The Value of Humanity 0198832648, 9780198832645

Citation preview

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

The Value of Humanity

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

The Value of Humanity L. Nandi Theunissen

1

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

3

Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © L. Nandi Theunissen 2019 The moral rights of the author have been asserted First Edition published in 2019 Impression: 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2019949438 ISBN 978–0–19–883264–5 DOI: 10.1093/oso/9780198832645.001.0001 Printed and bound in Great Britain by Clays Ltd, Elcograf S.p.A. Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

For my parents

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

List of Figures 1. End

59

2. Reverse priority

63

3. C1

75

4. C2

76

5. Reflexive relation

79

6. R*

81

7. Source

140

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

Acknowledgments Many people have been generous in reading and responding to this work at various stages. I am indebted to David Velleman for his mentorship over the years, and above all for bringing me to see that what I thought was a paper was really a book project—the current one. I am exceedingly grateful to Peter Momtchiloff for his early interest in and, together with April Peake, for his careful stewardship of this project, and to three anonymous readers for Oxford University Press under whose guidance I reframed, indeed rewrote, several chapters. The reviewers’ feedback and suggestions for further readings made a hugely significant difference to the manuscript, and I express my full-hearted gratitude. I hope I have been able to do justice to their thoughtful responses, and I am thankful to have still more to think about. I am singularly indebted to my teachers, Katja Vogt and Joseph Raz, without whose conversation, writing, and guidance I would have written something else and something much poorer. Katja’s seminar on Kant and his Critics at Columbia University in 2008 was quite formative for me, and ultimately brought me to write this book. While I read it at a very late stage, the ideas developed in Katja’s book Desiring the Good informed her instruction and our conversations spanning some years. Katja has seen countless chapters at various stages, and she generously read my manuscript in its entirety when it was close to completion. The feedback she provided on the fifth chapter was a spur to re-write a central line of argument, and I am truly grateful for her insightful responses. I cannot imagine philosophy without Joseph, and I am always profiting by something he is writing. Joseph was kind to invite me to present a version of the third chapter at the Columbia Law School in the fall of 2015. His singular written commentary, and the engaging discussion in his seminar, were exceedingly helpful. Joseph’s treatment of the value of humanity in his Value, Respect, and

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

xii acknowledgments Attachment is foundational for me, and his influence on my thinking will be quite visible on the page. The third chapter benefitted from questions on the part of audiences at Northwestern University, Johns Hopkins University, and the University of Wisconsin at Milwaukee. I wish to thank Kyla EbelsDuggan, and I am especially grateful to Richard Kraut whose work has been extremely important for me and influential. I appreciate the invitation to present an early version of the third chapter at the University of St. Andrews, and I thank Jens Timermann and Sarah Broadie in particular for their encouragement. I am most grateful to several anonymous reviewers, and to the Associate Editor Connie Rosati, for invaluable feedback on a part of the third chapter and Appendix 2 that appeared as L. Nandi Theunissen (2018). “Must We Be Just Plain Good? On Regress Arguments for the Value of Humanity”. In: Ethics 128, pages 346–372. I am thankful for permission to reprint portions of that article here. Richard Kraut kindly agreed to respond to this paper in an online discussion on Pea Soup, and his questions were most helpful in making refinements to the version that is included here. My sincere thanks to Michele Moody-Adams and Thimo Heisenberg for their sensitive commentary and thought-provoking questions at Katja’s Workshop in Ancient and Contemporary Philosophy in 2018, and to the audience at Columbia University for their sympathetic reception. Samuel Scheffler’s 2008 seminar on topics in the theory of value at New York University was influential in opening up a set of questions for me that eventually formed the basis of the fourth chapter. I was glad of the opportunity to present a version of it at the German Idealism Workshop in New York in 2014. I am indebted to Wolfgang Mann, Guy Fletcher, Sarah Buss, Matthew Smith, Felix Koch, and Connie Rosati for providing helpful written feedback on earlier iterations. I am also indebted to Wolfgang for kindly reading this and other chapters and for providing very helpful responses. Cora Diamond served as an examiner on a quite unrelated project for my MPhil at the University of Sydney some ten or more years ago. I find Diamond’s work on humanity powerful and provocative,

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

acknowledgments xiii and without it I would not have been drawn to this project. Brad Weslake was an exceedingly important interlocutor at an early stage; I owe much to his sharpness of mind. Jed Lewinsohn read a late version of the manuscript and provided very helpful comments on the fifth chapter, urging me not to “hide the ball.” I am grateful to Jed also for suggesting the Rabbinic teaching that I have included as an epigraph. Benjamin Parris has read drafts of chapters more than once. He has quickly seen to the heart of things, and has pushed me to trust my instincts. I am grateful to Ben for what I can only describe as his cosmic vision. To my phenomenal teachers in other areas of my life—Mary Flinn, Heather Hax, Martha McAlpine, Elizabeth Sitzler, Thomas Truss, Matt Markon—thank you for fortifying my spirit!

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

Hillel teaches, If I am not for myself, who will be for me? But if I am only for myself, who am I? And if not now, when? —Ethics of the Sages, 1:14

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

Introduction This book addresses the central ethical question of why human beings invite, or I might rather say demand, basic forms of recognition. I find it natural to approach this question in terms of value and its ground— hence the value of humanity. But I have come to think that one cannot address this topic with any seriousness without thinking about the nature of value in general. Questions about the nature of value tend to go together with questions about what is of value and why. One moves back and forth between thinking about what value is and what things are of value. Accordingly, this work addresses foundational topics concerning the value of human beings, and it has implications for quite general discussions of value and normativity. The classic Enlightenment proposal is that human beings are valuable in themselves. In terms that are familiar from Kant’s writings in moral philosophy, the value of humanity is absolute while the value of everything else is relational. The position enjoys wide contemporary support.1 But early on Kant expressed concern that philosophers would suspect it of high-flown fantasy. Philosophers would find the notion of absolute value fantastical because—as I will transpose Kant’s thought—when we explain the value of diverse kinds of object and activity, value is taken to be a relation between an object and a subject, a relation of being good for that subject. The form of value that Kant 1 Influential proponents include the early Christine Korsgaard (1986). “Kant’s Formula of Humanity”. In: Kant-Studien 77, pages 183–202; Joseph Raz (2001a). “Respecting People”. In: Value, Respect and Attachment. Cambridge: Cambridge University Press, pages 124–175; and David Velleman (1999a). “A Right to Self-Termination?” In: Ethics 109.3, pages 606–628.

The Value of Humanity. L. Nandi Theunissen, Oxford University Press (2019). © L. Nandi Theunissen. DOI: 10.1093/oso/9780198832645.001.0001

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

2 the value of humanity extends to human beings does not have this relational structure, so that it is unlike value as we know it. A central claim of this work is that Kant was right to be concerned. There is no such thing as absolute value. If we think that human beings are of value and for that reason such as to be treated in certain ways and not in others, then we must find another way to make this out. I develop a relational account of the value of humanity in this book. According to the proposal, the value of human beings is continuous with the value of other valuable things. I argue that there is something distinctive about living beings, and about living beings with a human form, but I do not think we illuminate this difference by way of absolute value. I offer lines of thought that make it plausible to say that goodness or value is affecting: such that it helps and serves and enriches. I take the Socratic starting point that good is a notion of benefit, or in a more contemporary idiom, that the good is the good for someone.2 If people are bearers of value, I propose that our value is no exception. I explore the possibility that our value is explained through reciprocal relations, or relations of interdependence, as when—as daughters, or teachers, or friends—we benefit others by being part or constitutive of relationships with them. I also investigate the possibility that we can be said to stand in a valuable relationship with ourselves. Ultimately I propose that people are of value because we are constituted in such a way that we are able to be good for ourselves in the sense that we are able to lead flourishing lives. The value of humanity is grounded in a relational conception of good as good for someone, and the resulting view captures what is distinctive about people and what is at stake in our ways of relating to one another. Intuitively, a person matters because she matters to herself in a very particular sort of way. She is the center of a life to which she bears a special relation.

2 Cf. Richard Kraut (2007). What is Good and Why. Cambridge, MA and London: Harvard University Press. See also Richard Kraut (2011). Against Absolute Goodness. Oxford and New York: Oxford University Press.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

introduction 3 The proposal takes a path between absolutism and eliminativism about human value, so let me say a word about both of the poles I mean to avoid. On the one side is Kant and the conception of human value as absolute, or—in a term I am using inter-changeably—nonrelational. As I will use it here, what is “non-relationally valuable” is of value independently of whether it can be good for someone—it is a benefit-independent form of value. What is “relationally valuable” is of value because it can be good for someone—it is a benefitdependent form of value. I note at the outset that questions about realism and anti-realism are quite orthogonal to this issue. I do not find it plausible to suppose that Kant was a normative realist about human value, as if our value were a mind-independent property of the kind that is theorized by G. E. Moore and those following him. This fits ill with Kant’s broader philosophical commitments. In calling human beings absolutely valuable, I take Kant to mean that human beings are such that they categorically must be valued: in other words, valued independently of their capacity to stand in relations of benefit to something or someone.3 To speak of how human beings are to be valued evidently implies that there are valuers around to do the valuing, so I am not issuing the familiar complaint against a would-be form of realism that it countenances a valuer-independent property that is not respectable from the point of view of modern science. As I will explain in Chapter 2, my complaint is rather that, since value is a relation, it cannot be as Kant proposes. To put it in terms of valuing, what is at stake between the relational and non-relational value theorist is a specific metaphysical dispute about whether the object of valuation is monadic or dyadic. Subject to refinements to come later, I am defending the view that it is always dyadic.

3 For Kant, categorical valuing also marks a contrast with desire—with valuing people because we want to and not otherwise—and that is because Kant takes all forms of relational value to be grounded in desire. I will deny this equivalence. When we are responsive to the value of people we are responsive to their capacity to stand in particular relations of benefit, but this licenses forms of practical response that are fully independent of our subjective motivational states.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

4 the value of humanity While moral philosophy inherits the idea of the absolute value of humanity from Kant, the notion has a far wider circulation. Broad appeal is made to the related idea of human dignity, especially in applied-philosophical and popular-intellectual discussions. Intellectual historians read a concern with dignity back into the history of philosophy to identify a conception of humanity that they see as perennial.4 Those interested in the middle decades of the twentieth century in particular use it to mark an urgent preoccupation in the lead-up to the Second World War.5 We notice the term dignity in the very first sentence of that century’s founding document for internationally sanctioned human rights. It is called upon to do the important work of underwriting non-defeasible forms of protection for human beings. I take the position that it is not up to that work because the conception of value is unstable. The literary critic Mark Greif has argued that discussions around dignity have the characteristics of “cant discourse.” In Greif ’s telling, a discourse is cant that employs terms which are not meant to be questioned or argued about or refined through disagreement. The terms originate in a theoretical framework that is no longer alive to its users, so they create an atmosphere of profundity and mystery in a purely symbolic way, and the talk “becomes a counterfeit that drives out the good.”6 This is not an unfamiliar form of argument. G. E. M. Anscombe made the case that outside of a divine law conception of ethics, the special moral ought has “mere mesmeric force.”7 Philippa Foot argued that Moore’s conception of goodness gives way to non-cognitivism as its natural heir.8 It must be said that these are polemical arguments. What I think 4 See Herschel Baker (1947). The Dignity of Man: Studies in the Persistence of an Idea. Cambridge, MA: Harvard University Press. 5 See Mark Greif (2015). The Age of the Crisis of Man. Princeton and Oxford: Princeton University Press. My thanks to an anonymous reader for Oxford University Press for suggesting Greif ’s book to me. Compare Samuel Moyn (2010). The Last Utopia: Human Rights in History. Cambridge, MA: The Belknap Press of Harvard University Press. 6 Greif, Age of the Crisis of Man, 11. 7 Elizabeth Anscombe (1958). “Modern Moral Philosophy”. In: Philosophy 33.124, pages 1–19. 8 Philippa Foot (2001). Natural Goodness. Oxford: Oxford University Press, ch. 1.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

introduction 5 is right about Greif ’s claim that discussions around human dignity are at risk of cant is that they tend to involve a too-easy appropriation of Kantian terminology with inadequate attention to foundational questions, including the question of what value is. The term becomes harmful because it becomes empty, and in becoming empty it drains the discussion. Greif is not deaf to the sense in which appeals to human dignity have a seriousness of intent. He is interested in appeals to dignity in the context of the grave transgressions of human rights that occurred in the middle of the last century. At that moment it was a gesture that said “there must be something [about human beings] that must be protected.”9 This is a gesture I take seriously. There is something about human beings that makes it appropriate to respond to one another in very particular ways. As I explain in Chapter 1, I take the view that practical reasons are grounded in values, so I theorize ethical responsiveness to human beings in terms of the value of human beings. While I do not offer a Kantian account of our value, I am also not denying that human beings are bearers of value. That means that though I am not an absolutist about the value of humanity, I am not an eliminativist either. I turn to eliminativism as the other pole I mean to avoid. The most interesting form of eliminativism, certainly the one that is most relevant to my own thinking about value, even as I do not draw its conclusions, is due to Peter Geach.10 Geach is not concerned with the value of human beings. His question is whether there is such a thing as the property good—or as we might rephrase his position, the property of being of value—which all good or valuable things have in common. Geach allows that we speak meaningfully of things as being good in some way, as this is a good action, or as that is a good book. But he denies that there is a property, the property of being good or of value, that all good or valuable things share, and it is in this sense that I am calling him an eliminativist: he is an eliminativist about 9 Greif, Age of the Crisis of Man, 12. 10 Peter Geach (1956). “Good and Evil”. In: Analysis 17.2, pages 33–42.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

6 the value of humanity the property good.11 Geach was struck by differences in the logical behavior of adjectives. The adjective “red” behaves differently from the adjective “big.” This is shown by the fact that while it is true to say of a red book that it is red and that it is a book, it is not true to say of a big flea that it is big and that it is a flea. Geach put the difference by saying that red is logically predicative while big is logically attributive. There is such a thing as being, simply, red, but there is no such thing as being, simply, big. Geach’s central claim was that “good” functions like big and not like red, for it would be false to say of a good knife that it is good and that it is a knife. It follows that there is no such thing as being good, only being a good so-and-so. Good is always an attributive adjective. Geach is taking aim at a rather wide set of philosophical practices. He is attacking the consequentialist’s use of “good” in “a good event” or “a good thing to happen,” arguing that we should not take it as obvious that anything is being expressed by these phrases.12 But Geach is also rejecting would-be philosophical uses of “good” in claims to the effect that “knowledge is good” or “pleasure is good.” Geach’s complaint is that the philosopher has not told us what it is for something to be good in this sense. In her recapitulation and defense of Geach’s argument, Judith Jarvis Thomson canvasses some options for what the philosopher has in mind. To say of some X that it is good, or equally, that it is of value, is to say that X has the property that makes it protanto ought-making, or in other words, that X has the property that makes it of practical relevance.13 So to say that “pleasure is good” is to say that pleasure has the property that licenses reasons to form various sorts of attitudes, including intentions to act. Thomson’s question is what the property that is pro-tanto ought-making is supposed to be. 11 As Katja Vogt formulates Geach’s point, he is against any unity to good-talk, or in scholastic terms, he is against the univocity of the good. Cf. Katja Maria Vogt (2017). Desiring the Good: Ancient Proposals and Contemporary Theory. Oxford and New York: Oxford University Press, ch. 2. 12 For comparable claims see Philipa Foot (1985). “Utilitarianism and the Virtues”. In: Mind 93.374, pages 196–209. 13 Judith Jarvis Thomson (2008). Normativity. The Paul Carus Lectures. Illinois: Carus Publishing Company, pages 14–15.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

introduction 7 She considers three possibilities: (1) it is the property of oughtness itself; (2) it is the property of being good in some respect; and (3) it is the property of being good in itself or simpliciter. Thomson rejects the candidates in turn. The property cannot be oughtness because goodness was meant to rationalize oughtness. The property cannot be goodness in some respect, because everything is good in some respect, so it is a trivial property that all things have. The property cannot be, simply, goodness because there is no such thing. It is Thomson’s case against 2 that interests me. To say of some thing that it is good or of value is to say that it is pro-tanto ought-making. Thomson rejects the suggestion that what it is for something to be good in this sense is for it to be constituted in such a way that it is good in some respect. Her candidates for being good in some respect are being good for England, being good for making cheesecake, being good to use in teaching elementary logic.14 Being good for England is a difficult and abstract sort of case, but the other examples—funny because banal—also await further specification. What is cheesecake good for? And what is the good of teaching elementary logic? There is a longstanding tradition according to which, to the extent these questions are answered in the affirmative, they are answered by saying that, either by themselves or in conjunction with other things, they are good for human beings. This is the classically humanist position. In Joseph Raz’s characterization, the humanist explains the goodness or value of things in terms of their actual or potential contribution to human life and its quality.15 Some will seek to widen the proposal to encompass other forms of sentient life so that the construction is in terms of beings or individuals.16 I am very sympathetic to this extension. But either way, being such as to contribute to the quality of the life of human beings or beings generally is not a trivial property, and not a property that everything has.

14 Thomson, Normativity, 6. 15 Joseph Raz (1986). The Morality of Freedom. Oxford: Clarendon Press, page 194. 16 Cf. Raz’s claim that “Morality is concerned with the well-being of individuals.” Raz, Morality of Freedom, 267.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

8 the value of humanity So here is my answer to Thomson’s question. When we say of some thing that it is of value we mean that it has a property that makes it practically relevant. This is the property of being constituted in such a way as to contribute, directly or indirectly, actually or potentially, to the quality of the life of human beings (or more broadly, of beings). This is to resist the eliminativist conclusion according to which there is no such thing as being good or of value, only being good for someone, or being good as some kind of thing. Instead, it is to propose that being good for human beings is a way of being valuable: it is an explanation of something’s status as normatively significant. It turns out that it is quite meaningful to say of some thing that it is of value: what is of value is or can be beneficial for human beings. The big claim of this book is that if people are bearers of value then our value is no exception. While it is true that there is something distinctive about human beings, and something distinctive about what it owed to human beings, the value of humanity is not different in kind from the value of other things but is continuous with it. Human beings are bearers of value because we are constituted in such a way that we give rise to reasons to be treated in certain ways and not in others, and the explanation of this reason-giving status is that we stand to be beneficial in specifiable ways. The proposal must grapple with some powerful objections from both the Kantian and the Moorean traditions in ethics, and my positive view emerges from a sustained engagement with ideas about value that animate these principal traditions in modern moral philosophy. The first challenge turns on the very humanist conception of value that I take as a starting point. It is not a challenge made by Kant himself, but it is one that is sometimes claimed to be Kantian in orientation. It is argued that the existence of what is indirectly and directly good for people entails that people are of value in themselves. Something must be the bearer of a special non-relational kind of value for anything to be relationally valuable, and people meet the criteria for being nonrelationally valuable. People are that for the sake of which all the other things are of value, but they are not of value for the sake of other things. In the book I give reason to doubt the model of relational value—a

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

introduction 9 “borrowing” model—that drives the form of argument, and I explore alternatives to terminating a regress in non-relational value. The second challenge stems from the intuition that the more we distinguish relational from non-relational value—the more we insist that the one does not depend on the other—the greater our difficulty in understanding the normativity of relational value. The objection is to the explanation of the value of any X in terms of being good for someone, S. If X is good for S, then it is clear enough how S has reasons in regard to X, but not so clear how X generates reasons for anyone else. If I propose that people are among the substitution instances for X, then I appear to be unable to demonstrate a person’s normative significance for subjects other than S. That would obviously be a worrisome outcome. The account would fail to meet an important desideratum for accounts of the value of people: namely, that they capture basic forms of ethical behavior owed to everyone. I address this and related Moorean objections to the nature and normative significance of relational value, and the discussion shows how a relational account of the value of people licenses basic forms of ethical behavior that are owed to everyone. Here is a road-map for what follows. In Chapter 1 I give my reasons for rejecting the view that being human is what it takes to enjoy special forms of moral consideration. I situate the ethical significance of human beings in a more general account of practical concern, defending the view that all valuable things should be treated as the valuable things they are. I provide a theoretical framework for the positive account to follow in which relationality and non-relationality emerges as an important distinction within evaluative explanation. I clarify my use of these terms in relation to other distinctions in goodness. Chapter 2 gives my reasons for departing from Kant’s influential treatment of the value of humanity as non-relational in the specific sense at issue, and my motivation for approaching value in relational terms. Chapter 3 provides a response to the regress challenge described above in the absence of a persuasive answer to which the view developed here would be an obvious non-starter. Chapter 4 develops a positive relational proposal in which the relational value of human beings

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

10 the value of humanity turns on the distinctive relationship we bear to objects and activities of value. The account contributes to recent discussions of valuing, final ends, and care. Chapter 5 addresses the Moorean challenge to the normativity of relational value described above, and the discussion clarifies the basic forms of ethical behavior that are licensed by the account of human value. The focus of the book is on theoretical as opposed to applied topics, and this reflects my contention that we cannot think well about the value of humanity without thinking about the nature of value in general.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

1 Common Humanity I take as a basic presumption that human beings are subjects of ethical concern: there are ways we should and should not treat one another. Though few would deny it, there is less agreement about how to place this idea in ethical theory, and about how much we mean to say with it. The philosophical territory here is well staked-out. Using the term in a way that is meant to do justice to a range of positions, some of them bearing a relation to the historical Kant, and some not, on the one side are proto-Kantians who propose that all and only people have special moral standing. On the other side are those who, identifying as more-or-less Consequentialist, reject this proposal as a form of human prejudice. By recalling some of what is at stake between them, and by learning from both, I take a position that is neither Kantian nor Consequentialist. I reject the view that humanity is the mark of moral standing. We should not begin ethical theory by asking what is owed to human beings. We should situate concern for human beings in a more general account of practical concern. I defend the view that we human beings make forms of ethical behavior appropriate because we instantiate something that is shared with other beings, objects, and activities—we are bearers of value. I take the view that normative reasons are grounded in values, and that all valuable things should be treated as the valuable things they are. I provide a framework for evaluative explanation that serves as the basis for the chapters to follow.

The Value of Humanity. L. Nandi Theunissen, Oxford University Press (2019). © L. Nandi Theunissen. DOI: 10.1093/oso/9780198832645.001.0001

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

12 the value of humanity

1.1 Questions of Explanation There is more than one way that the idea of human beings, and of humanity, comes up for investigation, and philosophers share an interest in these questions with the broader humanities.1 But the task of this chapter is to think through a specific cultural and philosophical inheritance. As I would tell the story of this inheritance, it begins with a widely shared conviction that there are ways we should and should not treat people, not because they happen to be rich, or famous, or friends of ours, or expedient, but because, as people, they make a claim on us. Call this the idea of common humanity. As readers of Anglo-American moral and political philosophy it is hard to escape the impression that the idea of common humanity was taken up with particular vigor and intensity in the second half of the twentieth century, and investigated systematically like never before. Political liberalism was formulated in a new way, and there was lively discussion of egalitarianism, fairness, and toleration.2 It would be tempting to read this work as growing out of the public reckoning over the Holocaust and Hiroshima, but that would be a claim for intellectual historians, and the claim would not be straightforward for several reasons, some of them having to do with Anglo-American philosophy as a practice and institution.3 What is clear is that the terms contemporary philosophers inherit are very much refracted through these twentieth-century discussions. Important recent work promises 1 To my knowledge, Bernard Williams is unique among philosophers in considering the broader humanistic context. See Bernard Williams (2006). “The Human Prejudice”. In: Philosophy as a Humanistic Discipline. Princeton, NJ: Princeton University Press, pages 135–152. I discuss aspects of this broader context, and of the various senses of “humanity,” in Appendix 1. 2 Key works include John Rawls (1971). A Theory of Justice. Cambridge, MA: Harvard University Press. Bernard Williams (1962). “The Idea of Equality”. In: Philosophy, Politics and Society, 2nd Series. Edited by P. Laslett and W. G. Runciman. Oxford: Blackwell, pages 110–131. Ronald Dworkin (1977). Taking Rights Seriously. Cambridge, MA: Harvard University Press. Harry G. Frankfurt (1999a). “Equality and Respect”. In: Necessity, Volition and Love. Cambridge: Cambridge University Press, pages 146–154. 3 Mark Greif has argued that these events served to focus and disseminate discussions that had begun a good deal earlier. See Greif, Age of Crisis of Man, ch. 3.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

1.1 questions of explanation 13 to reanimate the earlier contributions by John Rawls and others, but it is deeply and self-consciously embedded in the earlier framework.4 What unites an otherwise heterogeneous set of proposals is the conviction that all human beings have a special moral significance. In some formulations, human beings are uniquely subjects of “equal concern and respect,” in others we are subjects of “distinctive equality,” or we are alone bearers of rights.5 The conviction is motivated by the proper refusal to divide human beings into morally significant classes or ranks on the basis of race, gender, socio-economic status, or talent and skill.6 It is also prompted by a desire to qualify forms of familial or national allegiance with the more impartial spirit of cosmopolitanism.7 Importantly, however, there is thought to be a limit to the extension of impartial concern, so that there is an inside and an outside to the relevant form of moral consideration. On the inside are all the human beings, and on the outside are all the things that are other than human, and as the contrast is usually drawn, other animals. As Jeremy Waldron makes the point, “The principle of basic equality is opposed to any claim that there are moral distinctions and differentiations to be made among humans like unto or analogous in scale and content to this sort of differentiation among animals. [. . .] To repeat: there are no distinctions of the relevant kind between human and human, nothing like the distinctions commonly made between human and animal.”8

4 An important example is Jeremy Waldron (2017). One Another’s Equals. Cambridge, MA and London: The Belknap Press of Harvard University Press. While he is a critic notions of equality in particular, see also Harry G. Frankfurt (2015). Inequality. Princeton, NJ: Princeton University Press. 5 See Dworkin, Taking Rights Seriously, 180 and passim; Waldron, One Another’s Equals, 31 and passim. 6 Here is Waldron: “One way of thinking about basic equality is that it denies that there are differences of sortal status correlating to differences among humans. The principle of basic equality repudiates a position that once upon a time almost everyone embraced: that the law has to concern itself with different types of human beings. We now hold that there is just one sortal status: the status of being a human person.” Waldron, One Another’s Equals, 8. 7 Waldron, One Another’s Equals, 148. 8 Waldron, One Another’s Equals, 30.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

14 the value of humanity This style of proposal has long generated interest in questions of explanation, and these questions have arranged themselves in the form of a dilemma that is still regarded by many as “the deep problem” in this area. In my view there are more compelling and neglected topics, and the focus of this book will reflect this. But one cannot sidestep the issues posed by the dilemma. The dilemma can be formulated as follows: if human beings have a special moral significance, then there must be something about us that makes this so. But for any feature to which proponents make an appeal (self-consciousness, rationality, desires for the future, etc.), not all of us appear to possess it. We could deny that our standing is grounded in specifiable traits, but then there is the risk of prejudicial favoring of our own kind—of according special significance to all members of our species when there is nothing in the world to bear this out. So, either one must admit that not all human beings enjoy a special moral status, or one must adopt a position without right.9 There are various attempts to meet this objection. One strategy is to deny that the significance of being human needs to be grounded in something more basic or less contentious—some feature that all human beings can be shown to share.10 The suggestion is not that we cannot adduce relevant differences between human beings and animals. For Bernard Williams, a proponent of this style of approach, a particularly salient difference is that people are answerable to ethical norms while animals are not.11 But Williams resists the idea that we favor human beings over animals for the reason that we are answerable to ethical norms while they are not. Williams appeals to the actual

9 The challenge was influentially posed by Peter Singer (1975). Animal Liberation. New York: Harper. See also Peter Singer (1993). Practical Ethics. 2nd edition. Cambridge: Cambridge University Press, ch. 2. For a more recent statement and defense, see Peter Singer (2009). “Speciesism and Moral Status”. In: Metaphilosophy 40.3–4, pages 567–581. 10 This is the position taken by Williams, “The Human Prejudice,” 144 and passim. Cf. also Cora Diamond (1978). “Eating Meat and Eating People”. In: Philosophy 53.206, pages 465–479. And Cora Diamond (1991). “The Importance of Being Human”. In: Philosophy Royal Institute to Philosophy Supplement 29, pages 35–62. 11 Williams, “The Human Prejudice,” 148.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

1.1 questions of explanation 15 conditions of human life, and in human life as it is we are moved by a sense of fellowship with other members of our kind. The relevant thought for us is that we are we and that they are they, so that there is a basic sense of our ethical identity as a species.12 Arguments for extending our concern to other animals, as well as to wider groups of people, fail to be persuasive unless they begin from a recognition of this basic fact about us.13 In one way, philosophy came into being as an impulse towards the search for adequate grounds to our claims concerning matters of central importance in human life. It would be anachronistic to call this an impulse towards objectivity, but it is at least an impulse towards the formation of a “conception of the world,” a conception of how our conduct fits in with what we know of ourselves and the cosmos.14 On a related way of putting philosophical motivation, it is the desire to understand how things generally speaking hang together.15 For Thomas Nagel, who brings the Sellarsian conception of philosophy into ethics in particular, objectivity is the central problem.16 The ethical theorist is looking for a more impersonal standpoint on human beings and our view of things. She wants to know whether the normative appearances hold up after stepping back to form a more comprehensive view that does not abandon the view from inside but treats it as part of what needs to be understood.17 For Nagel the business of stepping back is fraught with risk, but the risks have to be negotiated the best we can.

12 Williams, “The Human Prejudice,” 152. 13 Williams, “The Human Prejudice,” 147. 14 I take this formulation from Dieter Henrich who is describing the difference between sophistry and philosophy. See Dieter Henrich (1992). Aesthetic Judgment and the Moral Image of the World. Edited by Eckart Forster. Stanford Series in Philosophy: Studies in Kant and German Idealism. Stanford, CA: Stanford University Press, page. 9. 15 Cf. Wilfrid Sellars (1962). “Philosophy and the Scientific Image of Man”. In: Frontiers of Science and Philosophy. Edited by Robert Colodny. Pittsburgh, PA: University of Pittsburgh Press. 16 Cf. Thomas Nagel (1989). The View from Nowhere. Oxford and New York: Oxford University Press, page 138. 17 Nagel, View from Nowhere, 138.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

16 the value of humanity It is not irrelevant that Williams takes there to be limits to what philosophy can do in the way of underwriting, or challenging, our ethical commitments.18 But Williams is not relevantly speaking without a conception of the world. Williams is relying on a set of commitments about what it is to be responsive to ethical considerations, and ultimately, about what it is to be moved to act, that he develops at some length.19 According to this conception, we are moved by patterns of emotional reaction, sympathy, personal loyalties, commitments, and allegiances that are themselves unmotivated. This has implications for how we theorize in ethics, and for how we articulate the normative reasons there are to do one thing rather than another. The reasons must be relevantly connected to the motivational states that are brute in us. Whether we find this persuasive depends on our stance on a host of foundational questions that have been the subject of intense discussion in theoretical ethics.20 The key question is whether it is true that actions are always most basically motivated by brutely conative states, and whether brutely conative states are necessary for the presence of normative reasons. For a prominent group of ethical realists, the answer is no in both cases. From their point of view, Williams’s strategy depends on some implausible assumptions about who we basically are. A different and more common style of response to the dilemma is to offer an account of the special moral standing of human beings in terms of a range or threshold property.21 What grounds the special moral standing of human beings is the capacity to exercise a morally relevant power or ability, where that capacity is understood

18 Cf. Bernard Williams (1985). Ethics and the Limits of Philosophy. Cambridge, MA: Harvard University Press. 19 Bernard Williams (1981). “Internal and External Reasons”. In: Moral Luck. Cambridge: Cambridge University Press, pages 101–13. 20 At least since Thomas Nagel (1970). The Possibility of Altruism. Princeton, NJ: Princeton University Press, ch. V. For an influential development of Nagel’s position, see T. M. Scanlon (1998). What We Owe to Each Other. Cambridge, MA: Harvard University Press, ch. 1. See also Joseph Raz (1999a). “Agency, Reason and the Good”. In: Engaging Reason. Oxford and New York: Oxford University Press, pages 22–45. 21 The approach was introduced by Rawls, Theory of Justice, 444.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

1.1 questions of explanation 17 to have practical significance in light of the practical significance of its exercise. As Waldron puts the point, “the thing about the possibility is that every so often it will explode into actuality. And we wouldn’t be interested in it if it didn’t.”22 There are acknowledged to be differences in how successfully people exercise the relevant capacity, and some of those differences are understood to have practical implications; they make it the case that a person is praised, is our friend, receives a prize, and so on. But, the thought is, possession of the capacity at a minimal threshold is sufficient to secure the basic forms of moral consideration that are at issue. The question is what to say about human beings who fall below the threshold: infants who do not yet but will one day possess the capacity; people who, in various stages of passage out of life, once did but no longer possess it; and people with profound disabilities who cannot do so.23 We meet various attempts to handle these cases. Take a recent view according to which all human beings have a special moral standing in virtue of the genetic basis for moral agency.24 The appeal to moral agency is common-place in these discussions; it is broadly Kantian, but it is more immediately a mark of Rawls’s influence.25 There are questions about what moral agency is, and questions about whether there is something distinctive about moral as against other forms of agency. These aside, the proposal is that all human beings possess the genetic basis for the capacity for moral agency including those who will grow up to exercise it, and those who once did but no longer exercise it, and those who, in a standardly cited case, are born with anencephaly—a non-genetic disorder that results in the failure of the upper brain to develop as a result of prenatal deficiencies. I trust we can accept the empirical claims. But the philosophical question is why the genetic basis for moral agency should have practical relevance.

22 Waldron, One Another’s Equals, 160; 250. 23 See Shelly Kagan (2016). “What is Wrong with Speciesism?” In: Journal of Applied Philosophy 33.1, pages 1–21. 24 See S. Matthew Liao (2010). “The Basis of Human Moral Status”. In: Journal of Moral Philosophy 7.2, pages 159–179. 25 See Rawls, Theory of Justice, 266–267.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

18 the value of humanity If we follow the above rationale, we could accept that the exercise of moral agency has ethical significance, and we could accept that the capacity for exercising moral agency has ethical significance— its significance derives from the live possibility of its exercise given that its exercise is ethically significant. But since the genetic basis for moral agency is possessed even by those for whom there is no live possibility that it can be exercised, its practical relevance remains opaque. I said that for Nagel the task of supporting the normative appearances is fraught with risk. Nagel is particularly sensitive to the modern tendency to over-privilege an objective perspective by according scientific forms of understanding complete authority about what can be said about us and the world.26 This is the mark of a temptation to ascend to a level of objectivity that is inappropriate when normative questions are under consideration. As I would adapt Nagel’s point, we want to be sensitive to the subject matter we are given to think about. Our question is the ethical significance of being human born, and the role of theory is to make sense of the subject so that we have a more reflective grasp on it. To be successful, a theoretical intervention must make sense of the subject matter—it must have a story to tell about how things hang together, and that story must be coherent and perspicuous, it must invoke ideas that we understand, it must explicate the ideas that are less clear, or that are invoked for explanatory purposes, etc.—and it must tell a story in such a way that it illuminates the subject matter. Our internal view of things may need to be refined in light of the demands of explanation. But equally, if the explanation abandons that perspective entirely, if it loses sight of the aim of helping us to understand a normative dimension of our lives, then it is an exercise without a point.27 A more compelling attempt to think through the normative significance of a shared organic basis for a morally relevant capacity is

26 He develops this thought more controversially in Thomas Nagel (2012). Mind and Cosmos. Oxford and New York: Oxford University Press. 27 Compare Raz, “Agency, Reason and the Good,” 44–45.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

1.1 questions of explanation 19 due to Waldron.28 In Waldron’s view, the relevant threshold capacity, whatever it is, should be understood as the realization of a potential that has an organic infrastructure the development of which occurs over time and in functionally directed ways. In Waldron’s example, we are born with a tongue and a larynx in order to speak. In the early stages, the organic basis of the capacity to speak does not yet support its function. It must develop over time, and this development will be conditioned by the way of all things in the realm of becoming—it is subject to break-down, injury, disrepair, decline, mutation and so on. We can identify the basis for a capacity when it is not yet, or not now, realized, and we can identify it even as something has prevented the structures that support it from developing so as to realize the purpose for which they were evolved. For Waldron, when capacities are understood more adequately in teleological, material, and diachronic terms, we can recover different relations that an individual may bear to the realized capacity and its exercise. The infant is at the early stage of a process the unfolding of which is singularly important. The person with a profound disability is someone who, since the unfolding of that process has gone wrong, bears a tragic relation to the capacity in question. Someone at the end of her life, who suffers from dementia, is a person whose capacities are in decline.29 But a human being is someone who has, or will have, or did have, or could have had, a certain power. There are more or less ambitious ways of construing the implications of Waldron’s insightful discussion. The more ambitious strategy would be to treat the organic basis for a capacity as sufficient for the relevant form of moral standing; but this would inherit the deficiency of the genetic account. As Waldron makes the point, the idea was to say that human beings have a right to liberty and selfdetermination and free-speech on account of their capacity to make their own mind up about things, or whatever. The application of these rights to a person who has the organic basis for the capacity, 28 Waldron, One Another’s Equals, ch. 6. 29 Waldron, One Another’s Equals, 251.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

20 the value of humanity but in whom that basis has not developed, is strained because it is the exercise of the relevant capacity that the rights are designed to protect.30 Waldron’s own proposal is to treat the threshold capacity as the basis for special moral concern, but to capture the grounds for regarding those who bear a more complex relationship to the capacity as fellows—“brothers and sisters in human dignity.”31 Here Waldron means to bring out what is peculiar in Peter Singer’s suggestion that a human infant is not relevantly different from a chimpanzee supposing they have the same cognitive capacity. Waldron’s suggestion is that capacities cannot be understood independently of the teleological differences that determine differences in kinds. The infant is different from the chimpanzee because while the chimpanzee cannot develop the capacity in question, the infant can (and in other cases, mutatis mutandis, did, or could have).32 These differences properly condition our understanding of and responses to the individuals in question. But as Waldron acknowledges, this insight falls short of supporting a strong status doctrine to the effect that all human beings have special moral standing. If the moral significance of human beings is thought to turn on a capacity that has some live possibility of being exercised, and not all human beings possess that capacity, then the proposal does not secure the special standing of all human beings. Moreover, if a being who is not a human being possesses the relevant capacity, then what would be the motivation for excluding them from the relevant form of consideration? That would be on a par with counting only the interests of the rich, or only those of one’s kin, or only those of people who are expedient to one. The feeling would be that defenders of the idea of common humanity have forsaken their own ideals of impartiality.33

30 Waldron, One Another’s Equals, 250. 31 Waldron, One Another’s Equals, 251. 32 Waldron, One Another’s Equals, 240. 33 I have learned from Raz’s discussion of doctrines of moral status in Joseph Raz (2002). “On Frankfurt’s explanation of Respect for People”. In: Contours of Agency: Essays on Themes from Harry Frankfurt. Edited by Sarah Buss and Lee Overton. Cambridge, MA: MIT Press, pages 299–315.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

1.2 a turn to value 21

1.2 A Turn to Value I began with the broad appeal of the idea of common humanity, the idea that human beings should be treated in certain ways and not others irrespective of their nationality, religion, sexual orientation, and so on. According to a characteristic development of this idea in late twentieth-century Anglo-American philosophy, a special moral status embraces all of us, distinguishing human beings from whatever is not human. So, as Waldron gives voice to the view, though he recognizes that his arguments do not take him all the way to the conclusion, there is a minimal threshold for special moral standing that is met by human beings at the lower end of a spectrum for the relevant form of functioning, e.g. moral reasoning. Below that threshold is a creature whose standing is different in kind, as Waldron gives an example, “a monkey:” It seems to me obvious that the threshold marks a massive difference here: a monkey may have some rudimentary sense of fairness—perhaps—but it is orders of magnitude inferior to that of even the morally incompetent human. Even if the human doesn’t use it, the human has access to a moral capacity which is quite unlike the monkey’s. That fact matters more for purposes of elementary human worth than where, exactly, the human is located or has chosen to locate himself on the scale of exercises of this capacity.34

However motivated by ideals of impartiality, this insistence on a boundary between human beings and animals strikes others as difficult, and there is a demand for explanation.35 As I have sought to bring out, in thinking about questions of explanation, one is led to basic issues about philosophical methodology. My own intermediate conclusion is that where the appeal to common humanity is formulated as an all and only claim, a claim to the effect that all and only human beings enjoy the relevant form of moral consideration, attempts to support it fail. We can learn much from defenders of this idea, but 34 Waldron, One Another’s Equals, 134–135. 35 For insightful recent discussion of the importance of questions about explanation, see Sarah Buss (2012). “The Value of Humanity”. In: The Journal of Philosophy CIX, pages 341–377.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

22 the value of humanity as I will indicate below, I think that way of understanding the import of common humanity is part of what they get wrong. For proponents of the dilemma, failure to explain why all human beings have special moral significance is grounds for theorizing in a different way. We should not begin ethical theory by asking what is owed to human beings. We should situate concern for human beings in a more general and impartial account of practical concern. For consequentialists the ground of practical concern is value and disvalue in general, and for utilitarians that means pleasure and pain in particular. Whether we follow the consequentialist way of thinking about value depends on our stance on a range of foundational issues. These are issues about the nature of pleasure and pain, about the primary bearers of value, and about the primary response to value. Familiarly, the basic axiom for utilitarians is that the capacity to suffer and to enjoy is what matters ethically because pain is bad and pleasure is good. As Nagel gives rein to the utilitarian impulse in ethics, pain and pleasure make a claim on anyone no matter whose pain or pleasure they are. Anyone has reason to want brute bodily pain to stop—injury, sickness, hunger, thirst, cold, and exhaustion—and in suitable circumstances, to help it to stop. Likewise, anyone has reason to want simple bodily pleasure to be experienced—the pleasures of food, drink, sleep, sex, warmth, and ease—and in suitable circumstances, to bring it about that they are.36 The duration and intensity of these states are the only factors that weigh in deliberation.37 Nagel is surely right that pleasure and pain are central topics in ethics. Plausibly, how our life goes has much to do with the extent to which we learn to respond to them well or badly. But, together with others, I find it peculiar to say that our primary relation to pleasure is that we want it to occur, even in our own lives. It seems to me that we structure our lives around the pursuit of certain pleasures and

36 Nagel, View From Nowhere, ch. VIII, §5. 37 Nagel does not support consequentialism outright. His own view is importantly and deliberately complicated by other-than-consequentialist considerations in ethics: for example, reasons of autonomy and deontological constraints.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

1.2 a turn to value 23 not others, giving deliberative priority to those we regard as higher or better so that they constrain our more particular actions. That we would take pleasure in something is often seen by us as no reason at all to pursue it. Its reason-giving force may be ruled out entirely by other considerations: for example, that it would come at someone else’s expense.38 Comparable claims may be made about pain. I am not sure that a conception of pain as bad-and-to-be-eliminated is entirely wise. If, as in many ethical traditions, life is full of suffering, then a conception of pain as bad sets us on a path of avoidance and distraction without end. If we could be brought to see that there is nothing inherently bad about these experiences, that the problem is fear or avoidance itself, then we might be rather liberated. We might not try as hard as we can to protect ourselves from experiences that are inevitable for us, and we might live in accordance with our being. The idea that production and elimination are the paradigm responses to values is something of a dogma in value theory. It is a natural complement to the view that the primary bearers of value are states of affairs. Both dogmas have come under intense scrutiny over the last thirty or more years by theorists who take and extend Kantian insights: in particular, the insight that people (and more generally, individuals or objects) are bearers of value. For these theorists, any kind of object or fact may be appropriately such as to constrain our deliberations, to give us reasons to behave in certain ways and not in others in regard to them. If states of experience are not the exclusive or the paradigmatic object of valuation, then it is not obvious that values should be met with the uniform response of maximization or production. Instead, values can be seen to invite a fuller range of appreciation, respect, love, engagement, non-destruction, protection, valuing, and so on.39 38 Cf. Scanlon, What We Owe To One Another, ch. 2. 39 On non-consequentialist approaches to values, cf. Elizabeth Anderson (1993). Value in Ethics and Economics. Cambridge, MA: Harvard University Press. Cf. Joseph Raz (2001b). Value, Respect and Attachment. Cambridge: Cambridge University Press, ch. 4, §7. Cf. also Samuel Scheffler (2010). “Valuing”. In: Equality and Tradition: Questions of Value in Moral and Political Theory. Oxford and New York: Oxford

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

24 the value of humanity I go this far with consequentialists. I think we should situate questions about the ethical significance of human beings in a broader account of practical concern. I share the consequentialist view that the right general approach to normative reasons is value-based. For reasons given, however, I do not adopt a consequentialist development of these ideas. I maintain that beings (and more generally objects) can be bearers of value, and that while our response to beings can include, say, the promotion of their ends, it may involve a plurality of responses besides. But this style of approach shares the deflationary spirit of the consequentialists in seeing human beings as like other things that are practically relevant in being bearers of value. The natural world, sentient beings, human beings, works of art, artifacts, public institutions are all possible bearers of value, and accordingly, they are all appropriately such as to constrain our deliberations. Though I will amend and even radicalize his approach in the chapters to follow, this way of setting things up owes much to Joseph Raz.40 Raz’s theory of the ethical significance of human beings builds on his quite general account of the relationship between normative reasons and values.41 According to Raz, we register something as an appropriate object of practical concern by saying, variously, that it is worth protecting, significant, necessary, attractive, important, pleasant, worthwhile, precious, etc. These are dimensions of goodness or value, so that value is the ground of practical concern. According to Raz, whatever is of value generates the minimal but stringent reason to preserve and protect it—to see to it that it is allowed to play its role, and to be realized, as the valuable thing it is. These are categorical reasons in the sense, both, that they do not depend on a person’s tastes, interests, or plans, and that flouting the reason is not just ill-advised

University Press, pages 15–40. On the differences between (broadly-speaking) Kantian and consequentialist approaches to value, cf. Ben Bradley (2006). “Two Concepts of Intrinsic Value”. In: Ethical Theory and Moral Practice 9, pages 111–130. 40 Joseph Raz (2001a). “Respecting People”. In: Value, Respect and Attachment. Cambridge: Cambridge University Press, pages 124–175. 41 See Raz, “Agency, Reason and the Good,” 30 and passim. Cf. also Joseph Raz (1986). The Morality of Freedom. Oxford: Clarendon Press.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

1.2 a turn to value 25 but wrong. Differences in how we respond to objects and activities of value turn on differences in the character of those values. For example, the fact that this creature can exercise the capacities that are part of its form of life imposes constraints on our responses to it. We should protect its ability to exercise those powers. In the case of animals with complex social relationships, or desires for the future, this may extend to a duty not to kill. In the case of animals who have no sense of the future, it may not.42 The specific content of the reasons will depend on the nature of the value-bearer in question. I am interested in human beings as one among other objects of ethical concern in this work. While the approach taken here is deflationary, in other respects it shares key ideas with the broad group of theorists I have labeled “proto-Kantian.” I share the view that the value of human beings should be theorized in terms of some relevant power or capacity. I also share the view that we take an interest in that power or capacity because the exercise of it is interesting. Waldron is right to conceive of this capacity (whatever it is) as something to which all human beings bear an important relation. But an account that looks to ground the value of human beings in a capacity holds only for those human beings in whom there is a live possibility that it can be exercised. A person who lives in conditions of deprivation may not now be in a position to exercise the relevant capacity, but she will if conditions change, or she moves elsewhere. The thought would be that it is precisely because a person living in conditions of deprivation cannot exercise the relevant capacity that we have reason to do what we can to change those conditions.43 But the account does not extend to people in whom there is not in this sense a live possibility that their powers can be exercised—it does not extend to people in irreversible coma, or people with profound disabilities. I am giving 42 Incidentally, this is not a point of contention between consequentialists and proto-Kantians. Singer is prepared to allow that it would be worse to kill beings with desires for the future than beings who do not have those desires. Cf. Singer, Animal Liberation, 18–21. 43 Cf. Joseph Raz (2004). “The Role of Well-being”. In: Philosophical Perspectives 18, pages 269–294, p. 290.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

26 the value of humanity an account of the value that is typical and characteristic of human beings, and I am deliberately using the language not of “persons” in the philosophical sense, but of “human beings.” Statements about the value of human beings have a generic character, and they are not thrown into question by the identification of exceptions. The account does not leave people who are exceptional out of ethical consideration. Views which treat humanity as the mark of moral concern are in a troubling position regarding human beings who lack the criteria for humanity as they define it; efforts to accommodate those human beings appear to be ad hoc. The approach taken here is not subject to this difficulty because the value that is typical of human beings is one among other values that appropriately constrain our behavior in ways that are sensitive to its nature and ground. All bearers of value should be treated as the bearers of value they are, and we should attend to the particularities of an individual’s reality—to the needs and the capacities of the being who is before us. In my view, not to do so is to fail to respond adequately to their situation in the world. The question of who is eligible for the value that is typical and characteristic of human beings and who is not is often taken to be the question in these discussions because it is assumed that having that value sets its bearer apart from everything else. This is not the view taken here. While I will argue that there is something distinctive about human beings and what is owed to human beings, as I have anticipated, I am not treating the value of humanity as different in kind from the value of other things. Part of the burden of Chapter 5 will be to show that we can capture what is owed to human beings without saying so. While there are a plurality of possible positions about the nature of the value, and great contention about it—this is the controversy at the center of this book—grounding ethical responses to human beings in the value of human beings is a classic approach to the topic. With some qualifications, it is Kant’s approach. But it does not garner universal assent, not even among Kantians. In his landmark discussion of “respect for persons,” Stephen Darwall raises an objection to value-based approaches. We respond to people in a way that tracks value-bearing features of them when we express veneration

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

1.2 a turn to value 27 for forms of excellence, including moral excellence (in Darwall’s terminology this is “appraisal respect for persons”). But we do not track value-bearing features of people when we constrain our behavior by acknowledging the fact that they are people in our deliberations concerning them (in Darwall’s terminology, when we show “recognition respect for persons”). For the whole point of recognition respect is that people should be treated in certain ways and not in others just because they are people, and not because they are people who have achieved forms of excellence—the virtuous among us, the good tennis players, the insightful critics, etc. So value cannot be at stake in explaining the sort of responses that are owed to people just because they are people.44 The argument assumes that for a person to be a bearer of value she needs to have made something of herself—of her gifts or talents; she needs to have cultivated some skill or exhibited positive forms of behavior that have become stable and enduring features of her. No doubt our primary encounter with the idea that people are good or of value is like this—“you should go to the lecture because she’s very good at explaining difficult ideas,” “he’s an inspired dancer,” etc. But as we have seen, we also have the idea of capacities as precious. We can experience sadness over our own (or another’s) wasted talents, and we can be moved to do something about, say, poverty because it stands to impact a person’s ability to realize a very real capacity. The discussion brings out that what Darwall is calling “recognition respect for persons” is not not a form of appraisal—not not valueinvolving. It is just that the object of appraisal—the ground of value— is a capacity rather than its exercise. Darwall’s own suggestion is that persons are persons in virtue of a capacity to act for reasons, which is to say, deliberately.45 A natural way to express this is to say that the fact that someone is a person properly constrains our behavior because the capacity to act for reasons is of value. Why else should it make a 44 Stephen Darwall (1997). “Two Kinds of Respect”. In: Ethics 88.1, pages 36–49, see esp. p. 46. Darwall’s target here is Carl Cranor (1975). “Toward A Theory of Respect for Persons”. In: American Philosophical Quarterly 12.4, pages 309–319. 45 Darwall, “Two Kinds of Respect,” 49.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

28 the value of humanity claim on us? This is fully compatible with thinking, indeed it implies, that the fact that someone has an ethically significant capacity makes certain sorts of responses appropriate, while the fact that someone has realized that capacity, and achieved some form of excellence, licenses others.

1.3 Distinctions in Value We human beings are subjects of ethical concern because we are bearers of value, and the question is how to conceptualize and explain this value. Accounts of the value of human beings tend to have two components. One is an account of the typical features in virtue of which we are of value. So we can ask: Are we of value in virtue of a capacity for rationality, or self-consciousness, or agency, or something else? This is a question about metaphysical dependence. Familiarly, to make claims about the features on which the value of something (wholly) depends is to identify features of the thing such that, necessarily, anything with those features has the value in question (and might not have it without them). Dependence has to do with the non-variability of value properties in the absence of variation in their subvening properties.46 Questions about metaphysical dependence have dominated discussions of the value of humanity. They drive one line of interpretive controversy about Kant’s position—does Kant take human beings to be of value because we are able to set ends, or legislate the moral law?—and questions about dependence tend to be what

46 G. E. M. Anscombe was early to think about this relation, in her terms, of relative bruteness. See Anscombe, “Modern Moral Philosophy,” 3–5. Anscombe was interested in the relationship of brute relativity between certain descriptions and certain sets of facts, but her points transpose to the relationship between descriptions and sets of properties. Anscombe investigated supervenience as part of a discussion of the fact– value distinction, and the would-be naturalistic fallacy. I am on the side of those who are not quite sure what the fallacy is supposed to be, and who do not see a difference in kind between facts and values. Here I am influenced by W. K. Frankena (1939). “The Naturalistic Fallacy”. In: Mind 48.192, pages 464–477. For me discussions of ethical naturalism begin with Philippa Foot (2001). Natural Goodness. Oxford: Oxford University Press.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

1.3 distinctions in value 29 motivate independent positive proposals. As Waldron describes his undertaking, he is “tracking down” properties that form the basis of our reasons to respond to people in very particular ways. Waldron’s is only the most recent book-length treatment of candidates, and of the range of their application. But there is a second component to accounts of the value of humanity, usually implicit, and often under-theorized. This is the provision of some explanation of how a candidate feature or set of features makes its bearer of value. Why should rationality, agency, or whatever make human beings valuable? How do they explain our value? What is the character of the value that they imbue us with? For Kant, the ability to legislate the moral law makes people absolutely valuable—absolutely as opposed to relatively valuable. Kant makes an important distinction here. It is a distinction that is crucially relevant to the chapters to follow. Either some feature makes people non-relationally valuable or it makes people relationally valuable, and an account of the value of humanity should take a position one way or the other. On the scheme that I will employ in this work, an account of the value of human beings involves some discussion of the nature of the features on which our value depends—call this the basis of value. It also involves an account of whether the properties that form the basis of our value make us relationally or non-relationally valuable—call this the explanation of the value of the basis. Both are relevant to how we should be responded to. My account of the value of human beings addresses these two questions, and the answers taken together provide the main contours of a theory of our value. In value theory the terminological issues are somehow especially vexing, so let me pause over questions of terminology here. I just invoked a distinction between absolute and relative value, or as I transposed it, between non-relational and relational value. The relation at stake in this distinction as I am employing it is the relation of being good for someone. So as I will use it here, “non-relational value” expresses the idea of a form of value that is not a function of being (actually or potentially) good for someone, while “relational value” expresses the idea of a form of value that is a function of

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

30 the value of humanity being (actually or potentially) good for someone. Taken this way, the distinction between non-relational and relational value is equivalent to that between good period and good for. This is a basic distinction in the metaphysics of value. There are other distinctions that look superficially similar to the one that centrally concerns me, so let me make the differences clear.47 Take the distinction between instrumental and non-instrumental value. As I will argue in Chapter 2, there is historical precedent for conflating “good for” with instrumental value, but this is a mistake. I find the notion of instrumental value less clear than it is often assumed to be, but for my limited purposes here it is enough to suppose that what is instrumentally valuable is of value because it contributes to something else that is of value. Being instrumentally valuable is a way something can be good for someone. What is instrumentally valuable is indirectly good for someone by contributing to something else that is directly good for them. Having a hammer permits someone to hang a painting the viewing of which is by itself good for them and others. That is to allow that things can be directly good for people—that is, non-instrumentally good for them—and we might put friendship, higher learning, the engagement with works of art (among other things) in this category. Arguably, it is an important evaluative category, one with which ethics is broadly concerned. So, instrumental value is not the only way something can be good for someone. Something can be non-instrumentally good for someone, 47 There is a large and fascinating literature on distinctions in goodness or value, and everyone has their own take on what the relevant distinctions are. To some extent the approach depends on the theoretical purposes for which the distinctions are being drawn. My own treatment is no exception. For important discussion of distinctions in goodness cf. Wlodek Rabinowicz and Toni Rønnow-Rasmussen (2000). “A Distinction in Value: Intrinsic and For Its Own Sake”. In: Proceedings of the Aristotelian Society, New Series 100, pages 33–51. Cf. Shelly Kagan (1998). “Rethinking Intrinsic Value”. In: The Journal of Ethics 2, pages 277–297. The classic paper on the distinctions in goodness is due to Christine Korsgaard (1983). “Two Distinctions in Goodness”. In: The Philosophical Review 92.2, pages 169–195. For significant commentary on Korsgaard’s paper, cf. Rae Langton (2007). “Objective and Unconditioned Value”. In: Philosophical Review 116.2, pages 157–185. I draw from these works in my discussion of the distinctions below. I am most grateful to an anonymous referee for prompting me to clarify the relationship between the distinction that interests me and a set of related distinctions.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

1.3 distinctions in value 31 and since that is a form of relational value—since its value turns on being good for someone—non-instrumental value is not equivalent to good period.48 A related further distinction is that between derivative and nonderivative value. The value of what is derivatively valuable is explained by the value of something else, while the value of what is nonderivatively valuable is not so explained. Instrumental value is a kind of derivative value, but it is not the only kind. The value of a symbol may derive from the value of what it is symbolic of, but it is not instrumentally valuable. Eudaimonia is a paradigm example of something that is non-derivatively valuable; for Aristotle, famously, it is the most complete end or final good. But the example brings out that the distinction between derivative and non-derivative value is not equivalent to that between relational and non-relational value. Eudaimonia is non-derivatively valuable, there is no higher value that it derives from, but it is arguably still relationally valuable. Plausibly, the good life is good for the person whose life it is. In that case, eudaimonia is a non-derivative form of relational value. The relationship between non-relational and relational value and intrinsic and extrinsic value also calls for comment. These terms are themselves used in a variety of ways, and on some uses the two distinctions come together. There are ongoing discussions about how we should use “intrinsic” and “extrinsic” value, but I will use them in a more or less traditional way.49 I will use the term “intrinsic value” to refer to those things that are of value in virtue of their intrinsic properties, and “extrinsic value” to those things that are of value on the basis of their extrinsic properties.50 Taken that way, the two distinctions concern importantly different ideas. Intrinsicality and extrinsicality concerns the nature of the properties on which something of value 48 Some argue that in order for something to be non-instrumentally good for someone it must also be good period, and I consider this proposal in Chapter 5. What matters here is that non-instrumental good for and good period are conceptually distinct, and this is accepted by those who take the one to entail the other. 49 For relevant discussion, see Michael J. Zimmerman (2005). “Introduction”. In: Recent Work on Intrinsic Value. Edited by Toni Rønnow-Rasmussen and Michael J. Zimmerman. Vol. 17. Library of Ethics and Applied Philosophy. Dordrecht: Springer. 50 Here I depart from Kagan, “Rethinking Intrinsic Value.”

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

32 the value of humanity depends, while non-relationality and relationality concerns a specific metaphysical question about the explanation of the value of a given thing (whatever the character of its subvenient properties), namely: Is its value indexed to being in some way good for someone, or not? The two distinctions come apart. Conceivably, something could be valuable on the basis of its extrinsic properties, though the explanation of the value of those properties is non-relational. On some views, whether plausible or not, the oldest tree in the world is of value independently of its relationship to any valuers. In that case, the value of the tree supervenes on an extrinsic property, being the oldest, though the explanation of the value of the oldest tree is non-relational. On the other hand, something could be valuable on the basis of its intrinsic properties, though the explanation of the value of those properties is relational. Some may take this to be a plausible conception of the value of a work of art. They may urge that the value of a painting necessarily depends (in part) on its intrinsic properties—on its coloration, on the density of the application of the paint, etc.—though what makes that particular constellation of intrinsic properties valuable is its propensity to enrich a viewer—a form of relational value. I will be especially concerned with forms of value that are of value on the basis of their intrinsic properties though they are relationally valuable in the sense that they are of value because they are or can be good for someone. Finally, the distinction between absolute and relative value may be used to mark a difference in the scope of the reasons to which a value gives rise. We may distinguish value that makes a claim on anyone, and is in that sense absolute, from value that makes a claim only on some, and is in that sense relative. What this difference amounts to will in part depend on who “anyone” or “someone” is a sample of. If the sample is human beings, then this is the difference between value that makes a claim on any human being and value that makes a claim only on some human beings.51 If the sample is all rational beings, then 51 This is a version of Nagel’s influential distinction between agent-neutral and agent-relative reasons and values. See Nagel, View from Nowhere, 152–153. I am grateful to Katja Vogt for discussion on this question, and I take the “sample” locution from her.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

1.3 distinctions in value 33 it is the difference between value that makes a claim on all rational beings (including God or gods) and value that makes a claim only on a class of rational beings (for example, only human beings).52 I have been discussing a set of evaluative distinctions that are related to but different from the distinction between non-relational and relational value as I am defining it. We clarify what we mean by a given term by saying what we do not mean by it. But plausibly, the value of any object or activity can be articulated in terms of the several distinctions that I have just outlined. Certainly, the account of the value of human beings that I develop in this work can be expressed in terms of them. To look ahead, I will develop the view that the value of human beings is: (i) relational; (ii) non-instrumental; (iii) nonderivative; (iv) depends on a set of intrinsic properties; and (v) makes a claim on (at least) all human agents. Chapters 2 and 3 develop (i). Chapter 4 defends (ii) and especially (iv). Chapter 3 gives my rationale for (iii). Chapter 5 gives an argument for my entitlement to (v).

52 See Kagan’s discussion of the contrast between (as I would put it) proposals according to which human beings have a value that is practically significant for other human beings, and proposals according to which human beings have a value that is practically significant for all rational beings including, in Kagan’s example, rational aliens. See Kagan, “What is Wrong with Speciesism?” I am grateful to an anonymous reader for referring me to Kagan’s article, and for most helpful discussion.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

2 Of Absolute and Relative Worth Relationality and non-relationality has emerged as an important distinction within evaluative explanation, but so far the distinction has been very schematically stated. In this chapter I endeavor to make the distinction determinate by thinking historically. The value of humanity is both an important topic in theoretical and normative ethics, and a key issue in the history of philosophy. Treatments have tended to proceed on both fronts, so that contemporary accounts are in conversation with the history of the tradition, and so that interpreters of the tradition are motivated by ethical and theoretical concerns. This book is not a work of philosophical history, but the concepts that interest me have a history, and it is important to bring this into focus for conceptual reasons. In what follows I recall key features of the form of value that Kant extends to humanity. I attend to Kant’s sense that this kind of worth is liable to seem puzzling, even peculiar. I bring out what is distinctive about Kant’s conception by locating points of difference from his predecessors, and in particular, from the ancient Greeks. I offer a reading of some key differences between modern and ancient conceptions of value in which there are undoubtedly larger socio-historical forces at work, and other significant players, but in which Kant is a central figure and architect. Here I seek to recover some ancient ideas about value that in one way Kant sees clearly, and in another he obscures. The discussion

The Value of Humanity. L. Nandi Theunissen, Oxford University Press (2019). © L. Nandi Theunissen. DOI: 10.1093/oso/9780198832645.001.0001

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

2.1 a kantian legacy 35 is important for my argument because it gives broader context and significance to the dispute over good and good for that is at the heart of this book.

2.1 A Kantian Legacy It can be difficult to take the temperature in a field, but if I were to take the temperature in contemporary moral philosophy, I would say that by the early part of the twenty-first century in North America, Kantianism has risen to a position of relative dominance. Kantianism is by no means a position universally held, but it is one in relation to which one is asked to articulate one’s terms. This may be part of the legacy of John Rawls and his influential students— Christine Korsgaard, Barbara Herman, Onora O’Neill, and others. These readers of Kant have tended to place particular emphasis on the “Formula of Ends” version of the Categorical Imperative according to which people have a special kind of value that must be taken into account in all of our interactions with them. This is arguably the most widely accepted dimension of Kant’s position, and for many the most appealing. It is no accident that Kant’s moral philosophy has gained traction this way—by emphasizing Kant’s views on humanity. Under this set of emphases Kant is read as a key figure in the human rights tradition: an arch proponent of the view that simply by being a creature with certain powers, which is to say, independently of whether a creature makes good on those powers, people have a distinctive worth that constrains what is permissible and impermissible in our dealings with them. Though the position has exerted considerable influence, textually speaking it is not without difficulty—all sorts of interpretive questions come up. It would not be an exaggeration to say that a literature has hemorrhaged around Kant’s discussion of humanity in the Groundwork of the Metaphysics of Morals. Several topics have particularly exercised the field. Interpreters have argued over how Kant conceives of humanity—whether it is setting ends, or

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

36 the value of humanity valuing, or freedom, or having good will.1 There is controversy over whether Kant takes the capacity for or the exercise of the morally relevant power to matter for absolute worth. Commentators have pointed to passages that suggest a starker-than-expected view in which Kant seems to be saying that it is the exercise of powers, and therefore merit, that makes all the difference to special standing.2 There is also great contention over how Kant argues for the claim that humanity is absolutely valuable. Interpretation has tended to be riven down meta-ethical lines, with some reading Kant as a realist, and others as a constructivist.3 Further discussion has centered on what the Formula of Ends entails for our treatment of human beings. On an influential reading, deception, force and coercion are ruled out as

1 This is a guiding question in interpretive disputes between, among others, Allen Wood (1998). “Humanity As End In Itself ”. In: Kant’s Groundwork of the Metaphysics of Morals: Critical Essays. Edited by Paul Guyer. Boston: Rowman and Littlefield; Paul Guyer (1998). “The Possibility of the Categorical Imperative”. In: Kant’s Groundwork of the Metaphysics of Morals: Critical Essays. Edited by Paul Guyer. Lanham: Rowman and Littlefield, pages 215–246; Richard Dean (2006). The Value of Humanity in Kant’s Moral Theory. Oxford: Oxford University Press. 2 Stephen Darwall (2008). “Kant on Respect, Dignity and the Duty of Respect”. In: Kant’s Ethics of Virtue. Edited by Monika Betzler. Berlin: De Gruyter, pages 175–200 has raised this issue on the basis of some under-read passages though he in the end reads them in the more traditional way. 3 An influential realist interpretation is given by Rae Langton (2007). “Objective and Unconditioned Value”. In: Philosophical Review 116.2, pages 157–185. See also Robert Stern (2013). “Moral Skepticism, Constructivism, and the Value of Humanity”. In: Constructivism in Ethics. Cambridge: Cambridge University Press, pages 22–62. As Stern brings out, realists and constructivists have tended to make different parts of the text central, with constructivists taking the core argument to arise at 4:429 with the claim that human beings necessarily represent themselves as ends in themselves. The most influential constructivist reading is due to Christine Korsgaard (1986). “Kant’s Formula of Humanity”. In: Kant-Studien 77, pages 183–202. But constructivist readings are independently proposed. See for example Oliver Sensen (2009). “Kant’s Conception of Inner Value”. In: European Journal of Philosophy, pages 1–19. It is a strength of Sensen’s paper that he raises the question of how realism about value of the kind we are familiar in the writings of G. E. Moore could be compatible with Kant’s broader systematic commitments. A review of the literature (which is vast, and more than I can take account of here) is given by LaraDenis (2007). “Kant’s Formula of the End in Itself: Some Recent Debates”. In: Philosophy Compass 2.2, pages 244–257.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

2.1 a kantian legacy 37 always impermissible. Others reject this entailment as implausible and implausibly Kantian.4 Fascinating though they are, these interpretive questions are orthogonal for my purposes. My interest is not principally in what Kant takes humanity to be. Nor is my question whether Kant thinks all or only some human beings have distinctive value. I am also not asking—or not in the first place—how we should classify Kant in light of contemporary meta-ethical categories of realism and antirealism. My interest is in the prior and more specifically valuetheoretic question of how Kant explicates the absolute form of worth that distinguishes humanity, and as a corollary of that, how Kant conceives of the value of humanity vis-à-vis other valuable things. Kant’s conception of absolute value is a composite—it contains a number of ideas. But a key feature is that the value of what is absolutely valuable is independent of benefit and a propensity to benefit. Kant characterizes absolute worth as benefit-independent in the course of his discussion of the value of the good will, a discussion that importantly prefigures his treatment of the value of humanity. Kant himself raises a question about the plausibility of value so conceived, and I explore his question here. What might be difficult about the idea of a form of worth that is in no way indexed to its propensity to help or serve or enrich? I do not propose to settle the question in this chapter. It is the question that presides over this book and motivates the investigation in the chapters to follow. The task of this chapter is to provide some basis for thinking that the question is worth asking and the investigation worth undertaking. As I will have occasion to articulate more than once, in thinking about the value of humanity we are called upon to think about the nature of value in general—about what sorts of things are good or of value, and why. Certainly Kant’s treatment of the value of humanity 4 For a reading that emphasizes non-coercion, see Onora O’Neill (1985). “Between Consenting Adults”. In: Philosophy and Public Affairs 14.3, pages 252–277. For a challenge to this traditional style of proposal, see Japa Pallikkathayil (2010). “Deriving Morality from Politics: Rethinking the Formula of Humanity”. In: Ethics 121, pages 116–147.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

38 the value of humanity emerges from this broader inquiry. Kant’s project begins, as moral philosophy traditionally begins, with claims—to put it colloquially— about the best thing there is. The classic strategy is to make a case for what the best kind of worth would need to be like to be counted as the best—hallmarks of superlative value—and to put forward a candidate that meets the criteria. Kant is no exception. The famous opening sequence of the Groundwork lays out the distinguishing features of the supreme good relative to other goods.5 While all other goods are fickle—good in some circumstances but not in others— the supreme good alone is always good. And while all other goods need the supreme good to be alloyed to them if they are to be good on some occasion, or conditionally, the supreme good does not need the presence of other things to be valuable and is for that reason unconditionally good. But the feature on which Kant lays particular emphasis as the sequence unfolds is that the supreme good is not merely always good, and is not merely necessary for other things to be valuable, but is good independently of what it does or is capable of doing. In a passage that is so frequently cited that the radical point Kant means to make risks being blunted by habit of repetition, the value of what is supreme is not a function of its effects or accomplishments: of what it brings about, or achieves, or is for. As Kant underscores the point, “Usefulness or fruitlessness can neither add anything to this worth nor take anything away from it.”6 Usefulness is simply not the evaluative mark of what is best.

5 I use the term “supreme” here advisedly. The traditional term in this context is the “summum bonum,” the highest good. Kant certainly makes use of this term but he regards it as ambiguous between the supreme good—the highest-ranking good— and the complete good—the most-inclusive good. The two sense are made explicit in Immanuel Kant (1997b). The Critique of Practical Reason. Edited by Mary Gregor. Cambridge: Cambridge University Press, 5:110. They are also anticipated in Immanuel Kant (1997c). The Groundwork of the Metaphysics of Morals. Edited by Mary Gregor. Cambridge Texts in the History of Philosophy. Cambridge: Cambridge University Press, 4:396. In the immediate context that interests me here, Kant is concerned with the supreme good. I turn to his treatment of the complete good later in the chapter. 6 Kant, Groundwork, 4:394.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

2.1 a kantian legacy 39 If these are Kant’s criteria for what is supreme, we are told at the outset that it is the good will that satisfies them, virtue in Kant’s distinctive sense. To have good will is expressly not—as the phrase tends to be used in ordinary English—to intend to do good for another, or to wish them well. It is to have accepted the moral law, colloquially, rectitude, as the basic coordinate in one’s life. The case is made for the good will over a variety of familiar candidates for what is supreme—empirical character and happiness among them— and its status as valuable independently of its effects is made out by the following conceit. We are to imagine a person’s volitional system that is so badly made that she is unable to enact or embody her plans in action, realizing nothing. It must be said that what we are to imagine is clear enough in cases of complex action that require cooperation from the world, and less clear where the gap between willing and succeeding is small. (Think of the difference between sending a charitable donation by mail and refusing to lie.) But the point is meant to be that even the agency that is limited in this way has a worth that is superior to anything else. Some read a concern with moral luck here, and matters of control. The discussion may have implications for these topics but they are orthogonal. Kant is not addressing questions of responsibility and blame, he is using some form of an isolation test to emphasize just how fully he means to distinguish the value of what is best from the value of what is brought about. As Stephen Engstrom has made the point, “Kant is denying, not merely that the goodness of the good will stems solely from the good effects the good will produces, but that its goodness stems from such effects at all.” As Engstrom adds: “To say of the good will that one values it (to whatever extent) for the good it produces would be ungereimt—it would be a kind of absurdity.”7 Outcomes are simply not the marker of the superlative form of value that interests Kant here. What is supreme does not need to do good and it does not need to be capable of doing good.

7 Stephen Engstrom (1996). “Happiness and the Highest Good”. In: Aristotle, Kant, and the Stoics. Edited by Stephen Engstrom and Jennifer Whiting. Cambridge: Cambridge University Press, ch. 4, page 111.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

40 the value of humanity Kant will introduce some terminology for this distinctive form of value, and it is terminology we need. Whatever is good because it is or can be good for something or someone—because of what it can do or bring about or serve—has relative worth, while whatever is good independently of whether it is or can be good for something or someone has absolute worth.8 Kant frames his discussion of the value of humanity in just these terms. He tells us that while other things— the things we seek, and equally, all of nature—have relative worth, rational beings alone are absolutely valuable.9 What is it about rational beings that gives them this distinctive form of worth? The expected answer, from what has been said, is that rational beings bear some relevant relationship to rectitude—to the good will. And though it has become customary to deflate Kant’s moralism with the suggestion that our value lies in a capacity to “set ends” in general, this expectation 8 Kant, Groundwork, 4:428. Kant makes the further claim that the ground of relative value is desire, since in his view aims or outcomes find their point in view of our inclinations (as opposed to our judgments about what would be worth doing). 9 As I read him, Kant’s argument here (4:427–429) proceeds in two stages. There must be something that is absolutely valuable for without it there could be no moral law (stage 1); humanity meets the criteria for bearing absolute value (stage 2). The argument for stage 1 is that action in accordance with the moral law requires an end, or good. Since the law is categorically binding, this good cannot be relative to inclination—it cannot be good because we want it and not otherwise, and for Kant that is equivalent to saying that it cannot be good because it is good for something or someone. The good in question must be absolute. The second stage of Kant’s argument is that, unlike the other candidates he considers—the objects of our inclinations, nonrational beings—the nature of human beings, as rational beings, marks them out as absolutely valuable. I have learned much from the reading given by Jens Timmermann (2006). “Value Without Regress: Kant’s ‘Formula of Humanity’ Revisited”. In: European Journal of Philosophy 14.1, pages 69–93. Importantly, Kant’s argument thus far is analytic in the sense that he is outlining what the notions of morality, of law, of duty, entail, and here we learn that they entail that humanity is absolutely valuable and an object of respect. Kant is not now arguing that there is such a thing as morality in his sense, or that humanity is indeed absolutely valuable; that would be the work of a deduction. And what we come to learn is that none of this can be deduced: we do not verify that human beings are absolutely valuable on the basis of experience, or study, or anything else. Rather, a conception of humanity as absolutely valuable is an implication of our self-conception as subjects who are bound by morality in Kant’s distinctive sense.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

2.2 a departure from the ancients 41 is borne out in what Kant goes on to say: “morality, and humanity insofar as it is capable of morality, is that which alone has dignity,” where “dignity” is another term for what has “inner” as opposed to relational worth.10 The suggestion is that human beings are absolutely valuable to the extent that we can have a share of morality—the best thing there is.

2.2 A Departure from the Ancients I have identified a key characteristic of the supreme good as Kant understands it. The value of the supreme good is not indexed to its actual or possible effects, and that is to say that it is absolutely valuable. I have recalled that the bearer of absolute value is in the first place the good will, colloquially, rectitude or righteousness of disposition. And I have brought out that Kant extends absolute value to humanity. Acknowledging that much interpretive energy has been expended on this extension, I have suggested that it is insofar as humanity is capable of rectitude that it has a share of absolute worth. The issue that really interests me—the issue I turn to now—is Kant’s expressed concern that the relevant form of value may be thought fantastical. In his words: There is, however, something so strange in this idea of the absolute worth of a mere will, in the estimation of which no allowance is made for any usefulness, that, despite all the agreement even of common understanding with this idea, a suspicion must yet arise that its covert basis is perhaps mere high-flown fantasy [. . .].11

Why would Kant say this? What is unusual or striking about a useindependent, or as Kant also puts it, effect-independent, form of worth? 10 Kant, Groundwork, 4:435. The “setting ends” interpretation was made famous by Korsgaard, “Formula of Humanity,” §II. 11 Kant, Groundwork, 4:394.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

42 the value of humanity Before I offer a response to that question, let me register a temptation to downplay what is striking about Kant’s proposal by thinking that he claims merely that the value in question is not instrumental. This temptation is encouraged by a tendency to define instrumental value as whatever is of value because of its valuable effects, or likely valuable effects. But this conception of instrumental value is too coarse-grained. Something can be valuable because of its valuable effects without being instrumentally valuable. Take the concept of what is non-instrumentally good for a person, and suppose that engaging with a work of art is an example of something that is non-instrumentally good for someone.12 Engaging with a work of art through reading, listening, or looking is an activity that involves emotional responsiveness, comprehension, memory, association, and so on. The activity involves thinking, and the thinking gives rise to thoughts. The subject’s imagination is set on fire. Her mind is pierced by a sudden realization that changes the state of her understanding. There is a change in the subject. There is movement from one state of being to another. There are states of transformation, and it is in these states that the value is being supposed to lie. These are not not occurrences. The form of value that is in question is not instrumental, but it is still affective. In making the distinction between what is instrumentally and non-instrumentally good for a person one would overshoot the mark to deny that the form of value that is in question in the second case is somehow outside the causal nexus.13

12 It is a commonly offered example, in my view plausibly, but it is not an uncontentious one. Should the reader have reservations she can substitute a different sort of case. What matters here is the conceptual point. 13 Those who urge that engaging with works of art does not have enrichment as an effect or consequence can be understood to be making a point against classical utilitarians who regard works of art as valuable insofar as they cause valuable experiences— pleasant emotional states or sensations—to arise in the one who engages, experiences that can just as well be caused in some other way. See Raz’s discussion of classical utilitarianism in The Morality of Freedom, 201. This is felt to be a crass view of art, and it is felt to misdescribe the phenomena. This seems right to me. But we need not cede affective notions to the utilitarian. It is not that the event of engaging with a work of art causes a separate and value-bearing state of experience to come into being. Rather, the

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

2.2 a departure from the ancients 43 So, when Kant says that the value of what is supreme is of value independently of its effects he is unquestionably denying that its value is indexed to its propensity to be instrumentally valuable. The point I have just made is that he is also denying that its value is indexed to its propensity to be non-instrumentally good for someone in anything like the above sense. The value of rectitude for Kant is not a function of its propensity to be part of what enriches anyone or anyone’s life, as engaging with a work of art may be thought to constitute such enrichment.14 For Kant, righteousness of disposition involves adopting principles of action in which we help to realize other people’s ends. But this is not because of considerations that have to do with the good for human beings—it is not because the good for human beings plays any kind of foundational role in Kant’s system. We are to adopt these principles because not doing so would involve a failure of pure practical reason, a so-called contradiction in the will. The distinctive form of value that interests Kant, the one he extends to morality and humanity, is independent of its propensity to be instrumentally good for someone, and it is independent of its propensity to be noninstrumentally good for someone. I will put this by saying that the relevant form of value is not indexed to the beneficial. And put this way it becomes clear what is striking about Kant’s proposal. Kant is writing against his predecessors in the tradition. The targets are avowedly the ancient Greeks whose criteria and candidates for the best kind of good are being opposed.15 Ancient conceptions value is in the engagement—it inheres in the activity of engaging with understanding and attention—an activity in which the work figures as an ineliminable part. 14 Questions about Kant’s treatment of the most comprehensive good—virtue together with happiness—naturally arise here, and I discuss them in the next section. 15 In fact, the first two criteria for what is best that are identified in the Groundwork discussion—being the condition of the value of other things, and being always valuable—are quite familiar from ancient discussions. For example, in Plato’s Meno the best good is something that makes other things valuable. Health, strength, wealth, and beauty can be good but they can also be bad. Courage, quickness, memory, and munificence can be good but they can also be bad. Whatever is best will be the factor that makes these things good when they are, and wisdom is identified as such a factor. So wisdom meets the criteria for being, in Kant’s terms, unconditionally and always valuable, while other things are conditionally and sometimes valuable. But where the

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

44 the value of humanity of the highest good are conceptions of the good as beneficial. The ancients are preoccupied with what it is for something to be good for someone, and with what it would be to have knowledge of the human good. They are interested in what kinds of activity, what kinds of lives, what kinds of external goods (if any), truly or ultimately deliver. For them, goodness is definitively a notion of benefit. That the good is beneficial or advantageous, and that the bad is harmful or disadvantageous, is a recurring assumption in Plato. That the good is beneficial is offered as a conceptual truth at, for example, Republic 379b. It figures as a key premise in arguments for tenets associated with Socrates: for example, that the good man cannot be harmed (Apology 41c); that everyone acts under guise of the good, and relatedly, that no-one willingly does wrong (Meno 77–78b); and that virtue (understood as knowledge) is sufficient for happiness (Euthydemos 280b–282). It figures as a topic for investigation in, for example, Protagoras 333d–334a. It is often thought that Plato abandons this conception of value when he introduces the Form of the Good, where the Form of the Good is taken to be a paradigmatic form of absolute value. One finds this conception of the Form of the Good not merely in casual references to Plato, but in prominent, you might say the

Meno proposal departs sharply from Kant way of setting things up is in taking as an explicit premise that whatever is good does good or is beneficial. The mark of the best kind of good for Plato here is that it ensures that other things are good in the sense that they are beneficial rather than harmful. Plato (1997b). “Meno”. In: Plato’s Complete Works. Edited by John M. Cooper. Indianapolis: Hackett Publishing, pages 870–897, 87d–89a. This is where Kant forcefully and explicitly takes a different view. For Kant, the best good is what makes other things valuable, and it is always good, but it does not make other things valuable in the sense that it makes them beneficial—in the sense that it allows other things to be used in such a way that they make people or their lives happy, well-going, etc. Rather, the presence of the good will ensures that gifts of fortune and qualities of temperament are not used in morally problematic ways, i.e. that maxims of action involving them are not impermissible. Here I am taking a different position from Stephen Engstrom (2016). “The Determination of the Concept of the Highest Good”. In: The Highest Good in Kant’s Philosophy. Edited by T. Howing. De Gruyter, pages 89–108 who sees fuller affinity between Kant and the Socrates of Plato’s Meno.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

2.2 a departure from the ancients 45 most prominent, scholarly discussions.16 It is not my project to engage with this interpretive legacy here. For my limited purposes it would be enough to say that I am appealing to more Socratic moments in Plato, but in fact I do not read the Form of the Good as a benefitindependent form of value at all; on the contrary, I think it is more plausibly viewed as that without knowledge of which nothing else can be of benefit (for the individual and for the city as a whole).17 Kant is not known for his insightful readings of the ancients, and indeed I will go on to urge that he flattens the character of value as the ancients understand it. But that his predecessors do not have his conception of a benefit-independent worth is a point about which he is perfectly clear-sighted. It is a difference on which Kant means to insist and in terms of which he sees himself as making an important intervention. And intervene he does. It would be hard to underestimate the legacy of Kant’s way of setting things up—of his signature conception of a benefit-independent form of moral worth.

16 See for example, John M. Cooper Cooper (1977). “The Psychology of Justice in Plato”. In: American Philosophical Quarterly 14.2, pages 151–157, and Terence Irwin (1995). Plato’s Ethics. New York and Oxford: Oxford University Press. 17 For this reading, see Terry Penner (2003). “The Forms, The Form of the Good, and the Desire for Good, in Plato’s Republic”. In: The Modern Schoolman LXXX, pages 191–233. See also Richard Kraut (2011). Against Absolute Goodness. Oxford and New York: Oxford University Press, Appendix F. Penner makes the intriguing suggestion that one can see the influence of Rawls in a prominent generation of Plato scholars, and a desire to rescue Plato from the wrong side of the Kantian divide between morality and self-interest. This results in a rather surprising reading of Plato as not a eudaimonist after all, and I share Penner’s sense that this is a surprising and difficult position to uphold. It would not be against the spirit of my remarks below to add that it concedes too much to Kant’s way of setting things up. Insofar as they conceive of themselves as Socratics, the Stoics arguably sharpen the conception of goodness that one finds in Plato’s writings. On the Stoic definition of the good as benefit, see Katja Maria Vogt (2008c). “The Good is Benefit: On the Stoic Definition of the Good”. In: Proceedings of the Boston Area Colloquium in Ancient Philosophy, pages 155–174. See also Michael Frede (1999). “On the Stoic Conception of the Good”. In: Topics in Stoic Philosophy. Edited by Katerina Ierodiakonou. Clarendon Press, pages 71–94. For contemporary approaches that self-consciously develop ancient insights, those of Plato and Aristotle, see Kraut, What is Good and Why, and Katja Maria Vogt (2017). Desiring the Good: Ancient Proposals and Contemporary Theory. Oxford and New York: Oxford University Press.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

46 the value of humanity Let me illustrate the distinctiveness of Kant’s position with an example. Like Kant, Plato of the Republic is interested in the nature of the value of rectitude—in Plato’s case, not the good will but justice. In Plato’s opening schema, the best good is one that is directly and indirectly beneficial: that is, by itself good for someone and good for someone by conducing to other things that are good for them.18 This is a version of the distinction I drew above between non-instrumental goods for people and instrumental goods for people. Plato is telling us that the best kind of good will be of value in both of these ways. Socrates is inclined to give justice as an example of the best kind of good: of something that is by itself good for us, and good for us because it conduces to other things that are good for us. But it is the first part of the proposal—the claim that justice is by itself beneficial— that seems contentious to his interlocutors. Indeed they report that most people are likely to find justice by itself onerous, even as they see that it leads to the goods of honor and their kin. So the task Plato gives Socrates is to defend the idea that justice is by itself good for the person who is just. If this is a version of an isolation test, the task is not to show that justice is valuable independently of what it does—independently of its effects. The task is to see what justice “itself does of its own power by its presence in the soul of the person who possesses it.”19 The point is to isolate the effects of justice on the mind—to show that it makes a person well-going independently of the external goods it helps to secure.20 The value of rectitude, as in the Dharmic traditions of the East, has to do with psychic harmony or the ordering of states of mind; indeed Platonic justice emerges as a form of psychic order. The contrast with Kant, for whom the value of rectitude in no way depends on its relationship to equanimity, could not be more pronounced.

18 Plato (1997c). “Republic”. In: Plato Complete Works. Edited by John M. Cooper. Indianapolis and Cambridge: Hackett Publishing, pages 971–1223, Bk II. I have learned from discussion by Nicholas P. White (1979). A Companion to Plato’s Republic. Indianapolis and Cambridge: Hackett Publishing, pages 74–79. 19 Plato, Republic, 366e. 20 For an insightful treatment of what it means to treat justice as a soul-good, see Foot, Natural Goodness, ch. 7.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

2.2 a departure from the ancients 47 Conceptions of the good are methodologically consequential, and here too we see foundational differences between Kant and his predecessors. A conception of good as beneficial drives the ancient commitment to some form of naturalism. If the good is the good for human beings, then there is interest in the nature of human beings— in the distinctive constitution of our mind or soul. Relational values must be understood in terms of their relata—of what is related to what, and how. What is it about some form of activity that affects a mind that is constituted in this way so as to be good or bad for it? And what are the markers of this relation—of being benefitted, or harmed? Again, recall the architecture of the Republic. The twin questions of What is justice? and Is it advantageous for the just person to be just? must be answered by accounts of what justice is, of what the human soul is like, and of what it is for the human soul to be faring well or badly. Here Plato is recognizably interested in what we now call moral psychology, though he offers not empirical or demographic studies, but portraits and descriptions. Kant’s way of proceeding could not be more different. For Kant, the starting point of moral philosophy is not the good for human beings. The concern is not with how we as one kind of animal are constituted and affected by this or that. Kant could not be more explicit that empirical considerations about human beings are to be excluded from the foundations of morality. Empirical generalities are not the stuff of a subject worthy of the name “morals,” for they are incapable of generating laws that hold of necessity for all rational beings. This is the center of Kant’s commitment to rationalism in ethics and it is what he sees as his greatest contribution to practical philosophy.21 It may be tempting to assign a relatively minor role to Kant’s conception of absolute value in his broader system, but I think this would be a mistake. Kant’s conception of absolute worth is merely the

21 I argue for this reading in L. Nandi Theunissen (2016). “Kant’s Commitment to Metaphysics of Morals”. In: European Journal of Philosophy 24.1, pages 103–128.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

48 the value of humanity evaluative side of his core commitments in moral philosophy.22 The central question of Kant’s theory—How is a categorical imperative possible?—may be put in evaluative terms—How is the moral good possible? In the words of Dieter Henrich, “it is not the content but the reality of the good that is the problem of ethics within the context of first philosophy,” and we should be hearing Kant’s concern that the moral good be thought fantastical in this connection.23 What reason do we have to believe in this kind of value? How could we show that it is actual? It is not known through sense perception, or induction, or introspection over time, nor is it present as a stimulus to empirical emotion.24 Here we come to the vexed question of the prospects for a deduction of the moral good, and as we know, Kant came to the conclusion that they are importantly limited—that theoretical reason oversteps its bounds in seeking to justify morality.25

2.3 A Suspicion of Fantasy I have been arguing that a mark of supreme worth for Kant—the worth he accords morality, and humanity by extension of morality— is benefit independence. I have sought to emphasize that this is a distinctive way of conceiving of value. It departs from ancient Greek conceptions of the good which, notwithstanding their many differences, take the beneficial to be its mark. So, come back to Kant’s suspicion of fantasy—to his anticipation of skepticism on the part of philosophically minded readers that the distinctive worth of the good will, and by extension, of human beings, can be as he says it is: such that the worth is not a function of the good it does or is 22 Lest it be thought that Kant cannot be interested in the good because by his own lights that would make his theory heteronomous, let it be observed that the moral good is not determined by anything other than practical reason. See Stephen Engstrom (2015). “The Complete Object of Practical Knowledge”. In: The Highest Good in Aristotle and Kant. Edited by Joachim Aufderheide and Ralf M. Bader. Oxford: Oxford University Press, pages 129–157, §1. 23 Dieter Henrich (1994). The Unity of Reason: Essays on Kant’s Philosophy. Edited by Richard L. Velkley. Cambridge, MA: Harvard University Press, page 57. 24 Henrich, Unity of Reason, 57. 25 Henrich, Unity of Reason, 83.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

2.3 a suspicion of fantasy 49 capable of doing. We might notice some complexity in the status of Kant’s remarks here. On the one hand he presents his findings as intuitive—as ideas about the good that arise even for the untutored; and on the other he regards them as quite radical and surprising. We can appreciate the felt intuitiveness of Kant’s conception of value by recalling the perspective of Kant’s Christian contemporaries. They would find the idea of the absolute worth of a good will, like the absolute worth of human beings, incontrovertible. They are wellversed in claims to the effect that we are made in the image of God, or that our distinctive worth is guaranteed by divine love. In the Groundwork, Kant less pays homage to the religious provenance of his conception of value than he takes it to be available to “common human reason.”26 That there is a purpose or use independent form of worth is supposed to be an idea that reason simply has (as opposed to an idea with a history, or an idea that arises in a certain culture, or an idea that is tied to religious beliefs). It may be that we, like Kant’s contemporaries, also find this conception both familiar and to our taste. We may be not so apt to balk at Kant’s proposal, and some might claim this as Kant’s legacy for modern moral sensibility, though likely sensibility is not so unified as to merit description in the singular, and a shared religious inheritance is surely part of the story. But it is also Kant’s aim to bring us to see that a vindication of this form of value, like a vindication of categorical bindingness, is deeply difficult. Kant fully appreciates the philosophical complexities that are involved in making use of a notion that is for him as for his contemporaries extremely familiar, and this is a signature strength. He issues the suspicion of fantasy from this point of view. The suspicion is not that absolute worth cannot be part of the fabric of the world—that it is a 26 “We have, then, to explicate the concept of a will that is to be esteemed in itself and that is good apart from any purpose, as it already dwells in natural understanding and needs not so much to be taught as only to be clarified.” Kant, Groundwork, 4:397. Contrast the frequent reference to the Gospels in Immanuel Kant (1997a). Lectures on Ethics. Edited by Peter Heath and J. B. Schneewind. Cambridge: The Cambridge Edition of the Works of Immanuel Kant.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

50 the value of humanity spectral property whose constitution is “queer” and whose knowability is unfathomable.27 These are charges traditionally brought against the realist’s conception of absolute value, and while it is true that absolute value has tended to be theorized in realist terms, it need not be. There is reason to doubt that Kant’s is a realist conception, so the suspicion must lie elsewhere.28 If, with Plato’s Socrates, we thought that good means beneficial—that it is part of the concept of good that it does good—then the worry would be that Kant’s conception of absolute worth is incoherent. I would rather put the concern in explanatory and metaphysical terms. When we explain the value of diverse kinds of object and activity, we point to their propensity to be beneficial in some way. The pencil allows us to make marginalia in a text whose central argument we are trying to keep clear; a gift allows us to celebrate a person in our life; a trip to the museum brings our studies of Tutankhamun to life and with it a chapter of human history whose imaginative resonance is extraordinary. These are colloquial ways of expressing the apparently affective nature of value—the different ways in which it helps and serves and enriches. The explanations can be taken further, and they want some filling out; but the point is that Kant’s notion of absolute value is wholly unlike the value that figures here. In these explanations value is taken to be a relation between an object and a subject, a relation of being (in different ways) good for someone. Kant is proposing a form of value that does not have this relational structure. Can value be as he proposes? In a clear sense, the possibility of a benefit-independent form of worth is something that Kant continues to be exercised by. The form of value that marks the good will is highest in rank, but it emerges that it is not such that nothing could be added to it to make it better. What could be added so that nothing is missing? Kant’s answer is happiness. 27 See J. L.Mackie (1977). Ethics: Inventing Right and Wrong. London: Penguin Books, ch. 1. 28 Again, it is a real strength of Sensen’s “Kant’s Conception of Inner Value” that he raises the question of how realism about value of the kind we are familiar in the writings of G. E. Moore could be compatible with Kant’s broader commitments.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

2.3 a suspicion of fantasy 51 In the second Critique, the good will together with happiness emerges as the most comprehensive good.29 How is the most comprehensive good to be realized? Kant’s answer bears his signature. In aiming for the complete good we must postulate the existence of a supreme supersensible being to see to it that people who are righteous are rewarded with happiness in the next life. That is, in aiming for the complete good we are constrained to postulate the existence of a God who will ensure that good people get what they deserve—the satisfaction of their permissible desires. This line of argument, together with the argument for the immortality of the soul that for simplicity sake I have left to one side, is one Kant continues to return to again and again in the canonical writings. To me it should incline us to think that, after all, even Kant is concerned with the beneficial (however differently he conceives of it from the ancients). He removes considerations that have to do with the good for human beings from the foundational purview of morality, but he adds them back in via a form of argument that involves the attainment of happiness, not through a person’s own efforts in this life, but through God’s grace in the next. As Katja Vogt has put it to me, Kant’s argument may be read as an indication that he is after all gripped by a version of Plato’s question of whether it is advantageous to be just. It turns out that even he is alive to the normative force of considerations that have to do with the good for human beings. So what if Kant was right to voice consternation, right to worry that supreme worth can be as he says it is in the Groundwork? What if, as Plato’s Socrates seems to find obvious, the good benefits and the bad harms? What if value is relational in the sense that it is always good for someone? Would it be possible to provide an account of the value of human beings that sees it as in this way continuous with the value of other valuable things? Could our value be captured by a propensity to stand in relations of benefit with other things or other beings? Would this be a coherent or plausible sort of proposal?

29 Kant, Critique of Practical Reason, 5: 110.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

52 the value of humanity A proposal along these lines will seem highly problematic if to conceive of value as relational—the good as good for someone— were to flatten the evaluative field to instrumentality. It would be troubling if it meant touting one side of Kant’s dualism between use value and moral worth and razing the other. So it is worth pointing out that we need not go that far with Kant’s way of proceeding. Kant is concerned with the possibility of absolute worth, and from that vantage point the varieties of relational value are not so interesting. Kant reduces relational value to instrumental value, or in a term he also uses, to price value. Instrumental and moral reason are taken to be the two forms of normativity that (together with beauty) constitute the practical field. Kant is at his most eloquent when he insists on the lowliness of the one, and the loftiness of the other, and this is likely something we inherit, if unreflectively. What the market is concerned with—use—is tied to the gratification of desire, and desire is in the service of our base constitution—seeking gratification, tending towards egoism, being transgressive, and ultimately burdensome.30 In these terms the ancient conception of value, in which there is not this difference in kind between moral and non-moral value, and no equivalent conception of the market, can have a hard time registering. For the ancients, the interesting questions about the good concern the different ways goods can be good for us. As we saw, Plato distinguishes goods that are by themselves good for people from goods that are good for people only in conjunction with other things. The former are those that directly affect the mind, as the performance of just actions is thought to harmonize the soul. In a teleological formulation, there 30 Kant’s assumption that relational value is inevitably grounded in desire has been overturned by others, for example Terence Irwin (1996). “Kant’s Criticisms of Eudaemonism”. In: Aristotle, Kant, and the Stoics. Edited by Stephen Engstrom and Jennifer Whiting. Cambridge: Cambridge University Press, ch. 3. What is good for you may make a claim on me independently of its connection to my antecedent motivational states, what is good for me likewise so. To put in terms that do not quite find a place in Kant’s moral framework, the reasons to which relational value gives rise may be perfectly external to our antecedent motivational states. I take up these issues in Chapter 5.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

2.3 a suspicion of fantasy 53 is the difference between goods that are good for someone for their own sake, and goods that are good for someone for the sake of other things. Aristotle’s concept of final value enters here as that for the sake of which all the other goods are good, though it is not good for the sake of other things. While final, and to that extent non-derivative, this value is relational: Aristotle’s good life is good for the person whose life it is (and of course not solely for them). In a larger discussion one would expound the Stoic distinction between what has value and what is good—between preferred indifferents and goodness proper— both importantly relational notions. This is the large field that holds eudaimonist traditions in all their variety, East and West. Relational value is not equivalent to instrumental value for the ancients, but neither does it bear the hallmarks of market value as Kant understands it—the marks, that is, of quantifiability, comparability, and fungibility. To recall the discussion from Chapter 1, that value is quantifiable, comparable, and fungible are dogmas in value theory, and they tend to be accepted by proponents of consequentialism. Unsurprisingly, it is the theorists who begin from Kantian insights— insights to the effect that objects can be bearers of value, and that production is not the paradigm response to value—who have been most influential in critiquing them. But their work takes us considerably beyond Kant’s basic distinction between dignity and price, and the dimension of value to which they draw attention does not pick out the non-relational good from the relational good for in the sense that is under discussion. Quantifiability, comparability, and fungibility are not the distinguishing marks of the beneficial. Kant should be fully credited for insulating forms of value from the relentless imperatives of capital, and in urging that it is a category mistake to put a price on human beings, Kant provides a powerful early argument against slavery. But presumably Kant does not go far enough in supposing that humanity alone enjoys protection from market forces. I would extend his protection much more widely. I would extend it to the basic constituents of life that support us as the creatures we are. I would extend it to other beings, to works of art, to love, to sex, and to manifold forms of activity. I share Kant’s view that it

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

54 the value of humanity does not make sense to ask how much value a person has, but equally, if we ask how much value there is in reading Proust, or how to quantify the value of the air we breathe, then something has also gone wrong. Equally, I think Kant was right to see that while we can say of someone that she is a better political scientist than someone else, or a better designer, we cannot say of someone that she has more value than someone else. But incomparabilities are widespread in the domain of value. “Is a shark better than a crocodile?” is a question I was fond of asking as a child, and my parents were rightly bemused. So too I think Kant was right to see human beings as irreplaceable—as subject to substitution with loss—but so it is with all things to which we appropriately form an attachment. Wittgenstein expressed admiration for a poem the central subject of which is a man’s relationship with the tree that he has grown from a sprig.31 This does not strike me as improperly sentimental.

2.4 Formulating a Hypothesis I began this chapter by recalling quite general features of Kant’s treatment of absolute value. I regard it as a strength of Kant’s discussion that he anticipates the philosophical difficulties with his proposal. The proposal is difficult if we think, with the ancients, that goodness or value is tied to the beneficial. I have emphasized what is distinctive about Kant’s proposal by drawing basic points of comparison with the eudaimonist tradition in ethics. Elizabeth Anscombe famously said that those who profess to find the notion of the moral ought in ancient texts must surely feel their jaws coming out of alignment. The teeth will not come together in the proper bite.32 As ever her formulations are striking, and they are polemical. Certainly, I am not defending her larger distinction between ancient and modern ethics here. But I will

31 Paul Engelmann (1968). Letters from Ludwig Wittgenstein with a Memoir. Edited by Brian McGuinness. New York: Horizon Press, pages 124–127. 32 Anscombe, “Modern Moral Philosophy,” 2.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

2.4 formulating a hypothesis 55 say that I do not find Kant’s notion of absolute value in ancient texts, and I find it important to be clear-sighted about this basic point of difference between Kant and his predecessors. The discussion is so-far inconclusive. I have not given an argument to the effect that value cannot be as Kant says it is: independent of the good it does or stands to do. I have marshaled lines of thought that, more modestly, raise a question about it. In doing so I have sought to motivate the exploration of a hypothesis. The hypothesis is as follows. If the beneficial is the mark of the evaluative, and we human beings are of value, then the value of human beings will lie in the relations we bear—relations of being of benefit to something or someone. Naturally the hypothesis will meet with philosophical difficulties of its own. I have given reason to doubt that a proposal along these lines is committed to reducing the value of human beings to market price, together with its attendant features. I have denied that fungibility, comparability, and replaceability are marks of the good as good for someone. But many questions remain. Who or what do we bear the relevant relation to? What if we stand in no such relation to anyone? Can a proposal of this kind capture basic forms of ethical behavior that are owed to everyone? These questions provide the subject matter for the chapters to follow. In taking up this hypothesis there is a sense in which I am recalling modern moral philosophy to ancient ethics, and in particular, to ancient preoccupations about value. Many of the writers whose work I admire the most—Elizabeth Anscombe, Philippa Foot, Bernard Williams—have made the call rather loudly, on various grounds. But I do not wish to overstate the ancient inheritance. I think there are crucial modern insights that any viable ethical theory needs to capture. Among them, and centrally for my purposes, is the idea that humanity is valuable. The ancients are greatly interested in what is good for human beings, and they are greatly interested in what it is to be an exemplary or excellent human being (these are importantly related ideas). But, while it nearly appears at certain moments, one does not find the idea that people are bearers of value clearly expressed

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

56 the value of humanity in ancient texts.33 Other things are good for people, and some— probably very few—people are virtuous, but it is not the case that people are of value. Arguably, ancient ethics misses something crucial to this extent. That human beings are bearers of value, and that this value should constrain us in our interactions with one another, has a real claim to being an ethical discovery. It is hard not to hear an indictment of slavery in Kant’s insistence that human beings do not have price value, and as we know, Aristotle compares badly in this connection.34 To me Kant provides a criterion of adequacy for accounts of human value. The power of Kant’s injunction to treat humanity as an end-in-itself lies in the idea that we are not to relate ourselves to people chiefly with a view to what they can do for us. Our conduct towards others should proceed from full awareness of their being the center of a life. A relational explanation of our value is adequate only if it captures this crucial idea. It is a mark of my respect for the Kantian tradition that I begin exploring the viability of a relational account of the value of human beings by considering an objection that is at least Kant-inspired. The objection is that for anything to be of benefit to someone, and in that way of value, something must be of value in a benefit-independent way. In other words, the existence of something that is good for someone entails that something is good in itself. Moreover, human beings are plausible candidates for being of value in themselves. I take up this objection in the next chapter. In the absence of a convincing response to it, a relational proposal will not get off the ground.

33 See Richard Bett (n.d.). “Do the Ancients See Value in Humanity?” In: ReEvaluating the Value of Humanity. Edited by Sarah Buss and Nandi Theunissen. Manuscript. 34 As recent scholars have brought out, Kant’s stated views about race, slavery, and colonial conquest are highly problematic. Lucy Allais (2016). “Kant’s Racism”. In: Philosophical Papers 45.1 and 2, pages 1–36 has argued that Kant’s stated views on race are impossible to square with his universalism about the value of human beings. For different readings, see Charles Mills (2005). “Kant’s Untermenschen”. In: Race and Racism in Modern Philosophy. Edited by A. Valls. Ithaca: Cornell University Press, pages 169–193. See Pauline Kleingeld (2007). “Kant’s Second Thoughts on Race”. In: The Philosophical Quarterly 57.229, pages 573–592.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

3 Must We Be Absolutely Valuable? I have been situating conceptions of value historically. I have sought to bring out key features of Kant’s absolute conception of the value of humanity, and I have given voice to a puzzle that anything could be valuable in that way. The discussion has so far been exploratory. In this chapter I investigate a powerful argument for the view that the value of humanity must be, as Kant says, absolute—indeed, that the very structure of value requires it. The argument is not made by Kant himself, though it is made by those who are more or less self-consciously working in a Kantian framework. The argument is that for anything to be of good for someone, something must be good in itself, and people meet the criteria. The value of what is good for people borrows its normativity—its reason-giving force— from the value of people whose value is not borrowed from it, or anything else. The discovery of the chapter is that human beings do not need to be absolutely valuable to countenance the existence of normatively significant goods for them. I consider various alternatives to terminating a chain of dependence between values in something that is absolutely valuable. By engaging with metaphysical questions about the structure of value, a substantive proposal emerges. I give reason to favor a proposal according to which the value of human beings is explained in terms of a relation we bear to ourselves. The chapter is a crucial first step in offering a positive relational conception of the value of humanity.

The Value of Humanity. L. Nandi Theunissen, Oxford University Press (2019). © L. Nandi Theunissen. DOI: 10.1093/oso/9780198832645.001.0001

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

58 the value of humanity

3.1 The Regress Argument The argument that interests me in this chapter has been defended by Joseph Raz and David Velleman.1 It concerns the relation that obtains whenever one thing is good for another. The argument starts from the assumption that being good for is not always a normatively significant relation. That is, when X is good for Y, it does not follow that X is good or valuable. For X to be of value it must be that Y is of value. Accordingly, the relation is one of dependence so that when X is good for Y the value of X depends on the value of Y. The dependence is illustrated with examples: watering the plant is part of what enables it to bear fruit, and the value of the watering depends on whether the fruit is worth having or eating (supposing it is not valuable in some other way). If there is nothing good in the fruit, there is no good in watering it.2 Being good for is a transitive relation, so that if watering is good for the plant in allowing it to bear fruit, and having the fruit is good for making a dessert, then watering also contributes to the good of making (and presumably, of eating) the dessert. This introduces the idea of a chain of dependence in which one thing is good for another. Given these assumptions, the argument emerges in two parts as follows. Part I: What is of value because it can be good for something of value must ultimately be good for something whose value is not a function of being good for something. For if there were an infinite chain of dependence between goods that are good because they can

1 Cf. Joseph Raz (1999c). “The Amoralist”. In: Engaging Reason. Oxford and New York: Oxford University Press, pages 273–302. And Joseph Raz (2001a). “Respecting People”. In: Value, Respect and Attachment. Cambridge: Cambridge University Press, pages 124–175. Cf. David Velleman (1999a). “A Right to Self-Termination?” In: Ethics 109.3, pages 606–628. And David Velleman (2008). “Beyond Price”. In: Ethics 118, pages 191–212. There are differences between Raz’s and Velleman’s presentations of the argument, and I will not be faithful to one or other of them on every point because I am interested in a common schema. I am most grateful to Joseph for detailed comments on the form of the argument, and I draw on these in what follows. For a much discussed but quite distinct version of the argument due to Christine Korsgaard, see Appendix 2. 2 The example is from Raz, “Respecting People,” 147.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

3.1 the regress argument 59 be good for something, nothing would be valuable (and some things are). Whatever is of value because it can be good for something of value must ultimately be of value because of something that is good in itself—because it in some way contributes to what is good in itself. For the final node in the chain of dependence to be good in itself, it must be of value independently of whether it can be good for something of value. It may be good for something too, but its value will not depend on being so. Part II: Valuers meet the criteria for being the final node in the chain of dependence because valuers meet the criteria for being of value in themselves. Valuers are such that other things can be good for them, and they are not such that they are of value because they can be good for other things. Valuers are of value independently of the good they do or stand to do—they are ends-in-themselves. See Figure 1. Raz uses the term “valuer” interchangeably with “person,” or “humanity,” and I will follow him in this, returning to the significance of the term “valuer” in the next chapter. To be a valuer in Raz’s sense is to be capable of engaging with (and thereby of benefitting from) objects and activities of value. Let me offer some clarifications on the terms of the argument. The form of relational value that is in question here is goodness for a person, and the form of non-relational value is goodness period, or as I will sometimes use the terms, absolute value or value in itself. It is assumed that something can be instrumentally or non-instrumentally good for a person: that is, indirectly or directly good for them. Watering the plant may be given as an example of something that is instrumentally good for a person (indirectly good for them), while eating the sweet

Valuer

Relational Figure 1 End

Non-relational

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

60 the value of humanity fruit of the plant may be offered as an example of a non-instrumental good for a person (something directly good for them). Typical examples of non-instrumental goods for human beings include simple pleasures, the engagement with works of art, philosophical conversations, a walk in the forest, and so on. Being “good for” a person is equivalent to being of benefit to or beneficial for them. The notion of noninstrumental or intrinsic benefit is meant to be intuitive—enrichment for its own sake is an elucidation—and different substantive accounts may be given. There is some controversy over whether the relation being good for is the same irrespective of what is substituted for Y when X is good for Y—artifacts, plants, animals, persons, etc.3 As the argument is formulated above, provided that Y is of value, the relation is neutral on substitutions of Y, though people emerge as (at least among) the final Y. Ultimately the argument concerns the nature of personal good: the existence of goods for people implies that people are of value in themselves. So for purposes of the argument, I will assume that the relation takes a person, or equivalently, a valuer, as the substitution for Y. Since proponents of the argument allow that something can be indirectly or directly good for a person, we may say that X can be “good for something,” meaning by this that X is indirectly good for someone.4 Finally, since the argument concerns the would-be structural dependence between goods that are good for someone and goods that are good in themselves, it does not depend on substantive accounts of good for and good in itself, and realist and anti-realist accounts of both are forthcoming.5 For purposes of the 3 Raz, “Respecting People” and Richard Kraut (2007). What is Good and Why. Cambridge, MA and London: Harvard University Press take the relation to be the same, or at least continuous, and Connie Rosati (2009). “Relational Good and the Multiplicity Problem”. In: Philosophical Issues 19 (Metaethics), pages 205–234 takes it to be importantly different. 4 There is an exception to this restriction in the discussion of “reverse priority” below, where people are understood to be good for things that are good period. The restriction otherwise holds, and the exception does not affect the larger argument. 5 On a prominent realist view, what it is for X to be intrinsically good for a subject, S, is for X to allow S to prosper or flourish as the kind of S she is, something about which there are attitude independent facts of the matter. Proponents of the view vary according to the permitted substitutions for S; and according to whether a further story

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

3.1 the regress argument 61 argument (at least in the initial stages) proponents need not take a stand on meta-ethical questions, and likely the argument has wide currency in part because it does not depend on contentious issues about realism and anti-realism.6 Here is a statement of the argument in sum: (i) Some things are good or valuable because they can be good for something or someone of value. (ii) There are chains of dependence between things that are valuable in this way. (iii) The chain of dependence between values must come to an end on pain of a vicious regress. (iv) What brings a chain of dependence to an end is something that is such that (a) there are things that are of value because they can be good for it, while (b) its value does not depend on being good for other things—it is good or valuable in itself. (v) Valuers meet conditions (a) and (b). (vi) Therefore, valuers can be the final node in a chain of dependence. about “prospering” is thought forthcoming. Prospering may be taken to be indefinable; it may be taken to be a natural or an irreducibly normative relation; it may be taken to be a second-order relational property of being productive of a set of enumerable features; or something else. For relevant discussion, see Kraut, What is Good, chs. 1 and 2, and Connie Rosati (2006). “Personal Good”. In: Metaethics After Moore. Edited by Mark Timmons Terry Horgan. Oxford: Oxford University Press, pages 107–131, pp. 118–123. See also Peter Railton (1986). “Moral Realism”. In: The Philosophical Review 95.2. On a prominent anti-realist view, what it is for X to be good for someone, S, is for X to conduce to S’s goals subject to some idealizing condition (full information, ideal rationality). See Judith Jarvis Thomson (1997). “The Right and the Good”. In: The Journal of Philosophy 94 (6), pages 273–298. An influential variant is given by Stephen Darwall (2004). Welfare and Rational Care. Princeton, NJ: Princeton University Press. Though it is less often seen, there are realist and anti-realist accounts of the nonrelational good. In the Moorean tradition, good in itself is a simple first-order nonnatural property; but it need not be. Kant takes persons to be of value in themselves though not in the sense that persons possess a self-standing first-order property of the kind posited by value realists. More likely Kant thinks that people are of value in themselves in the sense that they categorically must be valued (valued independently of whether they stand to be good for something or someone). 6 Raz and Velleman both defend versions of the argument schema, though they have divergent meta-ethical commitments. Contrast Korsgaard’s treatment of the issue of realism and anti-realism in “Two Distinctions,” 181–182.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

62 the value of humanity

3.2 Evaluating the Argument How strong is this argument? I work through possible responses to it in what follows. For ease of reference, I refer to those who defend the argument as proponents of the regress argument. An abrupt way to reject the argument would be to deny that being good for—the relation that figures in the very first premise—is a meaningful one. This was a charge long raised by G. E. Moore, and it finds contemporary defenders.7 I will follow proponents of the regress argument in assuming that being good for or of value to is an important and common relation between things of value. I will have more to say about the relation later, and I return to specifically Moorean objections in Chapter 5. The claim that there are chains of dependence between values strikes me as uncontentious (premise ii), though I will argue that the chain need not be linear. Premises (iii)–(v) strike me as more interesting. They invite us to consider whether a chain of dependence between values must come to an end with something that is good in itself, and whether human beings meet the criteria for being the final node in the series. Let’s consider some alternatives.

Reverse priority A structurally mild alternative accepts that there must be an end to the chain of dependence (premise iii), and accepts that the final node is good in itself (premise iv), but denies that human beings meet the criteria (premise v is false). Human beings do not meet the criteria because there is a reverse priority between some objects of value and human beings. Some objects are such that human beings are of value because they can be good for them, and such that they are not of value because they can be good for human beings. Human beings are valuable because they are able to give these objects their due. Call this option RP for “reverse priority.” See Figure 2. A version of RP is defended by Sarah Buss.8 To be clear, Buss is not defending 7 For example, Donald H. Regan (2004). “Why am I my Brother’s Keeper?” In: Reason and Value: Themes from the Moral Philosophy of Joseph Raz. Oxford: Oxford University Press, pages 202–230. 8 Buss, “The Value of Humanity,” 349.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

3.2 evaluating the argument 63

Valuer

Relational

Works of Art

Non-relational

Figure 2 Reverse priority

premises (i)–(iv). She is making the proposal that human beings, or more generally, valuers (those capable of engaging with objects and activities of value) are valuable because they can be good for some objects of value (in Buss’s example, works of art), while those objects are good in themselves or period. Start with the second part of the proposal: works of art are good period. This is so, Buss maintains, because the value of art is not exhausted by its role, or potential role, in enriching or giving meaning to people’s lives.9 By this Buss means that works of art deserve to be revered and honored as the valuable things they are, where this calls for a mode of engagement that focuses our attention on the works themselves (and not on our experience of them, or on the satisfaction we derive from them).10 Does the object-directed character of art appreciation show that works of art are good in themselves? Presumably those who take works of art to be intrinsically good for people— by themselves good for people, or good for people for their own sake—will, or should, acknowledge that they are so in virtue of features of the work. The sculpture has a striking luminosity and expresses the concepts of space, weight, and volume, in an imaginative way, etc. The work cannot be good for people unless they pay attention 9 For comparable claims, see Susan Wolf (2010). “Good-For Nothings”. In: Proceedings and Address of the American Philosophical Association 85.2, pages 47–64. 10 Nagel makes a similar point. For Nagel, some of our interests seem to give evidence that their object has a value that is not a function of being good for anyone, and this seems especially true in aesthetics where “the object of interest is external and the interest seems perpetually capable of criticism in light of further attention to the object.” Thomas Nagel (1989). The View from Nowhere. Oxford and New York: Oxford University Press, page 153.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

64 the value of humanity to these features, and pay attention without a narrow view to their own enrichment. “What is this doing for me?” “How much am I getting out of this?”—these questions are sure signs that a person is not engaging as she should. An explanation of the value of works of art in terms of their propensity to enrich our lives properly incorporates the features in virtue of which it does so and properly allows for object-directed attention.11 What of the first part of the proposal: human beings are of value because they can be good for works of art? One way a person can be good for works of art is by preserving and restoring them. Presumably these efforts find their point in facilitating appreciation, and perhaps appreciation is good for a work of art because without appreciation its value is wasted. But as a defender of premise (v) will make the case, it is valuers who stand to lose when a work of art goes unappreciated, and valuers for whom the realization of a value finds its point.12 Many find it peculiar to say that art is good whether or not there is or could be anyone around to engage with it.13 If there is to be an asymmetry between people and objects and activities of value (and the if is what I am in the end investigating in this chapter), then absent further argument, the asymmetry more plausibly goes the other way. To be good for a work of art is to be indirectly good for a person. The value of works of art is conditioned by their capacity to enrich us, to bring meaning to our lives, but our value is special because it is not so conditioned: we are ends in ourselves. To that extent, so far I think proponents of the special value of humanity win the argument. 11 I take up the value of works of art at greater length in “The New Mooreans: On Personal and Impersonal Good,” manuscript. 12 Here is Raz: “All this is a bit of a mouthful to say that paintings are there to be seen and appreciated, novels to be read, oranges to be eaten, mountains to be climbed, etc. They are there for these things to happen to them in the sense that their value to others remains unrealised until someone of value in himself relates to them in the right way.” Raz, “Respecting People,” 154. 13 Here is Nagel: “The problem is to account for external values in a way which avoids the implausible consequence that they retain their practical importance even if no one will ever be able to respond to them. (So that if all sentient life is destroyed, it will still be a good thing if the Frick collection survives).” Nagel, View from Nowhere, 153.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

3.2 evaluating the argument 65

Neither good nor bad A structurally more colorful alternative accepts that a chain of dependence between values must to come to an end (premise iii), and accepts that human beings are the final node in the series (vi is true), but denies that human beings are valuable in themselves (premise v is false). Instead, human beings are neither good nor bad. Human beings are the beings (or among the beings) for whom things are of value, but are not themselves that way. Other things are good because they can be good for people, but people themselves, or valuers qua valuers, are neither good nor bad (NGB). Plato investigates a form of this proposal in Lysis. Plato has Socrates present a model of benefit according to which what is good stands to benefit a subject who is neither good nor bad. A key assumption is that in order to be benefited a subject must be susceptible to ills that goods can cure. So the soul, which is susceptible to the ill of ignorance, stands to benefit from the good of knowledge. But importantly, to be benefited the subject cannot already be good, and the subject cannot already be bad. The subject cannot already be good because then the subject will be invulnerable to ills and will not be in need of improvement. On the other hand, while the subject must be susceptible to ills, the susceptibility must be such that it admits of a cure—the subject cannot be too far gone. For Socrates contends that what is already bad cannot be improved. It follows that to be benefitted, a subject must be neither good nor bad.14 Several assumptions may be called into question, But more interesting in this context is that Plato himself has Socrates envisage a regress argument by way of objection. Chains of dependence between values must come to an end with something for whose sake all the rest are valuable, and Socrates uses the example of a father who values his son more highly than all of his possessions, and values his possessions for the son’s sake. We may talk of how much we value gold and silver,

14 Plato (1997a). “Lysis”. In: Plato Complete Works. Edited by John M. Cooper. Indianapolis: Hackett Publishing, pages 687–707, S. 216c–218c.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

66 the value of humanity but that gets us no closer to the truth which is that we value above all else that for which gold and all other provisions are provided.15 Plato does not quite draw the conclusion that therefore the son is valuable after all, and in a special kind of way; but the conclusion does not look far away. This line of response is open to proponents of the regress argument. They will urge that the subjects who lie at the end of the chains of dependence cannot be neither good nor bad because it is for their sake that the values in the chain of dependence are sought, preserved, cultivated, and so on. We value objects and activities of value because we value valuers. If that is right, NGB is inadequate to the structure of valuation. NGB may also be inadequate to the experience of loving someone, an experience that arguably discloses them as a locus of value—as precious.16 On these grounds I will accept that proponents of the regress argument are right to reject NGB.

Devoid of value or harmful Could the chain of dependence come to an end with something that is devoid of value (DV), or harmful (H)?17 There may be an impulse to say that it can, for there is a use of “good for” according to which “x is good for y” and y is harmful, as in “a climate of prejudice is good for the perpetrators of genocide.” Clearly the phrase lacks normative significance. When we say that “works of art can be good for people” we have offered an explanation of the value of art, and we have provided a reason (all things being equal) to preserve and protect the works in question. But we have not explained the value of prejudice by pointing out that it conduces to the perpetration of genocide, and we have not provided a reason to protect or cultivate prejudice. Proponents 15 Plato, Lysis 218d–220a. 16 There is rich discussion of the value of humanity and love, and I cannot do justice to it here. One way or another, contemporary discussion tends to lead back to David Velleman (1999b). “Love as a Moral Emotion”. In: Ethics 109, pages 338–74. 17 What is devoid of value once was but has ceased to be of value, like a dead battery. Being devoid of value is different from being neither good nor bad for value terms apply to it, and it is different from being harmful, for it is not pernicious. I am grateful to an anonymous reviewer for asking me to clarify the difference between NGB and DV.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

3.2 evaluating the argument 67 of the argument conclude, I think rightly, that something cannot be valuable by being “good for” something worthless or harmful. It is worth pausing over the question of what we mean when we say that prejudice is “good for” the perpetrators of genocide, or that genocide itself is “good for” them. How should we understand these statements? Likely we mean that prejudice serves the perpetrators’ purposes—that it gets them what they want; or we mean that genocide is good from the perpetrators’ “point of view.” By contrast, when we say that engaging with works of art can be “good for” those who engage with them we mean that the work can be enriching to them or to their lives—that it is for their good, or even betterment. Benefit is certainly amenable to different analyses, and proponents of the argument can remain open to various options. In Chapters 4 and 5 I will endorse a realist account of the good for human beings. But at this stage of the argument I point out just that an analysis of benefit in terms of what contributes to a person’s goals whatever those goals may be finds few defenders among realists or anti-realists (notwithstanding their other differences). And while some have understood “good for” in perspectival terms, this reading is rightly rejected by contemporary theorists.18 Indeed, we may share the late Philippa Foot’s sense that it strains the concept to say that something like genocide is of benefit to its perpetrators, i.e. that it is for their good to commit genocide.19 Proponents of the argument are interested in “good for” understood

18 In an early paper Philipa Foot (1985). “Utilitarianism and the Virtues”. In: Mind 93.374, pages 196–209 understands “good for” in terms of point of view, and so understood, the notion is criticized by Thomas Hurka (1987). “‘Good’ and ‘Good For”’. In: Mind 96.381, pages 71–73. Against perspectival readings of “good for” see Kraut, What is Good, ch. 2. 19 Here is Foot: “Suppose we think of some really wicked persons such as the serial killers Frederick and Rosemary West, who did not spare their own children in their career of abuse and murder. For many years they were able to act out their sexual fantasies free from detection, and might well have continued to do so right until the end of their natural lives. What then would it have been right to say about the contribution of those whose behavior made this kind of thing possible? Would they have benefited the horrible Wests? It seems to me that in our natural refusal to say so we glimpse a conceptual truth that does not usually lie so clearly on the surface.” Foot means the conceptual connection between faring well and virtue, a connection she thinks

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

68 the value of humanity as benefit, a relation in which something does a person some good, makes them well, etc. So taken, we should not take it as obvious that genocide is good for its perpetrators.20

Infinite regress and the metaphysical principle The discussion brings out that each of the nodes in a chain of dependence must be of value for any part of the chain to make sense. Now the question becomes: valuable in what way? Could the nodes be of value in the sense that they are good for something or someone else, and so on ad infinitum (IR)? The problem with a regress here is presumably that without termination of a chain of dependence in something good period, nothing will be of value, for there will be nowhere for value to enter the chain of dependence.21 What underlies this worry is a conception of how X gains its value when X is good for V and X is of value. Metaphorically speaking, it is a borrowing conception: X takes its value on loan from V, and at the limit, V cannot itself be a borrower but must have a surfeit of value; V must be an arch lender.22 This conception may be formulated as a principle (hereafter “the metaphysical principle”) to the effect that when an object, activity, is evident in the concept of benefit. Philippa Foot (2001). Natural Goodness. Oxford: Oxford University Press, page 94 and ff. 20 Though I doubt that genocide provides such an example, there are complexities in cases where something is good for one person and not good (or harmful) for another. Insulin is good for me (to ingest) because I am diabetic, but it is not good for you (to ingest) since you are not. Does the insulin’s value for me explain its status as valuable? Does the insulin’s value for me explain its normative significance, including its normative significance for you? These are natural questions and I come to a full discussion of them in Chapter 5. Much turns on how we understand normative significance, and whether we allow that something can have general normative significance while figuring in different ways in the lives of different people depending on various subjective conditions. Plausibly, that insulin is not good for you (to ingest) does not entail that it lacks practical relevance for you. It is a kind of medicine, and medicine is good for human beings. Likely the insulin has practical relevance for you in the minimal but robust sense that you do not have reason to destroy it, or to prevent the diabetic from securing the doses she needs. 21 I am grateful to Aaron Abma for discussion of the problem with infinite regress. 22 To my knowledge, the borrowing metaphor can be traced to W. D. Ross (2002). The Right and the Good. Edited by Philip Stratton-Lake. Oxford and New York: Oxford University Press, page 75.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

3.2 evaluating the argument 69 or state of affairs, X, is instrumentally, constitutively, or in some other way, good for a valuer V, and X is of value, the value of X ultimately derives from the value in itself of V. Do we have reason to accept the metaphysical principle? It is instructive to consider an argument against the conception of relational value that is expressed in the principle in a different but related context. Earl Conee makes such an argument in the context of Moorean arguments from instrumental to intrinsic value.23 Conee shares the assumption that what conduces to a worthless end is not of value: for something to be instrumentally valuable it must conduce to something worthwhile on the whole. But he denies the implication that something intrinsically valuable exists. For something can contribute to what is worthwhile on the whole by preventing some evil, or making things better: Intuitively put the idea is that it is equally creditable to move the world upward some amount on the overall value scale from what would have been, whether or not the change introduces something better than neutral. [. . .] An event can have a valuable result, and thereby gain instrumental value, without taking out any ‘loan’ from the intrinsic value of another thing. It is enough to keep some evil from obtaining, or to make things better somehow. Of course this does not imply that an endless sequence of value dependents is possible. Rather, it shows that no such series is needed to have instrumental value without intrinsic goodness.24

So the event of a person’s taking pain-reliever causes there to be less suffering in the world. The pain medication has moved things up from a negative to a neutral value on the scale. What we have is not a great

23 Earl Conee (1982). “Instrumental Value without Intrinsic Value?” In: Philosophia 11, pages 345–359. Conee engages with an older skeptical literature about intrinsic value that claims inspiration from pragmatists like John Dewey. See Gilbert H. Harman (1967). “Toward a Theory of Intrinsic Value”. In: The Journal of Philosophy 64.23, pages 792–804, and especially Monroe C. Beardsley (1965). “Intrinsic Value”. In: Philosophy and Phenomenological Research 26, pages 1–17. As I bring out in footnote 25 below, Conee’s conclusions are importantly more limited and more qualified than that of his interlocutors, but for reasons I explain there, these qualifications are not relevant for my purposes. 24 Conee, “Instrumental and Intrinsic,” 354.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

70 the value of humanity world, but it is a better world than the one in which there was the pain. We can understand the value of taking the medicine, but we haven’t needed to posit intrinsic value. For, Conee contends, whatever else intrinsic value is, it is “positive value,” so instrumental value does not entail intrinsic value. Conee puts his conclusion by saying that in a world without intrinsic value there can be no value that is better than neutral. I would rather say that in a world without value in itself there can be no value that is independent of benefit. For a benefit can move things up, not just from bad to neutral, but from a neutral to a positive value on the scale. What matters is that the instrument makes some positive change, and that its value is a function of relative difference rather than of something that is intrinsically valuable. Conee’s argument purports to show that an outcome need not be intrinsically valuable for what conduces to it to be of value. To suppose that it must be is to overlook mitigation and improvement.25 How might proponents of the regress argument in question reply? For Conee the event in which a subject is experiencing pain is bad in itself, and the value of removing that pain is a function of its propensity, all things being equal, to alter the state of the world from bad to neutral on the scale. Proponents of the argument will urge that this way of thinking about the badness of pain loses its relational character: the sense in which pain is bad because it is bad for someone.26 The state in which a valuer is experiencing pain is bad because it is bad for the valuer, and the value of removing the pain is a function of the valuer’s 25 Here is the qualification that I mentioned in footnote 23 above. For Conee there is a successful regress argument for intrinsic value. It is an argument from the existence of what he calls “positive worthwhile value,” value that is more stringent or more “positive” than what is meritorious on balance, or worthwhile on the whole. A value is “positively worthwhile” in this sense if its value remains after its prevention and benefit values have been subtracted (and those of its causal successors). The argument is that for there to be value of the “positive worthwhile” kind, something must be intrinsically valuable. Whatever we make of this notion of “positive value,” and to my mind it is quite peculiar, the inference may well be right. But this is not the conclusion that proponents of the non-relational value of valuers are seeking, for non-relational value was meant to follow from benefit value—from personal good. 26 For a critique of the Moorean conception of pain, see Kraut, Against Absolute Goodness, chs. 8 and 9.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

3.2 evaluating the argument 71 value: their pain matters because they matter. From this point of view, Conee’s argument depends on some implausible assumptions about pain. Can we envisage Conee’s reply in turn? Conee, whose framework is Moorean in an extended sense, may equally contest the proponents’ conception of the badness of pain on grounds, to adapt Bentham’s adage, that the question is not, “Are they of value?” but, “Can they suffer?”27 Proponents of the argument may accept the implication that the pain of a sentient being may not be normatively significant, or they may deny the implication by making the case that all sentient beings are bearers of value.28 These are lively disputes, but to my mind neither of the objections regarding pain proves decisive, i.e. grounds for the other side to give up their starting assumptions. The differences are rather more axiomatic. For the Moorean, bearers of value are events or states of affairs, and value is summable, scalar, etc. For proponents of the argument, bearers of value include objects, and value is not by definition quantitative, scalar, to be maximized, etc.29 There are always dangers attending comparative work. One has to be clear-sighted about how to frame a point of comparison when the starting assumptions are quite different. To my mind each side should permit the other their basic foundational assumptions, and the positions evaluated on their own terms. Taken on its own terms, Conee’s argument against the inference from instrumental value (understood as whatever makes things 27 I am most grateful to an anonymous reviewer for raising Bentham’s reply. 28 They may urge that while all sentient beings are bearers of value (in a sense they would need to explain), our reasons concerning them are defeasible or limited, e.g., perhaps we have a duty not to cause pain to a rat, but no duty not to kill it, or no duty not to do everything we can to keep it alive. Or it may be that because of their tendency to spread disease, a general duty not to cause harm to sentient beings does not hold in the case of rats, or in cases where they are in plague proportions, or other circumstances etc. To that end, proponents of the argument may appeal to a principle to the effect that: “A general right is, therefore, only a prima facie ground for the existence of a particular right in circumstances to which it applies.” See Joseph Raz (1986). The Morality of Freedom. Oxford: Clarendon Press, page 184. 29 On the axiomatic differences between Kantians and Moorean, see Bradley, “Two Concepts of Intrinsic Value.”

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

72 the value of humanity worthwhile on the whole) to intrinsic value is entirely plausible. Conee shows that an outcome need not be intrinsically valuable for what conduces to it to be of value. To challenge the Kantian regress argument, Conee’s argument needs to be transposed to a framework in which valuers are permitted to be bearers of value, and in which the value of what is good for valuers is conditioned by their value. Then the question is whether a comparable argument can be made against the inference from the existence of relational goods for valuers to the non-relational value of valuers, so that from the observation that relational goods depend for their normative significance on the value of valuers—an observation made by all parties to the dispute (recall the failure of DV, H, and NGB)—it does not follow that valuers are of value in themselves. Can it be shown that what benefits valuers—whose benefit is good and reason-giving— does not take its credit for doing so on loan from the value in itself of valuers? Plausibly—yes. The argument is that it is not inevitable for the value of what is good for a valuer to be taken on loan from the value in itself of a valuer because it is enough for the valuer to be of value in the sense that she can take something bad away, or make things better somehow. Mary is of value because she can prevent some harm from coming to her patients, or make things better for them. In that case what is of benefit to Mary—a good night’s sleep, a course of study, a research grant—matters because Mary matters in the sense that she can benefit her patients. The value of what is good for Mary is not taken on loan from the value in itself of Mary. Must Mary’s value be taken on loan from the value in itself of her patients? The lesson of Conee’s argument, relevantly transposed, is that their value need be no different from Mary’s—it is enough for them to mitigate harms or to make things better somehow. Must Mary’s value, and Mary’s patients’ value, be taken on loan from something further? Perhaps we picture the value of valuers as a function of their ability to contribute to (or to be parts of) something larger than themselves—the public good. But, again, the lesson of the argument is that it is enough for the public good to be of value in the sense that it mitigates harms and

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

3.2 evaluating the argument 73 makes things better for the public that it serves. This does not imply an infinite chain of dependents, but a structure of mutual dependence. It emerges that the borrowing conception (the metaphysical principle) that drives the regress argument is not inevitable. It is also worth asking whether it is actually a compelling model for the value of what is good for someone. The model is relatively intuitive when we have to do with instrumental value—with what is instrumentally good for someone—though Conee shows that it is not inevitable here. But to my mind it is far less intuitive when we have to do with the value of what is directly or non-instrumentally good for a person. When an object (activity, state, event, etc.) is directly good for a subject, a certain relationship holds between them; in so many words it is a relation of benefit, fit, advantage, or suitability. Like “gift” these are two-place predicates relating objects and subjects.30 When a work of art is good for a valuer, we have a work of art, and we have a valuer, and we have a relation in which the valuer is benefitted by the work of art. We might say that when the work of art and the valuer meet (in the right conditions, the valuer has not been in the gallery too long, etc.) they partake of a relation of benefit with the work as the thing that confers the benefit and the valuer as the being who receives it. To my mind it looks strained to say that the value is located in the valuer and borrowed by the work of art. In that case there is no such thing as relational value; there is just non-relational value and what is related to it. I find it more natural to say that value is the encounter between them— an encounter that is more familiarly described as being enriched, nourished, nurtured, strengthened, enlightened, comforted, and so on. The encounter crucially depends on the character of the relata— on the properties of the work of art, on the capacities, taste (etc.) of the valuer, and on the value-bearing status of the valuer. This observation has been thought to force the conclusion that the locus of the value is the valuer after all, and that the value of the valuer must be non-

30 See Kraut, What is Good, 86. And see Kraut, Against Absolute Goodness, 70.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

74 the value of humanity relational.31 But we have learned that this conclusion is not inevitable. Among the features of V when X is good for V and X is normatively significant is an evaluative property, but this property need not be intrinsic and may itself be relational: V may be of value in the sense that V takes bad things away or make things better. The value of the relation is fully conditioned by the character of both relata, but its value need not “come from” one of the relata whose value does not come from something. Relational value need not be value that borrows from non-relational value, and the model is not obviously adequate to the phenomena.

Circular structure For proponents of the regress argument, a chain of dependence between relational values must terminate with something that is nonrelationally valuable for any of the links to make sense. Why must it terminate? It must terminate because it is assumed that the value of a prior node derives from the value of a posterior node and ultimately from a final node that has a surfeit of value; the prior nodes are empty of value without it. Relevantly transposed, the argument from Conee shows that the value of what is good for a valuer need not be taken

31 Connie Rosati (2008). “Objectivism and Relational Good”. In: Social Philosophy and Policy Foundation 25, pages 314–349, p. 329 has argued that “The relational complex, X is good for P, does not include the monadic property good at all. Instead, it includes the relational property is good for P: it has X and P as relata and is good for as a dyadic relation.” In this way, Rosati has argued that being good for is a genuinely dyadic form of value that is not built out of the monadic good. However—and here Rosati takes a position that is in line with proponents of the argument in question— Rosati suggests that the value of this relation derives from the value of a person: “the source of the ‘good-for value’ of any item that is good for a person is its relation to a being with value” (p. 343). It is not Rosati’s concern to offer a full account of the value of persons, and she memorably calls this “one of the million dollar questions in ethics” (p. 340). But I take the suggestion to be that the value in question is non-relational. As I understand, what borrows from the non-relational value of a person is the relation in which X is good for the person, so that, strictly speaking, the relation (the encounter, the being benefitted, the being enriched, etc.) is not where value lies. Instead, the value has its “source” in the non-relational value of the person, and the value of the relation derives from it. To that extent, Rosati’s conception is a borrowing conception.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

3.2 evaluating the argument 75 on loan from the value in itself of a valuer. It is enough for valuers to be of value in the sense that they mitigate harms or make things better. The discussion brings out that IR is not ruled out by the structure of value. But that may not be much to recommend it. It is something of a reflex to find infinite regresses intolerable, though what is amiss is not always made fully clear. I have assuaged one source of discomfort, but likely a sense of the intolerable remains. A clear worry is that if we are to imagine an infinite chain of dependence between values that is not circular but linear, so that the value of a node depends on some further node in the chain not yet appealed to, then there would need to be infinitely many things. And it seems implausible for the existence of value to depend on there being infinitely many things. Why should the existence of value, a most ordinary and ubiquitous thing, have so extraneous a condition?32 Consider, then, a further option (C) according to which the dependence between values has a circular structure of the kind anticipated by the figure of Mary above. There are at least two ways to conceive of C: According to C1: relational goods are valuable because they can be good for valuers, and valuers are valuable because they can be good for relational goods. See Figure 3. According to C2: relational goods are valuable because they can be good for valuers, and valuers are valuable because they can be good for other valuers. See Figure 4.

Valuer

Relational Goods Figure 3 C1

32 I am grateful to Brad Weslake for discussion.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

76 the value of humanity

Valuer

Valuer

Relational Goods Figure 4 C2

C1 may be spelled out in a number of ways. Valuers may be valuable because they can prevent (or mitigate or remove) a harm to a relational value—they can protect an antiquarian manuscript from humidity, or an institution from a bad piece of legislation, etc. Valuers may be valuable because they can make a positive difference to relational values—they can innovate a technique, or conduct an important study. Valuers may be good for relational values because they can be part or constitutive of them, as when they are part of a valuable relationship, or a member of a valuable community. Of course, these explanations need not exclude one another—we can easily envisage some combination. Since valuers are here held to be good for relational values, C1 is different from RP. But just to that extent—since relational values are good for valuers—C1 is actually a version of C2. C2 explains the value of valuers in terms of being good for other valuers. Again, there are various options. Valuers may be good for valuers because they can prevent or mitigate some harm from coming to them—as doctors cure their patients, and as parents console a child. Valuers may be good for other valuers because they can provide them with a more positive benefit—as healers enhance the well being of patients, and as teachers help students to become wise. Valuers may be good for other valuers because they can be part or constitutive of what is good for them—as one person is good for another in being her friend. So according to C2, one valuer is valuable because she can protect, enable, or partly constitute the good for another valuer, and that valuer is valuable because he can protect, contribute to, or partly constitute, the good for her. We can imagine complex forms of interdependence here, with many different kinds of relationship:

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

3.2 evaluating the argument 77 that between friends, neighbors, teachers, students, colleagues, fellow citizens, and so on. It has been shown that C2 is not ruled out by the structure of value. But is it a plausible model for the value of human beings (valuers)? A natural objection is that the value of human beings does not depend on being good for other things (premise vi). C2 makes the value of human beings derivative on something else in a way that is at odds with how human beings should be treated—as ends-in-themselves. Proponents of the argument may defend the asymmetry between values and human beings by raising this Kantian-style of objection. Now there may be normative grounds for resisting this objection. It may be thought that human beings should be valued in view of their relationships of fellowship and dependency with others. To value human beings separately from those relationships would be to insist on a false separation, and perhaps a false image of the self. So it is said that “I am because we are,” or “a person is a person through other people.”33 But suppose we take the point of the Kantian injunction to be that we should never relate to human beings with a view only to what they can do for us. In Raz’s formulation, “To care for someone for their own sake requires recognizing merit in them independently of their role in our life.”34 I find this normative claim quite plausible. As I would formulate it, we should relate to people always with a view to their being the center of a life to which they bear a special relation. To my mind a relational view is plausible only if it captures this crucial idea. Whether proponents of the argument have an exclusive claim on it, or whether it can be captured in another way—this is what I now look into. Many readers will be familiar with a Stoic image of how people are related to one another in terms of a series of concentric circles.35 33 These are ways of parsing the Zulu term “ubuntu.” For discussion of human relatedness in contemporary African philosophy, see Sandy Koullas (2019). “Love, Practical Reason, and African Philosophy”. In: The Routledge Handbook of Love in Philosophy. Edited by Adrienne M.Martin. Routledge, pages 313–324. 34 Raz, “The Amoralist,” 294. 35 The image is due to Hierocles. See A. A. Long and D. N. Sedley (1992). The Hellenistic Philosophers. Cambridge: Cambridge University Press, 57G. I have learned

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

78 the value of humanity At the center is the individual herself, for it is assumed that the closest relationship a person bears is a relationship to herself. In the next circle comes kin, then neighbors, then fellow citizens, then all of humanity, and then all beings. The Stoic ideal for interpersonal relationships is to bring those on the outer circles in, as a person is invited to relate to another as she relates to herself. The model raises a nice question. What is the relation a person bears to herself? As a person stands to be good for others by being her friend or sister or teacher, could she stand to be good for herself? Philosophers are not strangers to proposals that are against appearances, or common opinion. After all, we countenance backwards causation, or the robust existence of infinitely many possible worlds— including worlds with kangaroos without tails who are forever toppling over.36 If we have not come to expect peals of laughter from the uninitiated then perhaps we should. But there can be a lot behind a bold proposal. Contemporary moral philosophers tend to take the subject matter of ethics to be—in T. M. Scanlon’s apt phrase—what we owe to others. But there are older traditions in which ethics also has to do with our relationship with ourselves. For Kant there is the idea of duties to self. Though contemporary Kantians can be forgetful about this dimension of Kant’s view, it might be regarded as a strength.37 Since duties to others form a system with duties to ourselves, without duties to ourselves there is nothing to limit what we owe to others, and the theory gives rise to worries of over-demandingness.38 My own sympathy lies with views, not according to which there is some limit to what we owe to others, but according to which in order to be of from discussion by Katja Maria Vogt (2008b). Law, Reason and the Cosmic City: Political Philosophy in the Early Stoa. Oxford and New York: Oxford University Press, pages 103–104. 36 On kangaroos, see see David Lewis (1973). Counterfactuals. Oxford: Blackwell. I am grateful to Stephen Harrop for this reference and for discussion. 37 An exception to the trend is David Velleman (2006). “A Brief Introduction to Kantian Ethics”. In: Self to Self. Cambridge: Cambridge University Press. 38 See Katja Maria Vogt (2008a). “Duties to Others: Demands and Limits”. In: Kant’s Ethics of Virtue. Edited by Monika Betzler. Berlin and New York: Walter de Gruyter, pages 219–243.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

3.3 a reflexive relation 79 benefit to another we must truly have something to give. In Eastern and Western eudaimonistic traditions our relationship to ourselves is paramount for virtue. The good person is the person who benefits— who genuinely is of benefit. She is of benefit to others, of course, but also, and importantly, to herself. She has something to offer another because she is a master of one—she has knowledge of how to be well in her own life.

3.3 A Reflexive Relation The discussion suggests a further way of developing a relational explanation of the value of human beings. According to C2, relational goods are of value because they can benefit human beings (valuers), and human beings are of value because they can benefit other human beings, where standing in a relation of benefit to a human being is a way of being valuable. The present proposal starts from the observation that we can stand in a relation of benefit to ourselves, and that this is a way of being valuable. Relational goods are valuable because they can be good for human beings, and a person is of value because she can bear a relation to herself—can be of benefit to herself. A chain of dependence between relational values comes to an end with a reflexive relation (R). See Figure 5. R may take various forms. A person may benefit herself by preventing or mitigating some harm, and many of our actions are undertaken in this spirit—taking vitamins, having exercise regimes, and in general, taking care of what is onerous. We do not want to get sick, or have things be worse later, and so on. In that case we are of benefit to ourselves not by making things go well so much as preventing

Relational Figure 5 Reflexive relation

Valuer

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

80 the value of humanity things from going badly. But of course, a person can also benefit herself in a more positive way—by having a conversation with a friend, or by taking a trip to a museum, or indeed, by volunteering for a soup kitchen. There is a familiar argument according to which more local ends are undertaken for the sake of more remote ones—having meaningful relationships, taking an interest in culture and the arts, being a benevolent person—and ultimately for the sake of a single final end: that of having a good life. So at the limit, you might say, human beings are able to benefit themselves by having ends that are part or constitutive of a good life. Is the value of a good life foundational on this model? We seem to face a new regress argument, this time for the special value of a good life. The argument is that there must be some one thing for the sake of which we do everything we do, for without it our actions will seem pointless and vain. Our actions are not pointless and vain, and the thing for the sake of which we do all the rest is a good life.39 The argument would establish that the existence of what is instrumentally and intrinsically good for people entails the existence of final value (where what is finally valuable is a good life). But, as with what is instrumentally and intrinsically good for people, final value is best treated as a form of relational value—as good because it is good for someone. Aristotle’s regress argument should not be thought to establish something stronger: namely, that there must be something that is good in itself, and a good life meets the criteria. That involves further argument, and it is not on the face of it plausible. It fails to capture the sense in which a subject bears a special relationship to her own life. It is better to say that a good life is of value because, at a minimum, it is good for the person whose life it is.40 In that case R 39 The argument is recognizable from Aristotle, Nicomachean Ethics, 1.2. It is readily pointed out that 1.2 only secures the weaker conclusion that some of our ends must be more complete than others, not that there must be a single most complete end. I read 1.2, like 1.1, as prospective. It is supplemented by the discussion of self-sufficiency and completeness in 1.7 so that the regress argument announced in 1.2 is rounded out by the additional premises supplied in 1.7. 40 See Joseph Raz (2004). “The Role of Well-being”. In: Philosophical Perspectives 18, pages 269–294, p. 269.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

3.3 a reflexive relation 81

Valuer

Good Life

Relational Goods Figure 6 R*

should be formulated so as to make explicit that human beings can be of benefit to themselves in the sense that they can lead good lives, where the value of a good life is also explained relationally. Relational goods are of value because they can be good for human beings, and human beings are of value because they can lead good lives, and good lives are of value because, at a minimum, they are good for the human beings whose lives they are—call this R.* See Figure 6. Now return to the Kantian-style of objection. The objection to C2 was that we should value people for themselves and not for what they can do for another. R* has the advantage over C2 of making the value of people, if not non-relational, then non-derivative in the sense that a person’s value is not a function of being good for someone else. This makes the value of a person basic in a certain sense. Intuitively put, a person matters because she matters to herself in a very particular sort of way. To appropriate a phrase, she is a being for whom her life can be an issue. In this R* looks well-placed to capture what is at stake in our ways of relating to another. A person has a life to lead, and a good life is a wonderful thing. How a person’s life goes is something we stand to impact in our ways of relating to them. Our impact is of significance to us, to the people in their life, but perhaps most of all to them—for they are a very particular sort of center of a life. R* captures the sense in which people are of value independently of their role in someone else’s life—independently of what they can do for us or anyone else—and this is an important dimension of the idea that people are ends-in-themselves or of value in themselves. So the failure of the regress argument for the non-relational value of people does not leave us without resources to secure what is owed to people. R* is poised to make good on the intuition that we should relate to people

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

82 the value of humanity in such a way that we show appreciation for their being the center of a life to which they bear a special relation.

3.4 Worries About Self-Love The idea of standing in a valuable relationship with oneself— colloquially, a relationship of self-love or self-care—may be met with suspicion, as if the proposal were sanctioning a preoccupation with the self—with solipsism, or narcissism, or “getting one over on others”—that is antithetical to ethics and to other-concern. It is undeniable that concern with the self has a way of showing up in our culture as a form of indulgence or ruthlessness. As Michael Stocker has written: Here we might consider not so good people who see themselves as out for number one; [. . .] or who see the world in terms of constant struggle, with people pitted against each other as in the Social Darwinism of the captain in Jack London’s Sea Wolf ; [. . .] or who see others more or less as objects to be used and disposed of or to be played with; or who, for narcissistic reasons, require being the center of love and attention.41

Stocker writes memorably, too, of entitlement as a trope in advertising that coaxes us to think “we deserve or owe it to ourselves to get and have. (I mean get and have quite generally, not just what is being advertised).”42 As it is understood here, care of the self is part of the myth of capital whose principle is that we should have our heart’s desire. We find comparable ideas in Kant for whom self-love is the basically childish drive to get whatever we want—the motivational correlate to price value that is at the root of any inclination, and that is an endless threat to rectitude. As in Schiller’s famous satire—in which Kant is thought to enjoin us to despise our friends so that when we serve them out of duty our actions will be supremely valuable—

41 Michael Stocker (2008). “On the Intelligibility of Bad Acts”. In: Moral Psychology Today: Essays on Values, Rational Choice, and the Will. Edited by David K. Chan. Volume 110. Philosophical Studies Series. Springer, pages 123–140, pp. 137–8. 42 Stocker, “On the Intelligibility of Bad Acts,” 132.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

3.4 worries about self-love 83 Kant’s position can be overstated.43 But it is certainly true that actions motivated by inclination—by the concerns of “the dear self ”—are entirely without moral worth for him. While they take a distinctive form in modern life and moral philosophy, these are perennial concerns. It is part of Aristotle’s method to proceed with a survey of popular opinion, so we know that fourthcentury Athenians are reputed to have found self-love contrary to virtue: People criticize those who love themselves most, and call them self-lovers, using this as an epithet of disgrace, and a bad man seems to do everything for his own sake, and the more so the more wicked he is [. . .] while the good man acts for honor’s sake, and the more so the better he is, and acts for his friend’s sake, and sacrifices his own interest.44

As Aristotle unpacks this pre-conception, the good person acts for the sake of others and for the sake of what is admirable—to the extent that he sacrifices his life—while the self-lover competes with others over wealth, honor, and bodily pleasure as if they were the best of things. In popular opinion as Aristotle reports it, then, self-love is something vile and base. As we find in our day too, however, opinion tends to vary and it is instructive to consider the range. Aristotle also gives voice to a countervailing view according to which we should love our friends as we love ourselves—“that a friend is another self ”—indeed, that all friendly relations extend outwards from a person’s relationship with himself. Hence the proverb “charity begins at home,” and others like it. According to this line of thought, our relationship with ourselves is actually a model or ideal for interpersonal relations, not unlike the

43 Friedrich Schiller (1982). “The Philosophers”. In: Xenien. Edited by Erich Trunz. Volume 1. Munich: Beck, page 28. Kant does not require that we be disinclined to do what we must. In the first part of the Groundwork (4:397–4:399), we encounter examples of people who, for various reasons, do not want to do what they should and do it anyway. The point of these examples is to clearly identify actions that are motivated by duty, and since motivations are opaque, this is most obviously seen in cases where someone does what we know they have no inclination to do. 44 Aristotle, Nicomachean Ethics, IX. 8 1168a 28–35.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

84 the value of humanity Stoic image invoked above. So taken one way, a concern with the self is thought to be ruthless and terrible. But taken another way, it is thought to be exemplary—the starting point of ethics. It is characteristic of Aristotle that he finds a truth in both conceptions. As the phrase is commonly used, “self-love” is ascribed to those who pursue status and consumables at the expense of other things and other people. Since the pursuit of status and consumables is common, and since it is misguided, “self-love” tends to be used as a term of disapprobation. But Aristotle rejects this conventional usage as unsuitable. We should not say of a person who competes with others to gratify his appetites that he is good for himself, a self-lover, and so on. In saying that we concede too much to mistaken ideas about the good—as if to really be good for oneself is to acquire yachts and consume delicacies and champagne. On the contrary, it is the person who acts for the sake of others, and for the sake of what is admirable, who is truly good for themselves. In setting out to do what she should be doing—in seeking to secure the right course of action—this person assigns to herself the better good, for excellence, understood here as getting it right in practical matters, is better than hoarding and delicacies.45 According to Aristotle, then, it is the virtuous person who is the true lover of self—who is truly good for herself. A lover is one who wishes and does what is really good for the beloved’s sake, and the virtuous person wishes for herself what is really good (in the sense just described) and pursues it for her own sake.46 In a longer treatment of Aristotle one would consider remarks at the end of the Nichomachean Ethics to the effect that the ethical education of Aristotle’s audience will not be complete until they have studied the principles of politics. I share John Cooper’s view

45 Aristotle, Nicomachean Ethics, IX 4 and 8. 46 In his critique of Kant’s conception of self-love as really a conception of selfindulgence, Harry Frankfurt makes something close to Aristotle’s argument. For Frankfurt, love no more involves the gratification of brute desires when the object is another person than when the object is the self. In both cases, love properly involves concern for the beloved for their own sake. See Harry G. Frankfurt (2001). “The Dear Self ”. In: Philosopher’s Imprint 1.0, pages 1–14.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

3.4 worries about self-love 85 that, while Aristotle’s remarks may be taken in various deflationary keys, he means, quite strongly, that a genuine understanding of what community is is required for his hearers to lead individually good lives.47 Community is constituted by the common activities of its members that are sustained on an ongoing basis and continually. These activities are undertaken in different spheres of life—the family, the neighborhood, the city—and part of what it is to fare well in one’s own life is to participate in the relationships and to engage in the activities that constitute these larger spheres. Participating in the lives of others—being their friend or teacher or student—taking on the responsibilities that are part of those forms of participation, are crucial constituents of a person’s own good. Living well in one’s life involves relationships with others—and in that case what is beneficial for oneself and what is beneficial for others should not be sharply separated—but equally, since an individual is not imagined to be something other than the community, but a partial constituent of it, faring well in one’s own life has a wider context and import. If what it is for the whole to be well-going is for the parts that comprise it to be well-going in their spheres, then faring well in one’s own life is part of the communal or political good. The value of living well is not the possession of the individual herself, as if the benefit accrues to her and not to others. The point of this excursus is not to settle the question of what it is to stand in a relation of benefit to oneself. I have so far said nothing about what is involved in living a good life, or about what conception of a good life is licensed by R.* The point of the excursus is to bring out that, just as we should not take it as obvious that genocide is “good for” the perpetrators of genocide (recall the discussion of DV and H), so we should not take it as obvious that being good for oneself involves self-indulgence or isolation from others.

47 John Cooper (2010). “Political Community and the Highest Good”. In: Being, Nature and Life: Essays in Honor of Allan Gotthelf. Edited by James G. Lennox and Robert Bolton. Cambridge: Cambridge University Press, pages 212–264, pp. 18–19.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

86 the value of humanity In this chapter I examined an argument for the view that the value of human beings, or more generally, of valuers, must be non-relational. The argument is that the value of what is good for people ultimately derives from the value in itself of people. I offered a diagnosis of the conception of relational value that drives this argument, and I showed that the conception is not only not inevitable, but that it is implausible when we have to do with the value of what is non-instrumentally good for a person. By working through metaphysical questions about the structure of value, I canvassed several relational alternatives. I gave normative reasons to favor a reflexive model according to which people are of value because they can be good for themselves in the sense that they are able to lead good lives, lives that are of value (in the first place, but not solely) because they are good for the people whose lives they are. To offer an intuitive formulation, people are of value because they are beings for whom their life can be an issue.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

4 On Valuing and the Good Life The project undertaken in this book is to offer a relational explanation of the value of humanity. As I explained in Chapter 2, it is motivated by a hypothesis about what value is—a hypothesis to the effect that the evaluative is tied to the beneficial. Presumably a difficult testcase for this way of thinking about value is the case of we human beings ourselves. Indeed, according to a powerful argument, the very conception of value that is at stake in the hypothesis appears to imply that human beings are valuable in a distinctly different sort of way. In Chapter 3 I gave reason to reject this argument: the non-relational value of human beings is not forced on us by the structure of value. I also introduced a positive relational view according to which our value turns on the relation we can bear to ourselves, a relation of being “good for” ourselves. What remains is to develop this proposal so that it becomes, not just conceivable, but plausible. The proposal gains plausibility if it secures important desiderata for an account of the value of humanity, and crucially, core forms of ethical behavior that are owed to everyone. This will be the burden of Chapter 5. But first I need to address a lacuna in what has been said so far. I have yet to say anything about that in virtue of which human beings are bearers of relational value. In the course of a discussion of the distinctions in value in Chapter 1, I argued that theories of value have (at least) these two components: (i) features in virtue of which an object is of value, and (ii) an explanation of why those features make the object valuable. Both are involved in giving an account of how something should be responded to, and in this chapter I provide an account of (i).

The Value of Humanity. L. Nandi Theunissen, Oxford University Press (2019). © L. Nandi Theunissen. DOI: 10.1093/oso/9780198832645.001.0001

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

88 the value of humanity I address the question: What is it in virtue of which people have the form of relational value that is proposed on R*? What makes a person able to be of benefit to herself, and in that way of value? I develop the suggestion that the basis of our value is a capacity to value. It is because human beings are valuers that we can be “good for ourselves” and in that way of value. The discussion has implications for an account of the good life for human beings, and the chapter ends with an endorsement of a pluralistic form of realism.

4.1 Valuers Consider what we are asking about when we ask about that in virtue of which something is of value—what I am calling the basis of value. Imagine we are standing before a work of art—before an iteration of Constantin Brancusi’s Bird in Space—and we ask ourselves what makes it remarkable or worthwhile. When we try to say something about the sleekness of bronze on marble, or the sense we have of flight, and poise, we are pointing to features of Brancusi’s sculpture on which its value depends—at least in our estimation. By itself, this dependence relation tells us nothing about how the features make their bearer valuable. Conceivably, the features could make the sculpture such as to be non-instrumentally good for someone (that would be a relational explanation), or they could make the sculpture valuable in itself (that would be a non-relational explanation). These questions—questions of the basis and of the explanation of value— come apart because it is quite possible for Brancusi’s sculpture to be of value in virtue of its intrinsic properties, say, though those properties make the work relationally valuable in the sense that they make it non-instrumentally good for a person. And conceivably contrariwise. The point is that both axes are centrally relevant to questions of why something is good or of value, and together they constitute the ground of value.1 1 Naturally, there are further explanatory questions to ask, questions that belong more broadly to metaethics. On the relationship between metaphysics and metaethics

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

4.1 valuers 89 As we saw in Chapter 3, the working assumption of proponents of the regress argument is that it is valuers who are non-relationally valuable. The proposal of Raz and others is that people are of value in virtue of a capacity to value, and possession of this capacity makes people non-relationally valuable. I have given my reasons for rejecting the explanation in terms of non-relational value. But I have so far said nothing about the first part of their proposal: the value of human beings depends on a capacity to value. For an otherwise diverse group of thinkers—Sarah Buss, Robert Nozick, Christine Korsgaard— the basis of human value is thought to lie in one way or another in valuing—in the distinctive relationship we bear to objects and activities of value.2 It is an Aristotelian thought that there is nowhere to start but from where we are—from the thoughts we have already, or from the thoughts we understand of the wise.3 So let me follow the wise in taking as a starting point that the basis of our value is a capacity to value. The starting point finds support if it can be shown to generate a plausible explanation of our value. But first we must ask: what is it to value? What are the cognitive and affective dispositions involved in being a valuer? The topic of valuing has come into focus relatively recently in philosophy; the term “valuing” is certainly late to enter the philosophical lexicon. My sense is that philosophers have become interested in valuing in light of critical discussions of consequentialism. As ever, somehow the conversation begins with Bernard Williams, and Williams famously argued that in requiring us to give up what he called our “ground projects” if it would maximize general welfare to do so, consequentialism is implausible and at worst incoherent. Likewise, Elizabeth Anderson and Tim Scanlon argued against the idea in these discussions, see Rabinowicz and Ronnow-Rasmussen, “A Distinction in Value,” 37 and passim. 2 Cf. Buss, “The Value of Humanity.” Cf. Christine Korsgaard (1986). “Kant’s Formula of Humanity”. In: Kant-Studien 77, pages 183–202. And cf. Robert Nozick (1981). Philosophical Explanations. Cambridge, MA: Harvard University Press. 3 See Bk. I.I of the Topics. Aristotle (1987). A New Aristotle Reader. Edited by J. L. Ackrill. Princeton, NJ: Princeton University Press.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

90 the value of humanity that maximization is the primary response to value. There has since been interest in how to understand non-consequentialist responses to values, and a common thread is that when we value something we make it part of our lives. In what follows I investigate what is involved in the distinctive relationship we bear to objects and activities of value. In this way I hope to make the proposal that our value depends on that relationship somewhat fuller and more precise. Traditionally, our relationship to values has been analyzed in terms of belief or desire (or some combination).4 But you might think that our defining relationship to values has a richer structure than this. As Samuel Scheffler has observed, we can desire something for its own sake without valuing it (eating jelly-beans in his example); and we can believe that something is worthwhile without valuing it (believing that Bulgarian folk dancing is valuable without making that activity part of our lives).5 Whereas there seems to be no upper limit to the number of things we can desire, and no upper limit to the number of things we can believe are valuable or worthy, there does seem to be an upper limit to the number of things we can value. More is involved in making something part of our lives in the way that valuing something makes it part of our lives. Scheffler offers the most systematic account of valuing that I am aware of. He lays out several features. To value X involves: i) a belief that X is valuable or worthy; ii) a susceptibility to experience a range of context-dependent emotions regarding X; 4 This is is true of Korsgaard’s analysis. For Korsgaard, to value (or “set”) an end is to want something under the influence of reason, where that means to want something different from what instinct directs us towards. So it is to want a pear having tasted an apple and seen that they are visually similar (the work of comparison being the work of reason). Or it is to want love, beauty, and things in the future (desires characteristic of rational beings). On the other hand, Korsgaard suggests that to value an end is to believe or judge that it is valuable: worthy of pursuit or realization. These are different proposals, but we can easily imagine that Korsgaard envisages some combination of belief and desire. Conceivably, there are many variants of this basic model. See Korsgaard, “Kant’s Formula of Humanity,” 188–190. 5 Scheffler, “Valuing,” 21.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

4.1 valuers 91 iii) a disposition to experience these emotions as being merited or appropriate; iv) a disposition to treat certain kinds of X-related considerations as reasons for action in relevant deliberative contexts.6 Do these conditions capture what we want—and all we want—from an account of our defining relationship to values? One might think that valuing involves, but is not exhausted by, believing that what one values is worthwhile (condition i) to the extent that it is a rational attitude. An attitude is rational when it is responsive to reasons, and being responsive to reasons involves (at least tacit) belief in some positive feature of the object responded to. We form intentions on the basis of this belief, and we interpret and explain the actions of others by attributing this belief to them. I regard these as commonplaces about human action and intelligibility. They are also basic tenets of the guise of the good theory of action.7 But controversy over the guise of the good turns on how these commitments are understood— as contingent generalizations about human agency, or as necessary truths regarding the concepts of intention, reason, and value.8 I am not taking a stand on this question, and my rationale for condition (i) is intended to be relatively uncontentious.9 The attentive reader will already see that for the condition to be compatible with a relational

6 Scheffler, “Valuing,” 29. 7 See for example, Joseph Raz (2010). “On the Guise of the Good”. In: Desire, Practical Reason, and the Good. Edited by Sergio Tenenbaum. Oxford: Oxford University Press. 8 As Kieran Setiya has written (and Setiya rejects the guise of the good theory of action): “We tend to want what we perceive as being good in some way good, to choose what seems worthy of choice, and to act in ways we think we can justify, at least to some extent. The question for action theory is not whether to accept or deny these platitudes about human agency, but how to interpret and explain them.” Kieran Setiya (2010). “Sympathy for the Devil”. In: Desire, Practical Reason, and the Good. Edited by Sergio Tenenbaum. Oxford: Oxford University Press, pages 82–110, p. 82. See also Michael Stocker (2008). “On the Intelligibility of Bad Acts”. In: Moral Psychology Today: Essays on Values, Rational Choice, and the Will. Edited by David K. Chan. Volume 110. Philosophical Studies Series. Springer, pages 123–140. 9 It would be denied by those who see reasons as fundamental, and I will come back to this style of position in Chapter 5.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

92 the value of humanity proposal, the content of the belief cannot be that the end is good in itself or good period.10 One might also think that valuing involves a disposition to treat what we value as reason-giving, not merely in a single context but in contexts that are relevantly similar (condition iv). The condition should capture the sense in which we cannot value for an instant, but only over time, and with a degree of persistence. As it stands, however, the condition is too weak to capture the temporal dimension. For a person can have and then lose a disposition to treat something as reason-giving in relevant contexts. In a burst of enthusiasm, I can be disposed to treat writing as reason-giving in relevant contexts in the future, and lose that disposition as soon as I sit down to write. To value writing, the disposition to treat writing as reason-giving must itself be a stable feature of my motivations. Motivational stability is central to Harry Frankfurt’s account of the related notion of care, according to which to care about something is essentially to be committed to caring about it.11 I would say, more weakly, that valuing involves (but is not wholly constituted by) a commitment to valuing it. And though Frankfurt takes the commitment to feature in the content of the attitude of caring itself, it need not; the commitment could be a fact about the person: for example, a dispositional fact. For me motivational stability finds its point in view of the fact that we value things over time. But there is surely more to say about temporality. It matters not merely that values are ordered in time, but how they are ordered. Scheffler does not speak of ends, but it is an Aristotelian thought that goods are ends. For Aristotle ends form chains of dependence that are hierarchically structured, with more immediate ends undertaken for the sake of less immediate ones. It is the less immediate or more final ends that guide deliberations and furnish long-range plans—to write some experimental non-fiction,

10 On a related point, see Judith Jarvis Thomson (2013). “GOODNESS”. In: Philosophy and Phenomenological Research LXXXVII.2, pages 467–475, pp. 467–468. 11 Cf. Harry G. Frankfurt (1999b). Necessity, Volition and Love. Cambridge: Cambridge University Press, p. 161.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

4.1 valuers 93 to be a kind person, to become a dancer. To me the hierarchical ordering of ends in time seems to capture a distinctive feature of our relationship to values. Intuitively, we organize our lives around projects, interests, relationships, and self-ideals to which we are deeply committed. Our longer range or more final ends provide guidance for our more particular deliberations, so that my commitment to experimental non-fiction writing means that I set aside some hours to write everyday, including this day, and so on. As a final ends valuer there is a structure to our deliberations—to what we treat as reasongiving—with final ends shaping more local ones. I have appealed to Aristotle in motivating the idea of long-range deliberation, but I should immediately say that the notion of finality that I am employing departs from Aristotle’s in important respects.12 As we know, Aristotle is primarily interested in the idea that there is a most final end (though on some interpretations, the most final end is disjunctive, i.e., a list of various possible high-level final ends). Aristotle tells us that the most final end is “complete”—that for the sake of which everything else is undertaken, though it is not undertaken for the sake of anything else; and “self-sufficient”—such that nothing could be added to make it better.13 I do not assume that there is a most final end, and I do not assume that a final end will be complete and self-sufficient. My notion of finality is relatively weak in allowing that a final end can be pursued both for its own sake and for the sake of something else (as a vocation is pursued both for its own sake, and for the sake of gainful employment). A notion of final ends may also require: (a) that the end be an abstract value (for Aristotle it is happiness, but other candidates might be freedom, justice, and rightness); and (b) that the end be something an agent would not revoke, or put into question, when pressed. I am not committed to these further ideas.

12 For a reading of Aristotle’s conception of agency that places particular emphasis on long-range deliberation, see Vogt, Desiring the Good, ch. 5. 13 Aristotle, Nicomachean Ethics, 1.7.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

94 the value of humanity So, to Scheffler’s suggestion that valuing involves a disposition to treat what we value as reason-giving in relevant contexts I would add that the disposition must be stable, and I would add that we be guided by what we value in our long-range deliberations. I am understanding valuing in terms of having ends, where ends are things we aim to bring about by way of our actions. An end can be a goal, like becoming a writer of experimental non-fiction; an end can be the making of something, like a meal for seven people; an end can be an activity, like appreciating the beauty of the natural world; or an end can be a state, like the cultivation of ataraxia. I will take the relevant relation to ends to be one of having them. According to an influential proposal, the appropriate relationship to ends is one of “setting” them. But if setting an end implies knowingly deciding or choosing to pursue it, then setting is too restricted. For it seems possible to have an end without knowing that we have it as an end—our ends might reveal themselves in our behavior, rather than in our willed decisions and claims. On the other hand, if to set is not yet to have an end, then setting is too relaxed. For in the absence of having an end—in the absence of pursuing it—it is hard to say if we have really decided, or if that is clear enough, how valuable deciding is. So I will speak not of setting but of having ends.14 A further dimension of Scheffler’s account is that valuing is partly constituted by emotional susceptibility to what we value (condition ii), 14 Does Kant regard people as valuers in this sense—as having higher-level ends? Valuing is sometimes put forward as a Kantian account of the basis of human value. As I read the canonical writings I admit I do not quite see this. I see Kant saying that our value has everything to do with autonomy—moral legislation in his distinctive sense. It would be tempting to say that Kant is interested in moral rather than non-moral values or ends. But I am not sure that Kant quite sets things up that way. Kant is not making use of my notion of higher-level ends, but if we were to find a place for that notion in his system, it would likely come up in connection with imperfect duties to ourselves and others. In that case the ends would be moral through and through. To look ahead, I will argue that the capacity for having higher ends that grounds our value cannot be a capacity for nefarious ends—the ends of tyranny, oppression, racism, greed, and so on. My claim will be that these sorts of ends cannot make our life go well and to that extent cannot ground our value. So it will turn out that the ends that matter to me are not not moral. In that sense there are connections to a Kantian sort of proposal.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

4.1 valuers 95 so that if the photograph we value is destroyed then we are disposed to be distressed or saddened. This seems right to me. And yet the character of emotional susceptibility will have an additional dimension when values are ends. Ends are things we try to realize by way of our actions, either constitutively (riding a horse) or productively (making a table). When values are ends we will be emotionally vulnerable to the condition of the constituents of our ends, but also to whether our end is realized, and to what interim successes or failures mean for that realization. If spending time in state parks is one of our ends, the proposal is that we will be emotionally vulnerable to threats to the integrity of the parks, to the health of the habitat, and so on. But we will also be emotionally susceptible to threats to our ability to engage in the activity successfully: for example, to the prospect of moving to a place without access to green space. Success here is context dependent. In the case of spending time in nature, what counts as success may be no more than engaging in the activity in ways that are appropriate to it. In other cases it will depend on personal resolutions and plans. In others still, it will be determined by standards that are institutionally defined.15 In sum, I take valuing to involve having final ends, where having final ends includes but goes beyond Scheffler’s conditions. We have a final end iff we: i) have a belief that the end is worthwhile; ii) have a susceptibility to experience a range of appropriate emotions regarding the success or failure of our end; iii) have a stable disposition to treat the end as reason-giving in relevant deliberative contexts; iv) are guided by the end in long-range deliberation. 15 On Scheffler’s view, the person who is emotionally susceptible to what she values will be disposed to find that susceptibility appropriate (condition iii). Indeed, so much appears to be entailed by the first two conditions taken together—by belief that what she values is worthwhile together with the susceptibility to experience relevant emotions regarding it. The entailment requires the assumption that believing worthwhile entails (at least tacitly) believing that emotional susceptibility is appropriate. And it requires the assumption that being emotionally susceptible entails (at least tacitly) believing that we are. If these assumptions are plausible, then the appropriateness of emotional susceptibility need not be listed separately.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

96 the value of humanity So, if being a painter is a final end for M. T., then she believes that painting, or more likely, certain forms of painting, is worthwhile. Should she complete the body of work for which she received an interesting commission, and see it installed to her satisfaction, then she is disposed to take real pleasure in the work. Should she receive a comprehending critique from a person she respects, then she is disposed to feel gratitude. In the meantime, she is disposed to read or listen to relevant discussions about art, and to engage with other practitioners in her community. More than that, she is reliably disposed to establish a daily practice, perhaps planning travel around important exhibitions, and so on. Obviously, there is more to say about final ends. Some final ends will have a clear end-point or moment of completion. Others will be more diffuse, with no clear end-point. We may have several final ends, and they may be of varying importance to us. They may be related to one another in various ways (hierarchically, or as parts of wholes, or in some other way); or they may be largely unrelated. Though it is possible for the ends to be largely unrelated, there may be some need to unify them. There are likely other constraints. For example, a person’s final ends cannot be incompatible with one another—as the life of a professional ballerina may be incompatible with that of a student who is serious about becoming an academic.16

4.2 Valuing and the Good Human Life A common starting point in discussions of the value of humanity is that our value depends on the rich and complex relationship we bear to objects and activities of value. I have offered an analysis of that relationship—of valuing—in terms of having final ends, and the analysis makes the starting point more precise. If we join with others in accepting that we should address ourselves to a potential in someone, and I think we should, then what we are concerned with is the capacity to value. The starting point finds support if it can be

16 I am most grateful to Wolfgang Mann for a discussion of ends and finality.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

4.2 valuing and the good human life 97 shown to generate a plausible explanation of our value, and this is what I turn to now. How or why should the capacity to value make people of value? Some contend that valuing confers value on people because it makes people the non-relational “source” of value.17 Others contend that valuers meet the criteria for being non-relationally valuable. What is non-relationally valuable is such that other things are good because they can be good for it, though it is not such that it is good because it can be good for other things, and this is true of valuers. I have given my reasons for rejecting these arguments. The candidate explanation that I have been exploring is that our value turns on a relationship we bear to ourselves—R.* If valuing is the basis of our value, and bearing the relevant relationship to ourselves is the explanation of our value, then the proposal is that we are of value because, as valuers, we are able to be good for ourselves in the sense that we are able to lead good lives— lives that are good for us. More fully spelled out, people are of value in virtue of a capacity to value (intrinsic property), where that capacity, when exercised, makes people good for something (relational property), as on the view I favor, it makes people able to lead good lives. As I argued in Chapter 1, it is possible to conjoin a view according to which the nature of the properties on which something of value depends are intrinsic, while those properties admit of a relational explanation of the value of the object. It is a virtue of the proposal that it can be tested. The proposal entails that valuing indeed makes people able to lead good lives, and we can ask whether this is so. To substantiate the proposal it must be shown how each of the conditions on valuing described above is a plausible constituent of a good life. A proponent of R* is committed to treating the four conditions as necessary components of a good life, and as the sufficient components contributed by subjects. This allows that fortune also affects the course of our life while holding that the contributions of fortune are irrelevant to our value. Naturally it is not my aim to provide a full account of the good life here; that 17 See my discussion of this style of proposal in Appendix 2.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

98 the value of humanity would be a book length study of its own. The interest in outlining the connections between the capacity for valuing and the nature of a good life is twofold. On the one hand, it helps to show what being good for ourselves amounts to; on the other, it shows how valuing makes us good for ourselves. According to condition (i) we believe that what we value—what we have as a final end—is worthwhile. What it is for an object to be worthwhile is for it to possess features that make it such that it can be good for a valuer. To believe that an object is worthwhile is to believe (at least tacitly) that it possesses some such set of features. Is it plausible that a good life is a life in which a person pursues some project, self-ideal, or commitment that she believes is worthwhile in this sense? It would be bad for a person to pursue a project or commitment that she believed was harmful or of no benefit to anyone. If there is a rational connection between pursuing something and believing that it is worthy, then pursuing an end that is believed to be harmful or of no benefit to anyone would be irrational or incoherent. While certain aspects of our lives may be marked by irrationality or incoherence without harm, it would be harmful for our pursuit of final ends to be irrational, for ex hypothesi, final ends play a structuring role in our long-range deliberations.18 Condition (ii) involves appropriate emotional responses to the constituents of our ends and to how our ends are unfolding. Is it plausible that these forms of emotional responsiveness will be part of a good life? It is a longstanding Aristotelian presumption that appropriate

18 Contrast Harry Frankfurt’s view. A notion of care plays the central deliberative role in Frankfurt’s account that I am envisaging for final ends, but Frankfurt thinks we can appropriately care about something without believing that it is worthy. We can care, in Frankfurt’s idiom, for no reason, or at least, for no reason independently of our caring. Frankfurt cites care of one’s children as paradigmatic: one does not care about one’s children because one recognizes that they have value. Harry G. Frankfurt (2004). The Reasons of Love. Princeton, NJ and Oxford: Princeton University Press, page 39 and passim. But is this a clear example of caring for no reason? As Katja Vogt has suggested to me, perhaps the reason is simply that the children are one’s own. Indeed, it is difficult to find a case in which it can truly be said of someone that she cares for no reason.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

4.2 valuing and the good human life 99 emotional response is itself part of what is involved in engaging in action and activity in the right sort of way, and so itself part of what constitutes the good for people. I will not rehearse these arguments here. Let me simply make the further point that a person with the appropriate emotional responses towards her ends will find meaning and point in what she does; she will pursue her projects with energy. Dissatisfaction with how her ends turn out will lead her to adjust her plans, or to pursue them differently. The arguments for conditions (iii) and (iv) are related to one another. Consider the difference between what I will call simple valuing and final ends valuing. To be a simple valuer is to be guided by practically relevant features of the world in the formation of beliefs, emotions, intentions to act, and other attitudes. It is consistent with simple valuing that responses to practically relevant features of the world be relatively unsystematic; the simple valuer responds to this and then that, but the responses need not be continuous over time. Final ends valuing involves simple valuing—it involves responding to this and that practically relevant feature of the world—but it involves having simple valuation cohered or unified in specifiable ways. The responses of the valuer with final ends are enduring and directed over time. Local responses are guided by longer-range objectives, and this gives them a larger structure and point. That a good life involves deliberative unity or cohesion is an old thought, and it can be motivated with the following distinction. Consider the difference between something’s being good for someone, and something’s making someone (or their life) good. We might say: “It is, other things being equal, good for me to have this ice cream now. It does not follow that it will make my life, let alone me, better to have it.”19 The analogous claim would be that engaging with objects and activities of value in the manner of a simple valuer can be good for a person, but it cannot make a person’s life go well. For simple valuing lacks the robustness and complexity to be sufficiently affecting. It is compatible with this view that a person 19 The distinction and quotation are from Raz, “Respecting People,” 151–152, fn. 33.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

100 the value of humanity have a self-ideal of being spontaneous as a final end, or engagement with the manifold offerings of the world as they present themselves. What is denied is that the life of someone who pursues ends without a temporally ordered structure will be as well-going. According to the proposal that is under consideration, the capacity to value is that in virtue of which people are of value because its exercise makes people able to lead a good life. The proposal entails that valuing is essential to a good life for human beings. More particularly, it entails that a good life for human beings is a life in which a person has a stable motivation to pursue final ends that she finds worthwhile, that she is appropriately emotionally responsive to, and that play a structuring role in her deliberations over time. I have just made a provisional case that each of the conditions on valuing is indeed a necessary constituent of a good life. But it is worth asking how the resulting picture compares with accounts of the human good that are independently proposed. Joseph Raz has influentially argued that the good life for human beings consists in the successful pursuit of valuable relationships and projects that a person has adopted for herself.20 His proposal is close to the one the proponent of R* is committed to endorsing, but Raz makes two qualifications that strike me as important. The first qualification is that the pursuits in question must be adopted by the agent herself. This qualification finds its point in view of considerations that have to do with autonomy, but also with the fact that not all pursuits are suitable for all people. Which pursuits are suitable for a particular person depends on the nature of her constitution, interests, needs, tastes, loves, experience, sensibility, background, etc. How and what a person values, and the ways those values can be good for her, are variable. This seems right to me. It seems right even when two people share the same end. It took Leonard Cohen a good deal of time to write a song. When Bob Dylan asked him how long it took to write “Hallelujah,” a song Dylan greatly admired, Cohen told him that 20 See Raz, “Role of Well-Being,” Part 1. For a comparable view, see Scanlon, What We Owe, ch. 3.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

4.2 valuing and the good human life 101 it took two years—and he lied because it took him five. When Cohen asked Dylan how long it took to write “I and I,” a song Cohen admired in turn, Dylan’s answer was—fifteen minutes. This, Cohen said, is just the way the cards are dealt.21 The point of the story is not that the one way is better than the other. The point is that we all have own path, our own style, and our own capacity. We cannot live our lives with someone else’s capacity, or in view of the ends they have made part of their life. We can be instructed by others, we might even surrender to their conception of what our ends should be for a time, and that might be an important part of our training and growth. But if ends are to be assimilated into our lives, if they are to fit us given the particularities of our needs and experience, then they need to be made our own. And we are often in a privileged position to make judgments regarding the ends that are suitable to us. On these grounds I would suggest that the ends a person is capable of pursuing are ends that are appropriate to her given her interests, sensibility, and so on. Only the pursuit of ends that are fitting for her will make her life go well, and so only the capacity for fitting ends will serve to ground her value. The second part of Raz’s proposal which calls for some comment is that a person must not only believe that she is engaged in pursuits that are worthwhile—those beliefs must also be true. If a person engages with some object or activity in the false belief that it is worthwhile, then how could that be good for her (intrinsically, and as a rule)?22 This seems right: a valuer must be capable of pursuing final ends that are worthwhile. Raz’s own suggestion is that they need to be worthwhile in the sense that they are good in themselves or period. The proponent of R* will rather say that they need to be worthwhile in the sense that they are in fact good for herself and others. Not all ends will make a person’s life go well, and only the capacity for genuinely beneficial ends will ground her value. A person who in

21 This exchange, and Cohen’s comment about it, is recounted in David Remnick (2016). “Leonard Cohen Makes it Darker”. In: The New Yorker, October 17. My thanks to Wolfgang Mann for recommending Remnick’s article. 22 Raz, “Role of Well-Being,” 272.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

102 the value of humanity fact pursues harmful ends would not thereby relinquish the capacity to pursue beneficial ends, so she would remain a bearer of value on the present proposal. This would be true at the same time as someone who in fact pursues beneficial final ends stands under further normatively relevant descriptions (“is worthy of esteem;” “has good character;” “is a friend;” etc.). The refinement pushes a proponent of R* towards a form of realism about the good for human beings, and I turn to the issue of realism in the next section. I have offered two refinements to the proposal that valuing as it is here understood is necessary for a person to lead a good life. Before I say more about these refinements, let me ask whether valuing as I have analyzed it is also sufficient for a good life. It has been suggested to me that pleasure, playfulness, and a sense of humor make a person’s life go well in ways that go beyond their contribution to the pursuit of final ends.23 If that is right, then we should not conclude that the capacity for having final ends is the basis of the value of people, rather than one among other bases. Let me speak to these examples; I will do so on the assumption that my responses generalize to other kinds of case. Pleasure is rarely something that happens to us passively, as by grace. Pleasure is end-like. Some have taken pleasure to be the only or the ultimate end, though more likely pleasure is among the things we seek. But in seeking pleasure we must seek other things, for pleasure is an emergent feature of our engagement with those things. If pleasure contributes to a good life, it does not do so independently of our pursuit of ends. And there is reason to think the ends will need to be final. For while a moment of pleasure can be good for us, as with ice-cream, it will not make a life go well; for that it will need to be worked into our life with a certain structure and stability. Playfulness is a trait, and so is a sense of humor. Suppose possession of some trait or combination of traits is necessary for a good life. Either we think that traits are—like a prominent brow—fixed parts of our nature, or we think they can be cultivated. In the first case, possession of the trait makes a difference to our life but, as with gifts of fortune, it does not 23 I am grateful to an anonymous reader for this objection.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

4.3 realism about the good for human beings 103 affect our value. It is not among the components of a good life that are contributed by subjects. In the second case, since we can make it a project to cultivate a trait, as we might make it our project to be more spontaneous, it contributes to a good life in the manner of final ends.24

4.3 Realism about the Good for Human Beings I have been developing a particular kind of relational account of the value of human beings. I have provided discussion of the metaphysics of relational value—of what it is for something to be good for someone.25 I have made the case that the form of relational value that is at stake in the explanation of the value of human beings is what is good for a life taken as a whole. Human beings are of value because, as valuers, we are able to be “good for ourselves” in the sense that we are able to lead good lives.26 I have drawn out the conception of the good life that is licensed by the account, and I have argued that, with some important refinements, it is independently plausible. The proposal puts the notion of a good life front and center. In making the case that people are of value because, as valuers, we are able to lead good lives, it supposes that there is such a thing as a good or well-lived life. This is of course an ancient contention, and the picture that emerges is Platonic and Aristotelian in the following sense. The good for a being centrally involves engaging appropriately in the activities that are characteristic of that being. If what is characteristic of people is valuing, then faring well for people involves valuing in the appropriate ways. What remains is to probe my entitlement to the claim that not all ends are good for human beings—that some ends will not make our lives go well. Recall the worry that was left hanging at the end of Chapter 3. How do we know that we are not “good for ourselves” by being self-absorbed, isolated from, or ruthless with others? How do we know that the good life precludes having nefarious ends? Arguably, 24 I am grateful to Andreja Novakovic for discussion. 25 See Chapter 3. 26 On the significance of Aristotle’s conception of the good as the good life, see Vogt, Desiring the Good, ch. 2.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

104 the value of humanity this is the real burden of a function argument. We would all like to know what would allow us to live well, and the classic proposal is to look for an answer in the direction of our nature, and more particularly, our characteristic work or activity. That much may be easy enough to swallow. More difficult is the part where it is to be shown that responding to ethical considerations is among our characteristic activities, so that ethical virtue is involved in living well. How, as I will put it, do we establish a connection between ethical virtue and the beneficial? This is Plato’s question of whether it is advantageous for the just person to be just. As we know Plato’s immoralist contends that it is not—that it is injustice that is to one’s advantage. How do we respond to this challenge? For some Plato’s is the question in ethics—the hard question.27 The characteristic philosophical preference would be to resolve the question conceptually, and some have sought to guarantee a connection between ethical virtue and the beneficial by defining benefit in terms of virtue, so that what it is to fare well just is to behave in morally relevant ways.28 In the absence of this way of securing a conceptual connection between virtue and benefit, the fear is that the questions become empirical, and that here the philosopher has no relevant expertise. To my knowledge, no one has taken up these issues in contemporary discussions more searchingly than Philippa Foot.29 Foot does not wish to shrink from the difficulties that present themselves in this area, and that is a mark of her greatness as a philosopher. She would be the last to boast of anything like a conclusive treatment. Fortunately that would be more than what is needed here. In what follows I appeal to Foot’s discussion of the immoralist to say enough to give the question left hovering in Chapter 3 the direction for an

27 There are those who find the question ill-posed, notably H. A. Prichard (1912). “Does Moral Philosophy Rest on a Mistake?” In: Mind 21.81, pages 21–37. I discuss Prichard’s objection in the Conclusion of this book. 28 For this proposal, see John McDowell (1980). “The Role of Eudaimonia in Aristotle’s Ethics”. In: Essays on Aristotle’s Ethics. Edited by Amelie Oksenberg Rorty. Berkeley: University of California Press, pages 359–376. 29 Foot, Natural Goodness, chs. 6 and 7.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

4.3 realism about the good for human beings 105 answer. I do so in full awareness that the questions deserve more careful treatment.30 Foot responds to the immoralist’s challenge by marshaling the resources of her distinctive form of naturalism. Its key claim is that we make assessments of individuals as being good or defective in light of the life-form or kind to which the individual belongs—in light of what that being is, needs, distinctively does. Foot begins with this normative schema in plants and animals, and she makes a qualified case for its extension to human beings. Here is what matters for the extension to human beings. Ethical virtue in human beings is a way of being or doing that is necessary because and insofar as a human good hangs on it. As Foot adapts an example from G. E. M. Anscombe, keeping promises is a virtue in human beings because human beings need to be able to bind one another by their word. So, similarly, charity is a virtue in human beings because we are vulnerable and cannot manage without help and kindness from others. This way of thinking about the connection between goodness in human beings and the good for human beings provides a “test” for whether something counts as a virtue or not. We used to think that homosexuality and masturbation were vices and now we see that they are not. It is false that a human good hangs on sexual conformity, and false that a human good hangs on not giving oneself an orgasm. By this test charity emerges as a prime candidate for virtue because everyone needs compassion when the going gets tough (as it will). The immoralist, who for Foot is modeled in the figure of Nietzsche, gets the facts wrong about compassion. Nietzsche thought that being compassionate, charitable, and kind was crippling for human beings—that these ways of being make us weak. This is false, Foot contends, given the kind of being we are.

30 I am really grateful to Ned Howells-Whitaker for writing a thoughtful paper that urged me to think more about this topic. I am also indebted to Aaron Abma for identifying the issue of realism as critical, and for underscoring that ultimately one must take a step beyond metaphysics of value and into meta-ethics. I take up these questions at greater length in Nandi Theunissen (2021). “The Excellent and the Beneficial”. In: The Oxford Handbook of Moral Realism. Edited by Paul Bloomfield and David Copp. Oxford: Oxford University Press.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

106 the value of humanity On this way of seeing things, ethical virtue answers to human needs, needs which when unmet have appreciable consequences for us. Virtue forestalls things going badly for us. To put it baldly, and in a way that stands in need of much refinement, clarification, and defense, the basic thought is that the person who is pursuing nefarious ends is behaving badly by the norms that are determined by the kind of being he is, a human being. He is violating norms that find their point in the human good. He is not other than a human being, so these norms also find their point with respect to his good. It is not that we make this claim on the strength of our empirical observations of him— even as we are not surprised when we observe forms of disfunction. We make this claim given our understanding of human beings as a species or kind: given what we know of our needs, constitution, and soul-mind.31 This is also Plato’s eventual view. Plato offers an account of the human soul according to which the unjust person is in a state of psychic disharmony, a model he looks to bear out with vivid portraits of tyrants in various stages of conflict and disorder. Moral philosophers seldom know what to do with richly descriptive accounts of people and their lives. The call, in some quarters, for greater attention to literature in ethics may be well taken. Foot herself is careful to say that in this domain philosophy cannot claim a privileged voice: “facts about human life are in question and so no philosopher has a special right to speak.”32 Indeed, the questions here are also investigated by the non-fundamental sciences, and they are studied in a systematic way by highly sophisticated traditions in the East. According to Foot, then, or the central line of thought that I am drawing from her discussion, virtue and benefit are systematically related but distinct. The notions are distinct because we have an independent grasp of them. We have a picture of the virtues: of benevolence, justice, courage, and moderation. We have a picture of human deprivation, of needs that are unmet, and of things going

31 Cf. Foot’s claims about what the facts do and do not entitle us to say. Foot, Natural Goodness, 91. 32 Foot, Natural Goodness, 108.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

4.3 realism about the good for human beings 107 badly for us. But the two notions are systematically related because we explicate virtue by appealing to what human beings need in order to get along and so to what is beneficial and harmful for us. The systematic connection between them shows up in ways we talk—it shows up, unsurprisingly, in our concepts. So we speak of “good ways for human beings to be” in a manner that is interestingly equivocal between virtue and benefit. “The human good,” too, can mean both goodness in human beings, and goodness for human beings. For Foot the concept of benefit also takes the concepts of virtue and felicity together. In these ways we can speak of a conceptual connection between them. But the argument for that connection is not an argument by stipulation. That is why it is felt by Foot to be more satisfying than the claim that there is a way of thinking of the human good such that wickedness and happiness are a priori ruled out. The argument depends on claims about who and what we are, so that Foot’s approach is both conceptual and empirical. What emerges is a form of realism, and naturalism, according to which there are attitude-independent facts of the matter about the good for people. Foot is alive to the sense in which objectivism about the human good runs up against modern sensibilities, and in particular, the forms of subjectivism that emerged over the twentieth century in the wake of a felt collapse of Moorean non-naturalism. Foot makes few concessions to the spirit of subjectivism. If I were to make a concession it would be in the direction of pluralism. The account of valuing that I have offered deliberately leaves the character of ends open enough to include relationships, vocations, practices, activities, states, self-ideals, and so on. I include a diverse range of options because I think there are many ways for people to lead good lives. This form of pluralism about the good for people is still a form of realism. There is something to discover about our own good in view of the character of our ends and the nature of our characters, and we can be mistaken about both, and about the relationship between them. We can come to discover that what we valued was not quite what we thought it was, or that we do not have the motivation for it, or that it just doesn’t fit us right. Stories of this form of discovery are

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

108 the value of humanity among the oldest stories we have. On the other hand, while differences between us are important, they should not be exaggerated. We are all, alike, people—more human than not. It is here that I would join Foot in resuscitating the ancient contention that some ends are to no one’s benefit—the ends of injustice, and tyranny, and oppression, and discrimination, and others like them.

4.4 Taking Stock I have introduced an account of the value of humanity. According to the account, grounding consists of two elements: (a) the value of humanity depends on possession of a base property (the capacity for final ends); (b) the base property makes people of value by making people able to lead good lives. I have not given independent defense of (a) on the assumption that establishing (b) goes a long way towards establishing (a); (b) is the center of the argument. Some readers will find the explanation unexpected, even strikingly so. The explanation is that the capacity for having final ends explains the value of a person because having it is good for that person, and it is good for that person because its exercise enables that person to have a good life (a life that is good for them). People are of value because they can be good for something—they can lead good lives. The advantages of the account are two-fold: good for is free from presuppositions involved with the idea of something being good in itself ; there is a straightforward account of how possession of the capacity for having final ends is good for people through its relations to the good life. The discussion has identified some refinements to the account. First, a person must be capable of pursuing final ends that are appropriate to her given her interests, sensibility, and so on. Not all final ends are final ends for all people. Second, a person must be capable of pursuing beneficial final ends. To be sure, a person who in fact pursues harmful ends does not thereby relinquish the capacity to pursue beneficial ends, so she remains a bearer of value. Let me now make contact with questions raised in Chapter 1 about the scope of the account. In what sense is the account given here

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

4.4 taking stock 109 distinctive of human beings? Could the account be extended to other animals? For Kantians the value of human beings is metaphysically distinctive: it is unlike the value of what is good because it can be good for someone. For the proponent of R,* the value of human beings is metaphysically continuous with what is good because it can be good for someone. Whatever is distinctive about human beings would be captured by the special way we can be good for something: namely, our lives. Our life, as something we partly make, and as something that partly happens to us, is not identical with our biological life: that is, with our being and remaining alive. So some will say that under certain conditions, paradoxically as it may sound, it is better for us— better for our life—to be dead. How our life goes is something we are able to take as an object; it may be an object of regret, satisfaction, frustration, or hope. We might say that it has a narrative structure.33 But it is likely that certain animals bear a comparable relationship to their lives. Along with many others, I have enjoyed learning about how particular species of animals, for example elephants, are practically and emotionally responsive to how things turn out for them and their kin in time.34 I regard it as a strength of the account that the practical relevance of human beings is also not different in kind from that of other animals. Likely many animals are not final ends valuers, but valuers in the sense that they are able to respond to practically relevant features of the world as they present themselves; they are capable of intentional action. Plausibly, this makes many animals able to be good for themselves, not in the sense that their life as a whole is something they can take as an object—something with a narrative structure, something they can take pride in, or feel frustrated by—but in the sense that they can contribute to their remaining alive. To that extent,

33 See David Velleman (2003). “Narrative Explanation”. In: The Philosophical Review 112.1, pages 1–25. 34 I am grateful to Michele Theunissen for sending me an article from a local South African newspaper reporting that two herds of African elephants came to mourn the passing of Lawrence Anthony, their long-time friend and protector. On Anthony’s work with elephants, see Lawrence Anthony with Graham Spence (2009). The Elephant Whisperer. London: Pan Books.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

110 the value of humanity the value of many animals turns on being good for themselves in an extended sense. Does the account extend to all human beings? Are there some human beings for whom it does not provide guidance? I am theorizing the value of human beings in terms of a capacity to value. I am assuming that the value of a capacity depends on the value of its exercise so that a capacity is valuable only if there is a live possibility that it can be exercised.35 Someone who is in an irreversible coma, or someone with a profound disability like anencephaly, does not have the capacity in question, and my account does not provide guidance in their case. As I argued in Chapter 1, even as I find it important to say, with Jeremy Waldron, that all human beings bear an important relationship to the capacity in question, including a modal relationship, I am doubtful that any value-bearing feature is shared by all and only human beings. As I also argued in Chapter 1, this does not mean that people who lack the relevant capacity fall out of ethical consideration. I am not treating the value of humanity as marking the limits of moral concern or status. Instead, I contend that we treat all bearers of value as the bearers of value they are, and that not to do so is to fail to be attentive to their situation in the world. While my account does not provide guidance for human beings without exception, I would argue that it captures what is typical and characteristic of human beings. Life-long advocates for people with conditions such as Down syndrome emphasize the importance of recognizing their capacity for having final ends in the sense that is in question, including, centrally, stable reciprocal friendships and community.36 My account provides clear guidance in their case, and many others like them. It provides guidance for someone who, because she is depressed or for some other reason, neglects her capacity to value. There is certainly a live possibility that a person in this situation can exercise her capacity to value, and treatment is so important

35 Raz, “Role of Wellbeing,” 290–291. 36 This is central to Jean Vanier’s L’ Arche project for which he was awarded the Templeton Prize in 2015.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

4.4 taking stock 111 because the exercise of this capacity is so important for beings like us.37 In sum, the account proposed here is typical of human beings, it likely extends to some other animals, and it does not provide guidance in the case of some human beings, even as these cases are rare.

37 See Langton’s objection to Korsgaard’s reflexive proposal, and her discussion of the case of Maria Von Herbert, in “Objective and Unconditioned Value,” §6.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

5 The Normative Significance of Human Beings The full proposal is that human beings are of value because, as valuers, we bear a certain kind of relation to ourselves: we can be of benefit to ourselves. In developing this proposal I have responded to broadly speaking Kantian arguments concerning relational value. I examined an argument schema according to which for anything to be relationally valuable something must be non-relationally valuable, and I gave my reasons for rejecting the entailment. It remains to consider challenges from the other principal tradition in value theory—the tradition inaugurated by G. E. Moore. The Mooreans raise a rather foundational objection to the conception of value that lies at the center of the present proposal. They express puzzlement over just what relational value is meant to be, and more searchingly, they put pressure on the reason-giving force of relational value. Their challenges confront any theorist who invokes ideas of what is good for someone. In my case, the crucial question is how a relational account of the value of human beings is able to generate reasons to respond to people that are reasons for everyone: that is, for all valuers. If I am claiming that each one of us is “good for ourselves,” it may be clear enough how we each have reasons in regard to ourselves, but what reasons do others have in regard to us? And what reason do we have in regard to others? By engaging with the Mooreans I show how the present proposal licenses core forms of ethical behavior that are owed to everyone, and this is the

The Value of Humanity. L. Nandi Theunissen, Oxford University Press (2019). © L. Nandi Theunissen. DOI: 10.1093/oso/9780198832645.001.0001

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

5.1 the moorean challenge 113 remaining desideratum for the account. I conclude with a normative ethical schema for how we should respond to ourselves and others.

5.1 The Moorean Challenge The classic Moorean challenge to relational value is fully independent of questions about the value of humanity; indeed, Mooreans do not typically regard people as bearers of value at all.1 The Moorean challenge, in the first place, is that there is no such thing as being good for someone.2 To say of a thing that it is “good for someone” is to say that that the thing that occurs in someone’s life is good simpliciter; or it is to say that the fact that someone has a good in their possession is good in itself; or it is to say that someone thinks that possessing some thing is simply good. On none of these constructions do we find a genuinely relational kind of value. There is only non-relational value and relations of location, possession, or in the epistemic case, belief. This is a metaphysical point, but for Moore it has practical implications that he draws out in a discussion (and rejection) of egoism. The egoist appeals to an idea of relational value as part of an argument for the conclusion that each one of us has reason to promote what is “good for us” and no reason to promote what is “good for others.” Moore responds as follows. Since the only way for something to be of value is for it to be good simpliciter, and since there is a deep connection between value and reasons, what is of value generates reasons that are reasons for everyone. So egoism rests on a mistake about the nature of value. These lines of argument are what I now look into.

1 See Donald H. Regan (2002). “The Value of Rational Nature”. In: Ethics 112.2, pages 267–291. 2 See G. E. Moore (1903). Principia Ethica. Cambridge: Cambridge University Press, §59. For a contemporary statement of the argument, see Donald H. Regan (2004). “Why am I my Brother’s Keeper?” In: Reason and Value: Themes from the Moral Philosophy of Joseph Raz. Oxford: Oxford University Press, pages 202–230.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

114 the value of humanity

5.2 Being Good for P and Being Good and Occurring in P’s Life Start with the first part of Moore’s argument: the claim that there is no such thing as being good for someone, only being good simpliciter and occurring in someone’s life (and variations on that theme).3 So far, when I have said of some X that it is “good for someone” P, I have treated being good for as a relation in which P is enriched, or benefitted, or made better by, X, where the value is the being enriched, the being benefitted, etc. The question now is how the metaphysical structure of value, so understood, compares with that of what is good simpliciter and occurs in someone’s life. I join with others in suggesting that the structure is quite different. Here is Connie Rosati: The relational complex, X is good for P, does not include the monadic property good at all. Instead, it includes the relational property is good for P: it has X and P as relata and is good for as a dyadic relation.

So the logical form of X is good for P is not: (X is good) for P but, rather, xGp (using ‘G’ to express the relation is good for).4

Rosati argues that being good for someone is a dyadic form of value that is not built out of the monadic good, and the difference Rosati identifies appears to be borne out by examples. Penny attends an exquisite performance of Julius Eastman’s “Femenine,” but her mind wanders and she is not engaged. If it makes sense to say of something that it is good simpliciter and occurs in someone’s life, then this would be an example. But it is not an example of something’s being good for someone. For we have lost the sense in which there is a relation of fit 3 For reasons given in section 3.2, “Devoid of value or harmful,” I am setting aside the epistemic reading of good for someone. 4 See Connie Rosati (2008). “Objectivism and Relational Good”. In: Social Philosophy and Policy Foundation 25, pages 314–349, p. 329, including fn. 43.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

5.3 being good for p and being practically 115 between the object and the subject—an encounter in which the subject is improved, or enriched, or benefitted, or enhanced, or profited, by the object in question. This is the relation that has been treated up to now, and the discussion is enough to bring out the difference from a non-relational good that is located somewhere.5

5.3 Being Good for P and Being Practically Relevant to S Now consider the second part of Moore’s argument, his refutation of egoism. To my mind it contains the deeper challenge to proponents of relational value. If there is no such thing as relational value, and what is good is always non-relationally so, and there is a connection between values and reasons, then what is good generates reasons that are reasons for everyone—for all people. I have given grounds for rejecting the antecedent. Does it follow that relational values generate reasons that are only agent-relative (reasons only for some people)? To put the question in another way, if something is good for Penny, it seems clear enough how it generates reasons for her, but why should it generate reasons for anyone else? What reason does Sam have to care about what is good for Penny? In general, how does a value that bears a relation to one person make a claim on anyone else?6 The question is of some importance for my argument since I am proposing that people are relationally valuable even as they have a fully general practical

5 I set this part of Moores’s objection aside rather quickly because it has been dealt with successfully elsewhere. For further discussion, from which I have learned a great deal, and for further examples, cf. Rosati, “Objectivism and Relational Good,” §III. A. Cf. also Raz, “The Role of Well-Being,” 274. 6 My formulation of the challenge is drawn from Regan, “Brother’s Keeper.” In Regan’s restatement of Moore’s argument it emerges as the second horn of a dilemma: either the relational good for is no different from the non-relational good—so that being good for someone is being good simpliciter and occurring in their life; or the notion is different but it lacks normative force. In Regan’s example, if the good for Abel is peculiarly Abel’s, then why should Cain care? We think there is a deep connection between values and reasons, so if the good is relative to Abel, it would seem to create only Abel-relative reasons.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

116 the value of humanity significance—or so I would like to be able to claim. I investigate this question first by considering the reason-giving force of an object or activity X that is good for some P, and then by considering human beings as the substitution instances of X in the section to follow. Some may flatly deny that one person has any reasons in regard to what is good for another independently of its connection to their good. This need not be claimed as part of a crudely egoistic position. In Chapter 4 I canvassed lines of though to suggest that living a good life involves being practically responsive to what is good for others. If it is clear how my own good is practically relevant to me, and we accept that our good is interdependent, then there is no problem theorizing the ground of my responsiveness to you. I do not doubt that there are resources within this style of response, but I think it is possible to show how the good for one person has basic practical relevance for others independently of its connection to their own good. More generally, I propose to accept the Moorean assumption— an externalist assumption—that for a person to have a reason, the reason need not be relevantly connected to something in that person’s “subjective motivational set.”7 It should be clear how X’s being good for P has basic practical relevance for S (and others) independently of its connection to S’s good, and independently of S’s antecedent motivational states. I will come back to the topic of externalism later in the chapter. A solution that immediately suggests itself is to urge that X’s being good for P makes a claim on S because P is a bearer of non-relational value. According to this line of thought, the non-relationality of the value of people ensures that the relational good for an individual generates reasons for others.8 The solution may not satisfy the Moorean for whom people are not non-relationally valuable; but more importantly for present purposes, the solution is unavailable to a proponent of R.* A different, if comparable, solution looks to the object side of the relation between an object and a person. According to this style 7 This terminology is due to Williams, “Internal and External Reasons.” 8 This is the approach taken by Rosati. See Chapter 3, fn. 29.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

5.3 being good for p and being practically 117 of proposal, which is perfectionist after a fashion, what explains why something can be non-instrumentally good for a person is that it is good simpliciter. The view allows that being good simpliciter is not a sufficient condition on something’s being good for a person—not all that is good simpliciter will be good for all people. To be good for a person it must also fit into that person’s life, fit with their aptitudes and so on. But being good simpliciter is a necessary condition on something’s being good for anyone. Something can be good for a person because it is good simpliciter, and it is that very quality, being good simpliciter, that can make it good for others too.9 This way of understanding relational value supplies a response to the worry about normativity by deploying a Moorean premise. Since whatever is non-instrumentally good for an individual is good simpliciter, and since good simpliciter unproblematically generates normative reasons for everyone, whatever is good for one person has general practical relevance for others. To the extent that it rests on an appeal to non-relational value, this solution is also unavailable to a proponent of R.*10 My own proposal is that what is genuinely good for an individual makes a claim on others because it is good for that individual as a human being—it is an instance of something that is good for human beings—and being good for human beings explains its status as valuable, and thereby, its normative significance to human beings. The proposal draws on a set of premises that have been at play in this book as a whole. These are commitments to relational value theory, realism, 9 This view is defended by Raz, “The Role of Well-Being,” 274–275. Compare Thomson, “GOODNESS,” 469: “I’m sure that friends of GOODNESS [non-relational value] would say that both of the following are true: (1) The fact that our taking a holiday would be good for us is a reason for us to take a holiday, and (2) The fact that our taking a holiday would be GOOD is a reason for us to take a holiday. But I’m sure they’d say that (2) is fundamental, thus that (1) is true only because (3) If our taking a holiday would be good for us, then it would be GOOD, and (2) is true.” See also Roger Crisp (2013). “In Defense of Absolute Goodness”. In: Philosophy and Phenomenological Research LXXXVII.2, pages 476–482, p. 477 on non-eliminativist versions of welfarism. 10 I critique this model of relational value in “The New Mooreans: On Personal and Impersonal Good,” manuscript.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

118 the value of humanity and pluralism. Let me say a word about each of these commitments in turn. According to relational value theory, good is short for good for someone. Acknowledging that there are subjects for whom things are good who are not human beings, the subjects at issue in this study are human beings. I am concerned with what is good for human beings in the sense that it is good for their lives. In this way I am working with a humanist conception of value according to which it is sufficient for X to be good or of value that X contributes to human life and its quality. According to relational value theory, value properties are relational, and according to realism they are objective. Together, relational realism proposes that there are facts about what sorts of things are good for human beings, and facts about how we human beings can affect our good through our actions and activities. Claims about these matters are true to the extent they get the facts right. This schema for relational realism may be filled out in a number of ways, with different conceptions of objectivity, and different procedures for determining truth.11 I have proposed that facts about what is good for human beings are grounded in the nature of those things and the nature of human beings. I have accepted the externalist supposition that value properties give rise to reasons independently of their connection to our antecedent states, so that reasons are not derived from pre-existing motives. I have suggested that we discover truths about what is good for human beings through investigation of our nature and our world, allowing that this investigation is both conceptual and empirical in a broad sense. This proposal is compatible with thinking that there can be indeterminacy about whether something is good for us, and it is compatible with thinking that whether something is good for us may be a function of what we do, so that there is a sense in 11 In his classic paper, Peter Railton gives various dimensions along which a realist position in ethics is articulated. See Railton, “Moral Realism,” §1. On the basic commitments of the realist, see also Geoff Sayre-McCord (Fall 2017 Edition). “Moral Realism”. In: The Stanford Encyclopedia of Philosophy Edward N. Zalta (ed.) URL = . Forms of relational realism are defended by Railton, “Moral Realism;” Kraut, What is Good and Why and Against Absolute Goodness, and Vogt, Desiring the Good, ch. 4.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

5.3 being good for p and being practically 119 which we can make something good for ourselves.12 But what matters for the purposes of this argument is the schema of relational realism rather than any particular elaboration of it. Pluralism qualifies realism about the good for human beings. According to pluralism, there are real differences in what is good for each of us, so that what is good for one person may not be something that suits another. This is owing to differences in our constitutions, experiences, aspirations, aptitudes, and so on. For some object or activity to be good for an individual— for the relation of benefit to be instantiated—it must fit that individual given their particularities. Taken together, these commitments supply a response to the worry over the normativity of relational value, a response that I will unpack starting with relational value theory. The form of relational value theory that I endorse says that if something is genuinely good for human beings—for human life and its quality—then it is of value. So we can speak of goods or values—human goods or human values. As I have said from the outset, I am not an eliminativist about the property good or the property of being of value. I have set aside eliminativist views according to which there is no such thing as being of value, only being good for someone, or being good in a way.13 I think there is a perfectly clear sense in which something can be “good” or “of value,” and not because it is good simpliciter. Something is of value not because it possesses the property good (simpliciter), but because it possesses some other set of practically relevant properties. To that extent, being valuable is a second-order property: the property of having other properties that are practically relevant. What explains why those properties are practically relevant is that they are or can be good for human beings. To be of value is to possess reason-giving properties, where the properties are reason-giving because they bear the relevant kind of relation to human beings.14 This is the conception 12 For insightful discussion of the idea of “making something our good,” see Rosati, “Personal Good.” 13 See Thomson, Normativity, ch. 1. 14 The proposal is compatible with a “buck-passing” account of value according to which the normative burden is passed from goodness period to some set of properties.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

120 the value of humanity of the ground of value that has been developed in this work. According to it, value is both object and subject related: it involves a story about the basis of value that turns on the intrinsic and extrinsic properties of things, and a story about the explanation of the value of the basis that turns on the relation of being good for human beings.15 Realism fills out what is so-far a value-theoretic point. According to the form of realism that is at issue here, there are normative relational facts: facts about relations of fit between features of a thing and human beings such that the thing is good for human beings. As I said above, this schema for relational realism is compatible with different accounts of objectivity and procedures for determining truth. So Thomas Nagel has proposed that we discover truths about what is good for human beings by critically assessing what presents itself as good for us from an internal standpoint. We improve our understanding by detaching ourselves from a perspective that is internal to our own life and by seeking a more enlarged and impersonal view of things. We ask not, is this good for me, but is this good for this person? Does its appearance as good detach from the thought that it is mine? If it does then it is a candidate for being a human good. By this method, simple pleasure and the avoidance of pain emerge as basic goods for human beings, and Nagel adds freedom, self-respect, access to opportunities, and the basic resources of life.16 My focus has been on how we human beings can affect our good through our actions and activities, and I have drawn on the Aristotelian and Platonic traditions which make determinations about this by examining our nature. See Scanlon, What We Owe, 96 and passim. Where the proposal differs from a buckpassing account is in requiring a further explanation of how it is that those properties make the object practically relevant. In my view, the further explanation is supplied by the relation of being good for human beings. This is to divide up Scanlon’s buckpassing account into two independent proposals: (i) that the normative burden is passed from goodness to some other reason-giving property, e.g. pleasure; (ii) that the notion of a reason is explanatorily fundamental. Put this way, I accept (i) and deny (ii). 15 I take up these questions about grounding at greater length in “The New Mooreans: On Personal and Impersonal Good,” manuscript. 16 Nagel, View from Nowhere, chs. VIII and IX.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

5.3 being good for p and being practically 121 If human beings are valuers, then what is good for people is to value in appropriate ways. Whatever the precise form of relational realism, the schema says that there are truths about what satisfies the relation of being good for human beings that are independent of what someone may happen to think. Contrast an anti-realist view according to which there is no right answer to the question of what is good for a person beyond a subject’s own judgment. According to this style of proposal, a life devoted to trivial, demeaning, or even pernicious pursuits could be good for a person provided the subject judges it to be so, or takes satisfaction in it, and the subject is not pursuing these ends under duress or in ignorance of what they are doing.17 This way of conceiving of the good for a person does not explain why what is good for a person is thereby of value. Accordingly, it does not explain why what is good for one person makes a claim on others. Plausibly, I have no reason to support or in some other way respond positively to a person’s pursuit of trivial, demeaning, or pernicious ends because they are satisfied by them. In that sort of case, a person’s conception of what is good for them is sunk in privacy. It is not something that other people can think their way into or come to share. To borrow Nagel’s idiom, it does not detach from an internal or merely personal viewpoint. Pluralism anticipates differences in what is good for individuals— it self-consciously rejects the idea of a thoroughly uniform human nature.18 But a pluralistic form of realism holds that what is genuinely good for an individual is good for that individual as a human being, so that what is good for an individual is a form of the human good. What is good for a particular person, for an individual, specifies the human good in a concrete way. This is shown in the readiness with which we can describe what is good for an individual as the species of a genus. Reading experimental non-fiction is part of what enriches Penny’s life, and the life of readers with sensibilities like hers. Reading that kind of 17 For this style of proposal, see Wayne Sumner (1996). Welfare, Happiness, and Ethics. Oxford: Clarendon Press, pages 153–171 Compare Thomson, “The Right and the Good,” §VIII. Sumner’s proposal is a central target of Regan, “Brother’s Keeper,” 213–215. 18 Cf. McDowell, “Role of Eudaimonia,” 15.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

122 the value of humanity literature will not play this role for people with no taste for that sort of thing. But engaging with that sort of thing is readily described as a form of cultural activity—a genus that is good for human beings with many specific forms.19 So, similarly, insulin is good for this person to ingest because it is constituted in such a way that it is good for diabetics, and this person falls under that description. But insulin is good for diabetics as a form of medicine—something that is good for human beings. It is here that I would register disagreement with Nagel’s treatment of a classic example that has now taken on a life of its own. This is the example of someone’s end of climbing Mount Kilimanjaro.20 For Nagel it is important to distinguish something whose value for a person depends on that individual’s perspective and plans (“personal values” in Nagel’s terminology), from something whose value for a person is independent of that individual’s perspective and plans (“impersonal values” in Nagel’s terminology). Goods of the latter kind generate reasons that are reasons for everyone, and Nagel’s paradigm examples are simple pleasures and the absence of pains. On this way of setting things up, whether a value that bears a relation to one person makes a claim on anyone else depends on whether the value is impersonal or personal. Though he is sometimes interpreted differently, I read Nagel as treating both personal and impersonal values in his sense as forms of relational value. Nagel makes claims to the effect that pain is bad no matter whose pain it is, but I take it he is not denying that pain is bad because it is bad for someone. His question is whether we can think our way into something’s being bad for someone even as we detach from the perspective of the particular person who regards it that way. As Nagel understands a person’s end of climbing Mount Kilimanjaro, it is a merely personal value, and

19 The point should not be misunderstood so that anything that can be described as “cultural activity” is good for human beings. There are egregious forms of culture, ranging from art forms that are not worthwhile for people to engage with, to practices that are laced with prejudice. 20 Nagel, View from Nowhere, 167.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

5.3 being good for p and being practically 123 accordingly Nagel denies that it generates reasons for others. Barring particular ties to the person who has the end in question, no one has reason to help them realize it, or to care whether they do. This is not because Nagel thinks the end lacks intelligibility or depends on illusions—like the proverbial project of counting blades of grass. It is because the value of the end does not detach from a merely personal perspective. This is where I take a different view. Climbing mountains may not be something we ourselves do (it may not be one of our ends). But we can recognize it as a form of the human good. Presumably climbing Mount Kilimanjaro is an activity that involves the skillful navigation of difficult terrain. It is an activity people undertake in communities, or on their own, so that it involves relations of trust with others, or feats of self-reliance. It requires knowledge of a particular environment, its climate, animals, and species of plant. It holds opportunities for engaging with another culture, language and so on. It is the kind of thing one needs to train for over time. In these and other ways, it is a very good candidate for what I have called a final end. I have proposed that the pursuit of final ends is the primary way human beings can affect the quality of their lives. In that case, its value for an agent perfectly well detaches from their peculiar perspective. A person’s end of climbing Mount Kilimanjaro should be counted as an impersonal form of relational value in Nagel’s sense, and it should be counted as something that generates reasons for others. What sort of claim should a person’s end be thought to make on another? What reason does Penny’s end of climbing Mount Kilimanjaro make on Sam? Nagel denies that it makes any sort of claim. Even barring the identification of the end as a form of the human good, this is implausible. Surely Sam has reason, all things being equal, not to prevent Penny from taking the trip, and reason not to belittle her aspiration to go. He has reason not to sully the route, reason not to blow up the mountain, reason not to petition the company who leads the expedition. These are reasons for action and for the formation and expression of various cognitive attitudes. They are sometimes called reasons of non-destruction, non-interference,

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

124 the value of humanity preservation, and protection.21 But given the identification of this end as a form of the human good, I would urge that it merits more besides. As Nagel treats impersonal goods for people, they give everyone reason to support, or further, the good in question. Supposing that is true, everyone has reason to support Penny’s end of climbing Mount Kilimanjaro, and to care whether she does. Of course, just as one cannot attend to everyone’s access to the basic resources of life, or to everyone’s pains, so we cannot support the ends of every person in the world. Where we direct our efforts is a familiar issue regarding what Kant called imperfect duty. But as Kant will make the point, they are still duties, or in the contemporary idiom, things we have reason to do. In sum, my response to the problem about the normativity of relational value is that what is genuinely good for an individual is an instance of what is good for human beings, and being good for human beings explains why something is of value, and for that reason, generally practically significant—normatively significant for human beings. In Chapter 4 I endorsed a pluralistic form of realism about the good for human beings. The realism side of the proposal says that there are facts about what is good for human beings. What is good for a person is not a private matter. If it were then it would be hard to understand how what is good for one person makes a claim on someone else. If it were just that this thing satisfies that person’s preferences or gets them whatever they happen to want then it would be hard to see why others should care. The pluralist side of the proposal says that, while there is such a thing as the good for human beings, what is good for a particular person, for an individual, specifies the human good in a concrete way given a host of particular considerations. The pluralism foregrounds differences between what is good for you and what is good for me. Together with the realism, however, the proposal is that despite these differences, the goods are all, alike, forms of the human good. As I understand the force of the problem, an explanation of the value of something in terms of its being 21 For discussion, see Raz, “Respecting People,” §6 and 7.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

5.3 being good for p and being practically 125 good for an individual must show how the thing in question is really of value, and for that reason, reason-giving for others. If what is good for an individual is an instance of what is good for human beings, then, according to a humanist form of relational value theory, we can understand how it is genuinely of value. While my solution to the problem rests on a substantive conception of the human good, there is another sense in which it is deflationary. I am assuming that once it is shown that what is genuinely good for an individual is of value, there is no further question about how it is possible to act in light of our recognition of this fact. For many philosophers, questions about the action-guidingness of ethical claims are thought to be almost insuperable. For Kant the question of how we can act out of a recognition that something is rationally required of us—of how a categorical imperative is possible—is the philosopher’s stone: the deepest and most perplexing problem there is. As contemporary theorists have come to formulate the problem, it is how there can be external reasons for action: reasons that are not grounded in an agent’s antecedent motivational states. There is an interesting phenomenon in philosophy of making, or at least feeling that one has made, a discovery—something that renders soluble what previously appeared to be insoluble. It is hard not to hear this sense of discovery on the part of a generation of prominent normative realists who saw the felt difficulty about the possibility of practical reason as resting on a problematic assumption in the theory of action. In their diagnosis, the problematic assumption is that what most basically moves us to act are our desires, understood as brutely noncognitive states. Against this way of conceiving of our motivations, these theorists variously defended the view that desires standardly represent their objects in a favorable light, so that the typical case of action involves beliefs about reasons or values. If that is what action standardly involves, then there is no difficulty—or not the philosophical difficulty there is thought to be—of understanding how normative claims can have practical implications for us—can get a grip on our motivations. Normative claims can get a grip on us by our coming to believe that they are true. Beliefs are enough

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

126 the value of humanity to set our limbs in motion.22 In intellectual life, as we know, it is just when a discovery rises to the level of a consensus position that there begin to be rumbles of dissent, and the tide looks to turn again. This may be the situation for externalism about reasons as I write these words. I will not speak to any sources of discontent here. Partly for dialectical reasons (my interlocutors here are Mooreans), I am relying on the defenders of externalism to maintain that once it is shown that something is of value, there is no further problem of how its value can have practical relevance for people. I am assuming that recognition of value is sufficient for action.

5.4 Our Responses to Human Beings The worry concerned the normative significance of X when X is good for a person, P. If X is good for P, then it is clear enough how P has reasons in regard to X, but not so clear how X generates reasons for S, or anyone else. The solution to the problem is that X makes a claim on S because, when good for P, X is an instance of something that is good for human beings, and this explains X’s status as valuable, and therewith, practically relevant to human beings. How does the solution work when a person is the substitution instance of X? This is the question I turn to now. A proponent of R* claims that Penny is of value because, as a valuer, she can be of benefit to herself, and in that case the question is how Penny generates reasons for anyone else.23 The question arises because it can look as if R* gives an account of the value of each person herself, but not to others. So it can seem as if there needs to be a further story to take us from value in our own case to value in the case of others. But the discussion serves to bring out that being good 22 A representative sample of the theorists I am referring to are Scanlon, What We Owe To Each Other, ch. 1; Nagel, Possibility of Altruism, Part Two; Raz, Engaging Reason, ch. 2; and Foot, Natural Goodness, ch. 4. 23 A proponent of C2 claims that Penny is of value because she can be of benefit to Sam, and in that case the question is how Penny generates reasons for anyone other than Sam.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

5.4 our responses to human beings 127 for ourselves in virtue of the capacity to value is an explanation of our status as valuable, and for that reason, as of general practical significance. One way to make the point vivid is to underscore the sense in which Penny’s being good for herself is not a private matter. The very features that enable her to be good for herself—the complex cognitive, affective, and behavioral disposition that is constitutive of valuing—permit her to stand in community with others and be good for them. In being the best teacher she is able to be, Penny is most able to show up for her students. Being good for oneself is not a private form of goodness that lacks relevance for others. This is a symptom of the fact that valuing, and thereby living a good life, is an instance of something that is good for human beings. From the outset I have been working with a relational conception of value according to which whatever is of value is so because it is or can be good for human beings (or more generally, for beings). The big claim of this book is that the value of human beings is not different in kind from the value of other things. Other things are of value because they can be good for human beings, and human beings are of value because we can be good for human beings. Most basically, we can be good for ourselves. How does being good for ourselves explain our status as valuable, and thereby our relevance to others? The answer is that in having a capacity to value, and thereby, in being able to affect our lives, we are an interesting sort of instance of something that is good for human beings. Trees are good for human beings, and this is part of why they are valuable (it is not a complete explanation for they are good for things and beings apart from human beings). But we human beings can be good for ourselves by way of our agency and activity. To make a broadly Millian point about the importance of the active over the passive, we are able to contribute to our lives going well in ways that other things and other people cannot (even as, of course, we contribute to our lives through valuing other things and other people, and so on). As a valuer, a human being is a special instantiation of something that is of value because it can be good for human beings, and what is good for human beings generates reasons for human beings.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

128 the value of humanity What is the form of practical response that is owed to human beings? A full account of the value of human beings should make clear, not just that responses are owed, but the character of those responses themselves. At least, an account should provide a schema for the basic forms of ethical consideration that are owed to human beings. Fortunately, this much follows straightforwardly from what has been said. All valuable things should be responded to—acknowledged, treated, considered—as the valuable things they are. If people are of value because, as valuers, we are able to lead good lives, then we should respond to people as the value-bearers they are.24 In a word, we should respond to human beings as centers of a life to which they bear a special relation. Start with what each one of us owes to ourselves. The relationship a person bears to herself has been a recurrent theme of this book, and I have joined with the ancients, and with Kant, in treating it as a part of ethics. It is one-sided to restrict the purview of moral philosophy to what we owe to other things and other beings. It is also unwise. For if we fail to attend to ourselves in the sense that is under discussion, we are limited in our ability to show up for others. I accept the proposal canvassed above that the minimal or basic form of response to whatever is of value is one of non-destruction, and protection. In that case, each one of us has reason not to destroy, and to protect, our capacity to value. I have proposed that the value of our capacity to value is dependent on the value of its exercise, for it is the exercise of this capacity that contributes to our life going well, and this is what explains its value. If we have reason to safe-guard our capacity in view of the value of its exercise, then not only should we not damage 24 Those who contend that our ways of responding to values can come apart from the way those things are of value are not offering a normative account of our responsiveness to values. As Robert Nozick puts it: “the moral basis [of human beings] must be a characteristic that is relevant to certain behavior by others, so that either we can specify the ways in which that behavior by others is appropriate to the (bearer of) that characteristic, or at least we can ‘see’ that the behavior is a fitting response to the characteristic.” Nozick, Philosophical Explanations, 452. As Jospeh Raz argues: “the value of what has value, and the action it is a reason for, are intrinsically connected.” See Raz, “Respecting People,” 164.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

5.4 our responses to human beings 129 or interfere with our capacity to value, but, more positively, we should exercise that capacity. I have given an account of what this involves— of what it is to value. According to this account, we value in a way that is relevant to our living well when we pursue higher-level or final ends that are not harmful to ourselves or others, and that fit us given our individual sensibilities. I have given an analysis of the cognitive, practical, and affective states that are involved in final ends valuing. Three features of the analysis provide practical directives (once they are in place the remaining conditions take care of themselves). The first two have already been mentioned. We should not pursue ends that are harmful to ourselves and others, and we should pursue ends that are in harmony with, to borrow a metaphor from Iris Murdoch, the texture of our being.25 It may be that we take advice about how and what to value for a time, and this may be an inevitable part of human upbringing. But since our ends need to be assimilated into our lives, since they need to fit us given the particularities of our nature, taste, and experience, then they need to be made our own. We are often in a privileged position to make judgments regarding the ends that are suitable to us, and we should exercise this judgment. Third, and finally, we should take care to pursue ends in view of the fact that our existence is ordered in time. We should pursue ends that provide structure and direction for our more particular actions. There is a parallel account for what we owe to others, with one qualification. We have reason not to destroy, and more positively, to protect, other people’s capacity to value, and therewith, to live well. Since we cannot exercise another person’s capacity to value, however, our reason is to support them in their exercise of this capacity (provided that they are not pursuing harmful ends). As I said earlier, we cannot support the ends of all people. I share Kant’s view that we fulfill our standing responsibility to others by coming up with a coherent plan of action. Perhaps we choose a vocation where we are 25 Iris Murdoch (1956). “Vision and Choice in Morality”. In: Proceedings of the Aristotelian Society Supplementary Volume 30, pages 32–58, p. 39.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

130 the value of humanity teachers. Or we are a head of a household who provides for children and grandparents. There are many possibilities. In these capacities we have reason to help others find out what makes sense for them to value with a view to their life as a whole, and to support them in valuing those things. Here we do not seek to make judgments about which final ends are better (simpliciter) than others. Instead, we aim to help people judge for themselves what is better for them, and to support them in pursuing those things. We are responsive to others as individuals, thinking our way into their situation and character. While I have not given a Kantian account of the value of human beings, this view of what we owe to others is not out of step with Kant. I admit I am not sure I understand his arguments for this conclusion, but Kant thinks we are to adopt principles of action which further the ends of others. In the Metaphysics of Morals he introduces the framework in which we are to do so with metaphors of attraction and repulsion. We are to imagine that our external relations with others are governed by two forces: The principle of mutual love admonishes [human beings] constantly to come closer to one another; that of the respect they owe one another, to keep themselves at a distance from one another26

As Kant goes on to unpack the metaphors, mutual love requires us to adopt principles of active benevolence towards others. Active benevolence expresses itself in a requirement to “make others’ ends my own.” That means we are to help others realize their interests and projects. But the requirement to respect others means that we are to make the ends of others our own without encroaching on them. We are to further their ends in ways that are not at variance with their selfrespect. This takes non-paternalism on our part. If I depart from Kant it is in not making use of the term “respect.” In the Kantian context,

26 Immanuel Kant (1996). The Metaphysics of Morals. Edited by Mary Gregor. Cambridge: Cambridge University Press, §6:449.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

5.4 our responses to human beings 131 the rationale for non-paternalism is the character of the value of persons as autonomous—able to legislate the moral law. Respect is a distinctive posture that registers, by non-interference, an autonomous being’s ability to see its own way to moral principles. This is an ability that sets it apart from earthly things and gives it a share of divinity. These ideas go together with a very particular conception of our value and its ground. As Bernard Williams has written, respect is “a kind of secular analogue of the Christian conception of the respect owed to all men as equally children of God. Though secular, it is equally metaphysical: in neither case is it anything empirical about men that constitutes the ground of equal respect.”27 The account I have developed here is self-consciously non-Kantian, empirically grounded, and realist in orientation. Ideas of respect should be preserved for Kantian accounts of our value, and since this is not the view I have taken, “respect” is not one of my words. Of course, there are remaining questions about precisely what behaviors are compatible with this schema for ethical responses to human beings. For Kant the posture of non-destruction involves the avoidance of contempt, arrogance and ridicule, while that of active benevolence involves cultivating generosity and sympathetic feeling. Clearly there is more to say. There are questions about whether these further behaviors can be derived from the principle to not impede, and to support, the ability of people to pursue final ends. All this is part of a larger story about our responsiveness to human beings. My concern in this work has been more limited: to show how a relational account of our value is able to capture reasons that are owed to everyone, and I find it plausible to say that human beings make a claim on us to safeguard their capacity to lead good lives by not impeding or destroying it, and more positively, to support them in their exercise of this capacity. Human beings make this claim independently of their kinship to us, independently of our feelings about them, and independently of their social standing. Naturally there are further 27 Williams, “The Idea of Equality,” 116.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

132 the value of humanity questions about how this abstract kind of right generates specific obligations, just as there are questions about the role of conflicting considerations. These questions belong in a larger normative ethical study. But the interest of this work has been more foundational, and in a clear sense value-theoretic. In showing how the account developed here licenses a basic schema for the ethical responsibilities we owe to everyone, it has done what it set out to do.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

Conclusion At the end of his book-length investigation into treatments of human dignity in the decades surrounding the second world war, a significant period for these discussions, Mark Greif gives voice to a conflicted feeling: I want to tell my contemporaries: Stop! Anytime your inquiries lead you to say, “At this moment we must ask and decide who we fundamentally are, our solution and salvation must lie in a new picture of ourselves and humanity, this is our profound responsibility and a new opportunity”—just stop. You have begun asking the wrong analytic questions . . .1

It is an arresting passage in a memorable book. The feeling is that there is something feverish and a bit inscrutable in the intense collective effort to theorize the human being and the ground of our ethical significance. This is not an enterprise to which scholars should contribute or see continue. The sentiment is not quite one that Grief expresses in his own voice. He is imagining a layperson’s reaction— someone watching the movements of the academy from outside— but it is of course a perspective that the academy readily takes on itself. Speaking now of academic philosophy, a question that moral philosophers have traditionally posed for themselves is whether they should seek to answer a moral skeptic. And as the moral skeptic is influentially envisioned, she is one who denies, precisely, the value of human beings.2 So a question for moral philosophers, a question 1 Greif, Crisis of Man, 328. 2 See Raz. “The Amoralist,” section 1. See also Korsgaard, The Source of Normativity, 123 who is attempting to vindicate Enlightenment morality by showing that

The Value of Humanity. L. Nandi Theunissen, Oxford University Press (2019). © L. Nandi Theunissen. DOI: 10.1093/oso/9780198832645.001.0001

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

134 the value of humanity that is internal to the discipline, is whether we should attempt to secure grounds for the ethical status of human beings. The question is whether this enterprise is entirely well-conceived. H. A. Prichard famously made the case that attempts to address a skeptic are as misbegotten in moral philosophy as they are in epistemology, and his arguments were roundly influential.3 To be clear, Prichard was not speaking against attempts to vindicate the value of human beings in particular. To the extent that Prichard’s sights were clearly set they were set against Plato of the Republic. Plato had sought to answer the skeptical questions raised by the figure of Thrasymachus by showing that the unjust person is made miserable by his injustice, and Prichard found the shape of that answer quite wrong. If you try to argue that you ought to do the right thing because doing that will make you happy, then you are giving an answer of the wrong kind; a moral obligation cannot be grounded in this way. Though he goes on to rebuke Kant too, there is something decidedly Kantian about Prichard’s understanding of the normative concepts at issue for Plato, and one is not always sure that he is clear about the anachronism. But it is also fine with me to suppose that Prichard is right. We should not seek to persuade people to treat one another in very particular ways by showing them that this will make their life go well (even if, as I have argued, that is likely to be true). Do the resources of the discipline stand or fall with Plato’s form of argument? In the final chapter of a work that I continue to regard as a tour de force of moral philosophy, Philippa Foot wrests engagement with the skeptic back from Prichard.4 Foot’s skeptic is like Thrasymachus in rejecting the demands of virtue as so much simple-minded piety. What the skeptic fails to see, in Foot’s telling, is that the demands of virtue are set by facts about the life-form of human beings. Here Foot relies on the schema of natural normativity that she has developed human beings are valuable. I have learned from discussion of this form of moral skepticism by Kyla Ebels-Duggan (n.d.). “Love, Respect, and the Value of Humanity”. In: Re-Evaluating the Value of Humanity. Edited by Sarah Buss and Nandi Theunissen. Manuscript. 3 Prichard, “Does Moral Philosophy Rest on a Mistake?” 4 Foot, Natural Goodness, ch. 7.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

conclusion 135 at some length. Standards for individual conduct are determined by facts about the needs, limitations, and powers of the species to which an individual belongs. A person should be beneficent, say, because human beings need help and compassion when misfortune falls upon them.5 The details of Foot’s account do not concern me here. What is of interest is that she responds to challenges to our deeply held ethical commitments by thinking through the relationship between basic evaluative concepts like virtue, reason, and the good for human beings. In other words she does foundational theoretical work. This book has a comparable aspiration. Even as his objections are relevant downstream of my proposal, Foot’s immoralist is not my immediate interlocutor. I have sought to vindicate an ethical commitment that a different kind of moral skeptic does not see reason to accept. This is the claim that human beings are bearers of value. My motivation is a dissatisfaction with the predominant framework for theorizing our value, a framework that is more or less selfconsciously Kantian. My dissatisfaction with this framework arises from a suspicion that anything could be valuable in the way that Kantians take human beings to be of value. That is, human beings cannot be of value independently of their propensity to stand in relations of benefit to human beings or other beings because these relations are the very essence of value. It would overstate my case, and misdescribe the structure of my argument, to say that a relational account of the kind I have proposed in these pages follows by inference to the best explanation.6 I have not done the requisite comparative work to extant accounts to support that claim. For example, I have given my reasons for rejecting an influential argument for the non-relational value of humanity from the possibility of relational value, but I am not suggesting that this is the only way to secure the non-relational value of humanity, and I do not take myself to have foreclosed other possibilities. You might rather

5 Foot, Natural Goodness, 108. 6 I am grateful to Japa Pallikkathayil for asking me to clarify the structure of my argument in relation to IBE arguments.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

136 the value of humanity think of this work as proposing an experiment. The experiment is to offer a relational account of the value that is typical and characteristic of human beings, and that stands to capture crucial desiderata for an account: namely, basic forms of ethical behavior that are owed to human beings. The experiment is motivated by a Socratic conception of the good as the beneficial, and by a conviction that the value of human beings is not radically set apart from nature but continuous with it. If the experiment works then we have a relational explanation of human value, and its merits and deficiencies relative to other accounts can be made out. So come back to the conflicted feeling—the feeling that there is something both excessive and misguided in constructive efforts to provide grounds for the ethical significance of human beings of the kind that is given here. In making these efforts I take it that we strive to understand foundational ethical concepts and their relationship. We make ground if we bring to light puzzles or problems with received ways of talking, and if we offer imaginative resources for thinking differently. To me if there is a conflicted feeling it is that one’s philosophical investigation may not always lead one to where one was hoping to get. I have no illusions about having the answers to all the questions that come up for my proposal. My project, as I hope, ultimately relates to a rather wide range of issues in normative ethics, meta-ethics, and the metaphysics of value. Much of what will have to be said is future work. But I suppose I have the Socratic and Platonic intuition that on-going investigation is a philosophical state of being. I try to enjoy this state–this sense of not having settled everything, or even much at all.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

A P P E N DI X 1

The Senses of Humanity In one sense, the term “humanity,” from the Latin humanitus, means learnedness or cultivation, so that a person is “humane” who knows the important sculptors from antiquity.1 So we have “the humanities” (studia humanitatis) as the field of study that is concerned with human culture, and “humanism” as a movement in education that puts the classics at its center. Virginia Woolf gives voice to the spirit of humanism when she tells us that we are especially well-served to study the Greeks. For, “in spite of the labour and the difficulty,” there we see “the stable, the permanent, the original human being.”2 If we speak in this connection of the “humanity of human beings” then we speak of something elevated—a condition to which we aspire. There is early precedent for regarding it as divine. Influentially, Aristotle held that intellect is the best thing in us, and that to the extent we can contemplate and reason we can be close to God. We are not to heed the modest maxim of “Humans you are, think mortal thoughts,” but strain every nerve in our being to assimilate to the immortals. Even as the exercise of our intellect is shabby in comparison with God, in worth and power it surpasses everything.3 We should raise ourselves up! This is a familiar set of resonances, and it may be worked into a genealogical argument for how humanity should be seen and valued. But the argument would be one-sided. There are senses of “humanity” that point away from divinity, and Aristotle’s quarrel about mortality gives the clue. Aristotle is responding to a strand in Greek culture that sees humanity as essentially 1 I take this from the second-century Latin writer Aulus Gellius on whose authority humanity means “‘well-educated’ and ‘rather learned’: someone who would know, through books and history, who Praxiteles was.” Aulus Gellius (1998). Noctes Atticae. Edited by John C. Rolfe. Cambridge University Press, §13.17. Gellius is cited by Christopher S. Celenza (2008). “Humanism and the Classical Tradition”. In: Annali d’Italianistica 26, Humanisms, Posthumanisms, and Neohumanisms, pages 25–49, p. 27. I am grateful to Richard Bett for referring me to Calenza’s paper. 2 Virginia Woolf (1925). “On Not Knowing Greek”. In: The Common Reader: First Series. New York: Harcourt, Brace and Company, pages 10–17, p. 12. 3 Aristotle (2011). Nicomachean Ethics. Edited by Sarah Broadie and Christopher Rowe. Oxford and New York: Oxford University Press, 10.7 1177b 26–1178a 8.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

138 appendix 1 connected to mortality, something we are prone to forget, and forget at our peril.4 The Greek anthropinon is what belongs to human beings as contrasted with gods. We are the thnetoi anthropoi—the mortal or dying ones—essentially such as to come into and out of being. These resonances survive in vernacular English. We speak of “encountering our humanity” as a state of awakening to our inevitable slippage into death, and we speak of “showing humanity” as a compassionate response in the face of another’s difficulty. We see that someone is vulnerable and their vulnerability touches us as something we have shared or could share.5 Humanity in this sense may be extended to others, but equally it may be withheld, and it is in this context that we speak of “being humane” and of “humanitarianism.” A related term is “philanthropy,” from the Greek philanthropia, the kindly or benevolent attitude shown towards human beings on account of our imperfection. There is in this way a puzzling doubleness to our notion of humanity, and etymological arguments will not settle how to think of it. In a compelling essay, an essay that draws attention to this dual conception of humanity as both elevated and not, Wolfgang Schadewaldt makes the following constructive suggestion.6 Part of the idea of mortality is that mortal beings are subject to characteristic forms of vulnerability, limitation, and frailty. But we cannot understand what it is to be vulnerable in the way that human beings can be vulnerable, or to lead a mortal human life (as opposed to an immortal and godly one), without also thinking about what human beings characteristically do and are able to do well. In that case, to be a mortal being is not merely to be subject to harm and death, but to be capable of embodying human forms of excellence (as opposed to immortal and godly ones). In the traditions where this conception of humanity is at stake, an understanding of our mortality, which is to say, our human limitations, is the foundation for the development of whatever is best in us.

4 As Wolfgang Schadewaldt (1965). “Der Gott von Delphi und de Humanitätsidee”. In: Association Humaniste Grecque Centre D’Etudes Humanistes Classique, Première Série: Antiquité et Problèmes Contemporains 26, §3 discusses the Delphic sayings, the import of “Know thyself!” lies in “Think about your own mortality,” or better, “Let the awareness of your mortality become a constituent of all thinking, feeling, and action.” For discussion of the significance of the Delphic sayings for Plato’s Socrates in Apology, see Katja Maria Vogt (2012). Belief and Truth: A Skeptic Reading of Plato. Oxford and New York: Oxford University Press, ch. 1. 5 Cora Diamond (1991). “The Importance of Being Human”. In: Philosophy Royal Institute to Philosophy Supplement 29, pages 35–62 investigates the centrality of mortality to our conception of humanity. 6 See Schadewaldt, “Der Gott von Delphi.” I am most grateful to Eckart Forster for bringing my attention to this essay, and to Alexander Englert for his fine translation.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

A P P E N DI X 2

Korsgaard on Regress Arguments and Valuing Ourselves There is a version of the regress argument discussed in Chapter 3 that I have so far said nothing about. This version of the argument is much discussed, but it has been abandoned, or at least crucially modified, by its original proponent—Christine Korsgaard.1 Though Korsgaard is working with a quite different conception of relational and non-relational value, I discuss her version of the regress argument here on account of its enormous influence, and because my positive account of the value of human beings bears comparison with her eventual proposal. Korsgaard’s version of the regress argument is guided by the intuition that goodness has a source—that it comes from somewhere. Part I of the argument starts from the assumption that some ordinary objects of choice are good or valuable. It raises a question about the origin of their value, and contends that their value must ultimately come from something that is valuable in itself. For without a source that is valuable in itself, the value of one thing would derive from another, which would derive from another in turn, and so on ad infinitum. In that case nothing would be valuable, and some things are. Part II of the argument makes the case that rational beings, or valuers, or humanity (these terms are used interchangeably) meet the criteria for being the original source of value. Rational beings confer value on things by choosing them according to rational standards. Since value travels from a rational chooser to its object, and since the value of rational choosers has no external source, it follows that rational choosers, valuers, are valuable in themselves. On this way of setting things up, whatever has its value conferred by something else is “relationally good,” and whatever is good but not relationally so (that is, whose goodness is not conferred by something else) is “good in itself.” The argument is that the existence of relational goods establishes that something is good in itself, and that people meet the criteria. People meet the criteria because

1 See principally Korsgaard, “Two Distinctions,” 177–184, and Korsgaard, “Formula of Humanity,” 190–197.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

140 appendix 2

Valuer

Non-relational

Relational

Figure 7 Source

they are valuers, which is to say, conferrers of value on relational goods. See Figure 7. The argument has been seized at several joints.2 The idea that value has a source in the relevant sense is a supposition of meta-ethical constructivists, and is denied by realists.3 To that extent, the argument rests on a contentious theoretical premise. Realists have made the point that deriving the value of objects or activities of choice from our choosing, and not the value of our choosing from objects and activities of choice, does not have the appearances on its side, nor classic theories of agency.4 But granting that value has a source, and granting that value travels from rational choosers to objects, critics have denied the implication that rational choosers are of value. Here is Rae Langton: We have no more antecedent reason to expect the creators of goodness to be good than to expect painters of the blue to be blue, or the creators of babies to be babies. In general we don’t think the source of something valuable must itself be valuable. War can produce good poets, chicken manure can produce good roses, and in general the sources of good things can be bad.5 2 Less important for my purposes are criticisms from the point of view of Kant interpretation. On interpretive issues, see Jens Timmermann (2006). “Value Without Regress: Kant’s ‘Formula of Humanity’ Revisited”. In: European Journal of Philosophy 14.1, pages 69–93. 3 See Joseph Raz (2003). The Practice of Value. Edited by R. Jay Wallace. Oxford University Press, page 140, and T. M. Scanlon (2003). “Metaphysics and Morals”. In: Proceedings and Address of the American Philosophical Association 77.2, pages 7–22. 4 See Donald H. Regan (2002). “The Value of Rational Nature”. In: Ethics 112.2, pages 267–291. For discussion, see David Sussman (2003). “The Authority of Humanity”. In: Ethics 113.2, pages 350–366. 5 Rae Langton (2007). “Objective and Unconditioned Value”. In: Philosophical Review 116.2, pages 157–185, p. 175.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

appendix 2 141 To secure the implication that the ultimate source of the value of other things must itself be valuable, there needs to be a principle to the effect that there can be nothing in an effect that is not contained in its cause. But, as Jerry Schneewind pointed out, this principle is implausible.6 Korsgaard herself took Schneewind’s point to be decisive: [I]f we do not set a value on ourselves [. . .] then nothing else can have any value either. My argument for that point is not based on the traditional metaphysical argument that there is nothing in the effect that is not somehow in the cause, as Schneewind suggests. I am aware that in early papers I made it sound too much as if value were some sort of metaphysical substance that got transferred from us to our ends via the act of choice. As I’ve tried to clarify here, I don’t think that: I think that self-love, valuing oneself, is essentially involved in the valuing of our ordinary ends. In this sense the value of humanity is the unconditioned condition of all other value, while the other ends of practical reason are not.7 If the proposal was that valuers are valuable in themselves because they are the first cause of the value of other things, then Korsgaard rejects the proposal. By the lights of its primary proponent, the first version of the regress argument is unsuccessful. Korsgaard’s considered view, as stated above, is that while valuers confer value on other things, so that the value of other things is dependent on valuers while the value of valuers is not dependent on values, this form of asymmetrical dependence does not imbue valuers with distinctive value. Our value is also conferred, only reflexively. As the revised proposal is developed, the argument is not regressive but transcendental: it is a condition on the possibility of action that we must confer value on ourselves.8 That would appear to be a separate matter, requiring independent defense.9 I have offered an explanation of the value of valuers in terms of the relationship we bear to ourselves, and on the face of it the proposal bears 6 See J. B. Schneewind (1998). “Korsgaard and the Unconditional in Morality”. In: Ethics 109.1, pages 36–48, p. 39. 7 See Christine Korsgaard (1998). “Motivation, Metaphysics, and the Value of the Self: A Reply to Ginsborg, Guyer, and Schneewind”. In: Ethics 109.1, pages 49–66, pp. 63–64. 8 See Christine Korsgaard (1996). The Sources of Normativity. Cambridge: Cambridge University Press, §3.4.10. 9 For discussion of the relationship between the arguments early and late, and the suggestion that they are dependent, see William J. Fitzpatrick (2005). “The PracticalTurn in Ethical Theory: Korsgaard’s Constructivism, Realism and the Nature of Normativity”. In: Ethics 115, pages 651–691, p. 661 and passim.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

142 appendix 2 likeness to Korsgaard’s reformed position on which, as valuers, we confer value on ourselves. In fact the resemblance is superficial. For Korsgaard the relevant form of relational value is conferred value, not goodness for someone. And for Korsgaard a valuer is someone who is capable of conferring value, not of engaging with (and of being benefited by) relational goods. Korsgaard’s position is essentially constructivist, while R* is in principle compatible with both realist and anti-realist accounts.10 Korsgaard’s argument for why we must confer value on ourselves is that without doing so agency is impossible.11 R* makes no such commitment. Korsgaard states her thesis about the value of valuers as a thesis about how valuers are valued (for themselves, and by themselves). In claiming that human beings have a special kind of relational value, R* is a thesis about how human beings are of value. Whatever we make of the distinction between valuing and being of value, arguably Korsgaard herself takes the distinction to be significant.12 A final difference is that R* makes the notion of a good life central—what it is to be good for ourselves is to live well—and that is not a feature of Korsgaard’s proposal. For all these reasons, the positions are distinctly different.

10 R* is in principle neutral on meta-ethical questions though I gave my reasons for favoring a realist construction. 11 For relevant criticism, see David Enoch (2006). “Agency Shmagency: Why Normativity Won’t Come from What is Constitutive of Action”. In: The Philosophical Review 115.2, pages 169–198. 12 Korsgaard famously argued for two distinctions in goodness: “There are, therefore, two distinctions in goodness. One is the distinction between things valued for their own sakes and things valued for the sake of something else [. . .]. The other is the distinction between things which have their value in themselves and things which derive their value from some other source.” Korsgaard, “Two Distinctions,” 170. As Langton reads Korsgaard in “Objective and Unconditioned,” 160 and ff., the distinctions are between how we value things and how things are valuable. To my mind, the distinctions come apart only on condition that mistakes in valuing are being made. But I leave this to one side.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

Bibliography Allais, Lucy (2016). “Kant’s Racism”. In: Philosophical Papers 45.1 and 2, pages 1–36. Anderson, Elizabeth (1993). Value in Ethics and Economics. Cambridge, MA: Harvard University Press. Anscombe, Elizabeth (1958). “Modern Moral Philosophy”. In: Philosophy 33.124, pages 1–19. Anthony, Lawrence with Graham Spence (2009). The Elephant Whisperer. London: Pan Books. Aristotle (1987). A New Aristotle Reader. Edited by J. L. Ackrill. Princeton, NJ: Princeton University Press. Aristotle (2011). Nicomachean Ethics. Edited by Sarah Broadie and Christopher Rowe. Oxford and New York: Oxford University Press. Baker, Herschel (1947). The Dignity of Man: Studies in the Persistence of an Idea. Cambridge, MA: Harvard University Press. Beardsley, Monroe C. (1965). “Intrinsic Value”. In: Philosophy and Phenomenological Research 26, pages 1–17. Bett, Richard (n.d.). “Do the Ancients See Value in Humanity?” In: Re-Evaluating the Value of Humanity. Edited by Sarah Buss and Nandi Theunissen. Manuscript. Bradley, Ben (2006). “Two Concepts of Intrinsic Value”. In: Ethical Theory and Moral Practice 9, pages 111–130. Buss, Sarah (2012). “The Value of Humanity”. In: The Journal of Philosophy CIX, pages 341–377. Celenza, Christopher S. (2008). “Humanism and the Classical Tradition”. In: Annali d’Italianistica 26, Humanisms, Posthumanisms, & Neohumanisms, pages 25–49. Conee, Earl (1982). “Instrumental Value without Intrinsic Value?” In: Philosophia 11, pages 345–359. Cooper, John (2010). “Political Community and the Highest Good”. In: Being, Nature and Life: Essays in Honor of Allan Gotthelf. Edited by James G. Lennox and Robert Bolton. Cambridge: Cambridge University Press, pages 212–264. Cooper, John M. (1977). “The Psychology of Justice in Plato”. In: American Philosophical Quarterly 14.2, pages 151–157.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

144 bibliography Cranor, Carl (1975). “Toward A Theory of Respect for Persons”. In: American Philosophical Quarterly 12.4, pages 309–319. Crisp, Roger (2013). “In Defense of Absolute Goodness”. In: Philosophy and Phenomenological Research LXXXVII.2, pages 476–482. Darwall, Stephen (1997). “Two Kinds of Respect”. In: Ethics 88.1, pages 36–49. Darwall, Stephen (2004). Welfare and Rational Care. Princeton, NJ: Princeton University Press. Darwall, Stephen (2008). “Kant on Respect, Dignity and the Duty of Respect”. In: Kant’s Ethics of Virtue. Edited by Monika Betzler. Berlin: De Gruyter, pages 175–200. Dean, Richard (2006). The Value of Humanity in Kant’s Moral Theory. Oxford: Oxford University Press. Denis, Lara (2007). “Kant’s Formula of the End in Itself: Some Recent Debates”. In: Philosophy Compass 2.2, pages 244–257. Diamond, Cora (1978). “Eating Meat and Eating People”. In: Philosophy 53.206, pages 465–479. Diamond, Cora (1991). “The Importance of Being Human”. In: Philosophy Royal Institute to Philosophy Supplement 29, pages 35–62. Dworkin, Ronald (1977). Taking Rights Seriously. Cambridge, MA: Harvard University Press. Ebels-Duggan, Kyla (n.d.). “Love, Respect, and the Value of Humanity”. In: Re-Evaluating the Value of Humanity. Edited by Sarah Buss and Nandi Theunissen. Manuscript. Engelmann, Paul (1968). Letters from Ludwig Wittgenstein with a Memoir. Edited by Brian McGuinness. New York: Horizon Press. Engstrom, Stephen (1996). “Happiness and the Highest Good”. In: Aristotle, Kant, and the Stoics. Edited by Stephen Engstrom and Jennifer Whiting. Cambridge: Cambridge University Press. Chapter 4. Engstrom, Stephen (2015). “The Complete Object of Practical Knowledge”. In: The Highest Good in Aristotle and Kant. Edited by Joachim Aufderheide and Ralf M. Bader. Oxford: Oxford University Press, pages 129–157. Engstrom, Stephen (2016). “The Determination of the Concept of the Highest Good”. In: The Highest Good in Kant’s Philosophy. Edited by T. Howing. Berlin: De Gruyter, pages 89–108. Engstrom, Stephen and Jennifer Whiting (1996). Aristotle, Kant, and the Stoics: Rethinking Happiness and Duty. Edited by Stephen Engstrom and Jennifer Whiting. Cambridge: Cambridge University Press. Enoch, David (2006). “Agency Shmagency: Why Normativity Won’t Come from What is Constitutive of Action”. In: The Philosophical Review 115.2, pages 169–198.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

bibliography 145 Fitzpatrick, William J. (2005). “The Practical Turn in Ethical Theory: Korsgaard’s Constructivism, Realism and the Nature of Normativity”. In: Ethics 115, pages 651–691. Foot, Philipa (1985). “Utilitarianism and the Virtues”. In: Mind 93.374, pages 196–209. Foot, Philippa (2001). Natural Goodness. Oxford: Oxford University Press. Frankena, W. K. (1939). “The Naturalistic Fallacy”. In: Mind 48.192, pages 464–477. Frankfurt, Harry G. (1999a). “Equality and Respect”. In: Necessity, Volition and Love. Cambridge: Cambridge University Press, pages 146–154. Frankfurt, Harry G. (1999b). Necessity, Volition and Love. Cambridge: Cambridge University Press. Frankfurt, Harry G. (2001). “The Dear Self ”. In: Philosopher’s Imprint 1.0, pages 1–14. Frankfurt, Harry G. (2004). The Reasons of Love. Princeton, NJ and Oxford: Princeton University Press. Frankfurt, Harry G. (2015). Inequality. Princeton, NJ: Princeton University Press. Frede, Michael (1999). “On the Stoic Conception of the Good”. In: Topics in Stoic Philosophy. Edited by Katerina Ierodiakonou. Oxford: Clarendon Press, pages 71–94. Geach, Peter (1956). “Good and Evil”. In: Analysis 17.2, pages 33–42. Gellius, Aulus (1998). Noctes Atticae. Edited by John C. Rolfe. Cambridge: Cambridge University Press. Greif, Mark (2015). The Age of the Crisis of Man. Princeton, NJ and Oxford: Princeton University Press. Guyer, Paul (1998). “The Possibility of the Categorical Imperative”. In: Kant’s Groundwork of the Metaphysics of Morals: Critical Essays. Edited by Paul Guyer. Lanham: Rowman and Littlefield, pages 215–246. Harman, Gilbert H. (1967). “Toward a Theory of Intrinsic Value”. In: The Journal of Philosophy 64.23, pages 792–804. Henrich, Dieter (1992). Aesthetic Judgment and the Moral Image of the World. Edited by Eckart Forster. Stanford Series in Philosophy: Studies in Kant and German Idealism. Stanford, CA: Stanford University Press. Henrich, Dieter (1994). The Unity of Reason: Essays on Kant’s Philosophy. Edited by Richard L. Velkley. Cambridge, MA: Harvard University Press. Hurka, Thomas (1987). “ ‘Good’ and ‘Good For’ ”. In: Mind 96.381, pages 71–73. Irwin, Terence (1995). Plato’s Ethics. New York and Oxford: Oxford University Press.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

146 bibliography Irwin, Terence (1996). “Kant’s Criticisms of Eudaemonism”. In: Aristotle, Kant, and the Stoics. Edited by Stephen Engstrom and Jennifer Whiting. Cambridge: Cambridge University Press. Chapter 3. Kagan, Shelly (1998). “Rethinking Intrinsic Value”. In: The Journal of Ethics 2, pages 277–297. Kagan, Shelly (2016). “What is Wrong with Speciesism?” In: Journal of Applied Philosophy 33.1, page 2016. Kant, Immanuel (1996). The Metaphysics of Morals. Edited by Mary Gregor. Cambridge: Cambridge University Press. Kant, Immanuel (1997a). Lectures on Ethics. Edited by Peter Heath and J. B. Schneewind. Cambridge: The Cambridge Edition of the Works of Immanuel Kant. Kant, Immanuel (1997b). The Critique of Practical Reason. Edited by Mary Gregor. Cambridge: Cambridge University Press. Kant, Immanuel (1997c). The Groundwork of the Metaphysics of Morals. Edited by Mary Gregor. Cambridge Texts in the History of Philosophy. Cambridge: Cambridge University Press. Kleingeld, Pauline (2007). “Kant’s Second Thoughts on Race”. In: The Philosophical Quarterly 57.229, pages 573–592. Korsgaard, Christine (1983). “Two Distinctions in Goodness”. In: The Philosophical Review 92.2, pages 169–95. Korsgaard, Christine (1986). “Kant’s Formula of Humanity”. In: Kant-Studien 77, pages 183–202. Korsgaard, Christine (1996). The Sources of Normativity. Cambridge: Cambridge University Press. Korsgaard, Christine (1998).“Motivation, Metaphysics, and the Value of the Self: A Reply to Ginsborg, Guyer, and Schneewind”. In: Ethics 109.1, pages 49–66. Koullas, Sandy (2019). “Love, Practical Reason, and African Philosophy”. In: The Routledge Handbook of Love in Philosophy. Edited by Adrienne M.Martin. Routledge, pages 313–324. Kraut, Richard (2007). What is Good and Why. Cambridge, MA and London: Harvard University Press. Kraut, Richard (2011). Against Absolute Goodness. Oxford and New York: Oxford University Press. Langton, Rae (2007). “Objective and Unconditioned Value”. In: Philosophical Review 116.2, pages 157–185. Lewis, David (1973). Counterfactuals. Oxford: Blackwell. Liao, S. Matthew (2010). “The Basis of Human Moral Status”. In: Journal of Moral Philosophy 7.2, pages 159–179.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

bibliography 147 Long, A. A. and D. N. Sedley (1992). The Hellenistic Philosophers. Cambridge: Cambridge University Press. Mackie, J. L. (1977). Ethics: Inventing Right and Wrong. London: Penguin Books. McDowell, John (1980). “The Role of Eudaimonia in Aristotle’s Ethics”. In: Essays on Aristotle’s Ethics. Edited by Amelie Oksenberg Rorty. Berkeley: University of California Press, pages 359–76. Mills, Charles (2005). “Kant’s Untermenschen”. In: Race and Racism in Modern Philosophy. Edited by A. Valls. Ithaca: Cornell University Press, pages 169–193. Moore, G. E. (1903). Principia Ethica. Cambridge: Cambridge University Press. Moyn, Samuel (2010). The Last Utopia: Human Rights in History. Cambridge– MA: The Belknap Press of Harvard University Press. Murdoch, Iris (1956). “Vision and Choice in Morality”. In: Proceedings of the Aristotelian Society Supplementary Volume 30, pages 32–58. Nagel, Thomas (1970). The Possibility of Altruism. Princeton, NJ: Princeton University Press. Nagel, Thomas (1989). The View from Nowhere. Oxford and New York: Oxford University Press. Nagel, Thomas (2012). Mind and Cosmos. Oxford and New York: Oxford University Press. Nozick, Robert (1981). Philosophical Explanations. Cambridge – MA: Harvard University Press. O’Neill, Onora (1985). “Between Consenting Adults”. In: Philosophy and Public Affairs 14.3, pages 252–277. Pallikkathayil, Japa (2010). “Deriving Morality from Politics: Rethinking the Formula of Humanity”. In: Ethics 121, pages 116–147. Penner, Terry (2003). “The Forms, The Form of the Good, and the Desire for Good, in Plato’s Republic”. In: The Modern Schoolman LXXX, pages 191–233. Plato (1997a). “Lysis”. In: Plato Complete Works. Edited by John M. Cooper. Indianapolis: Hackett Publishing, pages 687–707. Plato (1997b). “Meno”. In: Plato’s Complete Works. Edited by John M. Cooper. Indianapolis: Hackett Publishing, pages 870–897. Plato (1997c). “Republic”. In: Plato Complete Works. Edited by John M. Cooper. Indianapolis and Cambridge: Hackett Publishing, pages 971–1223. Prichard, H. A. (1912). “Does Moral Philosophy Rest on a Mistake?” In: Mind 21.81, pages 21–37.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

148 bibliography Rabinowicz, Wlodek and Toni Rønnow-Rasmussen (2000). “A Distinction in Value: Intrinsic and For Its Own Sake”. In: Proceedings of the Aristotelian Society, New Series 100, pages 33–51. Railton, Peter (1986). “Moral Realism”. In: The Philosophical Review 95.2. Rawls, John (1971). A Theory of Justice. Cambridge – MA: Harvard University Press. Raz, Joseph (1986). The Morality of Freedom. Oxford: Clarendon Press. Raz, Joseph (1999a). “Agency, Reason and the Good”. In: Engaging Reason. Oxford and New York: Oxford University Press, pages 22–45. Raz, Joseph (1999b). Engaging Reason: On the Theory of Value and Action. Oxford and New York: Oxford University Press. Raz, Joseph (1999c). “The Amoralist”. In: Engaging Reason. Oxford and New York: Oxford University Press, pages 273–302. Raz, Joseph (2001a). “Respecting People”. In: Value, Respect and Attachment. Cambridge: Cambridge University Press, pages 124–175. Raz, Joseph (2001b). Value, Respect and Attachment. Cambridge: Cambridge University Press. Raz, Joseph (2002). “On Frankfurt’s explanation of Respect for People”. In: Contours of Agency: Essays on Themes from Harry Frankfurt. Edited by Sarah Buss and Lee Overton. Cambridge – MA: MIT Press, pages 299–315. Raz, Joseph (2003). The Practice of Value. Edited by R. Jay Wallace. Oxford: Oxford University Press. Raz, Joseph (2004). “The Role of Well-being”. In: Philosophical Perspectives 18, pages 269–294. Raz, Joseph (2010). “On the Guise of the Good”. In: Desire, Practical Reason, and the Good. Edited by Sergio Tenenbaum. Oxford: Oxford University Press. Regan, Donald H. (2002). “The Value of Rational Nature”. In: Ethics 112.2, pages 267–291. Regan,Donald H. (2004). “Why am I my Brother’s Keeper?” In: Reason and Value: Themes from the Moral Philosophy of Joseph Raz. Oxford: Oxford University Press, pages 202–230. Remnick, David (2016). “Leonard Cohen Makes it Darker”. In: The New Yorker, October 17. Rosati, Connie (2006). “Personal Good”. In: Metaethics After Moore. Edited by Mark Timmons Terry Horgan. Oxford: Oxford University Press, pages 107–131. Rosati, Connie (2008). “Objectivism and Relational Good”. In: Social Philosophy and Policy Foundation 25, pages 314–349.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

bibliography 149 Rosati, Connie (2009). “Relational Good and the Multiplicity Problem”. In: Philosophical Issues 19 (Metaethics), pages 205–234. Ross, W. D. (2002). The Right and the Good. Edited by Philip Stratton-Lake. Oxford and New York: Oxford University Press. Sayre-McCord, Geoff (Fall 2017 Edition). “Moral Realism”. In: The Stanford Encyclopedia of Philosophy Edward N. Zalta (ed.) URL=. Scanlon, T. M. (1998). What We Owe to Each Other. Cambridge, MA: Harvard University Press. Scanlon, T. M. (2003).“Metaphysics and Morals”. In: Proceedings and Address of the American Philosophical Association 77.2, pages 7–22. Schadewaldt, Wolfgang (1965). “Der Gott von Delphi und de Humanitätsidee”. In: Association Humaniste Grecque Centre D’Etudes Humanistes Classique, Première Série: Antiquité et Problèmes Contemporains 26. Scheffler, Samuel (2010). “Valuing”. In: Equality and Tradition: Questions of Value in Moral and Political Theory. Oxford and New York: Oxford University Press, pages 15–40. Schiller, Friedrich (1982). “The Philosophers”. In: Xenien. Edited by Erich Trunz. Volume 1. Munich: Beck. Schneewind, J. B. (1998). “Korsgaard and the Unconditional in Morality”. In: Ethics 109.1, pages 36–48. Sellars, Wilfrid (1962). “Philosophy and the Scientific Image of Man”. In: Frontiers of Science and Philosophy. Edited by Robert Colodny. Pittsburgh, PA: University of Pittsburgh Press. Sensen, Oliver (2009). “Kant’s Conception of Inner Value”. In: European Journal of Philosophy, pages 1–19. Setiya, Kieran (2010). “Sympathy for the Devil”. In: Desire, Practical Reason, and the Good. Edited by Sergio Tenenbaum. Oxford: Oxford University Press, pages 82–110. Singer, Peter (1975). Animal Liberation. New York: Harper. Singer, Peter (1993). Practical Ethics. 2nd edition. Cambridge: Cambridge University Press. Singer, Peter (2009). “Speciesism and Moral Status”. In: Metaphilosophy 40.3–4, pages 567–581. Stern, Robert (2013). “Moral Skepticism, Constructivism, and the Value of Humanity”. In: Constructivism in Ethics. Cambridge: Cambridge University Press, pages 22–62. Stocker, Michael (2008). “On the Intelligibility of Bad Acts”. In: Moral Psychology Today: Essays on Values, Rational Choice, and the Will. Edited by David K. Chan. Volume 110. Philosophical Studies Series. Springer, pages 123–140.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

150 bibliography Sumner, Wayne (1996). Welfare, Happiness, and Ethics. Oxford: Clarendon Press. Sussman, David (2003). “The Authority of Humanity”. In: Ethics 113.2, pages 350–366. Tenenbaum, Sergio (2010). Desire, Practical Reason, and the Good. Edited by Sergio Tenenbaum. Oxford: Oxford University Press. Theunissen, L. Nandi (2016). “Kant’s Commitment to Metaphysics of Morals”. In: European Journal of Philosophy 24.1, pages 103–128. Theunissen, L. Nandi (2018). “Must We Be Just Plain Good? On Regress Arguments for the Value of Humanity”. In: Ethics 128, pages 346–372. Theunissen, Nandi (2021). “The Excellent and the Beneficial”. In: The Oxford Handbook of Moral Realism. Edited by Paul Bloomfield and David Copp. Oxford: Oxford University Press. Thomson, Judith Jarvis (1997). “The Right and the Good”. In: The Journal of Philosophy 94 (6), pages 273–298. Thomson, Judith Jarvis (2008). Normativity. The Paul Carus Lectures. Illinois: Carus Publishing Company. Thomson, Judith Jarvis (2013). “GOODNESS”. In: Philosophy and Phenomenological Research LXXXVII.2, pages 467–475. Timmermann, Jens (2006). “Value Without Regress: Kant’s ‘Formula of Humanity’ Revisited”. In: European Journal of Philosophy 14.1, pages 69–93. Velleman, David (1999a). “A Right to Self-Termination?” In: Ethics 109.3, pages 606–628. Velleman, David (1999b). “Love as a Moral Emotion”. In: Ethics 109, pages 338–74. Velleman, David (2003). “Narrative Explanation”. In: The Philosophical Review 112.1, pages 1–25. Velleman, David (2006). “A Brief Introduction to Kantian Ethics”. In: Self to Self. Cambridge: Cambridge University Press. Velleman, David (2008). “Beyond Price”. In: Ethics 118, pages 191–212. Vogt, Katja Maria (2008a). “Duties to Others: Demands and Limits”. In: Kant’s Ethics of Virtue. Edited by Monika Betzler. Berlin and New York: Walter de Gruyter, pages 219–243. Vogt, Katja Maria (2008b). Law, Reason and the Cosmic City: Political Philosophy in the Early Stoa. Oxford and New York: Oxford University Press. Vogt, Katja Maria (2008c). “The Good is Benefit: On the Stoic Definition of the Good”. In: Proceedings of the Boston Area Colloquium in Ancient Philosophy, pages 155–174.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

bibliography 151 Vogt, Katja Maria (2012). Belief and Truth: A Skeptic Reading of Plato. Oxford and New York: Oxford University Press. Vogt, Katja Maria (2017). Desiring the Good: Ancient Proposals and Contemporary Theory. Oxford and New York: Oxford University Press. Waldron, Jeremy (2017). One Another’s Equals. Cambridge, MA and London: The Belknap Press of Harvard University Press. White, Nicholas P. (1979). A Companion to Plato’s Republic. Indianapolis and Cambridge: Hackett Publishing. Williams, Bernard (1962). “The Idea of Equality”. In: Philosophy, Politics and Society, 2nd Series. Edited by P. Laslett and W. G. Runciman. Oxford: Blackwell, pages 110–131. Williams, Bernard (1981). “Internal and External Reasons”. In: Moral Luck. Cambridge: Cambridge University Press, pages 101–13. Williams, Bernard (1985). Ethics and the Limits of Philosophy. Cambridge, MA: Harvard University Press. Williams, Bernard (2006). “The Human Prejudice”. In: Philosophy as a Humanistic Discipline. Princeton, NJ: Princeton University Press, pages 135–152. Wolf, Susan (2010). “Good-For Nothings”. In: Proceedings and Address of the American Philosophical Association 85.2, pages 47–64. Wood, Allen (1998). “Humanity As End In Itself ”. In: Kant’s Groundwork of the Metaphysics of Morals: Critical Essays. Edited by Paul Guyer. Boston: Rowman and Littlefield. Woolf, Virginia (1925). “On Not Knowing Greek”. In: The Common Reader: First Series. New York: Harcourt, Brace and Company, pages 10–17. Zimmerman, Michael J. (2005). “Introduction”. In: Recent Work on Intrinsic Value. Edited by Toni Rønnow-Rasmussen and Michael J. Zimmerman.Volume 17. Library of Ethics and Applied Philosophy. Dordrecht: Springer.

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

Index Note: Page numbers in italics refer to figures; ‘n’ after a page number indicates the footnote number.

A absolute value 1 absolute/relative value distinction 29, 32–3, 40 as benefit-independent 37, 41, 43, 45, 48–9, 50 capacity and 36 Kant, Immanuel 37, 40, 41, 47–8 realism 50 suspicion of fantasy of 2, 41, 48–54 see also non-relational value absolute value of humanity 57 as benefit-independent 48–9 human dignity 4 Kant, Immanuel 1–2, 3, 9, 26, 29, 36, 37, 40–1, 43, 48, 57, 61n5 non-relationally valuable 3, 8, 9, 29, 87 absolute worth see absolute value absolutism 3 ancient Greeks/ethics 55–6 good/supreme good as beneficial 43–4, 48, 52, 55 human being as bearer of value 55–6 Kant’s departure from 41–8, 55 Anderson, Elizabeth 89 Anglo-American philosophy 12, 21 animals (other animals) 15, 25, 105, 111 animals/human beings distinction 13–14, 21 capacity to be good for themselves 109–10 Anscombe, G. E. M. 4, 28n46, 54, 55, 105 Aristotle 56, 80, 89, 98–9, 120 ends 92, 93 eudaimonia 31 final value 53 happiness 93 humanity and morality 137–8

most complete end and final good 31, 80n39, 93 Nicomachean Ethics 80n39, 83, 84, 137 regress argument 80 self-love 83, 84 Aulus Gellius 137n1

B belief 90–2, 95, 101, 113, 125–6 benefit absolute value as benefit-independent 37, 41, 43, 45, 48–9, 50 ancient Greeks: supreme good as beneficial 43–4, 48, 52, 55 being of benefit to ourselves 79–80 good as benefit 44–7 good for as benefit 67–8 Kant: supreme good as benefit-independent 43, 48 non-instrumental/intrinsic benefit 60 Plato: good as benefit 44, 50, 52–3 regress argument: neither good nor bad 65–6 relational value and benefit-dependence 3 Socrates: good as notion of benefit 2, 44n15, 50, 136 value as benefit 8, 135 virtue and 67–8n19, 104, 106–7 see also good; good for; relational value ‘borrowing’ conception of value 8–9, 68, 73, 74n31, 74–5 Buss, Sarah 62–3, 89

C capacity 20, 110 absolute value and 36 exercise of 16–17, 19, 21, 25, 36, 128–9, 131 for final ends 102, 108

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

154 index capacity (cont.) for fitting ends 101 for having final ends 102, 108, 110 for having higher ends 94n14 for moral agency/moral capacity 17–18, 21 threshold capacity 19, 20 to lead good lives 81, 86, 100, 103, 108, 131 to pursue beneficial ends 102, 108 to value 88, 89, 96–8, 100, 110–11, 127, 128–9 valuing and 3, 25, 27–8, 40 care 77, 92, 98n18 self-care 82, 129 chain of dependence 57, 58, 59, 63, 66, 68, 75, 76, 79, 92 end of 61, 62, 65, 74, 79 infinite chain of dependence 58–9, 73, 75 valuer as final node in 61 see also dependence; regress argument Christianity 49, 131 common humanity 12–20, 25 value and 20–8 see also human being Conee, Earl 69–73, 74 positive worthwhile value 70n25 consequentialism 6, 11, 22, 24, 25n42, 53, 89 non-consequentialists 90 constructivism 36, 140, 142 Cooper, John 84–5

D Darwall, Stephen 26–7, 36n2 dependence 58, 60 asymmetrical dependence 141 metaphysical dependence 28 see also chain of dependence derivative value 31 desire 52, 82, 90, 125 relational value as grounded in desire 3n3, 40n8 dignity 41 human dignity 4–5, 20, 133 see also moral value

E egoism 52, 113, 115, 116 eliminativism 3, 5–6, 8, 119

ends 92, 93, 94, 129 capacity for final ends 102, 108 capacity for fitting ends 101 capacity for having final ends 102, 108, 110 capacity for having higher ends 94n14 capacity to pursue beneficial ends 102, 108 ends in harmony with ourselves 129 final ends valuing 99, 129 harmful ends 102, 103, 108, 129 having ends 80, 94 having final ends 95–6 higher-level ends 94n14 nefarious ends 94n14, 106 Engstrom, Stephen 39, 44n15 Enlightenment 1, 133–4n2 ethical behavior 9, 10, 11, 87, 112, 136 ethical theory 11, 15–16, 22, 55 eudaimonia 31 eudaimonist traditions 45n17, 53, 54 excellence 27, 28, 55, 84, 138 extrinsic value 31–2, 120

F Foot, Philippa 4, 28n46, 55, 67–8, 104–8, 134–5 Frankfurt, Harry 84n46, 92, 98n18

G Geach, Peter 5–6 genocide 66–8, 85 God 33, 49, 51, 131, 137 good as attributive adjective 6 being good in herself 59 property of being good 5–8 see also benefit; good for; good life; goodness; supreme good; value; virtue good for 108 being good for 1–2, 29–30, 58–9, 59, 60 being good for as dyadic form of value 74n31, 114 being good for ourselves 2, 78–9, 80, 84, 86, 87, 88, 97, 98, 103, 112, 126–7, 142 good as good for someone 2, 3, 8, 55, 118 good for as benefit 67–8

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

index 155 good for human beings 7, 47, 51, 55, 81, 118–22, 124–5, 127, 135 good for others 75, 76, 113, 116, 117 good period/good for distinction 30 Moorean challenge to relational value 112, 113 realism 60–1n5 realism about the good for human beings 102, 103–8 see also benefit; relational value good life 80, 103, 127 being good for ourselves 79, 80, 86, 142 capacity to lead good lives 81, 86, 100, 103, 108, 131 communal/political good and 85 as good for the person whose life it is 31, 53, 80, 86 living well 85, 103, 104, 116, 129 R* 80–1, 81, 85, 142 relational value 80–1 relationship with others 85 as successful pursuit of valuable relations/projects set for ourselves 100–2 valuing and 96–103 goodness 39, 44 as benefit 44, 54 distinctions in 9, 30n47 source of 139 see also good; value; virtue good period 31, 59, 59, 60n4, 68, 101, 119n14 good period/good for distinction 30 works of art as 63 see also non-relational value good simpliciter 7, 113, 114, 115n6, 117, 119, 130 see also non-relational value good theory of action 91 good will absolute value 41 Kant, Immanuel 37, 39, 40, 43, 51 as rectitude 39, 40, 41, 43, 46 as supreme good 39, 48 Greif, Mark 4–5, 12n3, 133

H happiness 39, 44, 107, 134 Aristotle 93 Kant, Immanuel 43n14, 50–1 most comprehensive good 43n14, 51

harm/harmful 44, 68n20, 98, 138 devoid of value or harmful 66–8 harmful ends 102, 103, 108, 129 preventing or mitigating harm 72, 76, 79 Henrich, Dieter 15n14, 48 Herman, Barbara 35 human being animals/human beings distinction 13–14, 21 anthropinon 138 as bearer of value 11, 23, 24, 29, 55–6, 108, 135 being of value in herself 56 ethical significance of 9, 25–6, 28 humanity, notion of 137–8 mortality 138 neither good nor bad 65–6 rational beings 32–3, 40, 47, 90n4, 139 see also common humanity; human relatedness; ourselves; valuer human good see good life; virtue humanitarianism 138 human relatedness 77–8 non-destruction and protection 128, 131 our responses to human beings 126–32 relating to people as centers of lives to which they bear special relation 77, 81–2, 128 relationship with ourselves 78–9, 83–4, 85 valuing people for themselves 77, 81 what we owe to others 78–9, 128, 129, 130, 131 human rights 4, 5, 35 Hurka, Thomas 67n18

I immoralist 104–5, 135 see also moral skeptic instrumental value 31, 59, 73 definition of 42 instrumental/non-instrumental value distinction 30, 42, 46 intrinsic value and 69–70, 71–2 price value 52, 56, 82 relational value and 30, 52, 53 use value 38, 52

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

156 index intrinsic value 69n23 instrumental value and 69–70, 71–2 intrinsic/extrinsic value distinction 31–2 intrinsic properties 88, 120 value of humanity as dependent on intrinsic properties 33

J justice Plato 46, 47, 51, 104, 106, 134 as supreme good 46 unjust person 106, 134

K Kagan, Shelly 33n52 Kant, Immanuel 28, 34, 36, 77, 129, 131 absolute value 37, 40, 41, 47–8 absolute value of humanity 1–2, 3, 9, 26, 29, 36, 37–8, 40–1, 43, 48, 56, 57, 61n5 categorical imperative 35, 48, 125 complete good 38n5, 51 Critique of Practical Reason 51 departure from the ancients 41–8, 55 duties to self 78, 94n14 Formula of Ends 35, 36 good will 37, 39, 40, 43, 51 Groundwork of the Metaphysics of Morals 35, 38, 41, 43n15, 49, 83n43 happiness 43n14, 50–1 higher-level ends 94n14 imperfect duty 94n14, 124 Kantian legacy 35–41, 48 Metaphysics of Morals 130 morality 40n9, 41, 48 moral law 29 race and slavery 56n34 relational value as grounded in desire 3n3 self-love 82–3 supreme good 38–40, 41, 43, 44n15, 48 Kantianism 35 Kantians 78, 109, 135 proto-Kantians 11, 25 Korsgaard, Christine 35, 89, 90n4 constructivism 36n3, 142 regress arguments 139–42 Kraut, Richard 2n2, 60n3, 118n11

L Langton, Rae 140, 142n12 love 66n16, 130 see also self-love

M market value/price 52, 53, 55 maximization 23, 90 meta-ethics 36, 37, 61, 88n1, 105n30, 136, 140, 142 metaphysics 3, 50, 86, 109, 113, 136 infinite regress and the metaphysical principle 68–74 metaphysical dependence 28 metaphysical structure of value 57, 114 Moore, G. E. 3, 4, 36n3, 50n28, 62, 71 instrumental and intrinsic value 69 Moorean challenge: normativity of relational value 9, 10, 112, 113–26 moral agency 17–18 morality 133–4n2, 137–8 Kant, Immanuel 40n9, 41, 48 moral law 29, 39, 40n9, 131 moral obligation 134 moral philosophy 4, 38, 128, 133, 134 Kant’s moral philosophy 35, 47, 48 modern moral philosophy 8, 55 moral psychology 47 moral skeptic 133–4, 135 see also immoralist moral status 14, 21, 110 doctrines of 20n33 moral value 52 see also dignity Murdoch, Iris 129

N Nagel, Thomas 15, 18, 22, 32n51, 120, 121, 122–4 aesthetics and works of art 63n10, 64n13 naturalism 28n46, 47, 105, 107 Nietzsche, Friedrich W. 105 non-derivative value derivative/non-derivative value distinction 31 value of humanity as 33 valuing people for themselves 81 non-instrumental good for 31, 42–3, 59–60, 73, 86, 88

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

index 157 non-instrumental value 42–3 instrumental/non-instrumental value distinction 30, 42, 46 value of humanity as 33 non-relational value 9, 113, 135, 139–40 definition 29 non-relationally valuable 3, 8, 9, 29, 74, 87, 89, 97, 116 realist and anti-realist accounts of 61n5 relational/non-relational value distinction 29–30 as ‘source’ of value 74n31 valuer 73–4, 86, 89 see also absolute value; absolute value of humanity; good period; good simpliciter; value in itself normativity 1, 16, 18, 24, 134–5 Kant, Immanuel 52 Moorean challenge 9, 10, 112, 113–19 normative reasons and values 11, 16, 24–5, 86, 117 normative significance 9, 66, 68n20, 72, 117, 126 normativity of relational value 119–25 value of humanity 33 Nozick, Robert 89, 128n24

O objectivity 15, 18 O’Neill, Onora 35 ourselves being good for ourselves 2, 78–9, 80, 84, 86, 87, 88, 97, 98, 103, 112, 126–7, 142 being beneficial to 79–80 duties to 78, 94n14 good life 2, 79, 80, 142 ‘Know thyself!’ 138n4 reflexive relation 79–82, 79, 81 relational value of humanity 79 relationship with ourselves 2, 27, 78–9, 83–4, 87, 112, 128, 141 self-care 82, 129 self-indulgence 84n46, 85 see also human being; R*; self-love

P pain 22–3, 69–71, 120, 122 Penner, Terry 45n17 philanthropy 138 philosophy 15–16

Plato 45n17, 120, 136 Apology 44 Euthydemos 44 Form of the Good 44–5 good as benefit 44, 50, 52–3 justice 46, 47, 51, 104 Lysis 65–6 Meno 43–4n15, 44 Protagoras 44 regress argument 65–6 Republic 44, 46, 47, 134 unjust person 106, 134 pleasure 6, 22–3, 60, 96, 102, 120 pluralism 107, 117–18, 119, 121, 124 political liberalism 12 political philosophy 12 positive worthwhile value 70n25 potential see capacity price value see instrumental value Prichard, H. A. 134 promote 113

R R* (reflexive relation*) 81, 81, 97, 101, 102, 126, 142 good life and 80–1, 81, 85, 142 Railton, Peter 118n11 Rawls, John 13, 16n21, 17, 21, 35, 45n17 Raz, Joseph 7, 24, 59, 61n6 caring for someone for their own sake 77 good life 100–2 regress argument 58 relational value 64n12 realism 3, 16, 36, 37, 117, 131, 142 absolute value 50 anti-realist views 37, 61n5, 121, 142 good for 60–1n5 good for human beings 102, 103–8 relational realism 118–19, 120–1 source of value 140 reasons 6, 91, 112, 115–16, 135 acting for reasons 27 agent-neutral reasons 32n51 agent-relative reasons 32n51, 115 externalism about reasons 116, 118, 125, 126 normative reasons and values 11, 16, 24–5, 86, 117 reason-giving and value 8, 57, 92, 94, 112, 119, 125

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

158 index reason (cont.) reasons for action 91, 123, 125 relational value and 52n30 values and 8, 23, 32, 113, 139 reflexive relation 79–82, 79, 81, 86, 141 see also R* Regan, Donald H. 115n6 regress argument 58–61, 59 Korsgaard, Christine 139–42 see also chain of dependence regress argument, evaluation of 62–79 circular structure 74–9, 75, 76 devoid of value or harmful 66–8 infinite regress and the metaphysical principle 68–74 neither good nor bad 65–6 opponents to the argument 62, 71 proponents of the argument 62, 66–7, 70, 71, 89 reverse priority 60n4, 62–4, 63 see also chain of dependence; reflexive relation relata 47, 73–4, 114 relational value 40, 74, 87, 127, 139–40 absolute/relative value distinction 29, 32–3, 40 benefit-dependence 3 ‘borrowing’ model 8–9, 68, 73, 74n31 definition 29–30 final value as relational value 53 grounded in desire 3n3, 40n8 instrumental value and 30, 52, 53 Kant, Immanuel 3n3, 40, 52 Moorean challenge 9, 10, 112, 113–19 normativity of relational value 119–25 relational/non-relational value distinction 29–30 valuer 75–6 see also benefit; good for; R*; regress argument, evaluation of relational value of humanity 2, 33, 79, 109, 136 being beneficial 8, 55 being beneficial to ourselves 79–80 being good for ourselves 127 good as good for someone 2, 3, 55 hypothesis on 55, 87 interdependence relations 2, 116 objections to 8–9 relationally valuable 3, 8, 29, 115 value of humans as continuous with value of other things 2, 8, 136

relational value theory 117–18, 119, 125 relative value see relational value relative worth see relational value respect 13, 23, 26–7, 40n9, 130–1 appraisal respect 27 recognition respect 27 self-respect 120, 130 Rosati, Connie 60n3, 74n31, 114 Ross, W. D. 68

S Scanlon, T. M. 78, 89–90 ‘buck-passing’ account of value 119–20n14 Schadewaldt, Wolfgang 138 Scheffler, Samuel 90–1, 92, 94, 95 Schiller, Friedrich 82–3 Schneewind, Jerry 141 self-love 82–5, 141 Aristotle 83, 84 Kant, Immanuel 82–3 as term of disapprobation 84 see also ourselves Sellars, Wilfrid 15 Sensen, Oliver 36n3, 50n28 Setiya, Kieran 91n8 Socrates 45, 46, 65, 136 good as notion of benefit 2, 44n15, 50, 136 Stocker, Michael 82 Stoics 45n17, 53, 77 subjectivism 107 supreme good ancient Greeks: supreme good as beneficial 43–4, 48, 52, 55 good will as 39, 48 justice as 46 Kant, Immanuel 38–40, 41, 43, 44n15, 48 morality as 41, 48 Plato 46 ‘summum bonum’ 38n5 see also good

T Thomson, Judith Jarvis 6–7, 8 Timmermann, Jens 40n9

U use value see instrumental value utilitarianism 22, 42n13

OUP CORRECTED PROOF – FINAL, 20/11/2019, SPi

index 159 V value 1, 10, 37 affecting nature of 2 basis of value 29, 88, 89, 103, 120 as benefit 8, 135 common humanity and 20–8 distinctions in 28–33, 87 final value 53, 80 ground of value 88, 119–20 normative reasons and 11, 24–5, 86 as object and subject related 120 quantifiability, comparability, fungibility of 53–4, 55, 71 as relation 3 response to 22, 23, 53, 90 source of 74n31, 139–41, 140 structure of 57, 75, 77, 86, 87, 114 value, explanation of 9, 32, 64, 66, 141–2 explanation of the value of the basis 29, 120 relational explanation of the value of humanity 79, 87, 103, 124–5 value in itself 59, 69, 70, 72, 75, 86 see also non-relational value value of humanity 9, 25–6, 28, 34, 108, 112, 137 capacity to value 110–11 concept of 1 human being as bearer of value 11, 23, 24 intrinsic properties 33 non-derivative value 33 non-instrumental value 33 normativity 33 securing grounds for 134 as unconditioned condition of all other value 141 see also absolute value of humanity; human being; relational value of humanity valuer 3, 73, 88–96, 99, 112 being good for other valuers 76 being good for ourselves 88 being of value in herself 59, 61 definition of 59 as end-in-herself 59, 59 as final node in chain of dependence 61 as locus of value 73, 127, 139, 141–2 non-relational value 73–4, 86, 89 rational being as valuer 139

regress argument 61 relational value 75–6 as source of value 139, 140 see also human being value theory 29, 87, 112, 135 dogmas in 23, 53 relational value theory 117–18, 119, 125 valuing 3, 89, 127, 142 as basis of human value 89, 97 capacity and 3, 25, 27–8, 40 capacity to value 88, 89, 96–8, 100, 110–11, 127, 128–9 conditions for 90–5 emotional susceptibility 94–5 final ends 92–3, 94–6 final ends valuing 99, 129 good life and 96–103 long-range deliberation 93, 94 motivational stability 92, 94 reason-giving 92, 94 valuing in appropriate ways 103, 121 Velleman, David 58, 61n6, 66n16 virtue 17, 44, 134–5 benefit and 67–8n19, 104, 106–7 ethical virtue 104–106 good will/supreme good as 39 happiness and 107 most comprehensive good 43n14 relationship with ourselves 79 self-love and 83 Vogt, Katja Maria 6n11, 51, 98n18

W Waldron, Jeremy 13, 13n6, 17, 19–21, 25, 29, 110 what we owe to others 78–9, 128, 129, 130, 131 Williams, Bernard 12n1, 14–15, 16, 55, 89, 131 Wittgenstein, Ludwig 54 Woolf, Virginia 137 work of art 53, 63–4, 66–7, 88 as bearer of value 24, 42–3n13 as good period 63 non-instrumental value of 30, 42, 60, 88 valuer and 73 World War II 4, 133