaedeutics,” are prior to moral choice raises serious problems for making sense of Kant’s descriptions of the relation between the free will and its manifestation in the world. These problems arise in a particularly striking way in Louden’s most detailed account of the relationship between inner moral transformation and external historical progress (Louden 2000:143–52). There, Louden explains that Moral character, at least on Kant’s view, is not about external events in time and space: indeed, it is not even about an internal process taking place in time and space. The latter is still only empirical character; and morality . . . is about intelligible character. . . . Human beings are cognitively limited creatures. Given these cognitive limitations, we must make do with what we have; we must try as best we can to decipher the inner from the outer. And since certain manifestations of the latter (e.g., law, art, culture, religious community) are themselves necessary preparatory steps for the former, it is correct to say that, from a human perspective, clear signs of outer progress also afford us clues regarding inner progress. . . . Human beings have no choice but to try to read inner moral character from its outer, empirical manifestations. (Louden 2000:152)

Louden seems not to appreciate the conflict between thinking of empirical aids as “manifestations” of inner moral character and thinking of them as “necessary preparations” for moral character. For Kant, looking to someone’s “deeds” – which include not only purely external processes but also internal temporal processes – is the best way that one can “try and read” that person’s moral character precisely

Moral Anthropology in Neokantian Ethics

87

because those visible processes are always the expression of moral character. Insofar as empirical influences are preparations for morality, however, they fail as expressions. A precondition of a good will, especially if it is a necessary precondition, cannot rightly be considered an expression of a good will. In sum, Louden’s account of the role of helps and hindrances fails on three counts. First, insofar as he suggests that empirical aids are necessary preconditions, he undermines Kant’s most basic convictions about the nature of freedom. Second, insofar as he merely points out that empirical aids are “preconditions,” he fails to enrich our understanding of their role sufficiently to show that there is no conflict between these aids and transcendental freedom. Third, by giving practical priority to these helps and hindrances, he undermines the relation wherein what is empirical manifests one’s moral status.

5. Munzel’s Alternative Account of Helps as Propaedeutics Like Louden, G. Felicitas Munzel offers an account of empirical influences as propaedeutics. And like Louden’s, her account is explicitly offered in the context of integrating Kant’s anthropology into his moral theory in general and his account of freedom in particular. At times, Munzel’s account of the propaedeutic function of helps and hindrances seems virtually identical to the account offered by Louden. She portrays the republican constitution, for example, as an “indispensable vehicle of moral cultivation” (Munzel 1999:259, emphasis added) and follows this up with an account of how this necessity does not interfere with freedom because it is not sufficient for establishing morality. She describes discipline in education as an “indispensable first step” (Munzel 1999:280, emphasis added). She names at least some empirical conditions “necessary (albeit not sufficient) conditions for producing character” (Munzel 1999:340). And, she claims, “the pedagogical process complete with propaedeutic functions which prepare and develop these aptitudes, even initially awakening them to a state of moral responsiveness, is deemed by Kant to be essential ” (Munzel 1999:340–1). All of this might lead one to think that Munzel, like Louden, offers an account of empirical aids as necessary but not sufficient propaedeutics to having a good will. But in other places Munzel’s account is more nuanced. She describes the cultivation of natural aptitudes not as necessary, but as a “central factor”

88

The Problem

for developing a good character. In the course of outlining hindrances to morality, she suggests that “life in even the so-called civilized state can present . . . nearly insurmountable obstacles” (Munzel 1999:177, emphasis added). In the presence of numerous hindrances to morality, “conditions in the state are not conducive for the successful cultivation of inner peace within . . . individuals” (Munzel 1999:179, emphasis added). Aptitudes for good “serve to promote the observance of the law” (Munzel 1999: 110, emphasis added). “Discipline and cultivation prepare the conditions most conducive to the formation of character in its strict sense” (Munzel 1999:332, emphasis added). In these cases, Munzel backs off from strong language of necessity, pointing out that helps merely help, rather than enable, and hindrances merely hinder, rather than prevent, developing a good will. If this were the only difference, Munzel’s account would be better than Louden’s in that it would not be susceptible to objections against the notion of a necessary propaedeutic. But her account would still seem to imply that natural influences “cut through the surface causal network” to influence one’s free will, and – like Louden – she does not explain how this can be reconciled with Kant’s asymmetry between freedom and nature. But Munzel provides several important illustrations of the relationship between the good will and empirical aids that suggest a different conception of the relation between the two. I will not discuss all the details of her account here but will focus on the two most fruitful analogies she offers for understanding how empirical influences relate to the free choice that ultimately establishes a morally good character in a person.25 The first analogy is medical. Munzel suggests that empirical aids can be viewed as treatments for the symptoms of a lack of morally good character. This analogy is particularly significant in the context of the helpful discussion of the relationship between empirical and intelligible character, a distinction that is important for my own account of the relationship between empirical aids and moral formation. Empirical character, Munzel points out, can be thought of as the effect of intelligible character. Insofar as one’s intelligible character is evil or unformed, one will have various symptoms at the empirical level, symptoms that can be treated by empirical means. Munzel says, Reflection on Kant’s own metaphor likening the virtuous comportment of mind to a moral state of health leads one to recognize that it is not at all unusual in cases of illness for treatment to begin with the mitigation of the symptoms, for fever and swelling to be reduced (for example), before surgery is

Moral Anthropology in Neokantian Ethics

89

undertaken to correct the actual source of these effects or symptoms. By extension, it is no more necessary to assume that the external discipline of the inclinations executed under just, civil government has become Kant’s basis for the formation of good moral character, than one would assume that the external application of ice pack and anti-inflammatory medications would correct or even begin to address the underlying cause in a diseased organ. (Munzel 1999:180)

This analogy is markedly different from Louden’s account of the role of propaedeutics in that it need not imply an actual influence of empirical aids on moral choice at the level of free choice. The “diseased organ” – the evil intelligible character – does not seem to be affected at all by the empirical influences, which merely treat the “symptoms”: the empirical character that is manifested in various evil deeds. Unfortunately, it is not at all clear in what sense this treatment of symptoms is morally significant. It is, of course, important that the number of evil deeds in the world be reduced. Because one ought to be concerned about the happiness of others, one must seek to reduce those deeds, performed by oneself or by others, that interfere with this happiness. But if treating symptoms is merely treating symptoms, if it does not at all relate to the diseased organ, then these empirical influences seem to have no moral significance at all. Munzel seems to recognize this problem, and she later recasts treating the symptoms in a more significant light. She continues to insist that treatment of symptoms is never the basis for moral improvement (see, e.g., Munzel 1999:322), but she suggests that The lawless exercise of human nature appears feverish and swollen and calls for discipline as the propaedeutic first step in bringing about a state of moral health by alleviating the symptoms. Only then are the human capacities fit for, able to undergo, their positive cultivation, which is the next step. (Munzel 1999: 280–1)

If alleviating the symptoms is a first step, such that a person is ready to be healed only after such alleviation, then helps and hindrances are morally significant. Unfortunately, this moral significance is bought with a price. In particular, an illustration that at first seemed to provide an important alternative to Louden’s account of necessary but not sufficient conditions collapses back into that account. This analogy gives a different interpretation of the way empirical influences are necessary conditions – an interpretation that takes much more seriously Kant’s distinction between intelligible and empirical character – but it is susceptible to the same objections.

90

The Problem

Munzel’s second illustration avoids this problem. She describes the analogy in terms of an oath of office. She says, The relation of the formation of character . . . to discipline and cultivation [i.e., empirical influences] may be understood in terms of an analogy with the relation of preparatory training to taking an oath of office. In this case, of course, the office is that of the vocation of our humanity. No amount of schooling equipping us (in our subjective capacities) to carry out the office well, even if essential for conducting ourselves well in office, can necessitate our taking the oath of office, our committing ourselves consciously and freely to assuming personal responsibility for fulfilling it. This choice to make it our own vocation, to make the promise (in effect) to ourselves and to others that we will duly perform it, can only be our own. It is not a degree ceremoniously conferred upon us that we may passively receive at the conclusion of our studies; we cannot be made or pronounced by another to be a person of good moral character. (Munzel 1999:331–2)

So far, this description of the relationship between the training and the oath seems to imply that the training is a necessary but not sufficient condition of taking the oath. Munzel’s emphasis is on the fact that no amount of training can be sufficient. But she goes on to insist that training is not necessary either: In principle, consistent with Kant’s description of the adoption of this resolve as a conversion or transformation in conduct of thought and with his fundamental premise of freedom, there is in fact nothing to prevent someone from taking this particular oath of office in the absence of preparatory schooling and then acquiring the latter as a kind of on-the-job training. Pragmatically speaking, to do so would be to take the hard road . . ., but it would not be per se impossible. (Munzel 1999:332)

Munzel’s attention to Kant’s “fundamental premise of freedom” as a constraint on her own account of propaedeutics leads to an important insight. It must be possible to be morally good – to take the “oath” – prior to any empirical aids. This passage also suggests what is certainly correct, that these empirical aids are still important subsequent helps, as “on-thejob training.” Unfortunately, Munzel does not adequately distinguish this training from the sorts of roles in Sherman’s accounts of empirical influences. Moreover, her account is still permeated with temporal language that confuses the sort of priority that the oath taking must have. As a result, she still seems to think of the sort of relationship between empirical influence and moral choice that she describes here as the exceptional case, the “hard road.” All of the authors discussed in this chapter offer profitable contributions to Kantian ethics, but none offers a satisfying account of how moral

Moral Anthropology in Neokantian Ethics

91

anthropology can be integrated with Kant’s views about human freedom. Sherman and Herman present important analyses of various roles that helps and hindrances – especially emotions – can play in moral life, but none of these roles fits the unique importance of empirical influences in moral anthropology. In particular, none shows how the presence or absence of helps and hindrances can make a difference for the moral status of someone already morally responsible. Louden attempts to address this issue head-on, but in the process he seems to sacrifice transcendental freedom to preserve the importance of impure ethics. Munzel offers the most promising approach, but she fails to clearly articulate in detail the relationship between one’s moral commitment and empirical influences. In the next chapter, I offer an account of the relationship between the good will and empirical influences that shows how these influences can have genuine moral significance without undermining the fundamental priority of freedom in Kant’s moral theory.

part ii THE SOLUTION

5 Transcendental Idealism, Radical Evil, and Moral Anthropology

Over the past several chapters, I have shown that there seems to be a deep problem in Kant’s moral philosophy, one that Schleiermacher recognized and to which he drew attention. This problem is not limited to Kant but is shared by any moral theory that takes seriously a strong, nondeterminist conception of human freedom as a condition of moral responsibility while at the same time recognizing the obvious importance of helps and hindrances to moral development. Essentially, Schleiermacher points out an apparent conflict between three fundamental assertions: that the will is transcendentally free, that anthropology is empirical, and that anthropology studies morally significant helps and hindrances. Because Kant so forcefully articulates the nature of freedom, the problem arises for him in a particularly stark way. The fact that Kant’s recognition of helps and hindrances has been ignored for so long only heightens the sense today that this recognition is incompatible with the rest of his theory. Moreover, as we saw in the previous chapter, recent attempts to make sense of these helps and hindrances either deny their full moral significance or sacrifice Kant’s strong conception of freedom. In this chapter, I offer a Kantian solution to Schleiermacher’s dilemma. In section 1, I show that Kant has the resources to distinguish between the empirical will, which can be affected by empirical influences, and the free will, which cannot.1 In section 2, I explain that changes in the empirical will are morally relevant as expressions of the moral status of the free will. In section 3, I argue that the Kantian notion of “radical evil” requires rethinking the nature of the good human will, and grace makes such a rethinking possible. In section 4, I show that proper expressions of a good human will involve attention to considerations of moral anthropology. 95

96

The Solution

Thus moral anthropology is morally relevant because it describes aspects of human life that are important expressions of a good human will.

1. Kant’s Crucial Distinction Unfortunately, Kant does not offer an extended, systematic treatment of the relationship between freedom and moral anthropology. However, his mature moral theory does provide the resources for solving the problem raised by Schleiermacher, and Kant provides scattered but important remarks that directly address it. The most direct treatment is a footnote in the Critique of Judgment. There Kant says, One of the various supposed contradictions in this complete distinction of natural causality from the causality through freedom is given in the following objection to it. It is held that when I talk about nature putting obstacles [Hindernissen] in the way of the causality governed by laws of freedom (moral laws), or about nature furthering it, I do after all grant that nature influences freedom. (5:196)

Here Kant sketches a criticism similar to the one that Schleiermacher raises several years later.2 The helps and hindrances that are a particular problem for freedom are those that are part of nature, which for Kant is just the empirical world. The problem of nature’s influencing freedom is the problem of empirical helps and hindrances’ influencing moral choice. To that problem, Kant offers the following solution: This is a misinterpretation, which is easily avoided merely by understanding what I have said. The resistance or furtherance is not between nature and freedom, but between nature as appearance and the effects of freedom as appearances in the world of sense; and even the causality of freedom (of pure and practical reason) is the causality of a natural cause (the subject, regarded as a human being and hence as an appearance) subject to the [laws of] nature.3 It is this causality’s determination whose basis is contained, in a way not otherwise explicable, in the intelligible that is thought under freedom ( just as in the case of the intelligible that is the supersensible substrate of nature). (5:196)

Kant’s explanation of the relationship between freedom and empirical influences depends on his distinction between phenomena and noumena, and specifically on his claim that each person exists both as a phenomenal series (nature) and a noumenal (free) ground of that series.4 The solution to Schleiermacher’s dilemma is that moral responsibility and hence freedom apply to oneself as a free ground of appearances, while helps and hindrances affect only one’s “natural” appearance.5

Radical Evil and Moral Anthropology

97

This solution, however, is no more than a preliminary gesture. In particular, it does not explain why, if empirical influences cannot affect the free self, they are morally relevant. One can be good or evil only insofar as one is free. A series of appearances is relevant to morality only insofar as that series expresses one’s free will. If empirical influences change the way one appears without changing the free will that grounds this appearance, then the appearances are not pure expressions of the free self – or so it would seem. Unless empirical influences can change the moral status of the free will, any effect these influences have on a series of appearances seems to be morally secondary. If empirical influences actually are relevant to one’s moral status, as Kant suggests in his anthropology, then they must have some important relation to oneself insofar as one is free. They cannot just influence how one appears. Thus this important footnote does not completely reconcile transcendental freedom with moral anthropology. It rightly points to the relationship between the free will and the appearance of that will as the core of the solution, but Kant still needs an account of how appearances relate to the free will. Given the inscrutability of freedom, this account cannot be a metaphysical one. But it also need not be metaphysical. All that Kant needs to show is why one should be morally concerned with those appearances that can be affected by empirical influences. In particular, he needs to show why one should be morally concerned about the sorts of changes involved in moral anthropology.

2. The Moral Importance of Appearances A. Connecting the Good Will to Its Appearance in the World A good will is a will that acts in accordance with the moral law. In the Groundwork, Kant focuses his attention on the concept of “duty, which includes that of a good will, though with certain subjective restrictions and hindrances” (4:397). For such a will, action in accordance with the moral law is action in accordance with a categorical imperative. From this categorical imperative are derived “all imperatives of duty” (4:421, cf. 4:422–3, 4:429–30, and 6:205–491). For the purposes of understanding the relationship between a good will and its expression, the details of these derivations are less important than the nature of the imperatives that are derived. Action for the sake of one’s duty characterizes the good human will, but the imperatives of duty are themselves directed toward appearances. In that sense, action in appearance that conforms to certain standards characterizes the good free will.

98

The Solution

For example, in the Groundwork Kant explains that one ought never act on a maxim of the form “When I believe myself to be in need of money, I will borrow money and promise to pay it back, although I know that I can never do so” (4:422). One is morally good only by refraining from false promises out of respect for the moral law. This respect for the moral law is the deliberate and constant commitment to act in such a way that one’s maxims can conform to the moral law. But although neither one’s maxim nor one’s commitment to the moral law is itself an appearance, both are directed toward appearances. A maxim is a rule for action that specifies a certain action or type of action as a means to an end or type of end in a certain situation or type of situation. For example, take the maxim “When I believe myself to be in need of money, I will borrow money and promise to pay it back, although I know that I can never do so” (4:422). Here the action is promising, the end is acquiring money, and the situation is having certain beliefs about one’s needs and inability to repay. Every aspect of this maxim is tied to appearances. The action is an action that will take place in appearances. A purely noumenal “act” is not likely to convince anyone to give money, because others cannot see or hear or experience this act. A promise is something that can be heard or otherwise perceived. Likewise, the acquisition of money, the desire to acquire money, the inclinations that are satisfied by having money, the pleasure of their satisfaction, and the beliefs about one’s needs and abilities are all appearances. Although inclinations, desires, pleasures, and beliefs may appear only in inner sense and might not be observable to anyone other than the one who has the desire, inclination, pleasure, or belief, they are nevertheless experiences. Now the moral status of an agent is not purely a matter of the action, the end, and the situation that together form the content of the agent’s maxims. Rather, Kant insists that the form of the maxim determines its moral worth. One might perform the same action, or act for the same end, or act in the same situation, without moral fault. Fault enters because of the relation between the action, end, and situation.6 And one might think that even if every aspect of the content of a maxim is appearance directed, the form of the maxim is not. In fact, however, concern with the form of a maxim is concern about the structure of one’s actions in the world. The conformity of a particular maxim to the moral law is not reducible to the content of the maxim, but neither is it separable from that content. To have a good will is to choose to act on certain sorts of maxims rather than on others. And the sorts of maxims on which one acts describe actions in the world for ends

Radical Evil and Moral Anthropology

99

in the world in situations that arise in the world. Morally upright agents are deeply concerned with appearances because effecting certain sorts of appearances is the way in which one expresses one’s moral goodness. It is important to note here that the term expression is used in the sense in which one expresses oneself through the clothes one wears, or the music to which one listens. This sort of self-expression is not the revealing of an already determined nature. One becomes the self one is through one’s self-expression.7 In the moral case this is true as well. When one expresses one’s goodness by refraining from false promising out of a sense of duty, one does not merely show a goodness that was “already there.” Instead, having a good will is a matter of choosing in certain ways. Refraining from false promises, when this flows from respect for the moral law, is a choice of the morally praiseworthy sort. By choosing in this way, one expresses one’s moral status in the sense that one chooses to be a morally good person. This emphasis on expressing one’s moral goodness may seem to shift the focus of moral attention from the moral law to oneself. Given that Kant says that one must act purely “from respect for [the moral] law” (4:400), this emphasis on oneself may seem inappropriate. In fact, however, acting from respect for the moral law involves performing deeds in the world (acting) in order to meet the demands of morality. And one always focuses on making one’s own actions, not just actions in general, conform to the moral law. In that sense, a person who acts so as to express his or her moral goodness is one who acts out of respect for the moral law. This model of the relationship between the good free will and its appearance helps to make sense of why one cares morally about one’s appearance in the world. As the expression of one’s noumenal will, one’s appearance in the world is morally relevant in the strongest possible sense. It is by being a certain sort of appearance that one shows oneself to be a certain sort of free agent. Not every feature of one’s appearance is relevant to the kind of free agent that one is. One’s physical appearance and natural temperament, for example, do not express one’s freedom. But all voluntary actions that are morally relevant8 express one’s free will, even though these actions are also the effects of empirical causes. This account still has two remaining problems, to which I will now turn. First, it is not clear how the sorts of concerns that are central to moral anthropology are relevant to expressing moral goodness. I explain this problem and its solution in more detail in sections 3 and 4. Second, and of more immediate concern, Kant often suggests that there is an insuperable barrier to inferring anything about one’s ultimate

100

The Solution

moral status from one’s appearance in the world. This barrier might seem to undermine any connection at all between the will and its expression. Before turning to the specific issues raised by moral anthropology, I devote the rest of this section to the general problem of the inscrutability of one’s moral status. B. The Inscrutability of One’s Moral Status In the midst of a discussion of freedom in the Groundwork, Kant claims that “we shall never be able to comprehend how freedom is possible” (4:456). This concern about inscrutability arises within both Kant’s theoretical and his practical philosophy. The theoretical limitations to knowledge about the relationship between the free self and its appearance were discussed in detail in Chapter 1. There we saw that although Kant can appropriately describe the free self as the cause or ground of its appearances, he cannot go further in describing the nature of this causal relation. The category of causation is invoked without a schematization. Thus it gives no true “knowledge” – by which Kant means theoretical knowledge – of the self. This theoretical limitation, however, need not interfere with practical knowledge of a nontheoretical kind about the relationship between the appearance of a self and the moral status of that self. However, it is crucial that any causal language about the relationship between the free will and its expression be purely practical. A further challenge comes from Kant’s practical philosophy. Given that the only “knowledge” of freedom that is possible is practical, it is crucial to know what limits Kant sets on knowledge of freedom from the standpoint of morality. We will find that one can know enough about the relationship between one’s appearance and one’s moral status to justify ascribing moral weight to influences on one’s appearance. In Kant’s practical philosophy, there is no general claim that one cannot know anything about one’s free self from the appearances of that self. But Kant sometimes seems to suggest that one cannot know the most important thing – one’s moral status – on the basis of appearances. In this section, I show that while Kant does insist that one can never know for certain that one is morally good, there is quite a bit that one can be justified in believing about one’s moral status from how one appears. The first Critique shows that one cannot have theoretical knowledge of the noumenal self, but Kant asserts on the basis of practical reason that one is free. Although this postulate is called mere “belief” or “faith” (Glaube), it is absolutely certain from the standpoint of practical reason. Thus it could be called “knowledge” in a practical sense, to distinguish it

Radical Evil and Moral Anthropology

101

from mere beliefs not held with practical certainty. The fact that one can know (in this sense) that one is free raises the question of how much more one can say about this free self on the basis of practical reason. One might think that one could at least know whether or not one is morally good, but Kant denies this. However, one can be certain that one is radically evil, and one can have reasons that justify a rational hope that one is good. Kant denies knowledge of one’s moral goodness for two reasons. The first is an epistemological point about moral awareness: We cannot know our own maxims, nor our volitional dispositions, with certainty. The second is more specifically moral: If one could be certain of having a good will, that certainty might lead to a moral complacency that would undermine such a will. Whereas a free will is necessary to make sense of moral action, knowledge of one’s good will actually undermines moral action. This practical limitation on self-knowledge will also turn out to be crucial for making sense of Kant’s account of helps and hindrances. Neither epistemic nor moral considerations, however, undermine knowledge of one’s own radical evil or of the fact that one can have some reasons for some confidence in one’s goodness. They limit only the extent to which one can be sure that one is good. Kant insists that we lack epistemic access to those aspects of ourselves that would allow for certainty regarding our moral goodness. The most famous declaration to this effect is in the Groundwork: In fact there is absolutely no possibility by means of experience to make out with complete certainty a single case in which a maxim of an action that may in other respects conform to duty has rested solely on moral grounds and on the representation of one’s duty. (4:407)

For Kant, neither the consequences of one’s actions nor the actions themselves, either of which could be empirically verified, determine one’s moral status. Instead, what matters are the reasons for acting, the maxims according to which one chooses. And there are no actions for which one cannot have a bad reason. As Kant makes clear in the Religion, even when one is evil, one’s “actions can still turn out to be as much in conformity to the law as if they had originated from true principles,” such that “the empirical character is then good but the intelligible character still evil” (6:36–7). Moreover, although observations of one’s inner life can reveal thoughts, feelings, and even patterns of deliberation, there is no guarantee that they will reveal the actual reasons for action. This is especially true,

102

The Solution

as Kant suggests in the rest of the passage from the Groundwork, because people have a very strong ulterior motive to present their actions as conforming to morality. We like to flatter ourselves with the false claim to a more noble motive; but in fact we can never, even by the strictest examination, completely plumb the depths of the secret incentives of our actions. For when moral value is being considered, the concern is not with the actions, which are seen, but rather with their inner principles, which are not seen. (4:407)

Human beings are adept at self-deception, especially when that deception can soften the bite of moral self-reproach.9 This deception takes many forms, and one of the most important is the presentation of one’s motives as more noble than they truly are. One is frequently able to mislead even oneself about the true maxims of one’s actions, covering one’s wickedness with a whitewash of illusory good intentions.10 Self-examination can never yield a final verdict on the moral status of one’s will because there may always be hidden or ignored motives at play. In the Religion, Kant explains a further difficulty with discerning moral virtue. A human being’s inner experience of himself does not allow him so to fathom the depths of his heart as to be able to attain, through self-observation, an entirely reliable cognition of the basis of the maxims which he professes, and of their purity and stability. (6:63, emphasis added)

To be truly good, at the level of the free will, one’s maxims must be not only pure but also stable. Even if self-examination could somehow overcome self-deception and complacency to reveal the purity of one’s maxims, no action can provide reliable assurance that one’s will is stably oriented toward the good. The moral status of one’s will thus remains in doubt. This inscrutability does not distress Kant, however. On the contrary, he notes an important practical benefit in remaining somewhat ignorant of one’s moral status. Kant cautions that “it seems never advisable to be encouraged to such a state of confidence [in one’s moral goodness] but much more beneficial (for morality) to ‘work out one’s salvation with fear and trembling’ ” (6:68). Good actions and good intentions offer agents some reasonable hope that they have a good will. Kant says only that one always lacks an “entirely reliable” cognition of the bases of one’s maxims. If one constantly acts in apparent accordance with and out of apparent respect for the moral law, one can reasonably hope that one is genuinely good. One might be deceiving oneself, but this hope is enough

Radical Evil and Moral Anthropology

103

to ground a connection in deliberation between moral status and actions in the world. But the ultimate inscrutability of moral status erodes overconfident complacency in one’s good will. The limit to self-knowledge actually promotes the maintenance of a good will. This consideration alone could not of course prove that one is epistemically limited, but it can help explain the importance for Kant of reminding his readers of the fact. And in this context, it is clear that inscrutability does not undermine the importance of appearances in moral action. One still seeks to effect the sorts of appearances that express a morally good will. But because no appearance is an incontrovertible expression of a good will, one never rests complacently in one’s moral goodness. Moreover, as we will see in the next section, there is at least one sense in which one’s moral status is not inscrutable. And it turns out that what one can know about one’s moral status provides a basis for being morally concerned about the sorts of considerations that arise in moral anthropology. Thus appearances are morally important as expressions of one’s free will. One evaluates and deliberates about one’s appearance in the world. Even though one can never be sure that any particular action is an expression of a good will, the possibility of justified hope is sufficient to motivate one to act well. This account of the moral importance of appearances has the advantage of maintaining Kant’s asymmetry between nature and freedom. Because appearances matter only as expressions of a free will but do not effect changes in that free will, the relationship is asymmetrical. Moreover, because this expression relation is purely a matter of moral responsibility, in that an expression is an expression of a moral status; it does not transgress the theoretical limitations on knowledge. Finally, because the expression itself is empirical, the account offered here shows how at least some empirical facts can be morally relevant. But I still have not shown how this account applies to moral anthropology in particular. The next two sections accomplish that task.

3. Toward the Moral Importance of Moral Anthropology: Radical Evil Kant’s account of the connection of the good will with appearancedirected maxims provides a reason for concern about appearances, and even the limitations on knowledge of one’s moral status do not undermine this concern. But this concern seems to be justified only in those cases when one’s appearance reflects a maxim that does not conform to the moral law, or a lack of respect for the moral law in the choice of

104

The Solution

one’s maxim. The account offered so far shows only that one should care about acting, in the realm of appearance, in accordance with the moral law. But the sorts of considerations that are part of moral anthropology are not themselves directly required by the moral law.11 In the next section, I explain why one has an obligation not only to act in the realm of appearance out of respect for the moral law but also to take into account moral anthropology to promote aids and avoid obstacles to morality. For that purpose, however, one must first take a closer look at what can be known about one’s moral condition. Kant’s most extensive discussion of this is in his Religion within the Boundaries of Mere Reason.12 A. The Proof of Radical Evil In section 2, we saw that Kant imposes stringent limits on what one can know about one’s moral condition. Kant’s remarks on the inscrutability of one’s moral status seem to be very general. It seems as though one of the most important facts about the relationship between a free will and the series of appearances that is its expression is that one can never be sure of anything about the free will from the details of its appearance. But Kant is careful to make a narrower claim. Kant’s discussions of inscrutability focus on the impossibility of knowing that one is morally good.13 Thus in the Groundwork he specifically argues that one might always have a maxim of self-love that underlies an action apparently done from duty alone. In the Religion, his claim is that inner experience does not allow one to know the “purity and stability” of one’s maxims. One cannot know that one’s maxims are good. And the state of confidence against which he warns is a state of confidence in one’s moral goodness. Human self-deception tries to make one seem better than one truly is. Kant seeks to silence those who assert their moral virtue, not to deny justification for any moral claims at all about oneself. In fact, in his Religion, Kant makes a rather startling claim, made even more startling by the fact that his support for it is empirical.14 He claims that human beings are “radically evil.” In the context of this discussion, Kant emphasizes the difficulty of judging moral status from the mere appearance of the will, but he goes on to explain how one can show moral evil from appearances. His reasoning is complicated, so it is worth looking at closely. Kant begins by explaining, We call a human being evil, however, not because he performs actions that are evil (contrary to law), but because these are so constituted that they allow the inference of evil maxims in him. (6:20)

Radical Evil and Moral Anthropology

105

This is the essence of the empirical argument. From certain kinds of actions, one can infer the presence of evil maxims. But then Kant makes a remark that suggests the impossibility of such an inference. Now through experience we can indeed notice unlawful actions, and also notice (at least within ourselves) that they are consciously contrary to the law. But we cannot observe maxims, we cannot do so unproblematically even within ourselves; hence the judgment that an agent is an evil human being cannot reliably be based on experience. (6:20)

Based on this passage, one might think that Kant will give an a priori argument for the evil in human beings, or just give up his assertion that one can know that they are radically evil. But Kant does neither. Rather, he goes on later in the section to insist that “according to the cognition we have of the human being through experience . . ., we may presuppose evil as subjectively necessary in every human being” and that “the existence of this propensity to evil in human nature can be established through experiential demonstrations” (6:32, 6:35, emphases added). And immediately following his apparent claim that one cannot know reliably from experience that a person is evil, Kant suggests that there might be an empirical basis after all. He says, In order, then, to call a human being evil, it must be possible to infer a priori from a number of consciously evil actions, or even from a single one, an underlying evil maxim. (6:20)

An a priori inference is necessary in order to “establish” the propensity to evil in human beings “through experiential demonstrations of the actual resistance in time of the human power of choice against the law” (6:35). Thus when Kant says, “[T]he judgment that an agent is an evil human being cannot be reliably based on experience” (6:20), he cannot mean that experience is irrelevant to such a judgment. Rather, he must mean that such a judgment cannot be based on experience alone, because an a priori argument is necessary in order to show a connection between actions and maxims. Kant must show a priori that there is a connection between actions and maxims such that one can conclude that a person has evil maxims on the basis of his or her actions. But this a priori argument alone will not prove that actual human beings are radically evil. To do that, Kant will have to make the further empirical argument that human beings actually perform the actions that justify inferring that they are evil. The a priori argument showing how actions can provide a basis for judgment about maxims is

106

The Solution

an important preliminary step before the empirical argument can get started. This inference from actions to maxims is based on the fact that certain actions are possible only because the moral law is overridden by other considerations. As Kant explains, If the law fails nevertheless to determine somebody’s free power of choice with respect to an action relating to it, an incentive opposed to it must have influence on the power of choice of the human being in question; . . . by hypothesis, this can only happen because this human being incorporates this incentive (and consequently also the deviation from the moral law) into his maxim (in which case he is an evil human being). (6:24)

From the failure to act in certain ways, one can infer the status of a person’s disposition. One might argue that what remains unknowable is precisely whether “the law fails to determine someone’s free power of choice,” and thus that while a person’s disposition could be inferred from knowledge of that, it cannot be inferred from actions (or internal motivational states) alone. In fact, however, some actions are contrary to duty and therefore cannot be done from duty. In the Groundwork, Kant briefly mentions this when he introduces his candidates for dutiful action: I here omit all actions already recognized as contrary to duty, even though they may be useful for this or that end; for in the case of these the question does not arise at all as to whether they might be done from duty, since they even conflict with duty. (4:397, emphasis added)

Some actions are recognized as in themselves contrary to duty, such that there is no maxim according to which they are justified.15 In the Metaphysics of Morals Kant takes this even further, articulating a whole doctrine of right based on the principle that “any action is right if it can coexist with everyone’s freedom in accordance with a universal law” (6:231). The doctrine of right deals with the moral status of actions, regardless of the maxim according to which they are performed. For Kant, governments can and should define those actions that are wrong. And although the state cannot punish all moral evil, those crimes that it can punish are included within the scope of moral evil. As Kant explains, “that I make it my maxim to act rightly is a demand that ethics makes on me” (6:231). Every act contrary to right implies a choice that is contrary to the moral law. Thus there is a whole set of actions, spelled out in the Doctrine of Right, the performance of which implies an evil will. Moreover, there is some reason to trust self-consciousness of evil even for those actions that could be consistent with some morally acceptable

Radical Evil and Moral Anthropology

107

maxim. Kant’s explanations of the difficulty of determining the true maxim of one’s conduct are directed against those who consider their conduct to be good because he recognizes the tendency of human beings to deceive themselves about their own goodness. However, there does not seem to be a comparable tendency to deceive oneself to one’s own detriment.16 If, on reflection, one recognizes deliberate violations of the moral law out of self-love, these are unlikely to be rooted in respect for the moral law at some deeper level. The awareness of action done with evil intention is good evidence of an evil disposition. Even if it is conceivable that one’s actual maxim is not the evil maxim of which one is conscious, it is likely to be a deeper, equally immoral maxim. And even if one’s true underlying maxim is consistent with the moral law, one who cannot – even on careful introspection – recognize the consistency of one’s maxims with the moral law is unlikely to have a stable commitment to acting only on those maxims that are morally permissible. In that context, Kant offers his empirical argument for evil. He has shown that at least for actions contrary to right, it is possible to infer with certainty the state of one’s maxims from appearances, and thus to the disposition of one’s free choice. More generally, for maxims that seem contrary to the moral law, it is reasonable to conclude from evil appearances that one has an evil disposition. To show that humans are evil, then, Kant must simply show that humans perform the sorts of actions from which the presence of evil can be inferred.17 And that is not particularly hard to do. Kant prefaces his litany of examples with the unfortunate truth that “we can spare ourselves the formal proof that there must be such a corrupt propensity rooted in the human being, in view of the multitude of woeful examples that the experience of human deeds parades before us” (6:32–3). This is followed by examples of unjust cruelty, betrayal, ingratitude, and finally war. The first and last are contrary to right, while the others are examples of which his readers cannot deny they have at times been guilty. All that it takes, for any individual, to be assured of one’s own evil is to reflect upon one’s actions. If one has ever knowingly violated the rights of others, one is evil. Kant’s argument for rigorism establishes that even a single evil action is enough to show that one has a will uncommitted to the moral law, which is to say an evil will.18 Even if one has never violated strict right, if one has ever acted from self-love in contradiction to or disregard of the moral law, one cannot reasonably deny that one is evil. Thus Kant explains that although “this quality [evil] may [not] be inferred from the concept of . . . a human being, . . . according to the cognition we have of

108

The Solution

the human being through experience, he cannot be judged otherwise” (6:32). B. The Nature of Radical Evil Once Kant sketches his argument that human beings are radically evil, he goes on to characterize the nature of that evil. And whereas the fact of radical evil is established on empirical grounds, the nature of radical evil must be explained a priori. Kant explains, “even though the existence of this propensity to evil in human nature can be established through experiential demonstrations . . ., these demonstrations still do not teach us the real nature of that propensity . . .; that nature . . . must be cognized a priori from the concept of evil” (6:35). On Kant’s account, the evil in human nature is the subordination of the moral law to maxims of self-love.19 If [a person] took [incentives of his sensuous nature] into his maxim as of themselves sufficient for the determination of his power of choice, without minding the moral law . . ., he would then become morally evil. (6:36)

Thus, The statement, “The human being is evil,” cannot mean anything else than that he is conscious of the moral law and yet has incorporated into his maxim the (occasional) deviation from it. (6:32)

When a person allows him- or herself to choose in such a way that the moral law can be overridden by sensuous incentives, that person is evil. In characterizing this evil, however, Kant also makes use of the important concept of a propensity to evil, a propensity that is a universal consequence of the radical evil in human nature.20 Kant explains his notion of a propensity as follows: Propensity is actually only the predisposition to desire an enjoyment which, when the subject has experienced it, arouses inclination to it. Thus all savages have a propensity for intoxicants; for although many of them have no acquiantance at all with intoxication, and hence absolutely no desire for the things that produce it, let them try these things but once, and there is aroused in them an almost inextinguishable desire for them. (6:29)

A propensity is distinguished from an inclination in that a propensity is merely a potential inclination. A propensity is merely potential, however, in an unusual sense. The propensity for intoxicants that Kant ascribes to “savages” implies not merely that when they “try these things” they can come to have an inclination. Rather, a propensity is potential in that it

Radical Evil and Moral Anthropology

109

depends upon certain conditions, but once those conditions are satisfied, “there is aroused in them a . . . desire.” Someone who does not desire intoxicants even after trying them lacks a propensity to them. Propensities cannot be ignored, although of course the desires to which they irresistibly give rise can be resisted. These aspects of Kant’s discussion are general, and it is difficult to apply this conception of propensity to the propensity for moral evil. After all, moral evil is not something that depends upon one’s inclinations but is just a matter of the way that one considers one’s inclination in one’s choices. So if propensities are just proto-inclinations, how can there be a propensity for evil itself ? Fortunately, Kant explains the special way that the notion of a propensity should be applied to moral evil, and in the process he drops all language of inclination. He says, A propensity can indeed be innate yet may be represented as not being such: it can rather also be thought of (if it is good) as acquired, or (if evil) as brought by the human being upon himself. – Here, however, we are only talking of a propensity to genuine evil, i.e. moral evil, which, since it is only possible as the determination of a free power of choice and this power for its part can be judged good or evil only on the basis of its maxims, must reside in the subjective ground of the possibility of the deviation of the maxims from the moral law. (6:29)

One with an ordinary propensity has a potential inclination (say, to drink intoxicants); one with a propensity to evil has a potential choice, that is, a potential “deviation of maxims from the moral law.” Thus, whereas an ordinary propensity can be overridden by the will because it affects only one’s desires, the propensity to evil affects the will itself. Moreover, whereas ordinary propensities may or may not be due to oneself, the propensity to evil is “brought by the human being upon himself.”21 This “possibility of deviation” has in common with other propensities, however, that it is not a mere possibility. The propensity to evil is not simply what makes it possible for human beings to choose evil. All that is necessary for it to be possible to be evil is that human beings have inclinations that can conflict with the moral law. A propensity is a potential that depends only upon circumstance for its actualization.22 Kant does not make clear exactly what these circumstances are. They might involve merely the opportunity for evil, or inclinations of a certain strength, or certain kinds of social pressures. Whatever the conditions, the point is that a human being will in fact choose evil when they are present. Thus just as one who does not desire intoxicants even when he experiences them must be said to lack the propensity to intoxicants, one who does not do evil given appropriate conditions lacks the propensity to evil. The propensity to evil

110

The Solution

in human beings is not identical to choosing evil, but it is the condition that implies that one will choose evil.23 This propensity need not imply that one always make choices that are contrary to the moral law. To choose evil is to make choices in such a way that one allows the moral law to be overridden given a sufficient sensuous incentive. Thus to have a propensity to evil is to have a breaking point, a degree of temptation to which one will submit. Kant illustrates this at the end of a section explaining that “the human being is by nature evil.” There he says, A member of the English Parliament exclaimed in the heat of debate: “Every man has his price, for which he sells himself.” If this is true (and everyone can decide for himself), if nowhere is a virtue which no level of temptation can overthrow, if whether good or evil wins us over only depends on which bids the most and affords the promptest pay-off, then what the Apostle says might indeed hold true of human beings universally, “There is no distinction here, they are all under sin – there is none righteous (in the spirit of the law), no, not one.” (6:39)

That all are under sin is precisely the claim that Kant puts forward in this section, and his basis for this claim is an empirical argument that everyone can decide about for him- or herself. What this passage adds is a clarification of the nature of that universal evil. The universal propensity to evil implies that everyone has a price, that none is in principle committed to the moral law above all sensuous inclination. This propensity to evil is due to the choice of the person affected by it because the person brings it upon him- or herself. Thus it expresses an already present24 moral evil. Kant explains, This propensity itself is morally evil, since it must ultimately be sought in a free power of choice, and hence is imputable. (6:37, cf. 6:32)

The propensity to moral evil is an expression of an evil will, but because it is a propensity, it is also a potential that will be actualized under certain conditions. And even if these conditions are never present, the propensity marks a willingness to transgress that is itself evil. The radical evil of human beings thus includes choosing both maxims that are themselves contrary to the moral law and other maxims that constitute or bring about a propensity to and persistence in evil choices. One who is evil – and all humans are – chooses evil and also ensures that evil will continue in one’s choices throughout life. This propensity to evil, constituted and reinforced by an evil will that refuses to submit to the moral law, employs three degrees of self-manipulation to ensure the continued dominance of sensuous

Radical Evil and Moral Anthropology

111

incentives over the will. Kant outlines these as the “frailty . . ., impurity . . ., [and] depravity of human nature” (6:29). All three degrees of evil are forms of a propensity to evil that is itself the consequence of evil selfmanipulation. Frailty, the first degree of evil, is when I incorporate the good (the law) into the maxim of my power of choice; but this good, which is an irresistible incentive objectively or ideally, is subjectively the weaker (in comparison with inclination) whenever the maxim is to be followed. (6:29)

This frailty is closely connected to the problem of a lack of character, as discussed in Chapter 3. In one’s cool, quiet moments, one articulates principles, but one is incapable of sticking to them. As Kant indicates in his anthropology, this lack of character precludes one’s having a good will, precisely because a good will depends on one’s consistently acting according to the moral law. Thus one way in which one’s evil will seeks to undermine even the possibility of future reform is by establishing a nature so frail that it is incapable of consistent action.25 This frailty ensures that ever-shifting sensuous incentives continue to get the upper hand over the principle of the moral law. A second means of ensuring the persistence of evil is impurity, according to which Although the maxim is good with respect to its object . . . and perhaps even powerful enough in practice, it is not purely moral, i.e. it . . . often (and perhaps always) needs other incentives besides [the moral law] in order to determine the power of choice for what duty requires. (6:30)

In this case, one ensures that even apparently morally praiseworthy actions will in fact be corrupted by sensuous motives. One of the most insidious forms of self-deception is the tendency to interpret as morally upright actions that have underlying sensuous motives, and Kant connects selfdeception of this sort with the impurity of the will.26 In a description of human depravity that parallels his account of the three propensities to evil, Kant explains that one origin of impurity is The frailty of human nature . . . coupled with its dishonesty in not screening incentives . . . in accordance with the moral guide, and hence . . . in seeing only to the conformity of those incentives with the law, not to whether they have been derived from the latter itself. (6:37)

Adding dishonesty to frailty helps produce impurity of will. One begins by ignoring certain incentives, focusing on the moral law as an incentive for action. In this way, one can act from self-love, along with the motive of duty, without even recognizing it. This is precisely the sort of deception

112

The Solution

that makes knowledge of one’s moral status so difficult, and here Kant points out that it also makes becoming truly good more difficult. Because one is likely to interpret even actions that are bad (in that they are not due to the moral law as a sufficient motive) as morally good, one’s resolution in remaining true to the demands of morality is weakened. One is likely to be complacent in impurity through self-imposed ignorance. Eventually, at the worst extent of impurity, one does not act from the incentive of duty at all, but only such that one’s actions are “consistent” with the moral law. One’s evil will maintains a foothold by eroding the resolution needed to have a truly good will. The final degree of the propensity to evil is the most extreme. The depravity of the human heart is “the propensity of the power of choice to maxims that subordinate the incentives of the moral law to others (not moral ones)” (6:30). The previous two degrees of the propensity to evil really represent the attempts of an evil will to undermine the possibility of good. The final degree of the propensity to evil is the direct promotion of evil. Insofar as an agent seeks to bring upon himself the propensity for evil in this sense, he directly promotes his future performance of evil actions. At this level too, self-deception is important. Kant explains, The third [degree of the propensity to evil can be judged] as deliberate guilt (dolus), and is characterized by a certain perfidy on the part of the human heart (dolus malus) in deceiving itself [emphasis added] as regards its own good or evil disposition. (6:38)

The final degree of the propensity to evil is pervaded by self-deception. This is a benefit from the standpoint of happiness in that it allows for a self-contentment that would be otherwise impossible. But it is extremely dangerous from the standpoint of morality, for it contributes to the deepening grip of evil on a person. Kant insists that This dishonesty . . . hinders the establishment in us of a genuine moral disposition. . . . It rests on the radical evil of human nature which (inasmuch as it puts out of tune the moral ability to judge what to think of a human being and renders any imputability entirely uncertain, whether internal or external) constitutes the foul stain of our species – and so long as we do not remove it, it hinders the germ of good from developing as it otherwise would. (6:38, emphases added)

Self-deception is a hindrance of the most drastic kind, and it is therefore an important part of the radical evil in human nature that promotes ever deeper and more persistent evil. For Kant, then, evil is present in human nature as a failure to subordinate sensuous inclination to duty and also as a propensity to continue

Radical Evil and Moral Anthropology

113

this failure, a propensity characterized by the cultivation – in large part through self-deception – of frailty, impurity, and ultimately wickedness. Radical evil in human nature is a fact. C. The Problem of Radical Evil The fact of radical evil poses an important problem, which Kant spends his Religion within the Boundaries of Mere Reason solving. Put simply, the problem is that evil seems inextirpable. As Kant says, This evil is radical, since it corrupts the grounds of all maxims; as a natural propensity, it is also not to be extirpated through human forces, for this could happen only through good maxims – something that cannot take place if the subjective supreme ground of all maxims is presupposed to be corrupted. (6:37, cf. 6:45)

People are radically evil, such that they not only act on evil maxims but also establish in themselves a propensity to evil which ensures that future choices will be made against the moral law. In that context, it seems impossible to choose to reject evil and pursue good. This ineradicability of evil does not imply any kind of moral determinism. Because the evil that is ineradicable is freely chosen, that evil is not determined prior to choice. If one chose to be good, one could be. The problem with radical evil is that one in fact does not choose to be good. Thus, the ineradicability of evil does not in any way decrease one’s culpability for that evil. Nonetheless, there is an important practical problem for an agent who recognizes the presence of radical evil. One is always obligated to do what is morally right. But if one is already committed to evil, then there seems to be a relevant sense in which one can no longer choose to obey the moral law. This is true for two very different reasons. First, given that one has subordinated the moral law to sensuous inclinations, it can never be the case that one completely prioritizes morality over inclination. One’s overall moral status depends on one’s life as a whole.27 To be morally good, one cannot ever compromise morality. But if one has already compromised morality, even if one always does one’s duty from now on, one is nonetheless a person who, given the right circumstances (which may include temporal conditions), violates the moral law. That is to say, one is nonetheless evil. In Kant’s terms, However steadfastly a human being may have persevered in such a [good] disposition in a life conduct conformable to it, he nevertheless started from evil, and this is a debt which is impossible for him to wipe out. (6:72)28

114

The Solution

The first reason that one cannot be good is that one has a stain of past evil that cannot be erased. The second reason is that one’s evil is not merely past evil deeds but includes a propensity to evil. The fundamental maxim governing one’s life is a commitment to prefer inclination over morality. And this fundamental maxim provides no ground for its own overturning. One will not reject evil because the basis of one’s decisions is evil. And although obstacles to choosing rightly do not erode one’s responsibility for one’s evil, they seem to undermine the real possibility of moral reform. But Kant insists that what is required by morality is that one do what is right, which includes improving the basis of one’s choices. As Kant explains, “In spite of the fall, the command that we ought to become better human beings still resounds unabated in our souls; consequently, we must also be capable of it” (6:45). Throughout Kant’s moral philosophy, he takes for granted that human beings have moral obligations. In the second Critique, he draws attention to a “fact of reason,” according to which one recognizes that one is bound by duty.29 On the basis of this “fact,” Kant argues that one is free, that God must exist, and that one is immortal.30 Just as those arguments begin with the assumption of moral responsibility, Kant here simply takes for granted that one who is radically evil is obligated. In this case, however, because of the problem of one’s propensity to evil, one is obligated not only to do good but even to become better, because this is a necessary condition of choosing rightly. Because one still ought to become better, one must still be capable of moral reform. But that means that the radical evil that seemed ineradicable must, somehow, be possible to overcome. D. The Solution to the Problem of Radical Evil: Grace Kant’s attempt to show how radical evil can be overcome involves two central concepts. The one that is crucial for understanding the place of anthropology in his moral theory is a new conception of the human good will. I discuss this concept in detail in the next section. Before turning to that, however, I must briefly mention another part of Kant’s solution to the problem of radical evil: grace.31 My goal in this section is not to give a full treatment of Kant’s account of grace, nor to enter into the Christian background of this account, but only to show how Kant’s conception of grace makes possible a new conception of the good will, one that is consistent with his claims about radical evil. Thus I do not address the many nuances and problems with Kant’s account of grace.32

Radical Evil and Moral Anthropology

115

Kant’s account of radical evil involves the claim that evil cannot be extirpated from one’s nature “through human forces” (6:37). Kant emphasizes both the depth of radical evil and the fact that the root of evil lies in one’s own will, but he still insists that we must be capable, in some sense, of moral improvement. Thus there must be some nonhuman force through which evil can be overcome: “Some supernatural cooperation is also needed to [a person’s] becoming good or better” (6:44).33 To understand the roles that grace plays, it is important to remember the two very different problems posed by radical evil. First, the stain of radical evil cannot be removed regardless of the extent to which one improves in the future because one has done wrong in the past. Second, radical evil seems to undermine the possibility of transformation by deliberately hindering one’s own moral development, promoting a “propensity to evil.” Kant addresses the first problem with a conception of atoning grace, whereby one is justified before God, “after the fact,” so to speak. The second problem is addressed by a conception of sanctifying grace according to which God actually facilitates moral transformation.34 Both sorts of grace reflect attempts by Kant to articulate the possibility of a will that is fundamentally good – hence good in the eyes of God – but the temporal expression of which reflects radical evil. Kant wants to show how such a will is possible without weakening the requirements of the moral law. The way that grace functions depends, in part, on the type of grace involved. With respect to atoning grace, Kant’s emphasis is on providing a supplement for actual (past) violations of the moral law. In the case of sanctifying grace, Kant is concerned with a supplement that can counteract one’s propensity to evil. I first focus on atoning grace, and then I briefly discuss sanctifying grace. In a series of ethics lectures given just a few years before writing the Religion, Kant explains the role of atoning grace in his moral theory: The holy law necessarily entails that punishments should be appropriate to actions. But is man, then, to be left without help, seeing that he is frail, after all, in regard to morality? He cannot, indeed, hope for any remission of punishment for his crimes from a benevolent ruler, since in that case the divine will would not be holy; but . . . if, for our part, we do everything we can, we may hope for a supplementation, such that we may stand before God’s justice and be found adequate to the holy laws. . . . In that case, then, instead of a lenient justice, we have a supplementation of justice. (27:331)

Kant’s challenge is to reconcile three fundamental commitments: (1) One can rationally hope to be fundamentally morally good. (2) One is radically evil, both in the sense that one has fallen short of strict obedience

116

The Solution

to the moral law and in the sense that one deliberately undermines one’s capacity for good. (3) The moral law is rigorous; there is no middle ground between moral good and evil because the moral law insists on unswerving obedience.35 So how does grace allow Kant to reconcile these commitments? Apparently the notion of a “supplementation of justice” is distinguished from “lenient justice” in that while either would preserve moral hope (1) and radical evil (2), “lenient justice” would do so at the cost of rigorism (3). What is not clear is how grace preserves all three claims. It looks as if any role that grace can play will compromise one of the claims it is meant to save. Perhaps grace can save the rigor of the moral law at least in one sense, in that there might not be any degrees of goodness: One either receives grace and is thus good, or one does not. But if the moral law demands perfect conformity with morality, then any grace that makes one good despite one’s imperfection is for all apparent practical purposes indistinguishable from lenient justice. Perhaps Kant thinks that the concept of grace provides a better basis on which to argue for a conception of the good will that does not allow excuses in the moment of deliberation. Exactly what this involves will become clearer in the next section. For now it is enough to note that for Kant, it is crucial that in deliberation one measure one’s proposed maxims against the moral law in all its rigor. To avoid despair, one must hope that one can be good despite past evil deeds. And so some concept is necessary to allow this hope. But whereas a lenient justice would taint one’s present deliberations by excusing one from strict obedience to the law, atoning grace should allow one to avoid despair without providing any excuse from the full demands of morality in deliberation. Ultimately, the issue of whether grace is more effective in this regard than a nuanced version of lenient justice would go beyond the limited needs of this discussion. But this is clearly a role that Kant wants grace to play. More important for the present purposes is the fact that one can invoke the concept of grace at all only if it can be given some coherent sense. To preserve his commitments to moral hope, radical evil, and rigorism, Kant needs a concept that is distinct from “lenient justice.” He gives this concept a name: “grace” or the “supplementation of justice.” But it is not clear that Kant shows that anything corresponds, or even could correspond, to the name. In general, Kant’s strategy at this juncture is to point out that the mechanisms of grace are inscrutable to human reason. In the middle of the passage quoted above from the lectures on ethics,

Radical Evil and Moral Anthropology

117

Kant says, “How God brings about this supplementation, and what sort of means he employs for it, we know not, nor do we have any need to know; but we can hope for it” (27:331, cf. 28.2.2: 1120–1, 1225, 1319). In the Religion, Kant says that of “supernatural assistance . . . we can have not the least cognition” (6:191).36 And in the Conflict of the Faculties, written after the Religion,37 Kant repeats this point: If worst comes to worst reason is entitled [befugt] to adopt on faith a supernatural supplement to fill what is lacking for his justification (though [reason is] not to specify in what this consists). . . . We need not be able to understand and state exactly what the means of this replenishment is (for in the final analysis this is transcendent and, despite all that God Himself might tell us about it, inconceivable to us); even to lay claim to this knowledge would, in fact, be presumptuous. (7:43–4, emphases added)

The mechanism of grace cannot be understood. To avoid despair, one must believe that there is some way to be morally good, but practical reason will not condone any lenience in the interpretation of the moral law. Thus there must be some “inconceivable” supplement for our failings, some supplement that reason does not fully specify. The status of grace in Kant’s practical philosophy can be understood by analogy with the status of the unconditioned in Kant’s theoretical philosophy. In Chapter 1, we saw that for Kant, speculative reason has a “need” that the concept of freedom satisfies. This need is due to the demand on the part of speculative reason for completeness, a demand that requires some unconditioned ground for empirical knowledge. We can see Kant’s writings on Religion as suggesting that practical reason has a similar “need” for a concept of grace. This need is due to a demand on the part of practical reason, a demand for uniting three fundamental claims about human nature, the reasonableness of moral hope, the fact of radical evil, and rigorism. Like the demand for an unconditioned ground, this need provides a merely conditional warrant for the belief in grace. Neither practical nor theoretical reason strictly proves the reality of grace. In this sense, the grounds for grace are less solid than even the grounds for the postulates of pure practical reason.38 In particular, practical reason does not have to remain committed to moral hope – the conviction that we are still free to be good. Although one must be transcendentally free to be morally responsible, practical reason by itself need not claim that one who does in fact choose contrary to the moral law can nonetheless be morally good.

118

The Solution

But given the role that practical reason must play in our temporally situated deliberation, we have a subjective need to believe in something like grace. By itself, this need is not enough to justify even a conditional warrant for grace. Recall that the speculative need for some unconditioned does not specify freedom or even God as the content of that unconditioned. Until the second Critique argues on nontheoretical grounds for freedom and God, there is merely a need for something unconditioned, not a need for freedom or God per se. Likewise, the need of practical reason, in the context of radical evil, is just for some concept that will unify moral hope, radical evil, and rigorism. This need is not specifically a need for grace. Moreover, Kant gives no details about the perspective from within which one can justify grace. In the realm of theoretical reason, it is not enough to show that theoretical reason has a “need” for some unconditioned ground of appearances. Kant also shows, in the first Critique, that such a ground would not conflict with the conditions of possibility of experience, as long as the ground is not itself an object of possible experience and not justified on theoretical grounds. Only once he articulates his transcendental idealism can Kant justify freedom on practical grounds or believe in it on the grounds of a subjective need of theoretical reason.39 In the case of grace, Kant must show that some concept of grace can be consistent with the constraints of practical and theoretical reason. He does not necessarily have to explain how it can be consistent, but he must say at least enough to convince us that there is no contradiction with the requirements of theoretical or practical reason. In the Religion, Kant gives some detail about the way in which grace can fill the need of practical reason, and he offers some argument that it is not ruled out by practical or theoretical reason. With respect to atoning grace, Kant’s account involves God’s superior knowledge of what is nontemporal and a rational reconstruction of the Christian doctrine of substitutionary atonement. The substitutionary doctrine is ingenious, in that Kant essentially posits that one can suffer on one’s own behalf. The temporal appearance that expresses a good will involves suffering, and this suffering wipes out the stain of one’s past wickedness. Philip Quinn has worked out Kant’s doctrine of atonement in detail in terms of a distinction between “pre-revolutionary and post-revolutionary persons” (Quinn 1986:450). I will not go over his arguments here. Quinn rightly notes that while Kant’s rational reconstruction improves on some aspects of the traditional Christian doctrine of Christ’s vicarious atonement for

Radical Evil and Moral Anthropology

119

sins, it has many problems of its own (Quinn 1986). Whether or not Kant can solve these problems, his account of atoning grace does at least provide some sense of how grace could be consistent with the demands of theoretical and practical reason. Grace in this context is really a matter of God’s recognizing an intelligible truth – one’s goodness – in the face of the complicated appearance of one’s whole life. As long as Kant does not claim any theoretical warrant or empirical interpretation of this grace, there do not seem to be theoretical reasons to reject it. Practically, it raises more problems. In particular, it is not entirely clear how this account of atonement provides a real alternative to “lenient justice” to save moral hope, radical evil, and rigorism. Nothing in Kant’s account of the moral law suggests that suffering can somehow make up for violations of the moral law. And it is crucial that this be the case, lest people excuse themselves from obedience only to follow up vice with further (monkish) vices of self-flagellation. This account of atonement seems to amount to lenience in the demands of morality, albeit a very complicated lenience. Kant seems to respond to this objection when he argues, It can . . . be asked whether this deduction of the idea of a justification of a human being who is indeed guilty . . . has any practical use at all, and what such use could be. It is hard to see what positive use can be made of it for religion and for the conduct of life. . . . The investigation is only an answer to a speculative question, but one that cannot therefore be passed over in silence, since reason could then be accused of being absolutely incapable of reconciling the human being’s hope of absolution from his guilt with divine justice, and this accusation might be disadvantageous to reason in many respects, most of all morally. (6:76)

Just as freedom meets a need of theoretical reason but is not itself to be put to use theoretically, atoning grace meets a need of practical reason but can be dangerous when put to use practically. Thus it is totally different from “lenient justice,” which would be a modification of the demands of practical reason itself. Unfortunately, Kant’s claim that the investigation is “speculative” is not helpful here. It is just not clear what kind of warrant the argument for grace is supposed to have, if it is not practical. The argument is not a theoretical one, so there must be some other sense – as a condition of possibility of hope, perhaps – within which grace makes sense. But the relationship between this sort of reason and practical reason is not specified, and it is unclear what the relationship would be. This lack of specificity

120

The Solution

is markedly different, for example, from the detailed description of the status of the postulates in the second Critique. Kant’s account of sanctifying grace raises even more problems. Sanctifying grace is necessary not to make up for past misdeeds but to make possible future good deeds, given that the basis of one’s choices is corrupt. Kant does not discuss this grace in detail prior to the Religion, and his discussion in the Religion emphasizes the problems that arise if one tries to account for sanctifying grace. Kant raises an apparent antinomy. The first prong of the antinomy is that one “cannot but regard [grace] as only conditional, that is, consider the improvement of his life conduct, as much as lies in his power, as having to come first, before he gives even the least credit to the hope that the favor from on high will redound to his good” (6:117). Grace must be conditional because grace is a benefit, which cannot rightly accrue to someone worthy only of punishment. The second prong of the antinomy is that because of radical evil, it seems impossible to first turn to the good and only later receive grace, because one’s will itself is corrupt. Faith in a merit which is not his own, but through which he is reconciled with God, would therefore have to precede any striving for good works, and this contradicts the previous proposition. (6:117)

Kant insists that the antinomy cannot be resolved at a theoretical level. This conflict cannot be mediated through insight into the causal determination of the freedom of a human being, i.e. into the causes that make a human being become good or bad: in other words, it cannot be resolved theoretically, for this question totally surpasses the speculative capacity of our reason. (6:117–18)

From a practical standpoint, there is an answer to the question of priority. When the question is not whether one’s action or God’s cooperation actually comes first, but which one should effect first, the answer is easy. Practically, however, where the question is . . . whence . . . we are to make our start, whether from faith in what God has done for our sake, or from what we ought to do in order to become worthy of it (whatever this may be), there is no hesitation in deciding for the second alternative. (6:117–18)

The second prong of the antinomy, that one must become worthy of grace before one can receive it, is true in a practical sense. Whether or not there is an actual priority (temporal or otherwise) of merit over grace, when deciding what to do, one must think not about first getting grace but rather first becoming worthy of it.

Radical Evil and Moral Anthropology

121

Again, “we can admit an effect of grace as something incomprehensible, but cannot incorporate it . . . for either theoretical or practical use” (6:53). But as in the case of atoning grace, Kant does not specify what alternative there is to theoretical or practical uses for the concept. And just as Kant could not actually provide a philosophical justification of freedom without the second Critique, despite the need of theoretical reason for some unconditioned, he cannot provide a philosophical justification of grace without some further Critique that gives a sense to the “grace” that is supposed to meet this need of practical reason.40 More troubling is that the arguments that Kant gives for this concept seem to imply that it should have some practical use, but Kant cannot allow such use without compromising his doctrines of radical evil and moral rigorism. In any event, a fully satisfying Kantian moral theory would have to either show how moral hope, radical evil, and/or rigorism can be given up without compromising Kant’s moral philosophy or provide some framework within which the need of practical reason to unify these commitments could be met. I will not attempt either of these tasks here. Kant’s account of grace fills an important need in his practical philosophy. There are several problems with his account of grace as it stands. First, Kant does not seem justified in saying that there is need of practical reason for grace in particular. Something is needed if moral hope, radical evil, and rigorism are to be maintained, but it is not clear that “grace” is the only option. Second, Kant’s description of grace does not show how it is relevantly different from the “lenient justice” he seeks to avoid. Third, Kant does not adequately show that grace is consistent with practical and theoretical reason. Especially with respect to sanctifying grace, there is an apparent contradiction with practical reason that is bypassed rather than resolved. Fourth, Kant gives insufficient details about the perspective within which grace would be justified. He provides a subjective need, but not a real justification, for the concept of grace. Finally, the details that Kant does provide raise problems of their own. Dealing with all of these problems is beyond the scope of this book.41 What is important for our purposes is the role that grace plays in Kant’s practical philosophy. In section 3, I have discussed Kant’s arguments for radical evil. Radical evil poses serious problems for giving an account of a good human will. But Kant claims that grace, by reconciling moral hope, radical evil, and rigorism, makes room for a conception of the human good will that is good but not identical to a will that would express itself in a life of perfect conformity to the moral law. Moreover, it does this without weakening the force of the moral law itself and without admitting degrees

122

The Solution

of goodness. In the next section, we turn to Kant’s detailed description of this new conception of the human good will.

4. Integrating Moral Anthropology: A New Conception of the Human Good Will Solving the problem of radical evil presents for Kant the challenge of articulating a new conception of what a human good will looks like in the world. Without awareness of the radical evil in human nature, one might think that a good will would express itself in a will that always chooses in accordance with the categorical imperative. The presence of radical evil makes such a will an impossibility for human beings because one can never completely escape the stain of past misdeeds and the enduring presence of a propensity to evil. But the possibility of atoning and sanctifying grace frees Kant to develop a new account of the good will. It must be possible to act and choose so that one is a proper recipient of grace, such that one’s free will can be considered good. There must be a sort of good will that is consistent with the reality of radical evil. Kant articulates this conception of a good will as a “new man,” a will that engages in a “revolution” against its radical evil. As Kant explains, That a human being should become not merely legally good, but morally good (pleasing to God), i.e. virtuous according to the intelligible character . . . – that, so long as the foundation of the maxims of the human being remains impure, cannot be effected through gradual reform but must rather be effected through a revolution in the disposition of the human being (a transition to the maxim of holiness of disposition). And so a “new man” can come about only through a kind of rebirth, as it were a new creation . . ., and a change of heart. (6:47)

Despite the Biblical language, Kant’s point is philosophical. Because of the nature of evil, a change in moral status cannot be gradual but must be an absolute transformation from good to evil. Evil is the willingness to subordinate the moral law to sensuous inclinations, even only occasionally. One does not become less evil by subordinating the moral law less often, or to stronger inclinations only. Only a complete shift, such that the moral law assumes absolute priority, constitutes genuine moral improvement on Kant’s account. Thus until and unless one makes this change, one remains evil. And this change cannot be made on the basis of the radical evil that one, as human, always already adopts. It instead depends on a revolution against one’s evil maxims and toward the good, a revolution made possible, though still not comprehensible, through grace.

Radical Evil and Moral Anthropology

123

As Kant makes clear in his discussion of the revolution, this description applies only to a person as free. He even uses the term virtus noumenon to describe the virtue that this revolution reflects (6:47). This discussion raises some important problems for Kant’s claim that one cannot have any knowledge of the noumenal realm, as well as for his insistence in the first Critique that noumena cannot be understood temporally. Kant does not directly address either of these problems, but one can make the most sense of the account by understanding the term revolution to refer not to something temporal but simply to a good will. Thus we can refer to this new man as a will “in revolution” to capture the fact that this revolution is not a temporal act but a noumenal reality.42 The will in revolution is a will that is fundamentally good, but good as a reaction against its own radical evil, rather than as an absence of any evil. This good will can be described by the term revolution to capture two important points. First, the will in revolution is a good will that is consistent with radical evil expressed in evil deeds and tendencies to perform such deeds. Second, this will is not temporal in the sense that it does not in itself involve gradual reform. This latter point is the main emphasis in Kant’s discussion of the revolution. He opposes any notion of gradualism which might suggest that one can be “partly good,” because goodness must be an absolute commitment to the moral law. Thus it is consistency with radical evil and resistance to gradualism that Kant captures with his notion of “revolution.” Rather than introduce a temporal notion, the whole point of the term revolution is to shift away from thinking of this change temporally. Whether there is a word that is better suited to such a nontemporal situation is less important than clarifying the necessity and possibility of the revolution and the nature of its appearance in the world. Instead of discussing the extent to which a “revolution” can be ascribed to a noumenal will, Kant focuses on the problem of explaining what this revolution would look like in appearance. This is important for reasons that have been discussed in section 2. Appearances are crucial for moral evaluation, because one evaluates agents on the basis of how their appearance measures up against a moral standard. To judge whether someone is good, one must have some sense of what the appearance of a good will is. More important, one makes moral decisions based upon effecting certain appearances. To understand what is required for one to receive grace by a noumenal revolution against evil, one must understand what such a revolution would look like in the world of appearances. Only then can it be understood how one can choose to pursue a pattern of

124

The Solution

behavior that corresponds to such an appearance of revolution. And it will turn out for Kant that this pattern of behavior includes pursuing helps and avoiding hindrances. Thus the problem of helps and hindrances is ultimately solved by Kant’s account of the appearance of the revolution that is necessary for a human will to be a good will. The first step in Kant’s articulation of the appearance of the good will is to apply the notion of atemporal revolution to the temporal series that constitutes the human being as appearance. In this context, Kant remarks, A revolution is necessary in the mode of thought but a gradual reformation in the mode of sense . . ., and [both] must therefore be possible also to a human being. That is: if by a single and unalterable decision a human being reverses the supreme ground of his maxims by which he was an evil human being (and thereby puts on a “new man”), he is to this extent, by principle and mode of thought,43 a subject receptive to the good; but he is a good human being only in incessant laboring and becoming; i.e. he can hope – in view of the purity of the principle which he has adopted as the supreme maxim of his power of choice, and in view of the stability of this principle – to find himself upon the good (though narrow) path of constant progress from bad to better. (6:47–8)

The “single and unalterable decision” is the revolution that makes one morally good at the level of one’s free self. But this decision must be expressed as progress in the appearance of that self. The revolution is directly relevant to morality, but it can never be known directly through experience. For God, who intuits one’s free will directly, “this [revolution] is the same as actually being a good person (pleasing to him)” (6:48). One can never know that one effects a revolution in one’s mode of thought; one can only perceive signs of that revolution through its expression in moral progress. As Kant explains, For the judgment of human beings . . . who can assess themselves and the strength of their maxims only by the upper hand they gain over the senses in time, the change is to be regarded only as . . . a gradual reformation of the propensity to evil. (6:48)

Just as one cannot know that one’s fundamental maxim is good, one cannot know that one is in revolution against evil. But just as one can reasonably hope for and pursue goodness, one can reasonably hope for and pursue a moral revolution. And the basis for hope, as well as the immediate object of one’s pursuit, will be the expression of moral revolution.

Radical Evil and Moral Anthropology

125

In that context, Kant considers several possibilities for the temporal expression of a nontemporal free agent undergoing a “revolution.” The first crucial requirement for a series of appearances to express a will in revolution is that the entire series express the free agent. In that context, Kant insists that the appearance that is relevant in evaluating whether or not one is good in the sense of revolting against evil be one’s entire life. Kant says, Reason, when it is a question of the law of our intelligible existence (the moral law), recognizes no distinction of time and asks only whether the event belongs to me as a deed and, if it does, then always connects the same feeling with it morally, whether it was done just now or long ago. (5:99)

When evaluating whether or not one truly revolts against evil, one must not simply look for some particular moment or phase of life but at one’s life as a whole. In this context, Kant considers several possibilities for how a good life would look. The most obvious option – a life free from any evil – is ruled out already by the presence of radical evil. Kant’s problem at this point is to consider what the most plausible remaining option is. Kant argues that a will in revolution must express itself in a complete life by contrasting morality with happiness. With respect to happiness, Griefs once endured, when we feel safe from them, leave no painful reminiscences behind but rather a feeling of gladness that makes the enjoyment of the supervening good fortune all the sweeter. For pleasure and pain (since they belong to the senses) are both included in the temporal series, and disappear with it; they do not constitute a totality with the present enjoyment of life but are rather displaced by it as it succeeds them. (6:70)

Insofar as one experiences proper pleasures and no pains at any moment, one can be described as happy, whatever pains one experienced in the past. In contrast, however, moral life is a nontemporal unity. The moral subjective principle of the disposition by which our life is to be judged is (as transcending senses) not of the kind that its existence can be thought as divisible into temporal segments but rather only as an absolute unity. (6:70)

Kant goes on to discuss, and to reject, one particular account of how a moral revolution might appear in the world. He sums this option up with the slogan “all’s well that ends well.” According to this view, what matters in life is the moral condition at which one arrives at the end of one’s life. Regardless of how wicked someone may be in youth, acting from duty alone in the minutes before

126

The Solution

death shows that such a person is a morally good person, that is, one who has undergone the moral revolution. In discussing this view, Kant focuses on the danger that it poses for morality: Give [a human being] encouragement (as with the proverb “All is well that ends well”) and from early on he will make his plans accordingly, with a view not to forfeit too much of life’s pleasures unnecessarily and, by life’s end, to settle accounts with speed and to his advantage. (6: 77–8, cf. 6:70)

The thought that one might be morally evil brings with it “moral sufferings, the reproaches of one’s conscience” (6:78 fn.), not to mention fears of suffering after death. Still, human beings have wants and needs that tempt them to violate the moral law. If one comes to believe that one can be morally good, can be spared from the sufferings (moral and otherwise) associated with having an evil will, and at the same time can spend all but the last few seconds of one’s life pursuing sensuous inclinations in disregard of morality, one seems able to have one’s cake and eat it too. One can actually be morally good while acting for most of one’s life in a way that disregards morality. For Kant, of course, such a possibility is a moral disaster. A view of moral goodness that justifies evil choices erodes the foundation of morality. It is no longer morally required, apparently, that every agent choose well. As long as an agent has a couple of well-timed good choices, the force of all other moral obligations erodes. And this can lead to moral complacency, which the doctrine of grace and the new conception of a good will are supposed to avoid. If it is impossible to be morally good, one can hardly make a serious effort at it – this is the problem of radical evil. But if it is too easy to be morally good, one will be almost as likely to ignore the moral law. When Kant contrasts the application of “all’s well that ends well” to happiness with its application to morality, he highlights this problem. Not only does this account fail to take into account one’s whole life, but it also has dangerous practical consequences for moral life. Still, Kant does not completely reject the possibility of conversion at the end of one’s life. He claims, “the common saying, ‘all’s well that ends well,’ can indeed be applied to moral cases, but only if by the ‘good ending’ we understand that a human being becomes a genuinely good human being” (6:70). This view fits well with his description of the sensible appearance of one’s noumenal revolution as “progress.” One measures progress, at least in part, by the end point at which one arrives. However, Kant is quick to point out a problem with resting one’s entire self-evaluation on a lastminute conversion. If one “becomes a genuinely good human being” at

Radical Evil and Moral Anthropology

127

the end of life, things will have turned out well. But, Kant asks rhetorically, “where is he to recognize himself to be such [a good human being], since he can draw this conclusion only from the constancy of his consequent good conduct, and, at the end of life, there is no time left for this?” (6:70). The inscrutability of one’s true disposition undermines shortcut attempts at moral reform. Because one has no direct access to the fundamental maxims of one’s actions, one can judge one’s disposition only on the basis of consistent struggle against evil in intentions and actions. If even the best appearances can be misleading, apparent dispositions at the end of life, without any corresponding consistency over time, are particularly suspect. And this suspicion will be especially pronounced when one has lived a life of wickedness. Of course, one can never rule out completely the possibility that a conversion at the end of life is genuine. Kant even remarks that it is particularly important to be careful how one interacts with those on their deathbeds. There is always hope for true reform, hope that can be undermined by bad clergy offering false comforts. It is important to hold out the need for genuine moral reform, for actually doing what is right and correcting past wrongs, even to those near death (see 6:78 fn). There is another sort of moral conversion that Kant does not discuss separately but that might seem more plausible as an account of the appearance of a good will. Rather than a conversion at the end of life, or a conversion in disposition only, one might imagine that a good will would appear as a person who, though wicked for an early portion of his or her life, at some point changes radically and from then on makes all choices in accordance with the moral law. Such a person would not base hope merely on an impression of a different resolution but would have a long record of deeds. On this account, one would always be justified in seeking to be good, because one can effect at any time the radical transformation that characterizes the good will. One could counteract any danger of postponing that transformation on the grounds that only when one turns to the good with enough time to exhibit that goodness as a stable aspect of one’s character does one truly express a will in revolution. This provision would eliminate many of the practical problems with deathbed conversions. At times, Kant seems to endorse a position similar to this one, although he ultimately rejects it in favor of an alternative account. His endorsement comes in the form of an emphasis on temporal conversion, on an establishment of good character at a certain time. In the Anthropology, Kant describes the establishment of character as a revolution. Because character

128

The Solution

is a precondition of having a will that is thoroughly good, this description seems to have some relation to the revolution described in the Religion.44 At times, Kant even uses similar language to describe the two revolutions. Thus in the Anthropology he says, The establishment of character is similar to a kind of rebirth, a certain solemn resolution which the person himself makes. . . . This stability and persistence in principles can generally not be effected . . . by degrees, but it can only be done by an explosion which suddenly occurs. . . . Perhaps there will be only a few who have attempted this revolution. . . . (7:294)

As in the Religion, the revolution that establishes character is distinguished from anything that happens gradually. The language of “rebirth” fits well with Kant’s description of the moral revolution as giving rise to a “new man.” But the revolution described here is very clearly a revolution that occurs at a particular time. Kant says, “this resolution and the moment [Zeitpunkt] at which the transformation took place remain unforgettable” (7:294, emphasis added). And his claim that few attempt this revolution continues with the following qualification: “Perhaps there will be few who have attempted this revolution before their thirtieth year, and fewer still who have firmly established it before their fortieth year” (7:294).45 The temporal nature of this revolution is sufficient to show that it is not identical to the moral revolution described in the Religion. Kant further distinguishes them when he explicitly describes character in the Anthropology as something of which one is “conscious” (7:294), whereas the revolution in the Religion is intelligible (6:48) and hence not an object of possible experience. But even if this revolution is not identical to the intelligible revolution in the Religion, it is still a revolution. And one might think that a revolution at some particular moment would be a good candidate for the temporal appearance of a moral revolution. Kant rejects this option. But while Kant makes very clear that the good will must appear as progress rather than revolution, he is less clear about why this is so. One obvious reason is that one’s entire life must express one’s revolution, but one might still think that the early parts of one’s life could represent the evil against which one revolts, and hence they are part of the revolt itself. To fully understand why every revolution must express itself as progress rather than as sudden conversion, one must remember the nature of the evil against which one revolts. If radical evil consisted only of acting on evil maxims, the revolution against it might consist simply of changing the maxims on which one

Radical Evil and Moral Anthropology

129

acts. But radical evil also involves a propensity to evil according to which one undermines one’s future choices, establishing a tendency to pursue further evil. Kant takes this aspect of radical evil to be an empirical fact. One recognizes not only one’s past evil deeds but also the present and future effects of past self-manipulation. Kant’s new conception of the good will must take into account the reality of this propensity to persist in evil. And this means that one has reason to predict, though never excuse, that one will do evil in the future. Whatever conversion one undergoes, one still suffers the effects of one’s past evil choices, including effects on one’s future behavior. One is fully responsible for these effects, because they reflect one’s free choice of evil rather than of good. But even if one commits to moral principles now, this commitment can only begin to undo the effects of one’s evil actions in the past. The temporal extension of past decisions does not undermine transcendental freedom, because the limitations of one’s ability to effect changes in oneself are self-imposed limitations. But the temporal extension of past decisions does affect the sort of expression that is possible for a will in revolution. It can at most appear as a constant struggle against evil.46 How then does an agent in revolution against evil appear in the world? This question can be applied in the context of either moral evaluation or deliberation. With respect to evaluation, although Kant does not think that one can ever be certain that one truly effects a revolution against evil, he offers some suggestions about the way to think of one’s moral status. He begins by asking, “What can a human being expect at the end of his life, or what can he fear, in virtue of his conduct during it?” In response to this question, Kant allows surprisingly large room for self-knowledge: A human being must first of all have cognition of his own character, at least to some extent. Thus, though he may believe that there has been an improvement in his disposition, he must be equally able to take the old (corrupted) one into consideration, the one from which he started, and examine what and how much of that disposition he has cast off, as well as the quality (whether pure or still impure) and the grade of the supposed new disposition for overcoming the old one and preventing relapse into it; he will thus have to look at his disposition throughout his whole life. (6:76)

Kant still insists that one has no direct access to one’s disposition. The barriers to self-knowledge have not been forgotten. Thus, one “must extract his disposition from the deed before him” (6:77). So what can one say about one’s moral status? First, one must consider one’s “whole life” (6:76), looking for progress throughout that life. Evils of one’s youth are

130

The Solution

important markers of where progress must occur, and there must be a consistent struggle against evils to which one is particularly prone along with a steadily building victory in that struggle. Even at the end of life, one may still in some respects be evil (that is what immortality is for, after all), but there must be a track record of constant and successful struggle. A sense of moral goodness comes only from “the assurance of the reality and constancy of a disposition that always advances in goodness” (6: 67, emphasis added). This self-evaluation must be based on actions, not on mere feelings or any supposed sense of oneself. One can never reliably know that one’s fundamental maxims are good, but unless one avoids those actions that cannot be done from duty and pursues those most congruent with it, one can be sure that one is evil. Only good actions and the constant struggle against one’s propensity to evil provide hope that one has a good disposition. This is as much as Kant says with respect to moral evaluation. In fact, however, Kant is more concerned with deliberation than with moral evaluation. In his objections to deathbed conversions, Kant focuses on the effects of various accounts of moral worth on moral deliberation. From the standpoint of deliberation, the question is not whether or not one is good but what one must do to be good. And it is from this standpoint that the practical role of helps and hindrances becomes prominent. The first and most important thing any agent must do to be morally good is obey the categorical imperative out of respect for the moral law. Kant points out that one “cannot allow the previously recognized disposition to take the place of the deed but . . . must extract his disposition from the deed before him” (6:77). In the context of evaluation, the “deed before him” refers to the sum of all the deeds of one’s life. From a deliberative point of view, however, the deed before one is the deed that one is to perform. Every moment of choice provides a new opportunity to express a will in revolution against evil. The bare minimum for expressing such a will in a given choice is that the moral law function as the governing maxim of one’s free will. Kant points out that “if [virtue] is not rising, it is unavoidably sinking” (6:409). Any missed opportunity to act on the basis of the moral law implies that one is sinking, that the progress which would reflect a good will is lacking. One who does evil “just this once” on the grounds that he or she can always show progress later errs seriously in moral deliberation. For although one always can choose rightly in the future and thereby show progress, what one actually will choose depends on whether or not one’s will is in revolution against evil. And the most

Radical Evil and Moral Anthropology

131

reliable way to express this revolution is through the action immediately before one. The only legitimate basis for the hope that one will choose well in the future is that one chooses well in the present. By choosing wickedness now, one leaves little hope that one will ever do any better. And this, of course, means that the primary responsibility of a moral agent plagued with radical evil but enabled through grace to be good is to actually be good, right now, with respect to each choice that presents itself. If that were the only responsibility of moral agents, Kant’s account of radical evil, although important for theology and interesting from the standpoint of practical reason, would not have any important implications for morality itself. For Kant, however, the fact of radical evil has important implications for what human beings actually need to do to express a good will. Kant explains, in response to the Stoics, that Those valiant men [the Stoics] mistook their enemy, who is not to be sought in the natural inclinations, which merely lack discipline and openly display themselves unconcealed to everyone’s consciousness, but is rather as it were an invisible enemy, one who hides behind reason and hence all the more dangerous. They send forth wisdom against folly, which lets itself be deceived by inclinations merely because of carelessness, instead of summoning it against the malice (of the human heart) which secretly undermines the disposition with soul-corrupting principles. (6:57)

In his footnote to this passage, he says even more explicitly that for the Stoics “everything was quite correctly apportioned . . . provided that one attributes to the human being an uncorrupted will” (6:58, emphasis added). For the Stoics, morality demands only that one prefer the moral law to sensuous incentives. And this is just what one might expect Kant to demand, because his initial description of the good will is just this kind of prioritization of predispositions. But Kant here insists that this is not sufficient, given the fact of radical evil. As he explains, To become a morally good human being it is not enough simply to let the germ of good which lies in our species develop unhindered; there is in us an active and opposing cause of evil which is also to be combated. (6:57) We cannot start out in the ethical training of our connatural moral predisposition to the good with an innocence which is natural to us but must rather begin from the presupposition of a depravity in our power of choice in adopting maxims contrary to the original ethical predisposition; and since the propensity to this [depravity] is inextirpable [in time], with unremitting counteraction against it. (6:51)

132

The Solution

What ultimately determines the status of one’s will is one’s life as a whole. And although the best indication of future progress is choosing rightly in the present, one can never ignore the presence, in oneself, of a propensity to choose evil. Thus at the same time that one must first and foremost always choose good, one must also actively combat this evil in oneself.47 This is especially important because a human being who chooses evil does not merely choose to do evil at a particular time but actively undermines the likelihood of choosing good with “soul-corrupting principles.” To effectively combat this evil, a good person must respond not merely with good choices but also with soul-correcting principles, with actions that serve to undermine the influence of evil and increase the strength of the moral law in one’s future deliberations. In the same way that evil human beings not only make evil choices but effect through their maxims a propensity to further evil, a human being who effects a revolution against evil must not only make good choices but also undermine this propensity to evil and develop the strength of will to continually pursue the good. As a result, Kant explains what one can do to effect progress in the purity and stability of one’s empirical commitment to the moral law. To make sense of this concern, it is necessary again to point out the relationship between the free will, the appearance of that will, and empirical influences on that appearance. From a metaphysical standpoint, the free will is a noumenal ground of the appearances; and these appearances, as objects of experience, are also capable of empirical influence. From a practical standpoint, one considers the free will as grounding the appearances not as an abstract truth but as the realized condition of deliberation. One’s commitment to oneself as free ground is manifested as one holds oneself accountable to the moral law in deliberation. Thus the crucial question is not the metaphysical possibility of a free self, but what that free self ought to do, what it ought to realize in the appearances of which it is the ground. But there is also a perspective of moral anthropology. The need for considering a moral–anthropological perspective arises from the recognition that one has a propensity to misuse one’s freedom. This leads, as we have seen, to a more complicated conception of the appearance in time of the good will. Whereas one might have thought that one need only do good deeds now to express a good will, one finds that one must act in such a way that one will combat the propensity to evil and promote good deeds in the future. One’s deliberation is always about what to do now, but moral anthropology provides a means for extending the effects of deliberation

Radical Evil and Moral Anthropology

133

beyond the present. Moral anthropology, then, studies the way in which natural causes lead to human actions that express most clearly, though never indubitably, moral goodness. This allows an agent to pursue in the present those causes which lead to actions that express a good will. And because the struggle against evil, expressed in consistent moral improvement, is the appearance that one must always seek to realize, one ought always to pursue these empirical causes. Moreover, Kant explains that ignorance of one’s moral status intensifies the importance of helps and hindrances. Given that one can never be sure whether one is making enough progress and that one can never be sure that one will continue to do good in the future, one must always seek to influence a future self that may return to evil. One must remain, as Kant says, “forever armed for battle” (6:93). In the cool hours in which one acts from good maxims, one must seek to increase the number and effectiveness of future cool hours, to combat the deceptions and manipulations of one’s own evil will, and to promote any empirical influences likely to result in further improvement of one’s actions and observable intentions. None of this will grant knowledge that one has a good will; nothing can. But the greater the degree of resistance to evil, both present and future, the more justified one’s hope that one is good. This resistance to evil, present and future, is in any event what the moral law requires of human beings, wicked as we are. And the resistance to future evil involves attention to empirical helps and hindrances. This ignorance also implies that one cannot simply compare one’s life with the lives of others. What is required to express a moral revolution may differ depending on one’s empirical conditions. Of course, one must always strive to act in accordance with the moral law and to promote aids to morality. But given the fact of radical evil, one will fall short of always acting in accordance with the moral law, and one will often fail to promote aids to morality. And one can never know for sure whether one has gone far enough. Thus any person committed to moral revolution ought to express that moral revolution at every opportunity. But when it comes to evaluating oneself or others, it is reasonable to have different levels of hope for the same behavior, depending on external conditions. One who has abundant advantages in terms of natural disposition, moral education, political stability, and a community that supports ethical behavior cannot reliably express a moral revolution merely by resisting some of the temptations to vice that occasionally arise. Given those advantages, such an individual can show his or her struggle against evil only by consistent obedience to the moral law and by promoting more subtle aids

134

The Solution

to future moral improvement. By contrast, one with a disadvantageous temperament who lacks moral education or the support of an ethical community might express moral revolution in the increasing effort to resist temptation and to promote at least a minimal respect for virtue in him- or herself and those by whom he or she is influenced. Neither one with advantages nor one without them is justified in moral complacency or despair. But the ignorance of what is needed to express a moral revolution does mean that one is not warranted in ascribing to external conditions a causal influence on one’s free choices.48 Finally, it is important to note that moral anthropology has limits. It is limited by the constraints of the empirical world, especially laws of human psychology. One can affect one’s future behavior, but only in ways that are consistent with the kinds of influence to which human beings are found to be susceptible. Moral anthropology is also limited by moral relevance, in the sense that only influences that are morally relevant are considered part of moral anthropology. One can influence one’s behavior in ways that are irrelevant to morality; Kant’s pragmatic anthropology includes a lot of advice that is primarily helpful for increasing future happiness. But only empirical influences on decisions that are morally relevant are included within the realm of moral anthropology. Finally, the range of moral anthropology is limited by transcendental freedom. One can affect future choices, where a “choice” is something empirical.49 But this “choice” is always only the expression of one’s free will, not the free will itself. Acting now can affect future choices and can express a revolution in the free will, but nothing empirical can affect the free will itself.

5. Conclusion Because one’s actions in the world are susceptible to empirical influence, and because one is prone to evil, one can and should study those influences that aid one’s good maxims in getting the upper hand over one’s evil maxims. This study, because it is undertaken with a practical goal (the overcoming of evil in oneself) and because it is a study of a person who is free (though not a study of that person insofar as he or she is free), is rightly considered a study of “what the human being makes, can, or should make of himself as a freely acting being” (7:119). At the same time, because it is a study of appearances and of the causal relations between appearances, it is a properly empirical science and can yield empirical

Radical Evil and Moral Anthropology

135

knowledge. This discipline, which teaches how one can effect the progress in appearance from evil to good, is moral anthropology. Kant’s anthropological observations about empirical helps and hindrances to the development of the good will do not undermine the priority of freedom over nature because the empirical helps and hindrances affect only the appearances of the good will in the world. And this, of course, is exactly what Kant means when he writes, “the resistance or furtherance is . . . between nature as appearance and the effects of freedom as appearances in the world of sense” (5:196). But Kant’s anthropology is also morally relevant because it studies appearances that express one’s moral status. The empirical influences described by moral anthropology are important parts of those expressions because of a propensity to evil that human beings freely adopt. This propensity affects one’s future choices because it is a propensity to choose evil in certain circumstances. The need for empirical aids to combat radical evil does not mark an exception to Kant’s principle of “ought implies can,” because this need is due only to choices freely made. The will itself is corrupt and thus needs help, but it is freely corrupt. But one will not simply overcome evil through isolated choices. Rather, one must effect changes in the empirical influences on one’s character. Effecting these changes expresses one’s moral revolution, and knowledge of them comes through moral anthropology. Moral anthropology thus provides human agents empirically informed advice about how best to express themselves morally, and this advice becomes morally relevant if such agents freely choose to act on it. One central problem remains with this account of moral anthropology. The explanation of anthropological influences as means for expressing one’s revolution against evil works well for those influences that one effects in oneself. But the account does not seem to work as well in those cases where one agent promotes helps or combats hindrances for another. This limitation seems particularly problematic given Kant’s claims about moral progress in human history and through moral education. The next chapter takes up this remaining problem.

6 Moral Influence on Others

In the previous chapter, I offered a Kantian account of the nature of the good will and its appearance in the world that makes sense of the moral significance of empirical helps and hindrances. The account offered there explains why promoting helps and combating hindrances to moral progress expresses a moral revolution, the only form of good will that remains possible for one who is radically evil. But many helps and hindrances do not seem to fit this model. When one struggles against passions because one recognizes that they are moral hindrances or when one enters into polite society because one knows that doing so can improve one’s own moral disposition, one expresses a will in revolution by struggling against the radical evil in one’s nature. But what is going on when a parent disciplines a child, or a teacher instructs a pupil morally, or when one is polite to present virtue in the most attractive light to others? In these cases one does not seem to struggle against one’s own evil, so one seems not to express a will in revolution. And the child, or pupil, or those to whom one is polite are not necessarily struggling against evil themselves, so they may not be expressing a will in revolution. In these cases, the account in the previous chapter – if it can be relevant at all – seems to grossly misplace the moral significance of helps and hindrances. When one promotes aids and resists hindrances for oneself, one’s activity is morally significant for the one who is helped and hindered. In the case of interpersonal interaction, the account in the previous chapter seems to make helps and hindrances to another morally significant for the helper or hinderer rather than for the one helped or hindered. That just seems to get the moral importance of helping and hindering all wrong. 136

Moral Influence on Others

137

One might think that the tension between freedom and nature is not as serious in the interpersonal case as in the individual case because one need not consider an agent that one seeks to improve from both the empirical and the practical perspectives at the same time. In one’s own case, one must consider oneself both from the practical perspective because one is making a choice and as susceptible to empirical influence because one is choosing an empirical influence on oneself. But in a case where one person deliberates about morally helping another, one might think that only the deliberating person needs to be considered free while the person helped is considered capable of empirical influence. Were that true, there would be no conflict in the case of interpersonal moral influence, because no person would need to be considered from both the empirical and the practical perspective at the same time. But freedom is threatened in the interpersonal case. The notion that one can consider oneself free while viewing the person to be helped as an appearance might be acceptable when one seeks merely to improve another’s physical or even intellectual capacities. But it is inadequate for cases in which one seeks to affect another for the other’s moral improvement. When one acts on a maxim like “I will do some action in order to promote the good will of my pupil,” either one holds that the action will promote the good will of the pupil or that it will not. If one denies that the action can have any effect on the will of the pupil, then one acts irrationally in following the maxim, because it violates the norms of technical reasoning. According to these norms, one should not perform an action for an end when one believes that the action has no effect on bringing about the end. But all of one’s actions are, from the standpoint of the other, empirical influences. Thus if one holds that the action can affect the will of the other, one grants that empirical influences affect the free will of another. This is the problem of interpersonal moral influence. In this chapter, I apply the model developed in the previous chapter to interpersonal moral influence. In section 1, I point out a seeming conflict between Kant’s account of virtue in the Metaphysics of Morals and his account of ethical community in the Religion, and I articulate a model of interpersonal influence in Kant that resolves this conflict. In section 2, I address the possible objection that my account of interpersonal influence ends up being self-centered. In section 3, I offer an interpretation of Kant’s writings on history in the light of my account of interpersonal moral influence.

138

The Solution

1. Kant on Interpersonal Moral Influence In the Metaphysics of Morals, Kant presents an explanation of our duties with respect to others which suggests that interpersonal moral influence is impossible. In a section entitled “What Are the Ends That Are Also Duties?” Kant lists only two ends that are duties, “[1] one’s own perfection and [2] the happiness of others” (6:385). The absence of “the perfection of others” or “the moral improvement of others” or something of the sort is noteworthy,1 but more important is the argument that Kant offers for why the perfection of another cannot be an end: It is a contradiction for me to make another’s perfection my end and consider myself under obligation to promote this. For the perfection of another human being, a person, consists just in this: that he himself is able to set his end in accordance with his own concepts of duty; and it is self-contradictory to require that I do (make it my duty to do) something that only the other can do himself. (6:386)

We ought not, because we cannot, do anything to morally improve another, because moral improvement is precisely something that one must do for oneself. The view presented in the Metaphysics of Morals seems to be quite clear, and it is consistent with Kant’s emphasis on freedom as well as with the account of helps and hindrances presented in the previous chapter. If one cannot influence others morally, then one can properly understand every case of empirical influence as an expression of one’s own struggle against one’s own radical evil, as described in the previous chapter. Interpersonal moral influence could be dealt with just by denying that it is possible. Such an approach suffers from one obvious disadvantage in that it seems implausible that people cannot have any morally relevant impact on one another. But more important, it seems to conflict with claims that Kant makes elsewhere. In the Religion, as well as in his writings and lectures on history and education,2 Kant suggests that interpersonal moral influence can play an important role in moral life, for either good or ill. Interpersonal influence can influence one negatively, because the main “assaults of the evil principle” for control of one’s will come “from the human beings to whom he stands in relation or association” (6:93). But others also play an important role in promoting morality, and one seems to have a responsibility to promote virtue in others because “the dominion of the good principle is not otherwise attainable . . . than through the setting up and the diffusion of a society in accordance with, and for the sake of, laws of virtue – a society which reason makes it a task and a duty . . . to establish” (6:94).

Moral Influence on Others

139

But how can one have a society for the sake of virtue if people are incapable of affecting one another morally? And how can one have a duty to establish such a society if the Metaphysics of Morals prohibits seeking to make the moral perfection of others an end for oneself ? In fact, by making clear the distinction between influencing the deeds of others and influencing their moral status itself, one can see several important roles for interpersonal “moral” influence that fit the strictures of the Metaphysics of Morals and the account of the good will offered in Chapter 5. Good deeds are the empirical expression of a good will. As empirical, these deeds are capable of influence. The good will that they express, however, is not. In light of this distinction, I present three roles for interpersonal influence with different degrees of moral significance. In none of these roles do one’s actions actually determine or even clearly affect the moral status of another. In that sense, they are all compatible with Kant’s claims in the Metaphysics of Morals about the limits of interpersonal influence. But in all cases one has a duty to promote good deeds in others. Two reasons for promoting good deeds are taken from the Metaphysics of Morals itself. The third is based on the discussion of ethical community in the Religion. In the context of discussing that role for interpersonal influence, I show how the promotion of good deeds in others is an important part of one’s own struggle against evil. A. Promoting Virtue to Promote the Happiness of Self-Esteem One reason to promote moral behavior in others comes from our imperfect duty to promote the happiness of others. In the Metaphysics of Morals, shortly after pointing out that one ought to have as an end the happiness of others but not their perfection, Kant adds, The happiness of others also includes their moral well-being, and we have a duty, but only a negative one, to promote this. . . . It is my duty to refrain from doing anything that, considering the nature of a human being, could tempt him to do something for which his conscience could afterwards pain him. (6:394)

Even if one cannot promote moral perfection in another, one can tempt others to actions that express an evil will. Kant’s description of the limitations on self-knowledge makes clear that although one can never know that one has a good will, certain acts and intentions are clear signs of evil. At least some of these actions, unless one is remarkably adept at selfdeception, will pain one’s conscience. This self-knowledge has beneficial effects, in that it gives rise to the recognition of radical evil and ushers

140

The Solution

in Kant’s new conception of what it means to have a good will. But the more direct effect of recognizing one’s evil is displeasure. Doing wicked things tends to make people unhappy, all else being equal. All people have a seed of goodness that cannot be extinguished, and that seed includes a conscience that causes pain at the recognition of moral evil in oneself. Thus the obligation to promote the happiness of others includes an obligation to refrain from tempting them to evil deeds for which they will afterward suffer. The promotion of happiness may include even the obligation to help others choose to perform more good deeds for which they will feel the pleasure of self-respect. It is important to point out that this reason for promoting good deeds does not imply that one helps others to pursue virtue for the sake of happiness. A father who helps his children do more good deeds out of a duty to promote their happiness ought to foster morally upright behavior at least in part for the sake of his children’s happiness, because he has a duty to promote their happiness. But his children ought to pursue morally upright behavior for its own sake.3 Insofar as his children perform good deeds for the happiness they will derive from such deeds, they fail to exhibit moral virtue, and thus they lose the proper pleasures of selfesteem that would have come from doing good for its own sake. Insofar as a good father cares about the happiness of his children, he will promote in these children a tendency to act in ways which reflect that they care more about the moral law than about their own happiness.4 In this context, it is also important to note that the promotion of good deeds includes good intentions. Intentions, like external actions, are appearances, and thus they are capable of empirical influence. And they are appearances that express, but are not identical to, one’s moral status. Thus one feels moral self-condemnation not only for external actions that are contrary to the moral law but even for internal states. And one who seeks to promote the happiness of another will seek to influence not only the other’s visible actions but also the intentions, thoughts, desires, and feelings of which only the other is directly conscious. None of this implies that one affects the free will of the other, because these mental states are merely objects of inner sense. But this does imply that the “deeds” with which one is concerned include mental states as well as external actions.5 B. Promoting Virtue as Indirect Virtue When one helps another to act in morally virtuous ways, one promotes not only the happiness of the one who is morally helped but also the happiness

Moral Influence on Others

141

of third parties. All people have a duty to promote the happiness of others. By encouraging those on whom one has influence to promote the happiness of others, one indirectly promotes that happiness oneself. But if one encourages someone to promote the happiness of others in ways that conflict with the moral law, then one’s encouragement itself violates the moral law. Thus one must encourage others to promote the happiness of others in ways that are morally right. And that is to say, one must encourage virtue. To see this more clearly, it is important to see under what circumstances it would be wrong to promote the happiness of another in a personal case. Take this maxim: M1: When I see some person X who can benefit from money or property belonging to another person Y, I will falsely promise to repay in order to borrow the money or property from Y and give it X, in order to benefit X.

On the one hand, this maxim is consistent with the requirement of virtue that one make the happiness of others an end. But one who acts on such a maxim is not virtuous, because this maxim is contrary to the moral law for the same reasons as the more selfish example of false promising that Kant discusses in his Groundwork (see 4:422, 4:429–30). Specifically, although the maxim would probably be to the benefit of X and may not significantly affect Y in any material sense, the maxim and its universalization cannot be consistently willed, and one who acts on such a maxim makes use of Y merely as means of promoting the good of X. But now imagine a virtuous teacher who considers teaching her students to act on maxim M1, or perhaps a refined version of maxim M1 that makes explicit that the transfer from Y to X will increase overall happiness. She acts on this maxim: M2: In order to promote the happiness of others, I will teach my students to falsely promise in order to borrow from some and give to others, in such a way that overall happiness is increased.

Assuming that the students are smart enough to make such judgments, acting on this maxim is likely to increase the overall happiness of others. Thus it would be a suitable maxim for someone who takes the happiness of others as an end. But like one who acts on M1, a person who acts on M2 is not virtuous, because this maxim is not compatible with the moral law. Given that human beings are susceptible to instruction, universalizing M2 will cause the same sort of contradiction as universalizing M1. Moreover, M2 makes use of others as mere means just as much as M1, even if the

142

The Solution

others are not as well specified in each situation. Insofar as a teacher is truly virtuous, she will teach her students to promote the happiness of others, because this instruction is an effective and legitimate means of promoting that happiness. But a virtuous teacher will be careful to teach her students to promote the happiness of others in morally right ways, because to teach them otherwise would itself be morally wrong. But teaching students to promote the happiness of others in morally appropriate ways just is teaching them virtue. In this context, promoting true virtue in others is an indirect way of being virtuous oneself. Both this reason for promoting virtue and the previous one express one’s own virtue in the sense that they show one’s concern for the happiness of others. Because taking the proper happiness of others as an end is morally obligatory and promoting good deeds in others is a way of promoting the happiness of others – whether the happiness of those who do the deeds or the happiness of benefited parties – promoting good deeds is a means to an obligatory end. C. Promoting Virtue in Others as an Expression of a Social Struggle Against Evil In addition to serving the happiness of others, promoting virtue in others also serves to promote the other obligatory end of virtue: one’s own perfection. Unlike the straightforward ways in which the virtue of others promotes their own happiness and the happiness of third parties, the effect of the virtue of others on one’s own perfection is complicated. The virtue of others can nonproblematically promote nonmoral perfections in oneself. A colleague’s fairness, diligence, and cooperation can help one cultivate one’s intellectual talents by pointing out fruitful new directions of thought or beneficial techniques for study and reflection. Because these perfections are nonmoral, the virtue of others does not in any sense interfere with one’s freedom but merely enables one to articulate and pursue one’s ends more effectively. But in the Religion Kant indicates that the virtue of others can promote one’s moral perfection. In fact, this is precisely the reason that Kant says one should strive for a moral community. Kant begins the third part of the Religion with a recapitulation of the central point of the previous two parts: [A human being] remains . . . exposed to the assaults of the evil principle; and, to assert his freedom, which is constantly under attack, he must henceforth remain forever armed for battle. (6:93)

Moral Influence on Others

143

In Chapter 5, we saw that this struggle against evil is the way in which human beings express a moral revolution in their free will. But in the third part of the Religion, Kant goes on to offer a specific source of evil and corresponding arena for struggle. If [a human being] searches for the causes and the circumstances that draw him into this danger [i.e., assault of the evil principle] and keep him there, he can easily convince himself that they do not come his way from his own raw nature . . . but rather from the human beings to whom he stands in relation or association. (6:93)6

Kant insists that one “is in this perilous state through his own fault,” and in the previous chapter we saw that the propensity to evil is itself an outgrowth of one’s own evil choices. But other human beings are among the most important empirical means by which one fosters one’s propensity to evil. Thus because one must in general oppose the propensity to evil using empirical influences, one must oppose the corrupting influence of other people in one’s society using particular sorts of empirical influences. One can struggle against the corrupting influence of society only through the moral reform of that society. The dominion of the good principle is not otherwise attainable . . . than through the setting up and the diffusion of a society in accordance with, and for the sake of, laws of virtue. (6:94)

To engage in a revolution against evil, one must foster a purely ethical community that has as its purpose not the cultivation of vice but the promotion of virtue.7 The empirical expression of a will in revolution, given the social nature of evil, is a will that engages in a social struggle against evil. Social influence, of course, does not affect one’s free will. The interdependence of human beings is an empirical claim, and thus the interdependence is at the level of human appearances. Kant even says that this corrupting influence is possible despite people’s wills’ remaining good: Even with the good will of each individual, because of the lack of a principle that unites them, they deviate through their dissensions from the common goal of goodness, as though they were instruments of evil, and expose one another to the danger of falling once again under its dominion. (6:97, emphasis added)

Others do not cause one’s radical evil, and they cannot cause one to undergo a moral revolution, but they can affect the way that moral revolution appears. Insofar as others act wickedly, they encourage one to act wickedly. And even the mere presence of others around someone

144

The Solution

makes him that person “anxious that other human beings will consider him poor and will despise him for it” (6:93). People appear in the order of nature, and they act causally on one another in such a way that desires are fostered on the basis of which people act contrary to the moral law. This evil social influence has important implications from the standpoint of someone wishing to express his or her struggle against evil. Such a person, recognizing the important interdependence of human beings at the level of appearance, must promote good deeds among those with whom he or she comes into contact. Good deeds in others, even if these deeds are merely appearances and do not express genuine moral goodness, promote good deeds in oneself. More important given the social hindrances to morality, one who is morally good will engage in a struggle against social evils by fostering an ethical community that explicitly aims at the promotion of morality.8 Only if others act virtuously and actively promote virtue can one’s moral revolution be fully expressed in the world. And thus insofar as one truly revolts against evil, one seeks a community within which that revolution will be expressed. Even if one can neither make others morally good nor even promote in others a genuine interest in promoting morality, one can affect the way others appear such that they at least seem to be good and to encourage goodness. This appearance promotes one’s own good deeds. The ethical community, even as a mere appearance, is a help to morality. Thus promoting an ethical community expresses one’s revolution against evil. This promotion of an ethical community cannot be limited to those who are likely to affect oneself, nor even merely to the generation in which one lives. Kant suggests that the promotion of ethical community involves sharing in a goal of which one will personally experience only some of the effects. Thus it might seem unreasonable to limit the significance of an ethical community to the expression of one’s own revolution against evil. And of course this community, insofar as it emerges, is not an expression merely of any individual revolution but the combined effect of many revolutions, all expressing themselves in part in striving for ethical community. Even from the standpoint of the individual, however, there are two important reasons for promoting ethical community beyond the likely range of its direct influence on one’s own actions. First, one can never be sure about what aspects of an ethical community will have a moral effect on one. The toddler in whom one encourages virtue today might become

Moral Influence on Others

145

an important influence on one later. Second, one’s radical evil promotes effects that persist beyond one’s death. Insofar as one is radically evil, one makes use of society to promote and encourage one’s wickedness. This use of society contributes to general societal corruption. And even if this corruption cannot actually cause others to be morally evil, it promotes injustice, unhappiness, and vice. These lingering social evils are at least partially attributable to one’s past evil choices. In order to fully struggle against the effects of one’s evil influence on society, one must struggle to improve society, even if such improvement will benefit only those who come later.9 In the previous chapter we saw that the presence of radical evil implies that one’s moral goodness is expressed not only in good actions but also in a constant struggle against evil. But it turns out that at the level of the appearance that expresses one’s moral status, one is inextricably linked with the appearances of others. Thus one’s own moral goodness expresses itself not merely in good actions combined with an individual struggle toward goodness but also in a social struggle toward good and against evil. Because my radical evil makes use of a social network of promoting evil, the revolution against evil must express itself in social transformation. This emphasis on social transformation and social evil does not imply that human societies or the human species as a whole has a moral status.10 Even when Kant talks about the human species being evil, he points out that what it means to say that “character applies to the species” is that “there is no cause for exempting anyone from it” (6:25–6). Similarly, moral improvement of a species is the moral improvement of its individuals. Social transformation is ultimately, at a moral level, widespread individual transformation.11 In the case of one’s own efforts for social transformation, one improves oneself morally; one expresses one’s own moral status. But this expression of moral revolution involves a society that improves in the performance of good deeds and in the explicit commitment to morality. The problem of interpersonal influence thus leads to a refinement in the conception of what the good will looks like in the world. Expressing a good will involves not only acting rightly and struggling to promote helps and combat hindrances to one’s own virtue but also participating in a social struggle by promoting helps and combating hindrances to virtue in others. This social struggle is not merely a means to an end. Because it is a means for combating evil in oneself, it becomes constitutive of the expression of a good will.

146

The Solution

2. Is Moral Influence Morally Self-Centered? In section 1, I offered three roles for promoting morality in others. All of them focus on others. The first role leads to happiness for the one in whom one promotes morality. The second focuses on happiness for third parties. The third is concerned with increasing the good deeds of others. But in another sense all three roles are fundamentally concerned with oneself. None of them actually promotes anything genuinely moral in anyone else. They all express one’s own moral goodness. For parents, teachers, moralists, or religious leaders who are deeply concerned with the moral status of their children, students, audience, or parishioners, the roles for moral influence offered here can seem grossly insufficient. In this section I consider how far Kant can go toward allowing for genuine and ultimate moral influence, and I partially defend the sort of concern for others that Kant allows. A. Why One Ought to Be Self-Centered The first point that Kant can and should make is to remind those who seek to help others become morally good that this goal reflects a concern that is morally suspect. Kant’s argument in the Metaphysics of Morals against seeking the moral perfection of others is based on the incapacity to do for another what only that other can do for him- or herself. In that argument Kant suggests a couple of dangers of seeking to help others become morally good. First, “the perfection of another consists in this: that he himself is able to” act in accordance with duty (6:386). True perfection comes only when it is due to oneself. Second, perfection of another depends on that other’s acting in accordance with “his own concepts of duty” (6:386, emphasis added). One’s concepts of duty should be universally valid. One does not arbitrarily choose what is one’s duty, so one’s own concepts of duty should be the same as any other person’s concepts of duty. But for Kant it is essential that duty be one’s own in the sense that one must be autonomous; moral obligations binding on one must have their source in one’s own practical reason. In this context, promoting virtue in another is dangerous because even if one avoids the temptation to promote what is merely one’s own idiosyncratic preference or fixation as though it were a duty, to promote one’s own concept of duty in the actions of another is to assign a heteronomous source – oneself – for the duties of the other. This influence need not be strictly heteronomous. One can teach another that the moral law must be legislated by him- or herself. But the relationship of teacher and learner always brings with it

Moral Influence on Others

147

at least the danger that the learner will submit to a heteronomous source of duty. A further danger is that the very encouragement by which a goodhearted parent or moralist seeks to raise another to moral perfection can in fact lower the other to the level of a mere thing. By seeking to affect the ultimate moral status of another, one treats the freedom of the other as something natural, susceptible to empirical influence and manipulation. What one wants to change in another – his or her moral status – is just the sort of thing that must be changed only by the other. To seek to promote ultimate moral change in another really just denies that other moral autonomy and hence moral dignity. Thus there is an important sense in which one ought to be morally selfcentered. Being morally self-centered is very different, of course, from being selfish. One ought to take the happiness and well-being of others as an end and ought always to behave justly and generously with others. But one ought not take it upon oneself to ensure that others undergo fundamental moral change. To desire to affect the ultimate moral status of another is simply to desire too much. Such a desire is a form of hubris that sets one’s own capacity to exercise influence in the world above the freedom of another person. B. When One Can Be Morally Centered on Another Of course, the prohibition on making one’s end the moral perfection of another does not mean that one cannot take a genuine moral concern in others, but this concern should limit itself to the good deeds of others. There is no reason why one’s concern for promoting good deeds in others need always be merely instrumental. I suggested three purposes for which one might promote good deeds in others, but one might also promote good deeds for their own sake. One is not obliged to make the good deeds of others an end. But if one has a fundamental interest in seeing one’s children or students or friends act in ways that are consistent with the moral law, then one can promote such action for its own sake. Given that actions express the ultimate moral status of an agent, the natural desire to see loved ones be morally good might reasonably take the form of a direct interest in the appearance of morality in those loved ones. Unlike the desire to promote moral change at the level of freedom, there is nothing morally dangerous about this interest. Without doing any harm to others, one can promote good deeds in them not merely for the sake of happiness or to express one’s own moral goodness but for the sake of the good deeds themselves or the good moral appearance of the agents performing them.

148

The Solution

Ultimately, this purpose too is self-interested in a sense, because one acts to promote one’s own interest in the good deeds of others. In this sense, a direct interest in the good deeds of others is like the natural benevolence that Kant describes in the Groundwork (4:398), where one takes a direct interest in the happiness of others. Such a direct interest is not morally praiseworthy, but it is not wrong when kept within the limits of the moral law. And it might seem less morally self-centered than other reasons for helping others, because one’s interest in others’ good deeds is direct rather than mediated by any concern with the effects of those deeds on one’s own moral status. C. Helping Others and Hope for Others One might add to this reason for promoting virtue in others a basis for hope that focuses directly on the ultimate moral status of others. In the previous chapter, the inscrutability of one’s ultimate moral basis for action led to a focus on what is observable – primarily one’s actions – as the ground for hope in one’s moral goodness. Because one can hope for grace and future improvement on the basis of present good deeds and a visible struggle against evil, one effects good deeds and the struggle against evil at every opportunity. In the case of others, one is similarly incapable of discerning the ultimate moral ground of the actions of others. But one is no less capable of discerning this ground than in one’s own case. One of Kant’s most radical claims in the first Critique, especially in the B edition, is that one does not have privileged knowledge of oneself. For Kant, one can observe one’s own desires and intentions – in inner sense – in a way that one cannot observe these in the case of others. It may be easier, at least in some cases, to recognize one’s own self-deception than deception in others. But this difference is a difference of degree, and often a slight one. In both one’s own case and the case of others, one can at most hope for – but never have knowledge of – moral goodness. And in both cases, one can hope only on the basis of actions. This similarity between the case of one’s hope for oneself and one’s hope for others is important in the context of acting to promote good deeds in others because it implies that one’s promotion of good deeds in others can be conjoined with a legitimate hope, as one sees another do more and more good, that that other is actually morally good. Just as a person who struggles against evil and does more and more good in life can legitimately hope for grace and thereby a good will, a parent who raises children who do more and more good in their lives can legitimately hope

Moral Influence on Others

149

that those children have fundamentally good wills. Acting to promote good deeds, insofar as this action is successful, can increase one’s hope in the ultimate moral goodness of those one helps. But there is a crucial difference when it comes to the implications of hope between hope in one’s own goodness and hope in the goodness of others. In the previous chapter we saw that hope in one’s goodness plays a role in choosing to perform good deeds and to struggle against evil. Hope in one’s goodness prevents moral despair and provides a reason for acting in such a way that this hope is legitimate. One acts in such a way that one can justify hope in one’s worthiness for grace. Because one is responsible for one’s own moral status, acting such that one justifies rational hope in one’s own moral goodness is a form of directly expressing that goodness at the level of one’s freedom. In the case of others, however, the justification of hope ought not provide a reason for action. When parents’ efforts to promote morality in their children succeed, the good deeds of their children can legitimately justify the parents’ hope that their children are morally good. But this legitimation of hope cannot be the basis for promoting morality in their children because for it to provide a basis the parents would have to reason that they have or can have some responsibility for shaping the ultimate moral status of their children. In the individual case, it is perfectly appropriate to think that I have responsibility for my ultimate status. But to think that I have any responsibility for determining the ultimate moral status of another is to deny the full freedom and autonomy of the other. To promote good deeds in another on the ground that I can actually determine – in whole or in part – the ultimate moral status of that other is to treat that person’s freedom as an effect in nature. And to treat freedom in that way is to treat the other as a mere thing. One ought to promote in others good deeds and even an explicit commitment to the moral law. One can promote these external signs of goodness through the empirical aids – such as politeness, moral education, and the ethical community – that Kant studies in his moral anthropology. And one can rightly hope that those who express a commitment to the moral law in word and deed are worthy of grace and thus morally good at the level of their transcendental freedom. But one cannot make it one’s intention to directly promote another’s ultimate moral goodness without disrespecting that other. One’s hope in ultimate moral goodness may support one’s efforts to promote empirical goodness but can never be the basis for those efforts. Still, insofar as actions provide the basis for hope in the ultimate moral goodness of others, and insofar as one promotes these

150

The Solution

actions, this hope reflects an important sense in which one’s efforts to promote moral goodness in others do reach the ultimate level of having real and ultimate moral relevance. In addition to promoting the happiness and good deeds of others, it is not impossible that one’s efforts on behalf of others actually have an effect on their moral status. One cannot know in a theoretical sense nor even postulate for practical purposes that one does have such an effect, but it need not be considered impossible. D. Gestures Toward a Possibility: Real, Ultimate, Moral Influence A final consideration can give some metaphysical weight to this possibility of real and ultimate moral relevance. In a discussion of grace in his Religion, Kant very briefly opens the possibility of God’s affecting human free choice. In his reconciliation of the antinomy of grace, where he argues that from a practical perspective grace must be considered consequent to good works, Kant says, We can admit an effect of grace [that is, a grace that is the ground rather than the consequence of good works] as something incomprehensible, but cannot incorporate it into our maxims for either theoretical or practical use. (6:53)

Later, Kant explains in more detail why we can admit this conception of grace as something incomprehensible: The concept of a supernatural intervention into our moral though deficient faculty . . . – this is a transcendent concept, merely an idea of whose reality no experience can assure us. – But even to accept it as an idea for a pure practical intent is very risky and hard to reconcile with reason; for what is to be accredited to us as morally good conduct must take place not through foreign influence but only through the use of our own powers. Yet its impossibility (that the two may not occur side by side) cannot be proven either, because freedom itself, though not containing anything supernatural in its concept, remains just as incomprehensible to us according to possibility as the supernatural [something] we might want to assume as surrogate for the independent yet deficient determination of freedom. (6:191)

From the standpoint of experience, it makes no sense to talk about a grace that pre- or co-determines free choice because neither grace nor freedom is comprehensible from this standpoint. From the practical standpoint, one can make sense of both grace and freedom, but giving any kind of practical priority to grace is “hard to reconcile with reason,” to say the least, for exactly the reasons already examined with respect to interpersonal moral influence. One’s moral condition is necessarily due to oneself. Insofar as someone or something else is a more ultimate ground

Moral Influence on Others

151

of one’s moral status, that other person or thing bears the moral weight. Only if a moral status is fundamentally the result of one’s own choice can it be a moral status of one’s own. A goodness that is fundamentally an effect of grace is not one’s own.12 However, Kant does not rule out the possibility13 of grace. He just rules out making use of grace from either a theoretical or a practical perspective. As noted in the previous chapter, Kant does not suggest any other perspective from which one might make use of grace. But there is conceptual space for such a new perspective. One might prefer to leave grace in a position of limbo, without a clear standpoint from which it makes sense to talk about prevenient grace. But one can use Kant’s argument against the impossibility of grace to open a similar space for genuine interpersonal moral influence. This goes beyond Kant’s own treatment of interpersonal influence, but it is consistent with his mature moral philosophy overall. Just as the freedom of a moral agent is “just as incomprehensible to us . . . as the supernatural,” the freedom of one moral agent is as incomprehensible as the freedom of the next. In the same way that the incomprehensibility of both grace and freedom precludes the denial of a sort of relation according to which grace affects freedom but freedom nonetheless remains the ultimate and fundamental ground of action, the incomprehensibility of different free agents precludes the denial of real and ultimate interpersonal moral influence. The relation between a parent’s incomprehensible free choice and the free choice of a child may be such that the choice of the parent somehow affects the choice of the child, but the choice of the child is nonetheless an uncaused and fundamental ground of the child’s actions. As in the case of grace, this “possibility” can be put to neither theoretical nor practical use. Perhaps one could articulate some third perspective – say, a perspective of hope – from which it would have use. For Kantian parents, teachers, moralists, and religious leaders who desperately want to hold open some possibility of genuine and fundamental moral influence, this possibility may be reassuring.

3. History and Moral Progress In the previous two sections, I have laid out a Kantian account of interpersonal moral influence as fundamentally a matter of promoting good deeds in others for the sake of increasing happiness in others and expressing one’s own moral revolution against radical evil. In this section, I show how Kant’s writings in history are compatible with this general account. Kant’s

152

The Solution

philosophy of history has received extensive discussion in recent years.14 Some commentators insist that Kant’s philosophy of history involves a sort of moral progress that would be incompatible with the account of moral influence that I have presented. Louden, for example, claims that “history, while concerned with visible events that we can see and touch, is nevertheless ultimately and most fundamentally concerned about moral progress” (Louden 2000:147).15 In this section, I briefly articulate my account of Kant’s philosophy of history. I then show how some of the passages most often taken to support claims that Kant advocates more radical moral progress can be read as consistent with my overall account.16 For Kant, culture is progressing, good deeds are becoming more prevalent, and good states will emerge. All of this takes place in the world of appearances. But there is no reason to think that there are or ever will be more good free wills than in the past. A. A Brief Survey of Kant’s Writings on History Kant writes on history throughout his Critical period, beginning with an essay in November 1784 for the Berlinische Monatsschrift, “one of the most aggressive of the secularizing-rationalist journals of the day” (Zammito 1992:329).17 This essay, “Idea for a Universal History with a Cosmopolitan Purpose,” offers nine propositions that constitute a preliminary philosophical history of the human species. The essay was followed the succeeding year with three reviews of Herder’s Ideas on the Philosophy of the History of Mankind. These reviews were published in installments (in January, March, and November 1785) in the Allgemeine Literatur Zeitung, a new journal run by C. G. Schutz. ¨ The reviews focus largely on methodological considerations and do not contribute significantly to Kant’s account of moral progress. Far more important is another essay published in the Berlinische Monatsschrift in January 1786. This essay, “Conjectures on the Beginning of Human History,” is a further response to Herder in the form of a philosophical interpretation of the Biblical narrative that describes the origin and fall of humankind. After this flurry of essays, Kant did not devote further works to philosophical history for almost a decade. During this time he worked on his Critique of Judgment (published in 1790) and his Religion within the Boundaries of Mere Reason (published in 1793). The former book was Kant’s reason for cutting off his reviews of Herder’s history,18 and it includes Kant’s most systematic description of the final and ultimate ends of all history. Thus it provides a conceptual framework within which Kant’s more explicitly historical writings can be situated. The latter includes Kant’s

Moral Influence on Others

153

discussion of radical evil and the ethical community that one promotes in one’s struggle against that evil. In the context of this discussion, Kant includes a brief account of the “historical representation of the gradual establishment of the dominion of the good principle on earth” (6:124). About the time that Kant published the Religion (in 1792–3), he also wrote a pair of essays that show further development in his reflection on human history. In September 1793, just after the publication of the Religion, he published in the Berlinische Monatsschrift the essay “On the Common Saying: ‘This May be True in Theory, but It Does not Apply in Practice,’” to which I refer simply as “Theory and Practice.” This essay includes several passages that seem to support those who ascribe to Kant the view that human history is a history of genuine moral progress. The following year (1794) Kant published “The End of All Things,” again in the Berlinische Monatsschrift, an essay that offers a philosophical interpretation of the final end of all things, including human history. In 1795, Kant published On Perpetual Peace, in which he outlines the role of war in promoting progress toward peace and political right in human history. Kant’s last major published work, The Conflict of the Faculties, published in 1798, includes a major section devoted to “A Renewed Attempt to Answer the Question: Is the Human Race Continually Improving?” This essay marks Kant’s most advanced reflections on the status of moral progress in history. The same year Kant published The Conflict of the Faculties, he published his Anthropology, which has its own brief discussion of historical progress, a discussion that draws from Kant’s lectures on anthropology. B. The Scope and Limitations of Moral Progress in History Kant’s Conflict of the Faculties gives his clearest statement of the scope of history. Kant asks whether “the human race is . . . continually regressing . . . continually progressing . . . or at a permanent standstill . . . at its present level of moral attainment” (7:81). He argues for progress, on the grounds of “an occurrence in our own times which proves this moral tendency of the human race” (7:84). The occurrence is the French Revolution, and it shows a moral tendency in the human race because of the enthusiastic support of the revolution by spectators. Although Kant describes this progress in terms of “moral attainment,” there are two important respects in which Kant clearly does not think that human history reflects genuine moral progress.19 First, while there is a “moral character” (7:85) that is revealed in people’s reaction to the French Revolution, this character is a fixed feature of the human species; one can “credit human beings with . . . a limited will of innate

154

The Solution

and unvarying goodness” (7:84, emphasis added). Second, the progress in human history is progress in good deeds and in more just political structures, such as a “national constitution [that is] just and morally good” (7:85). Kant clearly says, The profit that will accrue to the human race as it works its way forward will not be an ever increasing quantity of morality in its attitudes. Instead, the legality of its attitudes will produce an increasing number of actions governed by duty, whatever the particular motive behind these actions may be, . . . i.e. from the external phenomena of man’s moral nature. (7:91)

The empirical history of humans will involve progress, but history gives no reason to believe that this progress affects the free wills of human beings. Thus this sort of “moral” progress fits well with the account of interpersonal moral influence developed in this chapter. Although not primarily focused on history, the Critique of Judgment further clarifies this distinction between genuine moral change and the sorts of progress that can be studied in an empirical history.20 In particular, this Critique outlines two “purposes” of nature. One of these, the “final purpose,” is clearly moral. Kant says, “A final purpose is a purpose that requires no other purpose as a condition of its possibility” (5:434), and “only in man, and even in him only as moral subject,” do we find such a final purpose (5:435). By contrast, the “ultimate purpose” is “the purpose by reference to which all other natural things constitute a system of purposes” (5:429), that is, the end for the sake of which the rest of nature can make sense as a means. Humanity is the ultimate purpose of creation as well, but “this [ultimate] purpose must . . . be man’s aptitude and skill for various purposes for which he can use nature (outside or within him) . . ., his culture” (5:429–30). And although human beings constitute both the ultimate and the final purposes of nature, “we can even prove a priori that what might perhaps be an ultimate purpose for nature can still, insofar as it is a natural thing, never be a final purpose” (5:426). For Kant, it is crucial to distinguish moral goodness from culture in discussions of progress. In the Critique of Judgment, however, the final purpose is the proper subject of moral philosophy rather than of empirical history. History is the study of the ultimate purpose of nature, the culture of humanity. Kant defines culture as “man’s aptitude in general for setting himself purposes, and for using nature . . . as a means [for achieving them]” (5:431). This definition gives rise to a distinction between two types of culture. The culture of discipline “consists in liberation of the will from the despotism

Moral Influence on Others

155

of desires” so that one can set purposes for oneself (5:432). The culture of skill cultivates humanity’s abilities to use nature as a means (5:431– 2). Kant’s account of culture as the ultimate end of humanity includes brief versions of several of Kant’s more specifically historical writings. For example, Kant claims, Though war is an unintentional human endeavor (incited by our unbridled passions), yet it is also a deeply hidden and perhaps intentional endeavor of the supreme wisdom [Nature], if not to establish, then at least to prepare the way for lawfulness along with the freedom of states, and thereby for a unified system of them with a moral basis. . . . [Moreover,] war is one more incentive for us to develop to the utmost all the talents that serve culture. (5:433)

This claim that war is a means for cultural progress, and in particular for progress toward a union of states, is a theme throughout Kant’s writings on history (cf. “Idea” [8:24–6], “Conjectural Beginnings” [8:121], “Theory and Practice” [8:311], and Perpetual Peace). This connection between the discussion of culture in the Critique of Judgment and Kant’s more specifically historical essays suggests that these essays focus on history as the study of the ultimate and empirical, rather than the final and moral, purpose of human beings. But Kant’s historical essays also suggest that culture is not the only end of human progress.21 We have already seen that the Conflict of the Faculties focuses more on political progress and increasing good deeds than on cultural progress. In the Critique of Judgment, political progress is “the formal condition under which nature can alone achieve this final aim [of culture]” (5:432). Perpetual Peace (see 8:365–6) and “Theory and Practice” (8:310–13) also emphasize the importance of political progress. In Perpetual Peace, political progress is in the service of peace, rather than the development of culture itself, and in “Theory and Practice,” political progress is the ultimate end that “the very nature of things” will establish despite us (8:310–11). Thus these later writings are more interested in progress toward morally good ends – peace or justice – than is the Critique of Judgment. But this progress is still far from the final purpose of nature, the cultivation of morally good people. Some of Kant’s writings on history, however, seem to suggest that genuine moral progress does occur in human history. These seem to go beyond the limits set by the Conflict of the Faculties. These works seem to present a history where members of later generations not only do more good things and live in more just societies but are actually more

156

The Solution

likely to have morally good wills than members of earlier generations. In the rest of this section, I briefly address four of the most troubling passages in which Kant seems to suggest that history studies genuine moral progress.22 In “Theory and Practice,” Kant writes, “since the human race is continually progressing in cultural matters (in keeping with its natural purpose), it is also engaged in progressive improvement in relation to the moral end of its existence” (8:308–9). Later Kant adds that “various evidence suggests that in our age, as compared with all previous ages, the human race has made considerable moral progress” (8:310). Drawing on this passage, Robert Louden says, “Kant also maintains confidently in [“Theory and Practice”] that there exists strong empirical evidence for asserting that humanity is progressing morally” (Louden 2000:147). Louden is certainly correct that Kant seems to say that not only culture (the ultimate purpose) but also morality (the final purpose) progresses throughout human history.23 But Kant’s apparent focus on moral progress is not as straightforward as it seems. Kant never explicitly identifies the “moral end” toward which humanity progresses. And while it is natural to read this as a reference to the genuine moral improvement of individuals, the essay itself suggests that Kant has something else in mind. Shortly after mentioning this moral end, Kant associates it with Mendelssohn’s goal of promoting “the enlightenment and the welfare of the nation to which he belonged” (8:309). And the final paragraph of the essay offers the real key to interpreting the sense in which the human species progresses “morally.” There Kant explains that political right, not moral virtue, is the “moral end” of the human species. As he says, For my own part, I put my trust in the theory of what the relationships between men and states ought to be according to the principle of right. It recommends to us earthly gods the maxim that we should proceed in our disputes in such a way that a universal federal state may be inaugurated, so that we should therefore assume that it is possible (in praxi). I likewise rely (in subsidium) upon the very nature of things to force men to do what they do not willingly choose. (8:313)

The principle of right “recommends” – that is, imposes on us as a duty – that we promote political right in our interactions with one another, and the historical account in “Theory and Practice” encourages our belief that this “is possible.” In the Metaphysics of Morals, Kant reiterates that although one does not have any duty to promote fundamental moral change in

Moral Influence on Others

157

others, one does have a duty to promote political right (see 6:307–13). And in the Conflict of the Faculties as well, Kant uses language of moral progress to refer to a political change. Thus “Theory and Practice” fits well into Kant’s overall conception of historical progress. Politics as well as culture progresses throughout history. But none of this requires any deep interpersonal moral influence. “The End of All Things,” like “Theory and Practice,” makes reference to “the development of morality” (8:332) and “the moral disposition of humanity” (8:332) and explicitly refers to the conformity of rational beings to the “final end of their existence” (8:331), which the third Critique describes as moral rather than merely cultural or political. What is more, Kant connects this moral disposition to the rewards and punishments of the last day. If his conception of the highest good and the justice of God is to make any sense, “The End of All Things” has to be about the moral disposition of humanity in the deepest sense. The essay cannot be limited to mere political right, nor even merely to good deeds, but must touch on the ultimate moral status of human beings. Thus one might think this essay refers to moral progress in history.24 However, unlike Kant’s other historical essays, “The End of All Things” is not empirical history. The “end” is not a final moment in time, nor even an ultimate goal that the species asymptotically approaches. The end of all things is the end of all temporality, the “transition from time into eternity,” the thought of a “duratio Noumenon wholly incompatible with time” (8:327). This end is the thought of human beings “as supersensible, and consequently as not standing under conditions of time; thus that duration and its state will be capable of no determination of its nature other than a moral one” (8:327–8). Immediately after mentioning the “development of morality,” Kant insists that “here we have to do (or are playing) merely with ideas” (8:332). He goes further and claims that there can really be “no alteration” with respect to this moral end (8:333). Rather, one must think of the final end as an alteration, proceeding to infinity (in time) in a constant progression, in which the disposition (which is . . . something supersensible . . .) remains the same and is persisting. . . . [We must act as if] its disposition (the homo Noumenon, “whose change takes place in heaven”) were not subject to temporal change at all. (8:334)

This essay, in other words, discusses the ultimate moral “end” of all things as the unchanging timeless moral disposition that is the ultimate ground

158

The Solution

of a person’s appearance in the world. This analysis touches deeply on morality, to be sure, but adds nothing to Kant’s philosophy of empirical history and does not commit Kant to any kind of historical influence on one’s moral status. Like “Theory and Practice” and “The End of All Things,” Kant’s essay “Conjectures on the Beginning of Human History” seems to suggest that human beings undergo moral progress.25 Kant says, “I wish merely to consider the development of human behavior from the ethical point of view” (8:111) and refers to “man’s development as a moral being” (8:113). And the final step in Kant’s outline of the development of human beings is clearly a moral one: The fourth and last step which reason took, thereby raising man completely above animal society, was his (albeit obscure) realization that he is the true end of nature, and that nothing that lives on earth can compete with him in this respect. . . . This notion implies . . . an awareness of the following distinction: man should not address other human beings in the same way as animals, but should regard them as having an equal share in the gifts of nature. This was a distant preparation for those restrictions which reason would in the future impose on man’s will in relation to his fellows. (8:114)

This recognition that human beings are ends in themselves deserving of respect is a deeply moral realization. The “beginning of human history,” as Kant outlines it in this essay, describes moral progress. But it is important to note here that this history is a history of the beginning of human history, and Kant does not commit himself to the view that the moral progress which accounts for the birth of humans as moral beings will continue throughout the history of civilization. Kant explicitly says that “a history of the first development of freedom from its origins . . . is something quite different from a history of its subsequent course” (8:109). And although he discusses the development of human beings after their early beginnings, in those discussions he does not suggest any fundamental moral progress but only progress toward culture, art, a “civil constitution” (8:119), and perpetual peace (8:121). With respect to properly moral progress, he says that “the human race was . . . deflected from . . . cultivation of its capacities for goodness,” but we can remedy these evils by “improving ourselves” (8:120–1). Even when Kant discusses “moral” progress at the beginning of human history, this progress does not involve increasing moral goodness. Rather, Kant outlines how a pre-moral being came to be morally accountable at all. The progress is from nonmoral to morally responsible rather than from

Moral Influence on Others

159

morally worse to morally better. Thus Kant explains that this progress is a “release from the womb of nature” (8:114), not from the womb of wickedness, an expulsion “from a garden . . . into . . . unknown evils” (ibid.), not a release from evil into moral virtue. Prior to this moral progress, “there were no commandments or prohibitions, so that violations of these were also impossible” (8:115). Thus although “Conjectures on the Beginning of Human History” clearly outlines moral progress of a sort, the essay does not describe a moral progress that poses any problem for Kant’s account of freedom. There is no indication that empirical influences can affect one’s moral status, though they can affect whether one has a moral status. Incapable of doing good or evil, proto–human beings were happy, but without morality. After the progress into moral responsibility, human beings became capable of good and evil. But there is no indication in this essay that the history of this freedom is a history of progress from evil to good. All we get here is a conjecture on the beginning of human history.26 One final area in which Kant seems to discuss moral progress is in his Anthropology. There Kant suggests that historical progress has three parts: “man is destined . . . to live in a society of other people, and in this society he has to cultivate, civilize, and moralize himself” (7:324–5, cf. “Idea” 8:26, and 9:450, 15:641, and 25:847, 1198, 1420). The reference here to “moralization [moralisieren],” as something distinct from cultivation and civilization, suggests that the development of culture is not what Kant has in mind. Robert Louden, in fact, argues that moralization involves “the deep structure of an agent’s moral personality” (Louden 2000:42).27 Still, Louden rightly points out that with respect to the details of moralization, Kant is “exasperatingly brief” (Louden 2000:43).28 One possible interpretation of the moralization of the human species is that over time, more and more people come to be morally good. Unless this is the only reasonable interpretation, however, it should be rejected. It conflicts with Kant’s claims in the Conflict of the Faculties about what can be hoped for from history, and it suggests a sort of interpersonal moral influence that would be incompatible with Kant’s account of freedom. Fortunately, there are three other plausible ways to interpret “moralization” that do not require positing that history marks moral change in the free wills of human beings. Moralization could be interpreted to refer to the just political structures to which Kant refers in the Conflict of the Faculties and “Theory and Practice.” Kant says that we can

160

The Solution

recognize in the human species “a tendency . . . to bring about sometime in the future the development of the good out of the evil through its own efforts” (7:329). When Kant describes this development, which “can be expected with a moral certainty” (7:329), he says, As their [human beings’] culture grows, they become more and more aware of the wrongs which they themselves selfishly inflict upon each other. And since they see no other remedy for them than subordinating the private interest (of the individual) to the public interest (encompassing all), they subject themselves, reluctantly though, to the discipline (of civil restraint), which they only obey, however, by following laws which they themselves have given. They feel themselves ennobled by the knowledge that they belong to a species which fits the vocation of man as reason represents it to him in the ideal. (7:329–30)

Here Kant makes clear both why this progress is “moral,” and the limits of its influence. On the one hand, by becoming subject to just and representative political structures (“civil restraints”), human beings conform to a structure that reason prescribes as a moral ideal. On the other hand, because this conformity is possible even when people subject themselves “reluctantly [ungern],” this does not imply any fundamental moral change. Kant expands upon this political progress later, arguing that people “feel destined by nature to develop, through mutual compulsion and laws written by them, into a cosmopolitan society” (7:331), and the Anthropology ends with the hope that humanity can “advance constantly from evil to good . . . through the progressive organization of the citizens of the earth within and toward the species as a system which is united by cosmopolitical bonds” (7:333). However, in the Anthropology and “Idea,” Kant seems to think that moralization goes beyond mere political progress. The cosmopolitical bonds to which Kant summons his readers at the end of the Anthropology seem to be the means for a more significant moral progress. Likewise in the “Idea” Kant says that “the human race will remain in this condition [civilized and cultured but not moralized] until it has worked itself out of the chaotic state of political relations” (8:26). Of course, even the Conflict of the Faculties goes beyond merely political progress. There Kant claims that “an increasing number of actions governed by duty” will be an effect of political progress (7:91). Moralization could refer to this increase of good actions. The discussion of moralization in the Anthropology might even be read to suggest that moralization goes beyond mere progress in good deeds and includes the kind of resistance to evil discussed in the Religion.29 Kant introduces the notion of moralization after a discussion of “the moral

Moral Influence on Others

161

predisposition,” in which he very briefly recapitulates the first two books of the Religion, explaining that while every human being “sees himself . . . subject to a moral law . . ., experience also shows that in man there is a propensity to . . . evil” (7:324). Moralization is thus set as a task: “man [must] moralize himself” (7:324). Kant goes on to explain moralization as the process whereby man will “make himself worthy of humanity by actively struggling with the hindrances that cling to him from the crudity of his nature” (7:325).30 This reference to moralization as a struggle against obstacles echoes the language of struggle in the Religion, where this struggle is manifested in the promotion of an ethical community. And in the Religion too, this ethical community is distinguished from mere cultural or political progress (6:94–7), and even from progress in terms of good deeds. Moralization could refer, then, to that explicit commitment to and encouragement of virtue which characterizes the ethical community. But as we have seen, this ethical community need not refer to any fundamental moral change. Thus moralization can plausibly be read as progress toward more just political structures, more good deeds, or an ethical community. None of these implies fundamental moral change. Likewise Kant’s other historical writings all refer to one of these sorts of progress (in the case of “Idea,” “Theory and Practice,” Perpetual Peace, and the Conflict of the Faculties), or to the historical development of moral responsibility itself (in the case of “Conjectures”), or to morality itself, rather than history strictly speaking (in the case of “The End of all Things”). Some have read other passages in Kant’s historical writings as implying the kind of deep moral progress that would be incompatible with this view. Ultimately, these can all be dealt with in ways that are similar to my treatments of difficult passages in this section.31 Kant’s writings on history, like his more general accounts of interpersonal moral influence in the Metaphysics of Morals and the Religion, reflect a commitment to empirical helps and hindrances that are morally significant within the bounds of his theory of freedom.

4. Conclusion Kant’s account of interpersonal influence allows one to salvage many common-sense views of the importance of such influence. One ought to act for the moral improvement of others because such efforts can have positive results. One can increase the extent to which others do good things, which will often provide both them and others with better lives. Moreover, we are all part of communities such that each can profoundly

162

The Solution

affect the way others behave, and this provides a further reason to cultivate virtue not only in oneself but also in others. Still, on Kant’s account there is no practical or theoretical basis for believing that one person can determine, change, or even affect the fundamental moral status of another. Some might wish for such a possibility, but it is not clear that such a demand is made by the common-sense conviction that we should try to improve others morally, and that we can improve them. Kant allows that there is a sense, and a rather important sense, in which we can and should work toward the moral betterment of others. All people have a responsibility to promote morality in others in the sense that all people should promote the performance of good deeds and an explicit commitment to virtue. Kant can even go part of the way toward allowing for even more robust moral influence. One can rationally hope that those who perform good deeds and express commitments to morality are actually morally good, and this can give rise to a direct interest in the explicit goodness of others. And Kant may even have room for an inscrutable and practically important possibility of mutual influence at the deepest level. Thus there is room for serious social cooperation and corporate moral responsibility, and for the cultivation of moral community and mutual encouragement in virtue. And all of this is possible within the bounds of Kant’s account of transcendental freedom as a condition of the possibility of moral obligation.

Epilogue Incorporating Moral Anthropology and Defending Kantian Moral Philosophy

Near the end of Kant’s Impure Ethics, Robert Louden asks whether Kant is “saved by impurity” (Louden 2000:167). That is, he asks whether the anthropological and other “impure” elements of Kant’s moral philosophy save it from the objections to which it has been subject over the past two centuries. In this book, my task has been more limited. I am convinced that Kant’s moral anthropology strengthens his ethical theory. Kant rightly recognized the possibility of effecting changes in human agents, especially oneself. And he saw that these changes can be morally beneficial when they are directed to the cultivation of a good and strong character. One’s moral status is not merely a matter of the choices that one makes about the present but involves the sort of person that one is. And becoming a certain sort of person involves cultivating and promoting certain dispositions, emotional responses, interactions with others, habits of thought and action, and social structures. These aspects of human life are clearly important parts of morality, and Kant’s moral theory is stronger for having recognized them. But Kant’s moral theory is stronger for having recognized the importance of anthropology only if the theory remains consistent. The importance of freedom – both practical autonomy and the transcendental freedom which underlies that autonomy – is still the core insight of Kantian moral theory. Many moral philosophers have recognized the importance of anthropological considerations. What makes Kant both distinctive and attractive remains his emphasis on freedom from empirical influence in determining the content of the moral law and acting on the basis of it. Without these Kantian emphases, one may as well turn to Aristotle, Aquinas, Smith, Nietzsche, or any of a dozen other moral theorists. Unless 163

164

Epilogue

Kant’s account of moral anthropology is consistent with his account of freedom, the benefits for ethics of those anthropological insights cannot rightly be counted as advantages for Kant’s ethics. Any defense of Kant’s moral theory that depends on the value of Kant’s moral anthropology must show how that moral anthropology is consistent with Kant’s account of freedom. Showing this consistency has been the task of this book. This book thus serves as a prolegomenon to anthropological defenses of Kant’s moral theory. Schleiermacher objected to any such defense on the grounds that Kant’s anthropology and his moral philosophy cannot be made consistent. Although I showed that there are good prima facie reasons for this objection, I ultimately argued that Kant can respond to it. By showing that moral anthropology is consistent with Kant’s account of freedom, I thus legitimate attempts to defend Kant by drawing on his moral anthropology. The concrete examples of moral anthropology in Chapter 3 can serve as initial responses to those who argue that Kant fails to take into account the importance of politeness, the complexity of human emotions, or the role of character in moral life. There is much more work to be done fleshing out a robust Kantian anthropology, however. This book provides only a framework within which such work is possible. At the same time, I have shown that there are limits to how far Kant can go in allowing for morally significant empirical influences. Moreover, the framework within which these influences must fit involves a conception of the role of such influences that is not shared by all those who make use of anthropology in moral reflection. In contrast to Robert Louden’s account of Kant, for example, I insist that no empirical influence can “cut through the surface causal network” (Louden 2000:59) to effect a change of one’s moral status. This limitation does not imply that acts in the world are irrelevant to one’s moral status. As we have seen, one can effect changes in future choices through empirical influences, and effecting these sorts of changes is an important part of expressing a good will. But the empirical influences do not cause one to have a good will. At a practical level, this fact does not give one any reason to refrain from promoting good future choices through empirical influences. But it changes the philosophical understanding of the relationship between these influences and one’s moral status. One concrete implication of this different understanding, explained in detail in Chapter 6, is that one cannot seek to change the moral status of another through one’s actions in the world. This book leaves some unresolved issues. I have said very little about the accuracy of Kant’s particular anthropological observations. As I noted

Defending Kantian Moral Philosophy

165

in Chapter 2, Kant’s anthropology is fallible. He recognized mistakes over the years, and he made many mistakes that he did not recognize. As investigations in the human sciences have progressed over the past two hundred years, many of Kant’s claims have been called into question. Even in Kant’s own time, many of his anthropological claims were controversial. The relationship between politeness and virtue was a particularly hot topic in the eighteenth century, and claims about race and gender, the emotions, and the role of churches and political structures were and are still widely disputed. Some of these issues can now be thought through more carefully by philosophers. Many are being discussed among those involved in fields such as psychology, sociology, anthropology, and cognitive science. Today we have access to more data than Kant had about human behavior, motivation, and mutual influence. This new data will allow us to revise the details of Kant’s anthropology. But for Kant, no amount of data can determine the relationship between empirical data and the freedom that grounds moral responsibility. No data can show the ultimate significance of interpersonal influence. And no data can specify the proper structure of a moral theory. These philosophical insights about the limits of empirical inquiry remain a Kantian legacy for today. But the details of Kant’s anthropology may be in need of serious revision. Moreover, there are remaining philosophical problems with the account offered here. The limitation on interpersonal moral influence might be unsatisfying for many, even after my attempts in Chapter 6 to make this limitation more palatable. The role of grace in Kant’s resolution to the problem of radical evil still might seem to raise significant problems, and these may infect my integration of anthropology and freedom. Some might still have metaphysical worries about Kant’s apparent dualism. However one understands the relationship between nature and freedom, the dichotomy can seem troubling, and I have said little in this book to ease purely metaphysical troubles. Even if these worries are solved, Kant’s argument for transcendental freedom can still be challenged on grounds that do not relate to anthropology per se. Even if Kant’s account of transcendental freedom is consistent with his moral anthropology, that in itself does not provide any reason for thinking that one should accept transcendental freedom. Kant’s argument for such freedom, at least as I recounted it in Chapter 1, depends on a “fact of reason” about moral responsibility. Many might reasonably question whether there is such a “fact,” or whether such a fact involves any commitment to something as strong as Kant’s transcendental

166

Epilogue

freedom.1 What I have shown in this book is that two aspects of Kant’s moral theory – his account of freedom and his moral anthropology – are consistent. But even if Kant’s moral theory is consistent, it might still be wrong. Aristotelians, Humeans, Anti-theorists, Utilitarians, and other moral theorists need not hang up their hats and go back to Kant. But if the reason they reject Kant is his insufficient attention to moral anthropology, they have reason to give him another look. I have shown that one common objection to Kant is not decisive. Kant can reconcile his strong conception of freedom with a robust moral anthropology. This shifts the debate over Kant’s moral theory back to issues that Kant considered more central than anthropology, issues regarding the nature of moral obligation and the implications of moral obligation for our conception of human freedom.

Notes

Introduction 1. See the bibliography to Thomas Sturm’s “Kant und die Wissenschaft des Menschen,” unpublished manuscript. 2. Schleiermacher was not the only one to question Kant’s success in combining systematicity and popularity. One of Kant’s most important early interpreters, Karl Reinhold, sought to improve on Kant by providing a version of transcendental idealism that could be both popular and systematic. For more on the ways in which Reinhold misinterpreted Kant and thereby inaugurated the ambitious programs of later German Idealists, see Ameriks 2000 and Franks 2000. 3. Ultimately, Schleiermacher does not think that any particularly strong conception of freedom is necessary to make sense of normativity. In fact, his critique of Kant’s Anthropology is part of a larger project of replacing Kant’s notion of transcendental freedom with a soft determinism about human freedom. Still, Schleiermacher recognizes the need to give an account of the normative if one seeks to be a determinist about human choices. Much of On Freedom (Schleiermacher 1992) is designed to give just such an account, for both epistemological and moral normativity. 4. Even if Kant does throw out one or more aspects of his moral anthropology, the dilemma arises in other areas. The dilemma permeates Kant’s practical philosophy whenever he seeks to articulate means for promoting advancement toward greater autonomy. For example, in “What Is Enlightenment?” Kant insists that “Enlightenment is man’s emergence from his self-incurred immaturity” (8:35). But the essay also raises the question of how man is to emerge from self-incurred immaturity. “Laziness and cowardice” lead people to “gladly remain immature for life” (8:35). Kant goes on to shift responsibility for promoting enlightenment from the individuals whose immaturity is self -incurred to those “guardians who have kindly taken upon them the work of supervision [and who] see to it that by far the largest part of mankind . . . should consider the step forward to maturity not only as difficult 167

168

5. 6.

7.

8. 9.

10. 11.

12.

Notes to Pages 2–7 but as highly dangerous” (8:35). An essay that begins as a call to individuals to take responsibility for immaturity that is self-incurred turns out to be a call to leaders to promote maturity in others. It is apparently up to the leaders to help their subjects develop the courage to think for themselves, and subjects can apparently take responsibility for themselves only with the right kinds of people responsible for them (8:41). This case is not as problematic as moral anthropology in general. There is no strict inconsistency between the ideal of people thinking for themselves and the fact that they will realize this ideal only under certain leaders. The means toward enlightenment are not necessary parts of enlightenment. While members of an enlightened public can think freely regardless of their leader, for a public to become enlightened, a certain sort of leader may be necessary. Still, “What Is Enlightenment?” exposes two aspects of Kant’s notion of enlightenment that, though consistent, do not sit comfortably with each other. And in the case of moral anthropology, the conflict appears to be even more severe. See for instance Rehberg’s early criticism (in Schulz 1975) and Hegel’s famous criticisms in the Phenomenology. The best article for describing both objections to Kant and Kantian responses is Wood 1984. See, too, Allison 1983, 1990, and 1996; Beck 1987; and Korsgaard 1996a. Schleiermacher does pose his dilemma in general terms. He suggests that there is a problem with helps or hindrances to any “mental faculties [Gem¨uthsverm¨ogen]” (Schleiermacher 1984: v.1, p. 366). In fact, however, Kant is committed to transcendental freedom only for human choices, and then only for those choices for which one can be held morally responsible. Thus the real dilemma for Kant is raised not by his anthropology in general but by his specifically moral anthropology. Schleiermacher’s description of the dilemma as a conflict between claims about nature and choice shows his primary focus on moral helps and hindrances. Schleiermacher’s On Freedom (Schleiermacher 1992), written shortly before the review of Kant’s Anthropology, makes even clearer that moral issues are foremost in Schleiermacher’s mind at this time. This is discussed in detail in Chapter 3. Strictly speaking, one might still promote polite society for the sake of the happiness of others, even if doing so does not affect anyone’s moral status, but this is not the particular reason that Kant gives for polite society. Instead, he argues that politeness is a duty because the “illusion resembling virtue” can give rise to genuine virtue. See for example Baron 1995, Chapter 6. For neo-Aristotelian approaches, see MacIntyre 1984 and 1988, Hursthouse 1999, Foot 2001, and Anscombe 1958, just to mention a few. Humeans include Annette Baier (see Baier 1987, 1991, 1995) and, more recently, Simon Blackburn (Blackburn 1998). The most prominent Anti-theorist is Bernard Williams (see especially Williams 1981, 1985, 1995). One must be careful in reading too much of virtue ethics into Kant, however. Kant often describes very different aspects of moral life using terms that are

Notes to Pages 7–14

169

similar to those in virtue theories. When Kant discusses character, disposition, virtue, or emotion, he often does not mean by these terms what an Aristotelian or Humean would mean. For some specific examples, see the discussions of passions, affects, and character in Chapter 3. See too Munzel 1999.

Chapter 1 1. If one prefers more explicitly two-perspective language, one should read this to say that the categories and forms of intuition cannot be used in order to have knowledge of how things are when considered from a practical perspective. 2. Actually, Kant’s own language here is significantly different. The antinomy is between a thesis that “causality in accordance with laws of nature is not the only causality from which the appearances of the world can one and all be described. To explain these appearances it is necessary to assume that there is another causality, that of freedom” and an antithesis that claims that “there is no freedom; everything in the world takes place solely in accordance with laws of nature” (A444–5/B472–3). The tension between these claims is different from the tension between the claims that there is and that there is not freedom, but it is not necessary to discuss these differences for the purpose of the present discussion. This antinomy has been discussed in detail in the secondary literature. See for example Strawson 1966, Allison 1983, Pippin 1982, and Kemp Smith 1992. 3. There is some ambiguity in the third antinomy about the sort of priority the cause must have. Understanding demands a cause that is prior in some sense, and the only sense in which the understanding can describe that priority is temporal. Kant’s resolution of the antinomy ultimately depends on the possibility (as far as we can tell) of a nontemporal cause. Such a cause meets the demand for a cause, but not the limitations imposed by the understanding on the sort of cause that is available. In that sense, reason goes beyond the understanding. 4. I intend this formulation to be neutral between two-object and two-aspect theories of the relationship between things and their appearances. A two-object theorist should read this to refer to two distinct “things” – the thing in itself and the appearance – that are in some kind of causal interaction. The twoperspective theorist should just read this to refer to a single thing insofar as it is considered either “in itself” – that is, from some non-empirical perspective – or as an object of experience. For further discussion of the distinction between two-object theories and two-aspect theories, see Ameriks 1982, Pippin 1982 (especially pp. 194–215), Prauss 1983, Allison 1983, and Nelkin 2000. The best defense of a two-object view has been offered by Aquila (especially in Aquila 1979). The best defense of the two-aspect theory is presented by Allison (in Allison 1983 and Allison 1996). The two-perspective view has been applied to issues of freedom in moral philosophy in Allison 1999 and 1996 and in Korsgaard (especially 1996a). For an excellent overview of these accounts, see Nelkin 2000.

170

Notes to Pages 14–19

5. See section 4 of this chapter for more on the inscrutability of freedom and the analogical use of the categories. 6. I take the argument of the second Critique to reflect Kant’s most considered views about the justification of freedom. For an excellent discussion of the differences between the argument in the Groundwork and the argument in the second Critique, see Ameriks 1981. The clearest articulation of the argument I paraphrase here is given in the second Critique at 5:28–30. 7. It is easy to mistake this for the claim, which Kant is not making, that context is not relevant to what is morally required. Context is often a relevant feature of maxims that are tested for moral permissibility. But the context itself does not determine the way in which the maxims themselves will be evaluated. Rather, it is insofar as the content or context of a maxim affects its form that it becomes morally relevant. 8. Note that this does not mean that I am totally free of any influence. I am free of any empirical influence, which means that my desires and beliefs do not determine the way I act. But to act independent of these influences is to act on the basis of a moral law that provides a reason for action. I can be free of empirical causes, for Kant, but I always act for some reason. 9. This does not mean that one cannot give causal explanations of one’s actions. But these explanations are always only conditioned. The ultimate, unconditioned explanation of any act is the free agent, the self-in-itself. 10. Possibility must be understood loosely here. In the first Critique, Kant claims, “the fact that a concept does not contradict itself by no means proves the possibility of its object” (A596/B624). He adds in a footnote: A concept is always possible if it is not contradictory. . . . But it may none the less be an empty concept, unless the objective reality of the synthesis through which the concept is generated has been specifically proved; and such proof, as we have shown above, rests on principles of possible experience. . . . This is a warning against arguing directly from the logical possibility of concepts to the real possibility of things. (A596/ B624)

“Real possibility” is possibility as an object of possible experience, and freedom is not possible in that sense. But freedom is not merely logically consistent. It is “possible” in the sense that it is consistent with the world of experience, but only as something not included in that world. That is, it “does not conflict with” nature (A558/B586). 11. In fact, it is not clear that Korsgaard actually thinks there is any asymmetry. In the Sources of Normativity (Korsgaard 1996c:95–7), she seems to endorse a thoroughly symmetrical account. 12. It is important to recognize, however, that this is not a methodological priority. Theoretical reason is methodologically prior to practical reason. The conviction that people are free, a conviction established on the basis of practical reason, would be illegitimate without the possibility opened for freedom by theoretical reason. Thus theoretical reason holds a methodological priority over practical reason, such that the first Critique is a necessary preparation for the second. For Kant’s insistence on this sort of priority of the theoretical, see the first Critique, Bxxix.

Notes to Pages 21–23

171

13. Strictly speaking, theoretical reason holds open the logical consistency of and need for freedom as ground of experience. See footnote 10. 14. One passage in the first Critique may seem to raise a problem for this account. In his resolution of the Third Antinomy, Kant seems to suggest that the practical perspective – which in this context is the perspective from which one discusses the “intelligible ground” of appearances – need not be considered at all in empirical investigation. He says, This intelligible ground does not have to be considered in empirical enquiries;. . . . We have to take their strictly empirical character as the supreme ground of explanation, leaving entirely out of account their intelligible character (that is, the transcendental cause of their empirical character) as being completely unknown, save in so far as the empirical serves for its sensible sign. (A545–6/B573–4)

This passage might be read to imply that from the standpoint of theoretical understanding or empirical inquiry the status of the intelligible does not even arise. When Kant says that it “does not have to be considered” and should be left “entirely out of account,” he lends himself to this interpretation. But in the final part of the passage, Kant makes clear that what he in fact opposes is not the consideration of the relationship between the practical and theoretical perspectives but the use of insights from the practical perspective in empirical explanation. One can affirm, from a purely theoretical standpoint, that if there is a practical perspective, then it provides a more ultimate explanation than the empirical. But one cannot know that there is such a perspective, and one certainly cannot know anything about such a perspective, on the basis of theoretical reason. The practical perspective is useless for empirical explanation, even if theoretical reason can and even needs to establish that if there is such a perspective, it is prior to – in the sense of being the ultimate ground of – empirical explanation. 15. As mentioned earlier (see footnote 12), there is still a sense in which the theoretical perspective has methodological priority. The priority of the practical perspective is a priority of the moral ground over the empirically described ground as a proper explanation of the ultimate basis of an action. 16. This “realm of experience” is not limited to the realm of outer experience. One may evaluate desires, intentions, and plans that are experienced only by oneself, and only through introspection. But even this introspection gets merely at one’s appearance in experience, not at the noumenal ground of those experiences. See Kant’s discussions of self-knowledge in the first Critique (Bxl–xli, B68, B275f., A362–3). 17. There are some variations on these two ways of interpreting Kant that I have not considered. Most of these resolve themselves into either a two-object or a two-perspective interpretation. For example, Karl Ameriks has pointed out that even if Kant has a two-object view of the relationship between things in themselves and appearances in general, he may have a two-perspective view with respect to the self. Thus Ameriks suggests that at least some of the passages appealed to in defense of the two-perspective view involve the special context of freedom, where it is important (and proper . . .) for Kant to maintain that there can be an identity between the being who appears in phenomenal

172

Notes to Pages 23–26 and empirically determined action and the one who in a transcendent way is an absolute source of this action. Such a view does not require that in general all phenomena and associated noumena are to be said to involve the very same objects. (Ameriks 1982:5–6)

Such a view poses no special problem for the arguments presented here, because my arguments make a point only about the special case of freedom. For someone who holds that in general there is a distinction between phenomenal objects and their noumenal counterparts, but in the case of freedom the difference is merely one of perspective, my arguments about the asymmetry between perspectives should be sufficient. Another option, suggested by Ameriks in conversation and by Kant in his resolution to the Fourth Antinomy (see especially A561/B589), is that while the self is a single phenomenal object, its causality is free and therefore noumenal, but not in any sense an object. This view fits into a metaphysical middle ground between standard two-object and two-perspective accounts of the self, but its development is beyond the scope of my discussion here, and it is likely that the sorts of arguments developed to show asymmetry in the case of two-object and two-aspect accounts will apply in this case as well. 18. One passage in the Critical Kantian corpus suggests some form of soft determinism. In the Canon in the Critique of Pure Reason, at the end of a discussion of human freedom through reason, Kant says, Whether reason is not, in the actions through which it prescribes laws, itself again determined by other influences, and whether that which, in relation to sensuous influences, is entitled freedom may not, in relation to higher and more remote operating causes, be nature again, is a question which in the practical field does not concern us, since we are demanding of reason nothing but the rule of conduct; it is a merely speculative question, which we can leave aside so long as we are considering what ought or ought not to be done. (A803/B831)

This passage has inspired a flurry of creative interpretations. (See for example Beck 1960:190, Carnois 1987:29, Kemp Smith 1992:569–70, and Allison 1990:55–7 for some prominent examples.) The tricky phrase is “in relation to higher and more remote operating causes,” which makes it sound as though freedom may not be the ultimate explanation of an action. What is important for the purposes of this section is that whatever view Kant held at the time he wrote this passage – and it is not even clear when exactly that was – he came to strongly reject soft determinism by the time of the second Critique. 19. In the Religion, Kant seems to take a more restrained position with respect to the possibility of one’s freedom being determined by God at an ultimate level. See 6:191. I discuss this in Chapters 5 and 6. See, too, Quinn 1990 and Mari˜ na 1997. 20. I will take up some of these issues in my discussion of the timeless moral self in Chapter 5. Wood 1984 offers a detailed account of many of the most important problems that arise from Kant’s position on the free self. 21. I do not mean to imply here that he goes against these limits. Kant’s arguments for the reality of freedom in the first Critique do not contradict his limitations on theoretical knowledge precisely because his arguments are

Notes to Pages 26–34

173

practical. But as practical arguments, they go beyond speculative reason and fit awkwardly in the first Critique. 22. In “What Is Orientation in Thinking,” Kant takes the importance of imagination even further than I suggest here. There he explains, However exalted we may wish our concepts to be, and however abstract we may make them in relation to the realm of the sense, they will continue to be associated with figurative notions. The proper function of these is to make such concepts, which are not in other respects derived from experience, suitable for use in the experiential world. For how else could we endow our concepts with sense and significance if we did not attach them to some intuition (which must ultimately always be an example derived from some possible experience). (8:133)

23. As we will see in Chapter 5, this is particularly important with respect to the notion of the “moral revolution” that supposedly takes place for the noumenal free agent.

Chapter 2 1. Brandt 1997, for example, argues that the anthropology of the lectures and Anthropology is not moral anthropology, while Stark 1997, Louden 2000, and others argue that at least part of Kant’s “pragmatic” anthropology is properly moral. I take up this debate in the next chapter. 2. Robert Louden interprets differently the significance of both the emphasis on practical application and the problems Kant raises for anthropology. Whereas I point out the ways in which these confirm the empirical nature of anthropology, Louden suggests that they point to something non-empirical about anthropology. Thus he refers to “Kant’s conviction that accurate experiments and observations of human beings are not possible” (Louden 2000:68) and claims that “the goal of an observation-based science of behavior is not one he [Kant] embraces” (67). If Louden just means that Kant’s anthropology is not based on observation of human activity in the external world but that it also includes observations in inner sense, then his account and my own are consonant. But insofar as Louden makes the stronger claim – that something beyond any observation is useful or necessary in anthropology – Louden’s account varies from my understanding of Kant’s own assertions about his project. 3. There are, of course, important moral dimensions to Kant’s discussions of the character of the sexes and the nations. Kant points out moral dangers and advantages of various national characters, discusses various virtues and vices for each sex (25:719), and points out the role that women play in morally refining men. Still, the discussions do not have the systematic place that the accounts of the character of the individual and of the species do. Kant says that the character of the individual “shows what man is prepared to make of himself,” reiterating the distinguishing feature of his moral anthropology (7:285, cf. 7:119). By contrast, the character of the sexes is focused on “Nature’s foresight” (7:303) in the distinction between men and women and ends with an apology that “I have spent longer on the subject

174

4.

5. 6.

7. 8.

Notes to Pages 34–36

of characterization [of sexes] than may seem proportionate to the other divisions of anthropology” (7:310). Kant has two remarks that suggest a similar caution with respect to his treatment of the character of nations. He says near the beginning of that discussion that “hereditary maxims . . . are only . . . attempts to classify the varieties in the natural tendencies of entire peoples more empirically for the geographer than rationally for the philosopher” (7:312). And near the end of his discussion he says that “we are speaking here about innate natural character, which, so to speak, lies in the composition of a person’s blood” (7:319). Kant’s account of character is a “classification of varieties” and reads much like his geographical accounts. This suggests, though it does not decisively show, that Kant considered his account of the character of nations peripheral. There are several nonsystematic reasons for Kant to include these accounts of the character of the sexes and of the nations, however. For one, they are important for those who will travel, or for those involved in politics or the military, to teach them “what each [nation] has to know about the other, and how each could use the other to its own advantage” (7:312). Moreover, they fit Kant’s goal, expressed in his letter to Herz, of including in his anthropology lectures “so many observations . . . that listeners . . . find the lectures entertaining and never dry” (10:146). As dated as these passages are for us today, they would no doubt have been among the most entertaining parts of Kant’s lectures in his day. Erwin Hufnagel (Hufnagel 1988:47–53) does discuss the notion of “transcendental anthropology,” but his focus is on the role that the notion of Anlage plays in Kant’s philosophy of education. Hufnagel explicitly draws much of his account from the Religion, where Kant’s most systematic discussion of Anlagen is found. Although there are some differences between Allison and Hufnagel, both argue for the a priori status of universal claims on the basis of central concepts from the Religion rather than more general concerns from Kant’s Anthropology. Others have followed Allison’s example. See Herman 1996:42–3; AndersonGold 1984; Hufnagel 1988; and Munzel 1999:135f., 170, 278. In Chapter 5, I show in more detail why it is crucial that Kant’s claim of radical evil be both universal and empirical. Here my goal is simply to show that such judgments are possible. This is discussed in more detail in Chapter 5. Similarly, Kant writes in his Reflexionen that “in order to know whether someone has a character, many observations are needed; but to know that a certain character is not present, only a single observation is required. For, since character consists in Denkungsart, namely acting from principles, a single exception is proof that the basis of the action was not a universal maxim” (RA 1230, see Munzel 1999:86). Munzel has rightly pointed out in conversation that this brings an ineliminable transcendental component into Kant’s argument. In particular, one cannot directly perceive any propensity to evil in human nature. The inference from evil deeds to evil maxims, and even more so from evil maxims to an evil propensity, is a transcendental argument. There is a sense in which “radical evil” and the “propensity to evil” are answers to the

Notes to Pages 36–40

9.

10.

11.

12.

13.

14.

175

transcendental question about the conditions of the possibility of evil in human nature. But to even motivate that question, Kant needs to have an empirical basis for the claim that there is any evil in human nature that needs to be explained. For more, see Munzel 1999, Hufnagel 1988, and AndersonGold 2001. “The claim . . . is to be taken as a priori; indeed, as a postulate, it must be synthetic a priori. Consequently, it requires some sort of deduction or justification; and since Kant fails to provide one, we must attempt to do so for him” (Allison 1990:155). Given his excellent analysis of the way in which one must incorporate incentives into one’s maxims, it is surprising that Allison in this discussion describes the influences of self-love and the moral law in such passive terms. The moral law is a stronger incentive precisely because it is taken to be stronger – that is, because it is chosen. Another example of a universal empirical claim is Kant’s theory of matter in the Metaphysical Foundations of Natural Science. See Walker 1971, in which Walker argues against those who propose that this theory must be grounded a priori. For a very different account in the realm of natural science, see Friedman’s views about the status of gravity (Friedman 1992). Friedman argues that Kant’s theory of gravity is in a sense empirical, but unlike my account of radical evil, Friedman’s account of gravity involves the claim that Kant’s theory of gravity is meant to be held with strict – rather than merely comparative – universality. See Friedman 1992:157–210. To see this in the case of the Religion, see Wood 1970:224–6. Wood does not lay out the arguments as progressing along three different lines, but he does present arguments of all three sorts. I discuss Kant’s argument for radical evil in more detail in Chapter 5 of this book. This sort of universality applies only to the most general anthropological claims. Kant often makes narrower claims about particular sorts of people – people of different sexes, races, or temperaments. These claims have a more restricted range of application than properly “universal” anthropological claims. See too footnote 3. These two difficulties are the following: 2. If [someone] wants to observe himself, he will reach a critical point with regard to his emotional state where, generally, no further concealment is possible; that is to say he is not consciously watching himself when impelling forces are in action, and that he is observing himself when the impelling forces are at rest. 3. Conditions of time and place, when lasting, result in habits which, it is said, constitute second nature, which makes man’s judgment of himself more difficult. (7:121)

Kant reiterates these later, saying that “knowledge of man through inner experience” is difficult because “one who investigates his inner consciousness easily carries many things into his own self-consciousness instead of simply observing what is there” (see 7:143). In addition to these general concerns about the difficulty of knowing oneself, Kant also suggests that certain kinds of focus on inner sense can lead to dangerous fanaticism (cf. 7:132–4, 161–2). None of these concerns, however, takes from the fact

176

15.

16.

17. 18.

19.

20.

Notes to Pages 40–46 that for Kant introspection is “of great importance” (7:143) as the primary source of anthropological knowledge. The cautions simply point out how difficult anthropology is and warn his readers away from improper sorts of introspection. Thomas Sturm has recently argued that “from at least the 1780s on, [Kant] advances a methodological claim against introspection as the primary method of knowing the human mind” (Sturm 2001:174; cf. Sturm’s “Kant und die Wissenschaft der Menschen,” unpublished manuscript). While I agree that Kant is cautious in his use of inner sense as a source of anthropological knowledge and that Kant insists upon various corrections to inner sense, Sturm goes too far in his condemnation of inner sense as a source of Kantian anthropology. As we have seen, inner sense remains primary in the published Anthropology, even if it has important dangers and must be supplemented with other sources. Literature and experience are not the only ways in which one can correct and refine one’s judgment. Cross-cultural studies – the descendants of the travel writings of which Kant was so fond – are also important. Moreover, historical-genealogical projects – such as those of Nietzsche and Foucault – can be rightly appropriated within the context of a Kantian anthropology. Kant could not go as far as Foucault in treating these sorts of inquiries as unearthing a “historical a priori” (Foucault 1970:xxii), but he can and should endorse the important empirical project of clarifying what anthropological insights are historically local. Kantians should appreciate the ways in which Nietzsche, Foucault, and others help refine anthropological judgment about what features of human beings are universal, even if Kant would insist that there are at least some legitimate universal claims about human beings. See for example Chapter 3, footnote 17, where I point out a shift in Kant’s views about the ability of humans to sustain passions for long periods of time. At least two candidates come immediately to mind. Given the structure of the arguments in Perpetual Peace, one might think that he is using the term anthropological loosely here and really means that they need to adopt a moral vantage point. If he does mean anthropology as he normally does, this higher vantage point is probably the perspective of inner experience, within which an honest politician will admit his or her own capacity for doing good. Granted, Kant’s anthropology has its own errors. As has already been noted, he saw some of these and changed some of his views over the course of his life. Other errors (about women, other races, and the deaf, for example) he never corrected. One might think here of the failure to attend to women in traditional psychology, or the consideration of women as merely an afterthought. See for example Gilligan’s critiques of Kohlberg and others in Gilligan 1982. For a different approach, with more attention to the philosophical implications of a more inclusive focus in psychology, see Noddings 1984. The shrinking global village also presents more possibilities for harm from anthropological myopia, such as interventions of the U.N., the World Bank, or various NGOs in cultures that have been excluded from consideration in the development of the psychological theories put to use to help them. Still, while

Notes to Pages 46–50

177

globalization increases the dangers of acting on myopic views, it also makes those views more difficult to sustain. Neither extreme optimism nor extreme pessimism is warranted here. People have made some very serious mistakes about what is truly universal, sometimes with devastating consequences. But many people continue to come to deeper and deeper understandings of others, which often lead to fruitful and mutually beneficial interaction. 21. With respect to the postulates, Kant sometimes – especially in the second Critique – seems to consider them merely as rational requirements for making sense of moral obligation. In this context, they need not function as aids to morality at all, much less empirical aids. At other times – for instance, in the Critique of Judgment (see 5:451–3) – Kant seems to consider belief in the postulates to function as an aid to morality, going so far as to suggest that without such belief, one will be incapable of sustaining a commitment to the moral law. One might think, though, that even in this case, the postulates are not an empirical aid but a purely rational one, the psychological importance of which can be recognized by empirical means. My own view is that in these cases Kant focuses on an empirical aid, the belief itself as a possible object of inner sense, rather than a purely rational influence; but defending that claim is unnecessary for the purposes of this book. 22. Kant does think that people can make justified claims beyond the limits of experience, but these claims cannot be based on any experiential evidence. In particular, the fact of reason gives access to one’s non-empirical self precisely because it is not a fact of experience.

Chapter 3 1. Some recent neokantian appropriations of Kant’s moral anthropology have failed to make this distinction explicit. Thus Nancy Sherman, for example, says of emotions that they are “primarily a matter of moral anthropology for Kant – a way of applying the Categorical Imperative and its a priori motive congenially to the human case” (Sherman 1990:149). I agree with Sherman that emotions are an important part of moral anthropology but add that this is not merely a matter of applying the Categorical Imperative to the human case. (Sherman herself often has a more sophisticated account of the way in which emotions fit into anthropology. I discuss her account in detail in Chapter 4.) Robert Louden is more explicit about distinguishing the anthropology needed to apply the moral law from the specific task of moral anthropology. See Louden 2000:13–14. 2. Although Kant speaks specifically of a “power of judgment sharpened by experience,” the context of this passage is an explanation of the respective roles of pure moral philosophy and moral anthropology. Thus it is clear that Kant conceived of “experience” here not primarily as general life experience but as experience that is elucidated in anthropology. 3. For an example of his use of practical to describe parts of his anthropology, see 25:1367. 4. In his introduction to the lectures on anthropology, Brandt mentions some of the points to which I turn for evidence of Kant’s moral anthropology

178

5.

6. 7.

8.

9.

Notes to Pages 50–52

but says nonetheless that “pragmatic anthropology is in none of its phases of development identical with the anthropology that Kant repeatedly looks forward to as the completion of his moral doctrine after 1770” (25:xlvi). He goes further, claiming that “pragmatic anthropology is not systematically integrated into Kant’s moral philosophy” (25:xlvii). Both of these claims are true in a sense. Pragmatic anthropology includes more than mere moral anthropology, so it is not identical to it. And Kant never explicitly points out the systematic place of pragmatic anthropology in his moral theory. But moral anthropology is included in pragmatic anthropology. Moreover it does fit into his overall moral theory, and Kant makes specific mention in his lectures on both anthropology and ethics of how it fits, even if these mentions do not rise to the level of systematic integration. Thus insofar as Brandt suggests that moral anthropology is not to be found in the Anthropology, he goes too far. Others have recently begun to do more work on Kant’s pragmatic anthropology, and some of these studies (in particular Wood 1999 and Louden 2000) include accounts of different senses of pragmatic in Kant. Although there is some overlap between their accounts and my own, my ideas were developed independently and have some important differences that should become clear in the course of the chapter. Wood, in particular, misses one of the most important senses of pragmatic (my third) in his enumeration of four senses of pragmatic. One very important sense of pragmatic that I will not discuss in this chapter refers primarily to the civilizing influence of culture on the species as a whole. This sense is not discussed specifically in Wood or Louden, but it can be found in Kant’s works. See the Groundwork (4:417fn), Anthropology (7:323 and throughout the last section), and Kant’s writings on history. Even here, Kant attaches to his description of pragmatic a footnote that makes clear his reservations about considering this a strict definition of the term. For an interesting application of a similar anthropological principle in contemporary political philosophy, see Fleischacker 1999, especially Chapter 5 on “Proper Pleasures.” This is not to say that Kant actually thinks that we are determined in matters of prudence. But his argument for our freedom implies only that we must be free to act in accordance with the moral law and thus does not directly show any capacity to choose for or against dictates of prudence. Still, even when one is simply choosing between two morally acceptable alternatives, one ought (and therefore can) always make this choice only once one is assured that the alternatives are in fact morally acceptable. In that sense every choice is free. Moreover, once one accepts that one is free to choose the moral law over some particular preference, there is no reason to doubt that one can freely choose between sensuous interests, even when both are morally acceptable. But this latter sort of choice is not required by Kant’s argument. For more on Kant’s theory of prudence, see Kain (in press) and Kain 1998. As noted earlier (see footnote 4), Brandt does not take this to allow for a significant moral dimension in pragmatic anthropology. Stark, by contrast, insists that the anthropology is systematically linked with Kant’s moral philosophy. See Stark 1997.

Notes to Pages 52–58

179

10. For more on the relationship between Kant and Platner, see Sturm, “Kant on the Human Sciences” (unpublished manuscript), and Sturm, “Kant und die Wissenschaften vom Menschen,” unpublished manuscript. Sturm also discusses the relationship between Kant’s anthropology and the disciplines of rational and empirical psychology. For this discussion, see too Sturm 2001. It is important to recognize, in this context, that Kant’s distinction between pragmatic and merely theoretical anthropology does not merely distinguish his anthropology from Platner’s particular brand of psychology, but more generally from any version of rational psychology, and many versions of empirical psychology. 11. The other purposes are: 1. The better we know people, the better we know how to arrange our actions so as to be fitting with theirs. 2. It teaches how one should win [against] people. 3. It teaches self-satisfaction, when one has for oneself the good that one finds in others. (25:734)

12.

13.

14. 15.

All of these uses can relate to morality. The first clearly aids in the fulfillment of imperfect duties to promote the happiness of others and conduct oneself without interfering with others. The second relates to one’s ability to influence others and will be discussed next as a distinct sense of the term pragmatic. The third might seem to imply that by observing others one can come to see what promotes their happiness and thereby acquire that oneself. But given Kant’s association of self-satisfaction with moral goodness (see 5:38, 117–18; 6:391), one might reasonably read it instead as suggesting that anthropology teaches one how to acquire for oneself the moral good that one finds in others. Although the context here, where pragmatic seems to be distinguished from moral and technical, might lead one to read this prudence in the first sense (as related to happiness), the fuller context of the passage makes clear that it has more to do with influence on others, which could be used to any practical (including moral) ends. Kant also mentions another use of the term prudence. When used in the sense of “private wisdom,” prudence refers to “the sagacity to combine all these purposes for one’s own lasting advantage” (4:416fn.). This sense of prudence simply identifies it with skill in promoting one’s happiness. A pragmatic anthropology that developed this skill would be pragmatic in the first sense covered above. While I do not deny that this sense of pragmatic is at play in Kant’s anthropology, it is not the primary sense of prudence there, as the rest of this section will show. For more on politeness, see Brender 1997 and 1998. Kant gives numerous examples of how this illusion is supposed to work in the Anthropology and lectures, but his focus in those books is on distinguishing this “illusion” from deceit, in order to provide the anthropological resources for claiming that politeness can be morally acceptable. Basically, the distinction is that the effect of deceit depends on getting another to actually believe the content of the deception, whereas illusions depend for their effect on an

180

16. 17.

18.

19.

20.

21. 22.

23.

Notes to Pages 58–65 appearance, such that the effect is the same whether or not one believes the illusion to be true. For more on this, see Brender 1997 and 1998. This example has been discussed in much greater detail in Frierson 2000 and Sorenson (In process). Kant’s accounts of affects and passions change over the course of his lectures in anthropology. Specifically, in his early lectures (e.g., the 1776 Friedlander notes, 25:589–91, 612), neither passions nor affects are lasting conditions. In later lectures (e.g., the 1789 Busolt notes, 25:1526) and the published Anthropology, one of the most important differences between passions and affects is that passions last and affects pass quickly. In my account of passions and affects, I focus on the account in the Anthropology and supplement it with complementary details from the lectures. For more on the differences, see Frierson 2000. There are, of course, important moral dimensions to Kant’s discussion of the character of the sexes and the nations. The character of the races is simply too short for Kant to make much comment in the way of its moral relevance. But Kant points out the moral dangers and advantages of various national characters, and with respect to the character of the sexes, Kant discusses different virtues and vices for each sex (e.g., at 25:719), prescribes certain sorts of relations between the sexes, and points out the moral role of women in refining and morally improving men. This is a variant reading (see note 4 in the Academy edition, 25:631) that makes more sense in the context. The Friedlander lectures actually read “hard wax” here. Another important social help to the development of character is culture. Kant’s accounts of the importance of culture are linked with his discussions of politeness (mentioned earlier) and with his more general accounts of the development of the human species. Munzel 1999 examines this in detail. See Chapter 6 for more on Kant’s view of historical progress and its relation to his anthropology. He never lectured on ethics without also lecturing on anthropology during the same year. For more on the circumstances of Kant’s lectures, see Brandt 1997 and Stark 1997. In fact, in his review of Kant’s Anthropology, Schleiermacher suggests that there is little in it that is not merely peripheral. He calls the Anthropology a “collection of trivia” and a “clear portrayal of peculiar confusion.” Kant’s anthropology is particularly prone to this criticism in part because, as he proudly announces when he first writes about his course in anthropology, he includes “so many observations of ordinary life that listeners [of his lectures] have constant occasion to compare their ordinary experience with [his] remarks and thus, from beginning to end, find the lectures entertaining and never dry” (Kant’s letter to Herz, 10:146). This desire to make his lectures popular infects even Kant’s published Anthropology, and so one frequently finds tidbits of anthropological information that hold one’s attention but that do not really deserve the attention they receive. The extensive explanations of the character of nations and of the differences between men and women probably fall into this category. In addition to making the

Notes to Pages 65–70

181

Anthropology somewhat jumbled in general, these accounts can often obscure the particular systematic place of moral anthropology. 24. There is one clear sense in which all of pragmatic anthropology relates directly to morality. Pragmatic anthropology teaches one how to develop all of one’s faculties (of cognition, feeling, and desire). The cultivation of all of these is an imperfect duty, the duty of self-perfection. So all of pragmatic anthropology is morally relevant in that sense. And of course, one of the most important examples I gave – politeness – came from the middle of the section on cognition, so there is no need to rule all of Kant’s remarks there out of the scope of moral anthropology strictly speaking. But this can by no means be said of the whole of the Anthropology. 25. Kant’s own explanation of why he combines his discussion of affects and passions is in a note in the section on feeling: “Note: in this section we also ought to consider the affective excitement as a feeling of pleasure and displeasure, which transgresses the limits of man’s inner freedom. But since it is frequently confused with the passions, which will be discussed in another section, that is, the one on the faculty of desire, and since it is really closely related to desires, I shall address myself to the affective excitement at the proper time in the third section” (7:235).

Chapter 4 1. To be fair, neither Herman nor Sherman is specifically concerned with reconciling Kant’s anthropology with his theory of freedom. Nancy Sherman, in particular, explicitly says, in a comment at the beginning of Making a Necessity of Virtue that applies to all her work on Kant, that she will “not talk about the thorny problems of Kant on freedom” (Sherman 1997a:21). Moreover, although Sherman clearly sees her work on Kant as part of a revival of interest in “what Kant would call moral anthropology” (Sherman 1995:369), her conception of moral anthropology is considerably broader than Kant’s own. Sherman seems to realize the difference between what she means by moral anthropology and what Kant means by it. When she discusses Kant’s own explicit description in the Metaphysics of Morals of the distinction between moral anthropology and a metaphysics of morals, Sherman suggests that “this [distinction] misses the force of Kant’s own fuller theory” (1997a:134). Sherman finds fault with Kant here because Sherman’s own general conception of moral anthropology is so broad that it includes any anthropological considerations that are needed to apply the categorical imperative to human situations. I argued in Chapter 3 that it is more appropriate to read moral anthropology in a narrower sense, dealing only with helps and hindrances to having a good will. But Sherman generally uses the term more broadly. Sometimes, however, Sherman does seem to include the emotions within moral anthropology in the narrower sense (see footnote 3). 2. Louden even claims (see fn 57, p. 198) that Wood influenced his views. For further parallels, see below, footnote 22. 3. But see above, footnote 1. In my treatment of Sherman through the next three sections of this chapter, I look at her recommendations about the

182

Notes to Pages 70–73

roles that emotions can play in moral life as possibilities for the role of helps and hindrances more generally. As I argue in Chapter 3, Kant conceives of passions and affects, which Sherman sometimes includes among the emotions, as important hindrances to moral life. Sherman focuses, however, on emotions as empirical aids to moral choice. Moreover, although her conception of moral anthropology is often broader than mine, at times Sherman clearly does recognize the role of emotions in moral anthropology narrowly construed as providing subjective conditions that help or hinder having a good will. She says, for example, The emphasis in The Doctrine of Virtue is on who will act in morally worthy ways from a pure attitude of virtue. And the thought is that what we can do to increase our chances to be part of that pool is to cultivate emotions that do not battle with our duty motive and that positively promote it. Such emotions are not themselves expressive of the purest attitude of virtue in that they are not the ultimate source of adequate reasons for doing what is required or for determining what is morally permissible. But they are a layer of character that can, nonetheless, best support moral motivation. (1997a:144; cf. Sherman 1997b:272 and Herman 1993:13)

In this passage, emotions that support moral motivation are connected explicitly with increasing the “chances” that one will be morally good. Emotions help and hinder having a good will. Insofar as they have this role, it is reasonable to insist that Sherman explain how they can serve that role. In that context, it is proper to look to her recommendations regarding the place of emotions in moral life in order to find explanations of how one class of helps and hindrances promotes the good will. 4. See Sherman 1997a:144 and 1997b:273. I focus on the formulations in Sherman’s most recent work. Here she outlines three roles for the emotions but commits Kant to only two of them. She is more ambivalent about the third, the “motivational” role, as a matter of Kant interpretation, though she clearly thinks that it should play a role in a good moral theory. In an earlier essay, “The Place of Emotions in Kantian Morality” (Sherman 1990), Sherman outlines five different roles for the emotions. Three of these are the same as those in the later essays, though the motivational role is described in terms of emotions under a “regulative constraint” (Sherman 1990:161). The two roles discussed in that essay but dropped in her more recent work are the role of emotions as a “provisional morality” (Sherman 1990:158) and as an “aesthetic ornament” (160). Sherman does not reject either of these functions in her later work and even (in Sherman 1995) reiterates her commitment to the role of emotions in provisional morality, but they seem to play a less significant role. 5. Of course, this does not mean that all such failures are free of moral fault on the part of the agent. In the United States today, for example, it is conceivable that many people could innocently fail to recognize that referring to women as “girls” is demeaning, but it is hard to imagine a white adult who innocently thinks that there is nothing wrong with referring to African-Americans as “niggers.” 6. Sometimes Herman seems to think that greater epistemic awareness is all that is necessary for people to be morally good. Thus her account of character

Notes to Pages 73–78

7.

8. 9. 10.

11.

12. 13.

183

development is primarily an account of how one comes to judge rightly, and to have the right desires, and she explicitly denies that the cultivation of a good will involves developing any sort of moral muscle (see Herman 1996:43; for a response to Herman, see Brender 1997:39n17). Herman considers the possibility that emotions can play these roles (see Herman 1993:82), but her overall treatment is more general and focuses more centrally on moral education – another important empirical influence. See Chapter 3. In fact, as I argue in Chapter 5, Kant does think this! Again, I argue in Chapter 5 that Kant himself struggles with the latter option. Seeing that he is evil, Kant must find some way to combat the moral despair that seems to follow from this recognition. Sherman connects this role of the emotions with politeness in particular. (See 1997a:147f; 1997b:273f.) In this, she seems to confuse the importance of emotions for carrying out particular duties, say of benevolence, and the particular role of politeness as a part of moral anthropology. It is certainly true that service with a smile is more conducive to the happiness of others than begrudging fulfillment of one’s duty to help others. But when Kant discusses the particular importance of the graces in making virtue attractive, he has in mind (as indicated in the previous chapter) the way in which the graces promote the virtue, and not the happiness, of oneself and others. Sherman notes this difference in her attention to the “educative role” (1997a:148; 1997b:274) of emotions, but she fails to offer an account of how such an educative role is possible. Louden, by contrast, reads Kant’s account of temperament to imply sharp limits on one’s control over emotions. See Louden 2000:81. The importance of distinguishing between striving for certain emotions and actually having those emotions is emphasized in Baron 1995. She concludes with a wonderful summary of what an agent can be held accountable for, even if not for emotions themselves: In short, it may be that merely having the feeling is not something that Kant sees to be morally objectionable – or at least something for which the agent may legitimately be held responsible. But he does recognize responsibility for harboring, cultivating, or failing to cultivate certain desires or feelings and for taking desires to be reasons to act accordingly. (Baron 1995:199)

14. As noted in footnote 11, Sherman’s treatment of politeness in the context of the instrumental role of emotions is particularly confusing in this context. The “educative role” of the graces is extremely problematic, but Sherman seems to think that it can be explained in ways similar to the way in which the expression of sympathy and compassion serves to help comfort a crying child. 15. The cultivation of sensitivity to the pains of others could serve even evil ends in one who seeks to torture others for pragmatic purposes or who just enjoys the pain of others. 16. Of course, all kinds of caveats are necessary here. If one needs to work fourteen hours a day to feed one’s children without violating the rights of

184

Notes to Pages 78–81

those around one, and if one is naturally insensitive, such that the cultivation of sympathy would be very difficult and time consuming, it may be morally irresponsible to cultivate one’s emotional capacities. One who lacks basic necessities need not seek out the sickroom or the debtor’s prison in an attempt to share in the pains of others. 17. In the previous three sections, I have had to glean general accounts of the role of empirical influences from contemporary Kantians who are concerned with more particular issues. By contrast, Louden is explicitly concerned with problems of integrating moral anthropology into Kant’s account of freedom. Kant’s Impure Ethics (Louden 2000) offers an explanation and integration of Kant’s impure ethics into his moral philosophy. This integration is not the primary purpose of the book (as Louden has made clear to me in correspondence). Louden’s emphasis is on pointing out the moral importance of Kant’s work in anthropology, as well as related work in aesthetics, religion, history, and pedagogy. But even if the primary purpose of the book is to draw attention to aspects of Kant’s moral philosophy that have not been sufficiently discussed in other accounts of Kant’s ethics. Louden also explicitly addresses the philosophical problems with which this book is concerned. He explains that impure ethics has “the aim of locating factors that will help or hinder the development and spread of morality within human life” (Louden 2000:13), and he even explicitly raises the problem of reconciling moral anthropology with freedom: “A deeper doubt is whether the idea of a moral anthropology ‘to which empirical principles belong’ is itself consistent with Kant’s own dualistic views concerning transcendental freedom and nature” (Louden 2000:7). He asks, “What real impact can the various empirical precepts of practical anthropology and moral education have? Won’t they affect only our empirical character, and not our intelligible character?” (19). In response, Louden suggests that “Kant did not address these issues, and in order for me to do so it will be necessary to offer conjectures that occasionally go beyond his texts” (Louden 2000:19). Louden’s account, then, is explicitly designed to solve the problem that I address in this book and that Kant supposedly left unsolved. 18. Except for the term indispensable on page 116, all italics are my own. 19. One aspect of Louden’s view that I will not discuss in detail is a particular suggestion about how empirical aids act as necessary preparations for morality. Louden occasionally explains that certain empirical influences help by making the moral law more accessible to the senses. Thus he says that “aesthetic experience offers human beings tangible access to concepts of pure ethics, and we need this access if we are going to make ethics comprehensible” (Louden 2000:116), that “a large part of the project of impure ethics consists in finding concrete ways to make the message of pure ethics graspable to human beings” (127) and that “Kant’s account of historical progress is of a piece with the larger tapestry of impure ethics; it is yet another way of making morality graspable to human beings” (152). This notion of making morality “graspable” is similar to suggestions made by Paul Guyer (see Guyer 1993) about the way in which aesthetic feeling can facilitate moral feeling. With respect to this particular way in which empirical influences can play a

Notes to Pages 81–82

185

role as preparation for morality, there are two important problems. First, it is not clear why a clearer grasp of morality will make it more likely that one will choose good rather than evil. At most, it might be necessary to make it possible for one to recognize duties, which is necessary to be a moral agent at all (as noted in section 1, above). These influences could play an epistemological role that enables moral responsibility, but there is no explanation of how they facilitate virtue. In his discussions of aesthetics in particular, Kant seems more interested in describing how beauty functions to give people a better awareness of morality (something cognitive), rather than a greater commitment to it (something genuinely moral). Kant may – and almost certainly does – think there is a connection between these, but Louden never explains what this connection is nor how it is consistent with transcendental freedom. 20. One could of course hold that earlier generations are morally responsible and that these conditions really are necessary, if one denied that “ought implies can.” Louden might want to do this, but if so, he is giving up the core of Kant’s moral theory. 21. He mentions two other possible objections, but they relate less directly to the issue of the relationship between freedom and empirical influence: the problem of “perfecting others,” and the suggestion that there is a kind of hubris – hard to justify after the twentieth century – in claiming that we are morally better than previous generations. 22. Louden footnotes Allen Wood for “discussion on this topic” (fn 57, p. 198). This is no surprise, because Wood’s recent book (Wood 1999) falls into many of the same problems that Louden’s does. Throughout most of his discussion, Wood is more careful (or more ambiguous). He can be read as suggesting not that social and historical forces are necessary conditions of individuals’ coming to have good wills, but only of moral responsibility in general or the moral progress of the species as a whole. But at times he seems to suggest something stronger, as when he says that “Kant’s ethical thought is fundamentally about the human race’s collective, historical struggle to develop its rational faculties and then through them to combat the radical propensity to evil that alone makes their development possible” (Wood 1999:296). Following this bold summary of Kant’s ethical thought, Wood suggests that history can be divided into an “epoch of nature” that “continues today” (Wood 1999:296) and an “epoch of freedom,” to which “all ethical duties belong” (325). Thus Wood finally says, “Kant does not think that I can ever achieve this inner revolution [from evil to good] entirely on my own. The origin of evil is social, and so must be the struggle against it. . . . The pursuit of my own morality can be distinguished from the moral progress of the human species, but . . . the two ends are necessarily linked in their pursuit” (Wood 1999:314). In Chapter 6, we will see that there is a sense in which the moral revolution by which one has a good will is necessarily linked with one’s actions to promote progress in the species. But whereas Wood, like Louden, sees the origin of this link in the necessary social preconditions of “my own morality,” I argue that social influences must be considered expressions of a good will. Wood, like Louden, needs some account of how these

186

Notes to Pages 82–96

influences can cut through the causal network to affect one’s freedom, and this account will get him into the same problems that Louden encounters directly. 23. Kant does recognize that there are some historical prerequisites for moral responsibility. He discusses some of these in his essay “Conjectures on the Beginning of Human History,” discussed in Chapter 6. He also presumably can allow that some human beings, such as infants and the insane, are not morally responsible. But most of the aids to morality he discusses are not aids of this sort. 24. Munzel makes use of a similar argument. She says, for example, “the pedagogical process complete with propaedeutic functions that prepare and develop these aptitudes, even awakening them to a state of moral responsiveness, is deemed by Kant to be essential ” (Munzel 1999:340–1, emphasis added). Recognizing the potential threat to freedom, she then explains that “however many external agencies are involved (nature, parents, teacher, philosopher, constitution) on the behalf of human moral purpose, individual freedom is not jeopardized. The others’ efforts of discipline and cultivation stop short of the indispensable act of formation,” which can be effected only by free choice (Munzel 1999:341–2). Munzel even uses the language of a “necessary (albeit not sufficient) condition” (Munzel 1999:340). Insofar as Munzel takes these passages to save freedom in the light of empirical influences, my criticisms of Louden apply to her as well. In the next section, I argue that she need not fall into this danger. 25. Ironically, the main metaphor I do not discuss (because it is not as fruitful) is Munzel’s detailed horticultural metaphor. Munzel runs into several of the problems that I raised with Louden’s account in the course of her analysis of the horticultural metaphor. For one thing, empirical aids are represented by the horticulturist and his activity, which activity is absolutely essential for the success of the grafting process, which represents moral formation. This leads Munzel into thinking of empirical aids as necessary but not sufficient conditions of moral development, a mistake to which her other illustrations are less susceptible.

Chapter 5 1. For those who prefer two-aspect accounts of the will, this could be explained as the distinction between the empirical aspect of the will and the free aspect of the will. For those who prefer two-perspective language, the distinction is between the will considered as empirically determined and the will considered as free. 2. In part, this footnote is a response to criticisms raised in the first reviews of the second Critique. See for example reviews by Pistorius, Rehberg, and others. Several of these are collected in Bittner 1975. Rehberg’s criticism is found in Schulz 1975. 3. I follow Pluhar’s translation here, though with some reservations. Pluhar interprets jener in the German to refer to nature, but it could also refer to freedom. Ultimately, the reference of this term does not affect my overall

Notes to Pages 96–102

4.

5.

6. 7.

187

point, because Kant clearly says that the appearance of the self is subject to both the laws of nature and the laws of freedom. This partially explains why most contemporary “Kantian” solutions to the problem are quite different from Kant’s solution. Following the example of Rawls, many contemporary Kantians are deeply resistant to making use of this distinction (or any other aspects of Kant’s metaphysics) in their moral theories. Again, although Kant’s language here suggests a two-object account of the self, one could translate it into two-aspect or even two-perspective terms. But one cannot get around the sorts of metaphysical considerations I discuss in Chapter 1. In particular, one must take seriously the nature of the asymmetry between nature and freedom. There are metaphysical objections to this account of freedom and nature, and these have been discussed by Wood, Allison, and others. As noted in the Introduction, I am primarily concerned with the practical problems raised by moral anthropology. Christine Korsgaard has described this very clearly, though without sufficient attention to the role of situation, in Korsgaard 1996a:75. This notion of “expression” has some similarity to Charles Taylor’s notion of expression in his discussion of “The Expressivist Turn” in The Sources of the Self. Taylor explains, To express something is to make it manifest in a given medium. I express my feelings in my face; I express my thoughts in the words I speak or write. I express my vision of things in some work of art, perhaps a novel or a play. . . . But to talk of ‘making manifest’ doesn’t imply that what is so revealed was already fully formulated beforehand. (Taylor 1989:374)

There are two crucial differences between this Kantian notion of expression and Taylor’s Romantic notion. First, while expression in a Kantian sense does not imply that what is revealed was preexistent, this is not because the expression “shapes” what it expresses (Taylor 1989:376) but because what is expressed is necessarily something expressed, such that it “exists” only as something expressed. Second, Romantic “expression” is the expression of a highly individual self, such that one can express any of a vast number of possible “selves” and thereby define oneself in any of a vast number of different ways. By contrast, for Kant one is fundamentally either good or evil, and this moral status is what is expressed in one’s life. While the expressions may be different depending on the situations in which one finds oneself, the underlying moral disposition is, for each of us, either good or evil. 8. The restriction to voluntary actions is meant to exclude reflex actions or actions that are the results of affects. The restriction to morally relevant actions excludes considering eating one flavor of ice cream over another to be expressive of one’s free will. Such nonmoral choices might be free, but Kant is not committed to considering them expressions of one’s free will. 9. As Kant says, “the human being knows how to distort even inner declarations before his own conscience” (“On the Miscarriage of all Philosophical Trials in Theodicy,” 8:270). In this attention to self-deception, Kant fits into a tradition

188

10.

11.

12.

13.

14.

15.

16.

Notes to Pages 102–107 that includes Bishop Butler, Adam Smith, and Rousseau. For more on the influence of Smith on Kant, see Fleischacker 1999 and 1991. For more on the influence of Rousseau, see Velkley 1978 and 1989. It is particularly important to point out Kant’s awareness of this problem of self-deception in the context of those who criticize Kant for misunderstanding the complexities of maxim formation. In fact, Kant was well aware not only of the possibility that the same actions could be represented under different maxims but also of the human tendency to use this fact as a means of justifying evil deeds. Of course, one’s imperfect duty of self-perfection does require promoting helps and preventing hindrances to morality. But this requirement is due to the moral effects that these helps and hindrances have independent of their place as objects of an imperfect duty. Unless helps and hindrances affect morality in a way other than being objects of an imperfect duty, the imperfect duty to promote them cannot get off the ground. (See my criticism of Sherman in Chapter 4, section 3.) For other reasons as well, the Religion is a good place to look for an account of the relation between the good will and its appearances that can make sense of the importance of moral anthropology. The first half of the work involves an extensive discussion of the nature of the good will and its appearance, and Kant moves in the second half to his most extensive discussion of any single empirical aid to morality – ethical community. The first and second Critiques and the Metaphysics of Morals include significant accounts of the relationship between freedom and its appearance in the world. The Anthropology and essays on history include extended treatments of anthropological considerations. But the Religion most clearly combines an account of nature and freedom with an account of helps and hindrances to morality. This limitation of scope applies primarily to those discussions of inscrutability that occur in a practical context. In theoretical discussions, where Kant argues that one cannot have theoretical knowledge of the self, he does not restrict his scope to knowledge of one’s goodness. One cannot know theoretically whether one is good or evil. For more on Kant’s empirical argument for radical evil, see Chapter 2, section 2, where I respond to the objection that Kant’s argument cannot be empirical because radical evil is supposed to be universal. Here I argue against a different objection, that Kant’s account of radical evil runs afoul of limitations on theoretical knowledge of the self. One could read this passage in a slightly different way. In particular, when Kant says that he omits “all actions recognized as contrary to duty,” he could just mean that an action that is recognized by a particular agent as contrary to duty cannot be performed by that agent from duty. This would not require his positing any inherently wicked actions. Even in this case, however, whether or not an action is recognized by an agent as contrary to duty is an answerable empirical question, at least by that agent. That said, Kant is rightly concerned about the tendency to dwell on one’s weaknesses in a way that undermines hope in the possibility of moral reform. See Brender 1997 and 1998.

Notes to Pages 107–110

189

17. Kant’s claim that humans are radically evil applies to both individuals and to the species as a whole, but in this section I focus entirely on its application to individuals. Although Kant is interested in the role of evil at the level of the species, for his moral philosophy I take the individual case to be fundamental. What matters is that I, as a human being, am evil, and thus I need to struggle against that evil. Even his argument for the importance of social solutions to the problem of evil is rooted in the fact that each human being recognizes that evil is furthered through bad social influences. This is discussed further in Chapter 6. For accounts that focus on the social dimensions of evil, see Wood 1999 and Anderson-Gold 2001. 18. There is one exception. Insofar as one acts from affect, one’s will is not involved. Hence it is conceivable that one could perform “actions” contrary to right without having an evil will. Unless one has lived an exceptional life, however, it would be incredible to ascribe to this cause all one’s acts that are contrary to right. For most of us, to do so would verge on denying one’s own freedom. See Chapter 3 for more on affects. See too Frierson 2000 and Sorenson (in process). 19. Throughout, it is important to remember that for Kant self-love can include actions that are normally considered altruistic. The point is simply that one does what is in one’s sensuous interests, either in that it immediately satisfies some desire that one happens to have or that it is conducive to the satisfaction of desires at some later date. Thus someone who acts from a good-hearted desire to see others flourish is acting from “self-love” in this Kantian sense. 20. Several recent commentators (see, e.g., Allison 1990:146–61 and Wood 1999:283–5, but contrast Munzel 2000) misunderstand Kant’s insistence that all humans have a propensity to evil as the end of the story for Kant. In fact, the propensity for evil, because it is our own fault, is itself an indication of actual evil in human nature. Only in this context can one make sense of the problems that Kant raises in his discussions of grace, and in particular in his account of substitionary atonement. 21. Even in the Anthropology, when Kant points briefly to the fact that humanity is trapped in a “labyrinth of evil,” he quickly points out that it wanders into this labyrinth “through its own fault” (7:326). 22. Kant does not make clear the extent to which the notion of prior experience is relevant to the propensity to moral evil. That is, he does not say that one must have some prior experience of violating the moral law for such violation to be incorporated into one’s maxims. Moreover, it is hard to imagine that such an experience could be a necessary precondition. Whereas it is possible to experience an object for which one has no particular prior desire, one could not similarly experience moral evil, which is a matter of choice, without actually choosing it. Thus the conditions of actualization of the propensity to evil must be different from the conditions of actualization of other propensities (for objects of potential inclination). What is relevant, for Kant, is that human beings have a potential for deviation from the moral law, a potential that is actualized in them given appropriate conditions. 23. For more on this, recall the discussion of Allison in Chapter 2. The “necessity” that one with a propensity to evil will perform evil is not a logical necessity, nor

190

24. 25.

26.

27.

28.

29.

30.

31.

Notes to Pages 110–114 a morally relevant necessity, but akin to a natural necessity. Unlike a natural necessity, the propensity to evil reflects a commitment of choice rather than something out of one’s control, and in that sense one brings it upon oneself that one will do evil, in addition to merely bringing it upon oneself that one does evil. This is moral antecedence, not temporal. The choice of the person is responsible for the propensity. There may be degrees of this frailty, but for Kant the degrees are less significant than the fact of frailty itself. Thus some may be able to act on their adopted maxims much of the time but occasionally falter. Others may abandon principle for the slightest reasons. In both cases, one is incapable of being morally good, because goodness requires an unbending commitment to moral principles. However, frailty to a lesser degree may be less dangerous in that it is easier to overcome. The more consistent one is, the more one can make use of that consistency to further combat one’s frailty. If I cannot stick to any principles at all, it will be very difficult to improve. If I stick to principles for the most part, I can also stick for the most part to principles that will strengthen my character itself. See Ameriks 1989 for more. Self-deception is neither a necessary nor a sufficient condition of impurity. There are forms of self-deception that do not involve moral motives at all, and there are forms of impurity that are not rooted in self-deception. But self-deception is typical of the way in which one seeks to cultivate impurity in oneself. Kant explains in his arguments for rigorism in the Religion (6:22–5) that moral status depends upon one’s life as a whole. I discuss this throughout the rest of this chapter. There are obvious connections between Kant’s account of evil and the Christian doctrine of original sin. Although I do not read the Religion as merely a rational reconstruction of Christian doctrine, the similarities are undeniable. For more on this theme, see Quinn 1984, 1986, and 1990; Adams 1998; and Hare 1996. The one exception to this general approach is in the Groundwork, where Kant (in the first two sections) consistently brackets the question of whether one is bound by morality at all and (in the last section) “proves” that humans are morally obligated by showing that they are free. Between the Groundwork and his second Critique, however, Kant abandons this approach in favor of the “fact of reason,” according to which moral philosophy simply takes for granted the fact of moral obligation and explores its conditions of possibility. See Ameriks 1981 for an explanation of this transition. There are important differences between Kant’s defense of freedom and his defense of the postulates of God and immortality. See Rawls 2000:317–25 for a discussion of these differences. Although this account of grace is not central to my account of moral anthropology, it is important to be clear about the significance of grace for Kant. Without recognizing the importance of grace, it is easy to think that moral anthropology and a revised conception of the human good will provide Kant’s entire solution to the problem of radical evil. Natalie Brender,

Notes to Pages 114–116

32.

33. 34.

35.

191

for example, claims that “unsatisfactory as it is in the context of rational religion, the hypothesis of belief in divine aid is manifestly unavailable in the context of secular discussions” (Brender 1997:54). She seems to think that Kant is interested in carrying out a “secular discussion” in his solution of the problem of radical evil. This conception of Kant’s project leads Brender to make stronger claims on his behalf than Kant would make, such as the claim that “we must also remain convinced that we can overcome” our will’s corruption (Brender 1997:92, emphasis added). Thus she goes so far as to say that “Kant’s solution to this crisis of belief and motivation [caused by radical evil] is to direct an agent’s attention away from the impossibility of inner validation toward the sort of validation to be found in outer experience” (Brender 1997:118). Although this kind of attention to polite society provides an important part of moral anthropology, it is crucial to recognize that for Kant these sorts of anthropological considerations are only part of the story. Because he has a role for divine grace, Kant does not need to lean as heavily on his altered conception of the human good will to escape the problems raised by radical evil. Because his account is not “secular” in Brender’s sense, Kant has some problems that Brender does not have. But moral anthropology also carries less of a burden for Kant than for Brender. While it may be worthwhile to attempt a purely secular solution to the problem of radical evil – and my account in the next section provides resources for such a solution – it is important to realize that Kant did not make such an attempt. Several commentators have discussed Kant’s account of grace in more detail. See Mari˜ na 1997; Adams 1998; Quinn 1984, 1986, and 1990; and Hare 1996. Ultimately, I take seriously Kant’s claim that “only common morality is needed to understand the essentials of this text” (6:14), and so I see Kant’s account of grace as first and foremost “anthropological, not theological,” to use Allen Wood’s phrase in a slightly different context (see Wood 1999:291). Still, many of the issues that arise in the context of specifically Christian accounts of grace arise for Kant’s as well, as Mari˜ na, Adams, Quinn, and Hare point out. And this makes sense, given Kant’s claim that his religion is a “pure religion of reason” that is supposedly contained within the “wider sphere of [Christian] faith” (6:12). Because I am primarily concerned with how grace makes possible the account of the good human will that I develop in the next section, however, I do not discuss these theological issues in detail. See too 6:45, 6:61, and 7:328, where Kant makes clearer the connection between our duty to morally improve and our need for divine grace. Mari˜ na distinguishes these two different sorts of grace (as well as one other) in Mari˜ na 1997. Kant distinguishes between these notions of grace throughout the Religion (see, e.g., 6:143) and in the Conflict of the Faculties (see 7:43–4). John Hare nicely summarizes the challenge: If we want to keep morality as demanding as Kant says it is, and if we want to concede what Kant says about our natural propensity not to live by it, and if we want at the same time to reject these traditional Christian doctrines [of grace], then we will have to find some substitute for them. (Hare 1996:37)

192

36.

37.

38.

39.

Notes to Pages 116–123 Hare does not include (1) in this description, although denying (1) is a way to avoid Christian doctrines or their substitutes. One can simply admit the rigor of morality and our propensity not to live by it but claim that we have freely fallen short and now are morally evil without any further hope of reform. This would be an option, but it is not one that Kant is willing to accept. For some recent versions of this option, see Scheffler 1992 and Kagan 1998. The lectures on religion both before and after the Religion emphasize the inscrutability of atoning grace and give no further details. In the Religion, although Kant does point out that grace is ultimately inscrutable, he offers some details about the way in which atoning grace works. See Quinn 1986 for a discussion of these details. Even in the Religion, sanctifying grace is thoroughly mysterious. The Religion was published in parts during 1792–3. The Conflict of the Faculties was written, at least in part, by December 1793 (see letter to Kiesewetter, Ak. 11:456), but it was not published until 1798. This interpretation contrasts with Storr, who drew closer parallels between the postulates and grace, though see Rawls 2000, Chapter 10, for a defense of the claim that even the postulates of God and immorality depend on practical concerns that are not essential for pure practical reason. See Chapter 1, especially footnote 12, for more. Hare uses an argument of this sort to motivate Kant’s account of grace. As Hare explains, On Kant’s principles it is perfectly fair for him to refuse to give an account of how supernatural assistance might work to enable a person who is not naturally good to make herself a good person. It would indeed be inconsistent for him to presume to give such an account, which would go beyond the proper limits of our understanding. But it is not legitimate, on Kant’s principles, simply to say that a bad person’s becoming good is simply possible and leave it at that. This is not legitimate because it leaves us with an antinomy, with the apparent conclusion that the revolution of will both is possible (because obligatory) and impossible (because the ground of our maxims is corrupt). (Hare 1996)

A concept of grace is necessary to show that this apparent contradiction is only apparent. Of course, this concept of grace raises problems of its own and must be shown to be coherent before it can legitimately be used to resolve Kant’s antinomy. 40. Later philosophers and theologians do attempt to offer such a Critique. In different ways and with different debts to Kant, Fichte (especially in his Critique of All Revelation), Schleiermacher, Novalis, and Kierkegaard all articulate perspectives within which grace can be described. More recently, John Hare (Hare 1996) has offered a philosophical analysis of the problems and some attempted solutions, raised by the “moral gap” between the demands of morality and the problem of radical evil. 41. For more details, see Hare 1996, Quinn 1984, Mari˜ na 1997, and Adams 1998. 42. Kant does not use the phrase “will in revolution” in the Religion. However, the revolution that he describes is a revolution that takes place in the noumenal

Notes to Pages 123–134

43.

44.

45.

46.

47.

48.

193

will, whereby a will “becomes” a good will. Because this revolution is nontemporal, however, it is appropriate to think of it as the status of the will rather than as a change that happens to it. For some reason, Wood and DiGiovanni translate Denkungsart as both “mode of thought” and “attitude of mind,” even in the same sentence. This serves to obscure important connections in Kant’s line of thinking. I translate Denkungsart as “mode of thought” throughout. For an extensive discussion of Kant’s notion of Denkungsart, see Munzel 1999. G. Felicitas Munzel seems to take these passages to describe the same revolution that Kant discusses in the Religion. See Munzel 1999:27n13, 160–4, and 330. As will become clear, I disagree with this reading. This conception of establishing character is not something that Kant comes up with only after writing the Religion. Throughout his anthropology lectures Kant reiterates the theme. In the Friedl¨ander lectures of 1775–6, sixteen years before the Religion was published, Kant is reported to have said, “character is fixed very late, around [ohngefehr] the fortieth year” (25:654). Mrongovius, in 1784–5, reports Kant’s saying that “character comes with ripe old age” (25:1385). And by the time of the published Anthropology in 1798, Kant’s views on the late establishment of character appear not to have changed (see 7:294). There is another, less important reason that a will in revolution cannot be expressed as a life divided neatly into an evil phase and then a good one that follows a sudden conversion. Kant’s explanation of atoning grace involves the claim that one can suffer “punishment . . . executed in the situation of conversion itself” (6:73, cf. 6:73ff.). Kant depends on a temporal coexistence of the self that is radically evil and the self that is morally good, because only on this basis can the morally good self properly function as an atoning sacrifice for the wickedness of radical evil. For one way of explaining this account, see Quinn 1986:450f. Quinn does not work out in sufficient detail the relationship between what he calls a “lack of moral identity between the pre-revolutionary and the post-revolutionary persons” and the unity of the physical person. For both practical and metaphysical reasons, it is important to understand that the lack of moral identity between the radically evil and the morally good selves must be merely figurative, but this analysis is beyond the scope of this book. That combating one’s propensity to evil is possible is a condition of the possibility of our obligation to become better, an obligation that one recognizes by the fact of reason. How it is possible cannot be comprehended, though Kant makes use of the category of “grace” to describe the possibility of it, as noted in section 3. This also helps explain how Kant can respond to claims about moral unfairness or moral luck. Louden 2000 (pp. 55–6, 141) explains this problem in several ways. The most troubling form of the problem is the claim that whether or not one is morally good is a matter of luck or social conditions or historical accident. For Kant, luck, social conditions, and historical accident can affect the way in which one’s moral revolution is expressed, but they cannot affect the status of the will itself.

194

Notes to Pages 134–143

49. Kant allows for empirical choices in the Metaphysics of Morals, where his general notion of Willk¨uhr (choice) is a faculty of desire (something empirical, in this context) combined with “consciousness of the ability to bring about its object by one’s action” (6:213), where such consciousness is something that can be observed in inner sense. The empirical nature of this choice is clear later when Kant contrasts “human choice” with “animal choice” (6:213). There is no basis for positing anything about animals as things in themselves, so the notion of “animal choice” can apply only to something empirical, an empirical choice. Similarly, although there is a basis for positing a noumenal grounding choice in human beings, this choice grounds an appearance that includes numerous empirical “choices.”

Chapter 6 1. The absence of a duty to promote one’s own happiness is less important. Kant’s reason for omitting such a duty is that “what everyone wants unavoidably of his own accord does not come under the concept of duty” (6:386). 2. In the rest of this section, I focus on the Religion, because it provides the most general account of the importance of interpersonal moral influence. I deal with the writings on history in section 3. 3. The motivational structure here is more complicated than merely pursuing virtue for its own sake, or for the sake of duty. See Herman 1993 and Korsgaard 1996a (especially Chapter 2) and 1996b for discussions of the distinctions between acting from a particular motivation and for a particular purpose. 4. Moreover, it is important for parents to promote their children’s moral selfesteem in a way that does not do injustice or harm to others. This requirement is imposed by the categorical imperative, which prohibits the promotion even of the happiness of others if that end is achieved by immoral means. In the next section this argument is expanded with respect to the happiness of third parties. With specific reference to promoting the happiness gained by proper self-esteem, this requirement has the particular implication of ruling out any strategy according to which one dulls the conscience of another, rather than promoting their virtue, to prevent them from the pain of self-condemnation. 5. Kant uses the term deed in this broad sense when he refers to the sensible “deed contrary to law . . . that resists the law materially and is then called vice,” a deed that is “sensible, empirical, [and] given in time” (6:31). Just as sensible deeds contrary to the law are called “vice,” one can refer to deeds that reflect goodness as “virtue.” In general, Kant reserves the term virtue for good deeds that actually reflect a good will. In this section, however, I use the term virtue primarily to refer to good deeds themselves, whether or not they reflect an underlying good will. Thus one can promote “virtue,” in this sense, without promoting a good will itself. 6. In Chapter 3, we looked at passions and affects as hindrances to morality. In this section of the Religion, Kant specifically highlights passions as a hindrance that has a social origin. He writes, “The passions . . . assail his nature, which on its own is undemanding, as soon as he is among human beings” (6:94).

Notes to Pages 143–145

195

7. Of course, like all duties imposed by moral anthropology, the obligation to promote moral community is an imperfect duty. As such, the effort one should put into the promotion of such a community will depend on other duties – both perfect and imperfect – that one has. 8. Of course, one does not promote moral revolutions in others, just as they cannot determine that one undergoes a moral revolution. Kant makes this clear when he describes the “historical representation of the gradual establishment of the dominion of the good principle on earth” (6:124). In describing progress toward the ethical community, which he here associates with a universal Church, Kant insists that “pure moral faith . . . is not a public condition” and thus cannot be the object of progress. Instead, one can give only “a universal historical account of ecclesiastical faith” (6:124), the progress by which the Church gradually comes to endorse and promote “an autonomous principle [of morality] which is one and the same for all human beings for all times” (6:124). The details of Kant’s account of the relationship between the Church and the ethical community are unimportant here. The main point is that what one seeks to promote is a community focused on teaching and affirming the value of the moral law (see too 6:132). One does not seek and cannot expect to promote a moral revolution, or “pure moral faith,” in another. 9. Unlike Yovel (see Yovel 1980), I do not take Kant to advocate the pursuit of the moral community as a “highest good” that involves the moral goodness of all people united with their happiness. As a specifically moral end, the “highest good” as described in the Critique of Practical Reason (5:110) involves only one’s own virtue in a world in which happiness is distributed in proportion to virtue. As an end of theology, the founding of a “moral people of God is . . . a work whose execution cannot be hoped for from human beings” (6:100). But as an end of moral anthropology, the highest good is “an ethical community . . . greatly scaled down . . ., namely to an institution which [is] at best capable of representing with purity only the form of such a [truly moral] community” (6:100). And while Yovel may be right that this ethical community involves “an intersubjective system of attitudes that embraces, ideally, the whole human race” (Yovel 1980:139), these attitudes are, finally, only appearances and not good wills themselves, and this end is an end for me primarily because it expresses my own struggle against radical evil. 10. At times, Anderson-Gold (e.g., Anderson-Gold 2001:26, 45) seems to go too far in this direction. 11. Kant does occasionally use language that is misleading in this respect, as when he discusses the “duty sui generis, not of human beings toward human beings but of the human race toward itself ” (6:97). There is a sense, of course, in which an entire community, or even the entire human race, can fall short morally. A state can be unjust, for example, if its institutions are fundamentally unjust, if property rights are not respected, if criminals are not punished, and so on. Kant even goes so far as to say that a society that disbands would be morally required to kill any imprisoned murderers. These demands of justice are responsibilities that a political community (and only a political community) can and must fulfill. Falling short results in the state’s

196

12.

13.

14.

15.

16.

17.

18.

Notes to Pages 145–152 being unjust. Similarly, moral communities (churches, etc.) have the moral obligation to promote virtue. And the human race itself can be said to have the obligation to form political and moral communities. In all of these cases, however, the moral status of a community is not independent of the moral status of its members, and the ultimate responsibility always lies with the individual members of the state, the church, or the human race. There is one sense in which a goodness that is the result of grace can be one’s own, at least in part. Although Kant is adamant that from a practical perspective, one’s choice must precede grace, the form of this choice is a deliberate receptivity to grace. Grace can be a necessary condition of goodness, as long as the presence of this grace is dependent on some practically prior choice (temporal priority is not an issue here). In this case, one’s goodness is not entirely one’s own, because grace was required. This is one way in which a will in revolution differs from a morally untainted will. But because receptivity is both up to oneself and the ground of receiving grace, one’s goodness is in an important sense “ultimately” one’s own. See Mari˜ na 1997 for more on the importance of receptivity in Kant’s account of grace. Again, because of Kant’s peculiar conception of possibility, it would be better to say that he does not commit himself to the impossibility of grace. For Kant, to claim that something is “possible” is to make a very strong claim. All he is willing to say with regard to prevenient grace is that we cannot know that it is impossible. See especially Yovel 1980, Despland 1973, Fackenheim 1957, Galston 1975, Michaelson 1979, Kleingeld 1995 and 1999, and Anderson-Gold 2001. Wood 1999, Louden 2000, Munzel 1999, and Michaelson 1990 also include extensive treatments of history in the context of more general accounts of Kant’s moral philosophy. Louden 2000, in particular, has a very good overview of the way in which different accounts understand moral progress in Kant’s writings on history (see Louden 2000:144–9). For other accounts which to varying degrees argue that historical progress involves moral progress that cannot be reduced to appearances, see Guyer 2000: Chapter 11; van der Linden 1988; Wood 1999; and Yovel 1980. My criticism of Louden in Chapter 4 presents several philosophical reasons to reject readings of Kant that suggest any sort of historical influence on the moral status of free human beings. In this section, I focus on exegetical issues, dealing with what Kant actually said, which I show, in most cases, is just what he should have said. This survey covers those writings that I discuss in the following sections. “What Is Orientation in Thinking,” “On the Miscarriage of all Philosophical Trials in Theodicy,” and parts of Kant’s first and second Critique s have relevance to Kant’s general account of history, but none of these focuses primarily on history, and none contributes to the issue of whether or to what extent Kant is concerned with moral progress. For these reasons I omit them from my account here. Herder continued to publish his Ideas through 1791, but Kant said that he could no longer review the work because he was working on his Critique of Judgment: “The third part of Herder’s Ideen will now probably be reviewed by

Notes to Pages 152–158

19.

20.

21.

22.

23.

197

another . . . for I lack the time for it, since I must soon get to the Grounding of the Critique of Taste” (Letter to Christian Sch¨utz, 25 June 1787, 10:467). Robert Louden (Louden 2000:147–8), following Friedrich Kaulbach (Kaulbach 1975:83), claims that Kant’s reference in the Conflict of the Faculties to a “moral cause” influencing human events is a sign that Kant thinks there is “strong empirical evidence . . . that humanity is progressing morally” (Louden 2000:147). However, Kant immediately explains that this “moral cause is twofold: first, that of the right [to a civil constitution] . . . and second, that of the end . . . that that same constitution alone be just” (7:85). In both cases, the “moral cause” turns out to be a purely political one. See van der Linden 1988:135f; Wood 1999:310; and Louden 2000:141–2 for other examples of using the third Critique to clarify Kant’s conception of history. They also give considerably more detail about how culture (and other ends) is promoted in history. I do not provide the details of these discussions, because they are not directly relevant to Kant’s views about the end of historical progress. There are other passages that I do not discuss, though these can be dealt with in the way that I deal with one of the four passages below. For examples of such passages and their use to support a strong notion of moral progress, see Louden 2000:56, 147–8; Guyer 2000: Chapters 11 and 12; and Wood 1999: Chapter 9. There are other, more overarching interpretations of Kant’s treatment of history that raise challenges for my view, but a full discussion of these approaches is beyond the scope of this book. In this context, see especially Yovel 1980 and Kleingeld 1995. Kant’s justification for this assumption is even more provocative: I base my argument upon my inborn duty of influencing posterity in such a way that it will make constant progress (and I must thus assume that progress is possible), and that this duty may rightfully be handed down from one member of the series to the next. (8:309)

Kant seems to be saying that one can know – or at least postulate – that the human race will make moral progress because one is obligated to promote this progress. Were this the real import of this passage, it would provide a clear example of the importance of intersubjective moral assistance for Kant. 24. See Louden 2000:147 for an example of using this passage to support a reading of Kant according to which there is genuine moral progress. 25. It is not altogether clear how far this essay can be taken to be a serious representation of Kant’s own views. Herder’s Ideen has strong religious overtones, and Kant may have used a philosophical reconstruction of the Biblical narrative as merely a means for rebutting Herder’s own supposedly pious history. Thus even this essay, which deals directly with the development of the human species, may assume for the sake of argument that human beings have progressed and argue about how that progress occurred. Still, this section will show that even if one takes “Conjectures” to represent Kant’s own view about human progress, the essay does not show that the human species progresses morally, in the sense of its becoming more and more morally good.

198

Notes to Pages 159–161

26. This also helps makes sense of at least one way in which “nature . . . prepare[s] man for what he himself must do in order to be a final purpose” (5:431). For a different interpretation of this passage, see van der Linden 1988:142. 27. Louden draws heavily from Kant’s lectures on pedagogy (especially 9:450) for his interpretation of moralization. While these lectures are generally accepted as loosely based on Kantian ideas (see Louden 2000:33–6 and Beck 1978, but contrast Weisskopf 1975), they are unreliable as primary sources of Kant’s view of moralization. Still, even these lectures can be interpreted in a way that is compatible with the account of moralization offered here. 28. Louden says this specifically with regard to the “good ends” that one who is moralized will pursue, but the point holds for all of the details of moralization. 29. Likewise in the “Idea,” Kant specifically mentions the promotion of a “morally good character [moralisch-gute Gesinnung],” in contrast to mere “semblances of morality” (8:26). 30. Kant even makes oblique reference to the problem of evil and the role of grace, saying that “this human . . . is expected to bring about what he himself is still in need of” (7: 325) and thus “only expects [moralization] from Providence” (7:328). Yet just as in the Religion, Kant ends up claiming that practically one must take the stance that “the human species should and can create its own good fortune” and thus we should “further (each to his best ability) with all good sense and moral inspiration the approach to this goal” (7:329). 31. See footnote 29 for a troubling text from the “Idea.” Further suggestions of troubling texts are found in Louden 2000, Guyer 2000, van der Linden 1988, Wood 1999, and Yovel 1980. Beyond his strictly historical writings, Kant’s discussions of education in the Anthropology, anthropology lectures, the lecture on pedagogy, the Critique of Practical Reason, and the Metaphysics of Morals have been taken to imply interpersonal moral influence that goes beyond what my account allows. A full exploration of Kant’s philosophy of education is beyond the scope of my discussion here, but some indication of how I would deal with it is provided by my treatment of Kant’s historical writings. L. W. Beck points out, The ages of an individual life correspond to the stages in the history of the world. By reading this fuller treatment of the philosophy of history we discover a key to the less well organized treatment of the philosophy of education. (Beck 1978:197–8)

In particular, the reference to the “civilization, cultivation, and moralization” of the human species in the Anthropology is connected there with the claim that “man must be educated to the good” (7:325), and a similar tripartite structure of progress appears in Kant’s lectures on education (see Louden 2000:38–44). The same sorts of moral progress that are possible in history can be furthered through education. Just as Kant gives in the “Conjectures” a historical account of humankind’s emergence into moral responsibility, certain forms of education are necessary for children to become morally responsible at all. And just as Kant describes the cultural progress of human beings in the third Critique and “Idea,” some forms of education simply

Notes to Pages 161–166

199

further “cultural” development. Finally, just as Kant allows for the promotion of good deeds, and even the fostering of an ethical community explicitly committed to virtue, he points out in his writings on education how such commitment to virtue can be fostered in children.

Epilogue 1. Bernard Williams, for example, has argued that although Kant was right about the connection between morality as he understood it and freedom, he was wrong about morality, and thus about freedom as well (see Williams 1995). R. Jay Wallace (in Wallace 1994), Simon Blackburn (in Blackburn 1998), and others have worked out sentiment-based moral theories that do not seem to depend on the kind of transcendental freedom that Kant insists is necessary for moral responsibility. And even Schleiermacher himself (see especially On Freedom, Scheiermacher 1992) tried to define a notion of moral accountability according to which transcendental freedom was unnecessary.

References

Kant’s Works References to Kant’s works are cited by volume: page number of Kant’s Gesammelte Schriften. 1900–. Ed. The Royal Prussian (later German) Academy of Sciences. Berlin: Georg Reimer (later Walter de Gruyter). I have made use of the following translations, making changes where necessary: Anthropology from a Pragmatic Point of View, trans. Victor Lyle Dowdell. Carbondale: Southern Illinois University Press, 1978. Critique of Judgment, trans. Werner Pluhar. Indianapolis: Hackett Publishing Company, 1987. Critique of Practical Reason, trans. Mary Gregor. New York: Cambridge University Press, 1997. Critique of Pure Reason, trans. Norman Kemp Smith. London: Macmillan, 1929. Critique of Pure Reason, trans. Paul Guyer and Allen Wood. New York: Cambridge University Press, 1998. Groundwork of the Metaphysics of Morals, trans. H. J. Paton. New York: Harper Torchbooks, 1964. Kant’s Philosophical Correspondence 1759–99. Ed and trans. Arnulf Zweig. Chicago: University of Chicago Press, 1970. Lectures on Ethics/Immanuel Kant, Ed. J. B. Schneewind and trans. Peter Heath. Cambridge: Cambridge University Press, 1992. Lectures on Metaphysics/Immanuel Kant, Ed. and trans. Karl Ameriks and Steven Naragon. Cambridge: Cambridge University Press, 1997. Political Writings, Ed. Hans Reiss. Cambridge: Cambridge University Press, 1970. Religion within the Boundaries of Mere Reason, ed. Allen Wood and George di Giovanni. Cambridge: Cambridge University Press, 1998.

200

References

201

Secondary Sources Adams, Robert Merrihew. 1998. “Introduction” to Religion within the Boundaries of Mere Reason. Cambridge University Press. Allison, Henry. 1996. Idealism and Freedom. Cambridge University Press. 1990. Kant’s Theory of Freedom. Cambridge University Press. 1983. Kant’s Transcendental Idealism: An Interpretation and Defense. Yale University Press. Ameriks, Karl. 2000. Kant and the Fate of Autonomy: Problems in the Appropriation of the Critical Philosophy. Cambridge University Press. 1981. “Kant’s Deduction of Freedom and Morality.” Journal of the History of Philosophy 19:53–79. 1989. “Kant on the Good Will.” Der Begriff der Politik: Bedingungen und Gr¨unde politischen Handelns. Ed. Otfried H¨offe. Stuttgart: J. B. Metzler. 1982. “Recent Work on Kant’s Theoretical Philosophy.” American Philosophical Quarterly 19:1–24. Anderson-Gold, Sharon. 1986. “Kant’s Ethical Commonwealth: The Highest Good as a Social Goal.” International Philosophical Quarterly 26:23–35. 1984. “Kant’s Rejection of Devilishness: The Limits of Human Volition.” Idealistic Studies 14:35–48. 2001. Unnecessary Evil. SUNY Press. Anscombe, G. E. M. 1958. “Modern Moral Philosophy.” Philosophy 33:124. Aquila, Richard E. 1979. “Things in Themselves and Appearances: Intentionality and Reality in Kant.” Archiv f¨ur Geschichte der Philosophie 61:293–307. Baier, Annette. 1995. Moral Prejudices. Harvard University Press. 1991. A Progress of Sentiments: Reflections on Hume’s Treatise. Harvard University Press. 1987. “Hume, The Women’s Moral Theorist.” Women and Moral Theory. Ed. Eva Feder Kittay. Rowman-Littlefield. Baron, Marcia. 1995. Kantian Ethics (Almost) without Apology. Cornell University Press. Beck, Lewis White. 1960. A Commentary on Kant’s Critique of Practical Reason. University of Chicago Press. 1978. “On Education.” Essays on Kant and Hume. Yale University Press. 1987. “Five Concepts of Freedom in Kant.” Philosophical Analysis and Reconstruction. Martinus Nijhoff. Bittner, R¨udiger, and Konrad Cramer, eds. 1975. Materialien zu Kants >Kritik der praktischen Vernunft