Hume on Motivation and Virtue: New Essays (Philosophers in Depth)

Hume on Motivation and Virtue

Philosophers in Depth Series Editors: Stephen Boulter and Constantine Sandis Philosophers in Depth is a series of themed edited collections focusing on particular aspects of the thought of major figures from the history of philosophy. The volumes showcase a combination of newly commissioned and previously published work with the aim of deepening our understanding of the topics covered. Each book stands alone, but taken together the series will amount to a vast collection of critical essays covering the history of philosophy, exploring issues that are central to the ideas of individual philosophers. This project was launched with the financial support of the Institute for Historical and Cultural Research at Oxford Brookes University, for which we are very grateful. Titles include: Charles R. Pigden (editor) HUME ON MOTIVATION AND VIRTUE

Forthcoming titles: Leonard Kahn MILL ON JUSTICE Arto Laitinen and Constantine Sandis HEGEL ON ACTION Katherine Morris SARTRE ON THE BODY Sabine Roeser REID ON ETHICS Daniel Whiting LATER WITTGENSTEIN ON LANGUAGE

Philosophers in Depth Series Standing Order ISBN 978-0-230-55411-5 Hardback 978-0-230-55412-2 Paperback (outside North America only) You can receive future titles in this series as they are published by placing a standing order. Please contact your bookseller or, in case of difficulty, write to us at the address below with your name and address, the title of the series and the ISBN quoted above. Customer Services Department, Macmillan Distribution Ltd, Houndmills, Basingstoke, Hampshire RG21 6XS, England

Hume on Motivation and Virtue Edited by

Charles R. Pigden University of Otago, New Zealand

Selection and editorial matter © Charles R. Pigden 2009 Chapters © their individual authors 2009 All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No portion of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, Saffron House, 6-10 Kirby Street, London EC1N 8TS. Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The authors have asserted their rights to be identified as the authors of this work in accordance with the Copyright, Designs and Patents Act 1988. First published 2009 by PALGRAVE MACMILLAN Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS. Palgrave Macmillan in the US is a division of St Martin’s Press LLC, 175 Fifth Avenue, New York, NY 10010. Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world. Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries ISBN-13: 978-0-230-20527-7

hardback

This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. Logging, pulping and manufacturing processes are expected to conform to the environmental regulations of the country of origin. A catalogue record for this book is available from the British Library. A catalog record for this book is available from the Library of Congress. 10 9 8 7 6 5 4 3 2 1 18 17 16 15 14 13 12 11 10 09 Printed and bound in Great Britain by CPI Antony Rowe, Chippenham and Eastbourne

TO ZENA For putting up with a philosopher for thirty-two years

This page intentionally left blank

Contents Preface and Acknowledgments

ix

A Note on References to Hume and Locke

x

Notes on the Contributors

xi

Introduction Charles R. Pigden

1

1

Expressivism, Motivation Internalism, and Hume Richard Joyce

30

2

Is Hume Inconsistent? – Motivation and Morals Norva Y. S. Lo

57

3

If Not Non-Cognitivism, Then What? Charles R. Pigden

80

4

The Motivation Argument for Non-Cognitivism Michael Smith

105

5

Experiences of Value Graham Oddie

121

6

Hume and the Debate on ‘Motivating Reasons’ Constantine Sandis

142

7

Against All Reason? Scepticism about the Instrumental Norm Stephen Finlay

155

Why Internalists about Reasons should be Humeans about Motivation Kent Hurtig

179

8

9

Humean Sources of Normativity Herlinde Pauer-Studer

186

10 Two Kinds of Normativity: Korsgaard v. Hume Luke Russell

208

11 What Kind of Virtue Theorist is Hume? Christine Swanton

226

12 Kinds of Virtue Theorist: A Response to Christine Swanton Annette Baier

249

vii

viii

Contents

13 Reply to Baier Christine Swanton

259

14 Hume on Justice Rosalind Hursthouse

264

Bibliography

277

Index

287

Preface and Acknowledgments There must be mistakes that it is possible for an editor to make that I haven’t made and misfortunes that can befall an editor that haven’t befallen me, but if so, I am not anxious to make those mistakes or experience those misfortunes. So first of all, I would like to thank my contributors collectively, both for their work, and for their patience in putting up with a project that has taken such a very long time to come to fruition, partly because of my blunders. Some, who were just starting out in 2003, are now not only rising but risen stars, and I apologize to them in particular for occluding their sparkle for such a long time. As for my older contributors with plenty of publications to their credit, it is still a bit annoying to have some of that credit deferred, and I would like to apologize to them too. I was forced by reasons of space to omit a couple of papers I would have liked to have included. My apologies to those I have disappointed and inconvenienced. My thanks to Palgrave Macmillan for rescuing the project when it seemed about to fall through, and to my colleagues at Otago – particularly Peter Anstey and James Maclaurin for their judicious combination of nagging and encouragement. I would like to thank my research assistant Rebecca Thompson for her editorial services and the research cluster on Early Modern Thought at the University of Otago for providing the wherewithal to pay for them. Thanks to Annette Baier, Josh Parsons and Helen Beebee for useful discussions. Finally, my family. Even those bachelor philosophers of long ago – such as David Hume himself – were often heavily reliant on family support and I am not one of those bachelor philosophers. So thanks to my wife, Zena, my mother, Jean, and my (now grown-up) children, Guy Jemima and Abigail, for their love, help and encouragement.

ix

A Note on References to Hume and Locke Generally, we employ the name/date system for references, but with Hume himself and one or two others we employ a minor variant of the Hume Studies system, now widely used by Hume scholars. With Hume’s Treatise of Human Nature, the letter T is followed by the book, part, section and paragraph numbers with a page reference to the famous Selby-Bigge/Nidditch edition of 1978 following a forward slash. Thus the reference for the famous is/ought paragraph is T, 3.1.1.27/469–70. References to the Abstract will be by paragraph number followed by a forward slash and page reference to the reprint at the back of Selby-Bigge/Nidditch. With the Enquiry Concerning Human Nature and the Enquiry Concerning the Principles of Morals, the letters EHU or EPM will be followed by the section (part) and paragraph numbers, with a page reference to the Selby-Bigge/ Nidditch edition of 1975 following a forward slash. The appendices to the EPM will be referred to as EPM, App 1 etc. and the Dialogue as EPM, Dialogue, with paragraph numbers and page references to Selby-Bigge/Nidditich. Thus the reference for the famous “consign them then to the flames” passage is ECU, 12.3.34/165; the reference for “Have the gods forbid self-murder?” is EPM, Dialogue, 35/335. Since most modern editions of Hume’s works have either part, section and paragraph numbers or the Selby-Bigge/Nidditch page numbers in the margins, I hope this will make things easy for readers. References to the Dialogues Concerning Natural Religion (Dialogues) and the Natural History of Religion (NHR) follow the same format with page references to Gaskin’s Oxford Classics edition of 1993, since Oxford is the only publisher with the wit to serve up both of Hume’s principal works on religion in the one volume (though unfortunately the readers must put in the paragraph numbers themselves). References to Letters are to the two-volume 1932 edition of Hume’s letters, edited by Greig, by volume, letter and page number (following the forward slash). References to Hume’s Essays are to Miller’s revised edition of 1987, published by the Liberty Fund, by part, essay and, where appropriate, page number, following the forward slash. For Hume’s History of England we employ W. D. Todd’s excellent Liberty Fund edition of 1983 with the format History, 1.3/137 (Volume I, Chapter 3, page 137.) For the Dissertation on the Passions we use the Clarendon edition of 2007 edited by Beauchamp. References to Locke’s Essay Concerning Humane Understanding are to the 1975 Clarendon edition, edited by Nidditch, and follow the rubric Essay, 4.3.18/549 (Book 4, Chapter 3, paragraph 18/page 549). x

Notes on the Contributors Annette C. Baier was educated at the universities of Otago and Oxford. She writes on Hume and on trust-centered ethics. She has taught at the universities of Aberdeen, Auckland, Sydney and Pittsburgh. In retirement she lives in her native New Zealand. Stephen Finlay is an assistant professor at the University of Southern California and specializes in metaethics. Originally from New Zealand, he completed his PhD at the University of Illinois at Urbana-Champaign in 2001, and now lives in Los Angeles with his wife, Sarah, and two daughters. Rosalind Hursthouse did her undergraduate degree and MA in philosophy at the University of Auckland, and a BPhil and DPhil at Oxford, where she taught for six years before becoming a lecturer at the Open University. She found this so rewarding and inspiring that she stayed there happily for the next 25 years, until even more happily, she returned to the Auckland Department as Professor. She is the author of Beginning Lives and a series of articles on ethical questions and the virtues, culminating in her book On Virtue Ethics. Kent Hurtig was born in Sweden and studied philosophy at Wichita State University in Kansas, before going on to do an MLitt and a PhD at the University of St Andrews, Scotland. Since 2005 he has taught philosophy at the University of Stirling, Scotland. He has published on topics in metaethics, practical reason and rationality. Richard Joyce was born in England and raised in New Zealand. After studying at the University of Auckland he went on to graduate school at Princeton University. He has been a lecturer at the University of Sheffield and a Research Fellow at the Australian National University. He is presently a Research Fellow in the Philosophy Department at the University of Sydney. His principal areas of research are metaethics and moral psychology, though he is also interested in practical reason, philosophy of biology and the emotions. He is author of The Myth of Morality (CUP, 2001) and The Evolution of Morality (MIT Press, 2006). Norva Y. S. Lo was born in Hong Kong and received her PhD in Philosophy from the University of Western Australia in 2002. After two years as a postdoctoral fellow she became a lecturer in the Department of Philosophy, La Trobe University, Melbourne, Australia, in 2006. She has published in Inquiry, Proceedings of Aristotelian Society, Environmental Ethics, Environmental Values xi

xii Notes on the Contributors

and the Stanford Encyclopedia of Philosophy. Her research interests include Moral Philosophy, Experimental Philosophy, Environmental Philosophy and David Hume. Graham Oddie was born and raised in Timaru, in Aotearoa (New Zealand). After stints as a shepherd and a truck driver, he left for the University of Otago where he was entranced by Pavel Tichy’s demolition of Popper on truthlikeness. He wrote a PhD on verisimilitude at the University of London (LSE) and which morphed into his first book G. Oddie (1986) Likeness to Truth (Dordrecht Reidel). He has taught at Otago (as lecturer), Massey (as professor) and is currently Professor of Philosophy (and Associate Dean of Humanities) at the University of Colorado at Boulder. He is interested in metaphysics and value theory with the odd foray into the philosophy of science and philosophical logic. He is the author of Value, Reality and Desire (Oxford University Press, 2005). Herlinde Pauer-Studer is Associate Professor for Philosophy at the University of Vienna. She is the author of Das Andere der Gerechtigkeit (Akademie Verlag, 1996); Autonom Leben (Suhrkamp, 2000); Einführung in die Ethik (UTB, 2003) and Commentary on David Hume, Of Morals (Suhrkamp, 2007). Charles R Pigden was born in England and studied philosophy at Cambridge before going on to do a PhD at La Trobe University, Melbourne, Australia. Since 1988 he has taught philosophy at Otago University, Dunedin, New Zealand. He has published on a wide range of topics from conspiracy theories to the reality of numbers, but, if pressed, will admit to being a metaethicist with special interests in Russell, Moore and Hume. He is one of the very few academics to have published a philosophical dialog in blank verse (“Complots of Mischief”). Luke Russell completed both a BA and a PhD at the University of Sydney, where he is now Lecturer in the Department of Philosophy. He has published papers on various topics in ethics, including normativity, virtue and evil. Constantine Sandis is a lecturer in Philosophy at Oxford Brookes University and NYU in London. He works primarily on the philosophy of action and its explanation, and is the author of numerous related articles and a book, The Things We do and Why We Do Them (Palgrave Macmillan). He is the editor of New Essays on the Explanation of Action (Palgrave Macmillan) and is currently co-editing (with Timothy O’ Connor) A Companion to the Philosophy of Action (Wiley-Blackwell). Michael Smith is Professor of Philosophy at Princeton University. His books include The Moral Problem (1994), Mind, Morality, and Explanation: Selected Collaborations (2004) (with Frank Jackson and Philip Pettit), and Ethics and the A Priori: Selected Essays on Moral Psychology and Meta-Ethics (2004).

Notes on the Contributors

xiii

Christine Swanton is a professor in the philosophy department at the University of Auckland. Her book Virtue Ethics: A Pluralistic View (Oxford University Press, 2005) is in paperback. She now works on virtue ethics (including the ethics of Nietzsche and Hume), role ethics, moral particularism and right action. She has published widely in these areas.

This page intentionally left blank

Introduction Charles R. Pigden

0.1 Prologue We come to Hume to argue, not to praise him For Hume’s philosophy lives after him And since it’s not interrèd with his bones Like dead men’s thought that lives, it lives because We think that it contains important truths (Truths that illuminate the nature of Morality, the mind and reason too, The “sentiments” that make us do our duty, The role of passion and the sense of beauty); Or think that Hume is wrong, but that his faults Are kissing cousins to important truths (So, loving truth, we love the truths they’re like); Or think him wrong, but think that when he errs They’re mighty errors, great and grand mistakes, That represent temptations of the mind Which must be wrestled with and overcome That we may learn to see these things aright. But if we seek to learn from David Hume, We cannot hope to learn without debate, His ideas must be tested in the fire Of criticism if we are to find The philosophic gold that lies within. For philosophic truth is our concern Above all else; less so, historic facts About a thinker’s thought, however great. For though we seek to know great David’s mind, We want to know what David thought so we Can find if we agree or disagree Or better still, if David’s wrong or right. 1

2 Introduction

But right or wrong, we hope that Hume repays The homage that we pay – attention first, (“What does he say? What does he mean by it? What was the context of that barbed remark? Who were his targets? And who influenced him? What larger ends did David have in view When arguing for this doctrine or for that?”) Next comes analysis (“Exactly how Is this specific inference meant to work? Does Hume appeal to missing premises, Obvious to him but not so much to us?”) Next comes critique (“Well then, is David right? Does this conclusion follow from that claim? And what about these claims – are they all true?”) Sometimes that’s it, since Hume, it seems, is right, (“So, Hume’s correct and reason’s passion’s slave!”) Sometimes amendment follows (“It is clear, That in this case the argument is lame, But nonetheless another argument Can be constructed which would prove the point From something like great David’s premises. ‘No Ought From Is’ – amended – we can prove, Employing logic quite unknown to Hume, So he’s not right, but yet not wholly wrong.”) Sometimes what follows is an inference: (“It seems that in this instance Hume’s correct – At least if he’s corrected he’s correct – But what then follows, what does this suggest About the status of our moral claims?”) Sometimes amendment seems impossible But still we learn by seeing why that’s so (“So David’s wrong – the argument’s a dud, And rational reconstruction is in vain, But if he’s wrong and can’t be put to rights Still, Hume’s missteps can lead us to the light, For from his errors we can lessons take – This is what we’ve learned from his mistake …”). History is a necromantic art At least when history serves philosophy; We aim to resurrect the mighty dead, To grapple with their thought and thus to learn, Contesting boldly their contentious claims, Whether to lose the contest or to win.

Charles R. Pigden

No Faustus we, whose Mephistopheles Could only summon silent ghosts, for we Must needs have ghosts that speak, since it’s their thoughts That interest us. We need no Helen dumb To make us all immortal with a kiss– Immortal thoughts suffice, but since those thoughts Still live, although the thinker is long dead, Because they live, they move, migrate and change Thus undergoing metamorphosis. The thoughts which have an influence on men’s minds (And women’s too – let’s not be sexist here!) Allegedly from some philosopher May not be quite the thoughts that stirred his brain. Thus “Humean” doctrines, “Humean” arguments, That dominate the philosophic scene, Are sometimes not the thoughts of David Hume – That does not mean they should not be discussed, For thoughts, mutating, may perhaps become More like the truth, or if mistaken still, More like the happy errors from the which We learn the most by thinking why they’re wrong. Is Reason passion’s slave in David’s sense? Perhaps not – but perhaps in Michael Smith’s – As true philosophers, we’d like to know! Are moral judgments meant to be truth-apt, The sorts of things that can be true or false? So David thought (or so I think he thought) But that is not the view of many now Who think Hume’s arguments prove otherwise. If arguments derived from David Hume Can prove what Hume perhaps did not believe, That may still be a matter of some pith And moment that deserves to be discussed. Ideas, mutating, sometimes are improved And even if the change is for the worse May still acquire a major influence, And consequently merit some debate. This is a book addressed to “Humean” themes; Sometimes they’re themes direct from David’s brain, His arguments or theses he advanced; Sometimes they’re themes derived from David’s thought – Arguments based on arguments of his, Theses quite like the theses he advanced,

3

4 Introduction

Sometimes they’re consequences of his thought, Deductions he may not have followed through; Sometimes we can’t be sure, when David’s texts Bear two interpretations (if not more!) And we cannot be certain which he meant; But whether they’re his thoughts unmodified Or thoughts derived at some removes from his They’re thoughts which dominate today’s debates. Ther’re few philosophers are cited more. At least in ethics; for there’s scarce a book In ethics, meta-ethics, action theory Or what is called “moral psychology” That does not doff its hat to David Hume. That’s our excuse (if an excuse is needed) For turning once again to his ideas. There’s gold in Humean hills or thereabouts And these are our attempts to dig it out. O Hume, great David, thou art mighty yet! Thy spirit walks abroad; now thy thoughts thrive They live more now than when thou wast alive!

0.2 Three themes and a conference This book deals with three themes from Hume’s moral philosophy which loom large in contemporary ethical thought: (1) Hume’s famous argument that “Reason is, and ought only to be, the slave of the passions, and can never pretend to any other office than to serve and obey them” (T, 2.3.3/413–18); (2) the Motivation (or Influence) Argument (T, 3.1.1.5–10/456–8) which purports to prove that “since morals … have an influence on the actions and affections” and since “reason alone [as Hume claims to have already proved] can never have any such influence,” it follows that [moral distinctions] cannot be derived from reason; and (3) Hume as a virtue theorist. The various essays deal with Hume himself and with “matters arising” from the Humean agenda. The book is loosely based on a mini-conference held at the University of Otago in January 2003, Hume, Motivation, “Is” and “Ought”; “loosely” because the collection includes several essays not presented at the conference and excludes some which were. Most of the excluded essays – the No-Ought-From-Is papers – have been hived off in a separate collection, Hume

Charles R. Pigden

5

on “Is” and “Ought” (Pigden, 2010). This is partly a matter of convenience and partly a matter of conviction. For it is my editorial belief that the Motivation Argument has a lot less to do with the famous No-Ought-From-Is paragraph (T, 3.1.1.27/469–70) than is commonly supposed. Of course they both appear in the same section of Hume’s Treatise (3.1.1) and are both designed to prove (or to help to prove) that “moral distinctions are not derived from reason” (whatever precisely that means). But they are distinct arguments and the one does not depend on the other. But this is far from being the conventional wisdom among philosophers, so before discussing what this book is about I am going to devote a couple of pages to what it is not about to get the issue out of the way. I shall have to be dogmatic as I aim to be brief. Readers who want the gory details can turn to the other collection.

0.3 Clearing the decks: No-Ought-From-Is “In every system of morality which I have hitherto met with”, says Hume, “the author proceeds for some time in the ordinary way of reasoning, and establishes the being of a God, or makes observations concerning human affairs; when of a sudden I am surprized to find, that instead of the usual copulations of propositions, is, and is not, I meet with no proposition that is not connected with an ought, or an ought not. … But as this ought, or ought not, expresses some new relation or affirmation … [it] seems altogether inconceivable, how this new relation can be a deduction from others, which are entirely different from it.” And the implication would appear to be that this not only seems inconceivable but actually is inconceivable, and that you cannot deduce moral conclusions from non-moral premises. Note that Hume is speaking the language of logic. In every other instance, when Hume talks of “copulation” it is sex that he has in mind. Here alone he is talking about the “copula,” a technical term in the logic of the day which was seldom employed in any other connection. This suggests that in the No-Ought-From-Is passage, Hume is making a logical point. Furthermore he is appealing to a commonplace of eighteenth century logical theory – that the conclusion of a valid inference is contained in the premises. So if there is new matter in the conclusion that is not contained in the premises – namely the copulations of propositions ought and ought not – then the inference cannot be “just and true,” that is logically valid. Notice that all that is required for this point to hold (at least so far as eighteenth century logical theory is concerned) is a prima facie distinction between moral words and others. There need be no deep difference between semantic kinds. And this is fortunate since Hume probably did not believe in such a difference. After all, he explicitly compares vice and virtue “to sounds, colours, heat, and cold,” (T, 3.1.1.26/469) which, suggests that in saying “X is virtuous” we are saying something similar to “X is red.” This is confirmed by a famous passage in the EPM, App.1.10/289: “The hypothesis which we

6 Introduction

embrace is plain. It maintains, that morality is determined by sentiment. It defines virtue to be whatever mental action or quality gives to a spectator the pleasing sentiment of approbation; and vice the contrary. We then proceed to examine a plain matter of fact, to wit, what actions have this influence.” Here Hume gives an analysis of “virtue” as a causal concept, defining it via its causal role, with a view to establishing, by empirical research, that virtues [mental actions or qualities which give to a [suitably qualified] spectator the pleasing sentiment of approbation] are also qualities “useful or agreeable to the person himself, or to others” (EMP, 9.1.1/268). Thus Hume seems to think (to use post-Humean terminology) that it is analytic that a virtue is a quality that arouses in a [suitably qualified] spectator the pleasing sentiment of approbation, and furthermore he needs it to be analytic since it is hard to see how he could establish the point by empirical enquiry. Rather he has to take it as read before the empirical enquiry into “what actions have this influence” (that is, of exciting approbation) can get off the ground. Indeed, following his sometime philosophical hero Hutcheson (Hutcheson, 2002, p. 146), Hume actually defines the word “obligation,” if not the word “ought”: “All morality depends upon our sentiments; and when any action or quality of the mind pleases us after a certain manner we say it is virtuous; and when the neglect or nonperformance of it displeases us after a like manner, we say that we lie under an obligation to perform it” (T, 3.2.5.4/517). Thus to say that A ought to do B is to say that if A does not do B, this act of omission would excite the sentiment of disapprobation in a suitably qualified spectator. The upshot is that in the Is/Ought passage, Hume is probably not trying to argue that non-moral premises do not entail moral conclusions, where a set of premises A entails a conclusion B if B can be logically derived from A with the aid of (perhaps) unstated analytic bridge principles. Thus NoOught-From-Is is not a thesis about entailment at all, and neither implies nor presupposes a fundamental distinction between the moral and the non-moral. In particular, it neither implies nor presupposes any kind of non-cognitivism. So what is the point of it then? Well, Hume’s over-arching aim in Treatise, 3.1.1 is to argue that moral distinctions are not derived from reason in any sense. But to prove this larger claim he has to prove the subsidiary thesis, that the basic principles of morality cannot be demonstrated. That they could be so demonstrated was a popular view in the late Seventeenth and early Eighteenth Centuries. Now for moral truths to be demonstrable they would have to be deduced from self evident truths. By the end of 3.1.1, Hume has already proved, to his own satisfaction at any rate, that there are no self evident moral truths. This still leaves open the possibility that the truths of ethics might be derived from self evident truths of some other kind, a possibility suggested by Locke (1975, 4.3.1.18, p. 549). The point of No-Ought-From-Is is to foreclose this option. If there are no self evident

Charles R. Pigden

7

moral truths and if moral conclusions cannot be deduced from non-moral premises, then moral truths are not demonstrable, whatever Locke, Clarke and Spinoza may have to say to the contrary.

0.4 Reason, the slave of the passions So much for what this book is not about, now for the main themes. One of the most talked about sections in Hume’s Treatise is 2.3.3 “Of the Influencing Motives of the Will.” This is the locus for Hume’s famous claim “that reason alone can never be a motive to any action of the will” (T, 2.3.3.1/413) or (more picturesquely) that “Reason is, and ought only to be, the slave of the passions, and can never pretend to any other office than to serve and obey them” (T, 2.3.3.4/415). Call this “the Slavery of Reason Thesis” or the SRT. What does he mean by this claim? What are his arguments? What part does this thesis play in his overall polemic against moral rationalism? And what is its relevance to current concerns? 0.4.1 The slavery of reason thesis: Meaning Let’s start with meaning. What Hume means by “reason alone can never produce any action, or give rise to volition” depends on what he means by “reason” and “alone.” With respect to “reason” there are three possibilities: [1] “Reason” either stands for our belief-forming faculty or is a collective name for our belief-forming faculties taken together (especially when they are properly exercised or are operating without error). Thus the Slavery of Reason Thesis amounts to the claim that “beliefs by themselves can never produce any action or volition” or that “beliefs alone cannot motivate.” [2] Reason is a sub-faculty of [1], namely the intellectual organ by which we infer new beliefs (whether the inference is inductive or deductive). Thus the Slavery of Reason Thesis amounts to the claim that “inferred beliefs by themselves can never produce any action or volition” or that “inferred beliefs alone cannot motivate.” [3] Reason is a sub-faculty of [2], namely the intellectual organ which derives new beliefs from old beliefs, or, at least, that derives new beliefs from ideas rather than impressions (where ideas are copies or combinations of copies of prior sensations or passions). Thus the Slavery of Reason Thesis amounts to the claim that “beliefs inferred from ideas can never produce any action or volition by themselves” or that “beliefs inferred from ideas cannot motivate alone.” There is textual evidence for all three interpretations. For the moment we can leave the matter open. What about “alone”? Here there is a lot less ambiguity. When Hume says that “reason alone can never produce any action, or give rise to volition,” what

8 Introduction

he means is that reason-derived beliefs cannot produce an action or a new desire without the aid of a pre-existing passion. Passions – that is pre-existing passions – are necessary if the products of reason are to result in action. As Rawls puts it in his Lectures on Moral Philosophy, “the passions specifying the final ends in deliberation … must be passions we have and may be moved by now, at the time of deliberation and before we carry out the action.” This means that “the bare knowledge of our future passions does not move us now, unless that knowledge guides, or connects with, some passion that we have now” (Rawls, 2000, p. 34). The belief that unless I start saving now I am likely to have a bad time in my old age (at which point I will passionately want to have more money) isn’t going to move me unless I currently care about my future self. I must now want that my future wants be satisfied, if the knowledge of those wants is to move me to action in the present. Luckily, most of us are in fact endowed with a “general appetite to good, and aversion to evil” (T, 2.3.3.8/417), a desire for our long-term survival and satisfaction, which means that reason-derived beliefs about our future wants sometimes translate into immediate action. But this a basic instinct: it is not a want that is derived from reason but an appetite that we must presuppose if reason-based beliefs about our future wants are to have any effect on our current conduct. A reason-based belief, even a belief about future passions, can neither prompt us to action nor create a new desire without the aid of a passion that is already in existence. But what kind of impossibility are we talking about here? For Hume, I think, the impossibility has to be causal. It can hardly be conceptual since it is part of his official theory that it is conceptually possible for anything to cause anything. So the idea is that it is causally impossible (given the basic structure of the human mind) for a reason-based belief to produce an action or a volition without the cooperation of a passion. But what does Hume mean by a “passion”? That is a long and tangled tale. Fortunately for us, the details of Hume’s theory are largely irrelevant to the concerns of this collection. For Hume’s latter-day disciples tend to equate passions with desires, and when Hume himself talks about the passions in this connection, it is often desires that he is thinking of. Reason produces an action or gives rise to a new volition (that is a movement of the will such as a new desire) by pointing the way to realize an end set by passion. Thus passions for Hume are end-setting entities. And the obvious example of an end-setting mental entity is a desire or a want. There are two basic ways that reason-based beliefs can cause a new desire or action in tandem with a pre-existing desire: (1) The belief can suggest the means to a desired end; or (2) the belief can suggest a constituent or a specification of the desired end. I want to get to work without increasing my carbon footprint; reason tells me that I can do this by going by bike; and the want in tandem with the belief causes a new desire to cycle to work. I want to have a good time tonight and reason informs me that seeing the Dark Knight conforms to my idea

Charles R. Pigden

9

of a good time. But seeing the Dark Knight is not a means to having a good time tonight – rather my good time consists in seeing the The Dark Knight. But although, for Hume, all desires are passions, some passions are not desires. Consider, for example, “kindness to children” which is certainly a widespread tendency (T, 2.3.3.8/417). Is this a desire? Not if desires are propositional attitudes, that is to say attitudes toward propositions (that of wanting them to be so). For being kind to children is not a matter of wanting something to be the case. There are no doubt people (such as Save the Children activists) who want everything to be well with the world’s children. But with most of us, our kindness to children, in so far as it exists, is not like that. It is a “tendency” rather than a desire, a disposition to acquire desires for the welfare of the individual children, not a generalized desire that children should be happy. Kindness to children is not the only motivational state that falls into this category. If I have a taste for chocolate, the desires that I feel for Mars bars and Kitkats are not desires for the means to achieve some larger end (such as keeping up a daily intake of chocolate). Rather I am unusually prone to acquire desires for chocolate, especially when it stimulates the senses. An amorous person is not (or need not be) someone with the Valmont-like project or maximizing his or her sexual conquests. Rather he or she is unusually prone to sexual desire, that is, unusually disposed to acquire desires for sexual relations with the attractive people that they happen to meet. Even “curiosity or the love of truth” would appear to fall into this category. When people complain that George Bush lacks curiosity, they don’t mean that he lacks the desire to approximate omniscience. They mean that when exposed to information about this or that (such as the existence of foreign countries or the workings of the economic system) Bush is not disposed to acquire desires to know more. So although some passions are desires, some are dispositions to acquire desires (or DTADs), a point not often noticed by Hume’s latter-day disciples. Indeed, it is not clear that Hume himself realized that he was postulating DTADs as well as desires, and that the two are distinct. For once we allow for DTADs as well as desires, the Motivation Argument begins to look distinctly shaky (see Essays 3 and 4). Hume is careful to stress that the passions which move us are often “calm,” and that a passion can exercise a powerful causal influence even though it is principally known from its effects and does not excite much “emotion” in the mind. Thus for Hume a strong, or causally influential, passion (such as the desire to lead a long and healthy life) can be almost invisible from a phenomenological point of view, while a passion which excites a psychic stir (such as the craving for just one more cigarette) can be relatively weak in its effects. It is this that leads the reformed smoker to believe that when she forgoes the cigarette she is actuated by reason alone, whereas she is in fact motivated by the reason-based belief that smoking can kill plus the calm desire to live long and prosper. Thus the Slavery of Reason Thesis amounts to this: given the constitution of the human mind, beliefs, inferred beliefs or beliefs derived from ideas are

10

Introduction

causally incapable of producing either actions or new desires without the aid of pre-existing passions, which themselves are usually to be understood either as desires or as dispositions to acquire desires (DTADs). That’s the doctrine: what are the arguments? 0.4.2 The slavery of reason thesis: The arguments Hume’s chief argument for the Slavery of Reason Thesis is a simple one. It can be summarized thus: Slavery of Reason Argument: Non-Committal Version (i) Beliefs derived from reason are either analytic (to do with the relations of ideas) or synthetic (to do with causal relations). [“The understanding exerts itself after two different ways, as it judges from demonstration or probability; as it regards the abstract relations of our ideas, or those relations of objects, of which experience only gives us information.”] (ii) Analytic beliefs cannot motivate except in so far as they lead to synthetic beliefs. [“Abstract or demonstrative reasoning, therefore, never influences any of our actions, but only as it directs our judgment concerning causes and effects.”] (iii) Synthetic/causal beliefs cannot motivate if the causes and effects believed in are indifferent to us. [“It can never in the least concern us to know, that such objects are causes, and such others effects, if both the causes and effects be indifferent to us.”] (iv) Hence synthetic/causal beliefs cannot motivate unless they point the way to realize some pre-existing passion (from iii). [“’Tis evident in this case that the impulse arises not from reason, but is only directed by it.”] (v) Beliefs derived from reason cannot motivate by themselves, that is, without the aid of some pre-existing passion (from (i), (ii) & (iv)). [“Reason alone can never be a motive to any action of the will.” “Reason is, and ought only to be the slave of the passions, and can never pretend to any other office than to serve and obey them.”] This is a carefully non-committal exposition of the argument. The conclusion is that beliefs derived from reason cannot motivate by themselves, but whether this means that all beliefs, inferred beliefs or beliefs derived from ideas cannot motivate is simply left open for further debate. Hume’s fundamental thought is this. Given the basic structure of the human psyche it is causally possible for a human being to entertain any reason-based belief (and any conjunction of reason-based beliefs) while remaining psychologically unmoved. The belief, by itself, is causally inert. So, if a reason-based belief is to cause an action or volition, it can only do so with the aid of some other psychological entity. Another reason-based belief

Charles R. Pigden

11

could not do the trick. For it is psychologically possible (given the basic structure of the human mind) to remain unmotivated by the conjunction of the first reason-based belief with the second. So this extra cause must be a psychological entity of some other kind. But any entity that could cause either an action or a new desire in conjunction with a reason-based belief would fit the functional profile of a passion. Hence, if a belief is to motivate, a passion of some kind is causally necessary. So much for the thesis that “reason alone can never be a motive to any action of the will.” But what about the corresponding claim that reason “can never oppose passion in the direction of the will”? Hume insists that the second thesis follows from the first by a necessary consequence (T, 2.3.3.4/414–415). But how is this supposed to work? The argument, I take it this: (1) The final determinants of the will are desires (which are a kind of passion). [Premise.] (2) Nothing can conflict with a desire Φ except a) another desire Ψ or b) a belief Γ negating a belief Δ on which the desire Φ causally depends. [Premise.] (3) The products of reason are all beliefs. [Premise.] (4) Product-of-reason beliefs cannot give rise to new desires by themselves (that is, without the aid of a pre-existing passion). [Premise: a variant of conclusion v) above.] (5) Hence the products of reason by themselves cannot give rise to anything that conflicts with an existing desire Φ, except a belief (or beliefs) Γ negating a belief (or beliefs) Δ on which the desire Φ causally depends. [From 2), 3) and 4).] (6) Hence reason (by itself) cannot give rise to anything that conflicts with an existing desire Φ except a belief (or beliefs) Γ negating a belief (or beliefs) Δ on which the desire Φ causally depends [From (5).] (7) Hence reason (by itself) cannot give rise to anything that conflicts with an existing desire (or passion) Φ in determining (that is, directing) the will except a belief (or beliefs) Γ negating a belief (or beliefs) Δ on which the desire Φ causally depends. [From 1) and 6).] With these arguments in place, and reason defined as either our beliefforming faculty or a sub-faculty of that faculty, Hume is in a position to state his rather blood-curdling conclusion: ’Tis not contrary to reason to prefer the destruction of the whole world to the scratching of my finger [unless that preference is based on a false belief].’Tis not contrary to reason for me to chuse my total ruin, to prevent the least uneasiness of an Indian or person wholly unknown to me [Ditto]. ’Tis as little contrary to reason to prefer even my own

12

Introduction

acknowledg’d lesser good to my greater, and have a more ardent affection for the former than the latter. (T, 2.3.3.6/416)

0.4.3 The SRT and the polemic against rationalism What is the point of the SRT? Well, as we shall see, it provides one of the premises for the Motivation (or Influence) Argument which is the chief weapon in Hume’s anti-rationalist armory in Treatise 3.1.1. (“Moral Distinctions not deriv’d from Reason”). But there is a bit more to it than that. For there are remarkable affinities between Treatise 2.3.3, and Section I of Hutcheson’s Illustrations on the Moral Sense (Hutcheson, 2002, pp. 137– 55). But there is no trace of the Motivation Argument in Hutcheson, presumably because he did not believe the second premise, that moral beliefs motivate. (See Darwall, 1995, Chapter 8.) Thus for Hutcheson at least, and presumably for Hume, the SRT has an independent part to play in the refutation of rationalism. In my view, its role is to counter one of the rationalists’ chief arguments against sentimentalism. The sentimentalists – Hutcheson, Hume – construed the moral properties as akin to secondary qualities. Slurring over some important complications, their view was that to say an action is right or virtuous is to say that it would arouse the approbation of a suitably qualified human spectator, someone impartial, relevantly informed and dispassionate (that is, devoid of any distorting hangups). The rationalists did not like this view. Here, for instance, is John Balguy: “if God had not framed our natures with such a propensity [i.e. the moral sense], and given us this benevolent instinct, we should have been altogether incapable of virtue, notwithstanding intelligence, reason and liberty.” Balguy finds this incredible. “How a notion can be true that labours under such a consequence as this, I cannot understand.” Surely if we are rational we can recognize and be moved by the right whatever our instinctual inclinations. “Supposing us void of natural compassion as well as benevolence; might we not possibly be induced to attempt the relief of some person in distress, merely from the reason of the thing? … In short, if we made use of our understandings, they would not fail, I think, to discover our duty in such a case. Nay they would prompt us to undertake it, and condemn us if we omitted it” (Balguy, 1733, pp. 390–1). His argument, I take it, goes something like this: (1R) It is a conceptual truth that the moral facts, if any, would have to be both accessible to reason and (in today’s terminology) objectively prescriptive (that is, such as to defeasibly motivate any conceivable rational agent that became aware of them, whatever that agent’s desires or inclinations). (2R) Facts about would what arouse the approbation or disapprobation of a suitably qualified human observer are not accessible to reason (at least

Charles R. Pigden

13

not in the right kind of way) and are not objectively prescriptive (that is, such as to defeasibly motivate any conceivable rational agent that became aware of them, whatever that agent’s desires or inclinations). (3R) So the moral facts, if any, are not facts about what would arouse the approbation or disapprobation of a suitably qualified human observer. The sentimentalists replied, or can be reconstructed as replying, as follows: (1S) There are no objectively prescriptive facts (facts that would be defeasibly motivating to any conceivable rational agent that became aware of them, whatever that agent’s desires or inclinations). (2S) So if there are to be any moral facts, it is not a conceptual truth that the moral facts, if any, would have to be objectively prescriptive (that is, such as to defeasibly motivate any rational agent that became aware of them). They may be facts that are only motivating to human beings or creatures with similar dispositions (and not necessarily to all of them). The point of the SRT is that it provides the backing for the sentimentalists’ premise (1S). Reason is either our belief-forming faculty or a sub-faculty of that faculty. Beliefs derived from reason cannot motivate without the aid of a pre-existing passion, either a DTAD or a desire. Hence facts accessible to reason cannot motivate human beings without the aid of a pre-existing desire or DTAD, since to be aware of a fact is, in part, to believe in it. No desires or DTADs are constitutive of rationality, except, perhaps, a penchant for true, as opposed to false, beliefs and a taste for the right kind of beliefforming strategies. (“’Tis not contrary to reason to prefer the destruction of the whole world to the scratching of my finger” (T, 2.3.3.6/416) but “a wise [and hence, presumably, a rational] man … proportions his belief to the evidence” (EHU, 10.1.4/110).) Indeed, this pretty much follows from the definition of Reason as either our belief-forming faculty or a sub-faculty of that faculty. But if facts accessible to reason cannot even motivate human agents without the aid of a pre-existing desire or DTAD, and if no morally salient desires or DTADs are constitutive of rationality, then there are no objectively prescriptive facts (facts that would be defeasibly motivating to any conceivable rational agent that became aware of them whatever that agent’s desires or inclinations). The upshot for Hume is that morality is rationally optional. Moral considerations will appeal to those with the right kind of psychological make-up but not to those without. Such people may be repellent but they need not be irrational. The point becomes plain in his famous discussion of the “sensible knave” a sort of mild and discrete Thrasymachus (perhaps like Mr Elliot in Jane Austen’s Persuasion) who remains unmoved by considerations of

14

Introduction

justice, except in so far as they serve his self-interest: “A sensible knave … may think, that honesty is the best policy, [is] a good general rule; but [that it] is liable to many exceptions: And [that] he conducts himself with most wisdom, who observes the general rule, and takes advantage of all the exceptions” (EPM, 9.2.22/282–3). Hume admits that he has no rational answer to the knave: “If a man think, that this reasoning much requires an answer, it will be a little difficult to find any, which will to him appear satisfactory and convincing. If his heart rebel not against such pernicious maxims, if he feel no reluctance to the thoughts of villainy or baseness, he has indeed lost a considerable motive to virtue; and we may expect, that his practice will be answerable to his speculation” (EPM, 9.2.23/283). The only consolation that Hume seems to suggest is that such sensible knaves are relatively rare, and that many of them are not as sensible as they like to suppose, and get caught out because of over-confidence. 0.4.4 The SRT and modern moral philosophy So much for Hume, but what about us? Why is the SRT important for modern moral philosophy? (1) The SRT provides one of the main premises for the Motivation Argument which is taken by many to be the chief argument for noncognitivism or expressivism. (Non-cognitivists think that, strictly speaking, moral judgments are neither true nor false since their function is to convey commands or express attitudes. Expressivists think that although the primary function of moral judgments is to convey commands or express attitudes, this is compatible with the capacity for a rather watered down kind of truth. Thus expressivism is non-cognitivism for wimps.) (2) The SRT – at least some incarnations – poses a problem for modern moral rationalists for much the same reasons that the original SRT posed a problem for their eighteenth century counterparts. If you want to base morality on facts which are objectively prescriptive – either for rational beings generally or for human beings in particular – then the SRT stands in your way since it seems to show that there can be no such facts. Suppose (A) that beliefs cannot motivate without the aid of a pre-existing desire or DTAD. And suppose (B) that there are no morally salient desires or DTADs that are constitutive of rationality. Then for any fact there is a rational being that can remain unmoved by it without prejudice to its rationality (except for the motivation to draw certain inferences in accordance with the agent’s belief-forming propensities). Hence there are no facts, either accessible to, or constructible by, reason, such that they would be defeasibly motivating to any conceivable rational agent that became aware of them, whatever that agent’s desires or inclinations. Thus a philosopher who wants to formulate an objectively prescriptive ethic has three options (though they are not mutually exclusive): She can deny (A) that beliefs cannot motivate without the aid of a pre-existing desire or DTAD; she can deny

Charles R. Pigden

15

(B) that there are no morally salient desires or DTADs that are constitutive of rationality; or she can insist that there is a special kind of awareness of the moral facts that does guarantee defeasible motivation but that this awareness transcends mere belief. Thomas Nagel in The Possibility of Altruism (1970) adopts the first strategy. For him, if I do X in the belief that X leads to Y, it follows that I have a desire for Y, but this desire need be nothing more than a conceptual shadow, cast by the fact that I did X in the belief that X leads to Y. (See Pigden, 2009a, for a critique of this argument.) Korsgaard as a Kantian also adopts the first strategy (See Korsgaard, 1986, 1996b, 1996d). Michael Smith, by contrast, adopts the second, since he is a Humean about motivation though a Kantian in ethics. For him a desire to desire what all fully rational agents would desire, and a disposition to desire what the agent believes that all fully rational agents would desire are both constitutive of rationality (Smith, 1994). Korsgaard, who is something of a motivational double-dipper, also seems to think that there are motivational dispositions that are constitutive of rationality. McDowell, in so far as I understand him, adopts both strategy 1 and strategy 3, since he endorses Nagel’s view that desires need be nothing more than conceptual shadows cast by deliberate action while insisting that seeing things in a certain way can be sufficient for rational motivation (McDowell, 1978). Indeed, it would not be going too far to say that there is a whole philosophical industry of practical reasoners devoted to downplaying or denying the SRT in the interests of moral objectivity. It might be thought that Hume and his disciples are committed to at least one action-oriented DTAD as constitutive of rationality, namely the disposition to acquire desires for (what you believe to be) the means to your ends. This is, so to speak, a function that takes us from desires and beliefs to new desires and subsequently actions, and it would be hard to recognize a being without it as any kind of rational agent. But if it would be irrational not to have this disposition, surely it is a disposition that we rationally ought to have? Thus Hume has been widely read as an instrumentalist about practical reason, subscribing to the norm that we rationally ought to be disposed to perform those actions which are (or are believed to be) the means to our desired ends. So although Hume believed that it is not contrary to reason to prefer the destruction of the whole world to the scratching of my finger, he is sometimes supposed to have thought that (given such a preference) it would be contrary to reason for me not to push the red button if I correctly believed that by pushing the button I could destroy the world while saving the finger. Others, most notably Millgram, have argued that there is not much sign of any such norm in Hume, and that he seems to think that we automatically do the things that we believe to be conducive to our ends, in so far as we really desire the ends and in so far as we really believe (Millgram, 1995). I think that Millgram is correct about Hume and that this view can be vindicated in the light of a modern “Humean” conception of

16

Introduction

beliefs and desires. For the very fact that we have beliefs and desires in the first place guarantees that we will have the disposition to desire the means to our ends. And since we have this disposition anyway simply in virtue of having beliefs and desires, there is no place for a norm of rationality recommending rational agents to realize this disposition, where a norm is, minimally, something you might fail to meet. According to Kant it is analytic that “whoever wills the end, wills also (so far as reason decides his conduct) the means” (Kant, 1994, 4: 417), and though his reasons for saying this are somewhat obscure, I think he is basically correct. But this does not preclude a causal disposition to desire the means to our ends, internal to our natures as believing/desiring beings. Following Michael Smith (1994) I characterize a desire as a complex dispositional state grounding a range of counterfactuals. To desire a state of affairs X is to be defeasibly disposed to desire Y given the belief that Y leads to X, to desire Z given the belief that Z leads to Y and to be averse to W given the belief that W would hinder or prevent X. (The qualification “defeasibly” is necessary since I may have other desires which are incompatible with the desire for the means to something that I genuinely desire. If I believe I cannot win my mistress’s love without swimming the Hellespont but know that I am a poor swimmer who would die if I made the attempt, prudence – my “general appetite to good, and aversion to evil” – is likely to trump my instrumental desire to attempt the swim.) To believe that Z leads to Y is to be in a complex dispositional state, grounding a related range of counterfactuals such as a disposition to desire Z given a desire for Y and to be averse to Z given an aversion to Y. If this is roughly right, it follows automatically that we tend to desire as a means what we believe will facilitate the things we desire as ends. If we are not even defeasibly disposed to desire the means to what we profess to desire as an end, that is evidence that we do not really desire it, or perhaps that we do not really believe that the supposed means will foster the end. It is analytic that if we have beliefs and desires we will have this disposition, but the disposition itself is grounded in the causal configuration of the human mind. But since we have a tendency to desire (and hence to choose) the means to our ends simply in virtue of having beliefs and desires in the first place, there is no need for a norm of practical reason telling us that we rationally ought to do this. (See Finlay, Essay 7 for a detailed and sophisticated defense of this kind of view.) Even philosophers with the modest ambition to ground morality in reasons for action are likely to face difficulties with the SRT. A fact Δ constitutes a reason for an agent Γ to ϕ if a belief in the fact Δ would motivate Γ to ϕ in virtue of Γ’s rationality. But if there are no desires or DTADs, and hence no ends, that are internal to rationality, then there are no reasons for action in this sense. Perhaps we can save the day by treating being a reason for action as a four-place rather than a three-place relation. A fact Δ constitutes a reason for an agent Γ to ϕ given ends Ψ if a belief in the fact Δ would motivate Γ to

Charles R. Pigden

17

ϕ in virtue of Γ’s agency and Γ’s ends Ψ. This gives us a Humean conception of reasons for action but makes them unsuitable as the basis for ethics. If a fact Δ only constitutes a reason for an agent Γ to ϕ given Γ’s ends Ψ, then Δ’s being a reason for Γ to ϕ cannot be the basis of Γ’s obligation to ϕ since (generally speaking) that is supposed to hold irrespective of Γ’s ends. (You can’t say that you are not obliged to do the right thing because you happen to have evil preferences.) Perhaps we can solve the problem by appealing to a norm of rationality which judges between ends. There is no doubt that we appeal to a variety of such norms in daily life. Some condemn actions which are self-destructive or which cut across the agent’s long-term self-interest. The Common Law was long dominated by a norm which condemned suicide as irrational. If somebody killed himself this meant that “the balance of his mind” must have been “seriously disturbed.” Economists, at least when they are on duty, tend to subscribe to a self-interested norm of rationality which would condemn the choice of “my total ruin” for the sake of “an Indian or a person wholly unknown to me” as, at best, non-rational or a-rational, if not positively irrational. Thus economists tend to build a “general [and selfish] appetite to good, and aversion to evil” into their conceptions of rationality. But even here there is considerable disagreement about such things as the discount rate for future pleasures and pains. On the other hand, we sometimes condemn as irrational actions which don’t take other people into account. “He is too dumb [that is, too irrational] to realize that the Party is more important than his big ego.” “She’s too stupid to realize that other people’s feelings count.” So perhaps we can say that a fact Δ constitutes an R-reason for an agent Γ to ϕ of if a belief in the fact Δ would motivate Γ to ϕ if Γ were R-rational, that is, if Γ conformed to some norm of rationality R. Indeed, if we could establish the correctness of a norm of rationality, we could simply say that a fact Δ constitutes a reason for an agent Γ to ϕ of if a belief in the fact Δ would motivate Γ to ϕ if Γ were really rational, where to be really rational is to conform to the correct norm. The problem is that it is hard to see what could justify the claim that this or that norm of rationality is correct. Since we can skip gaily from one norm to another in the course of a single conversation, we certainly cannot say that it is analytic that it either is or is not rational to be motivated by false but intelligently acquired beliefs, to discount the future at this rate or at that, to be motivated or not by other people’s interests, or to be motivated or not by moral considerations. We commonly use “rational” and its cognates in all these senses and more. But if it is not analytic, what is the truthmaker for the claim that a given norm of rationality R1 is correct as opposed to one of its rivals R2, R3 etc? It is hard to see how there could be a fact of the matter here. Hence there is something a bit naive about Nozick’s book title The Nature of Rationality. There is not a thing, rationality, with a nature waiting to be investigated. Rather practical rationality, once it transcends the dispositions internal to being a believing/ desiring agent, is a norm-relative affair. But if this is right, we cannot say that

18

Introduction

a fact Δ constitutes a reason for an agent Γ to ϕ but only that Δ constitutes a reason for an agent Γ to ϕ according to the norm of rationality R when it might not constitute a reason for Γ to ϕ according to some other norm. And that looks a bit too relativistic to afford a basis for ethics. (3) There is a third reason why the SRT is important for modern moral philosophy. It provides one of the premises for Mackie’s well-known Argument from Queerness, (the AFQ) which is designed to support the “error theory,” the thesis that moral judgments are systematically false and morality a myth (though perhaps a necessary one). As I understand it, the AFQ takes a premise apiece from the arguments of the eighteenth century rationalists such as Balguy (and I think Kant) and their sentimentalist opponents, Hutcheson and Hume: AFQ(CRP) (1R) It is a conceptual truth that the moral facts, if any, would have to be both accessible to reason and (in today’s terminology) objectively prescriptive (that is, such as to defeasibly motivate any conceivable rational agent that became aware of them, whatever that agent’s desires or inclinations). (S1) There are no objectively prescriptive facts (facts that would be motivating to any rational agent that became aware of them whatever the agent’s desires or inclinations). [As noted above, the SRT provides the basis or this claim in both Hutcheson and Hume.] (M3) So there are no moral facts. (M4) Non-cognitivism is false. Moral judgments are truth-apt, true-orfalse, in the true/false game. [Premise] (M5) For a substantive non-negative proposition to be true there must be a fact to make it true. [Premise] (M6) But since there are no moral facts to make them true, there are no substantive moral truths (what I elsewhere describe as non-negative atomic moral truths). (M7) Hence all systems of morality (and not just the ones I have “hitherto met with”) are systematically false. (Mackie, 1977, Chapter 1) It is because modern rationalists are inclined to accept premises (R1), (M4) and (M5) while rejecting the conclusion of the AFQ, that they are so bitterly opposed to the SRT. It is, however, possible to accept (1R) and some versions of the SRT while denying (S1). If you define a rational agent as an agent which meets a certain norm of rationality, and if this norm involves being motivated by certain ends or considerations – for example moral considerations – then there can be objectively prescriptive facts even if you accept principle (A), that beliefs cannot motivate without the aid of pre-existing desires or DTADs, and

Charles R. Pigden

19

the further thesis that talking and thinking beings are conceivable whose DTADs and desires would leave them unmoved by the moral facts. A talking mantis from Mars, with a taste for human brains and no interest in the wellbeing of non-mantises, won’t count as rational whatever her intellectual attainments, since she won’t be motivated by the right kind of considerations. So you get objectively prescriptive facts in a sense. There will be facts that are motivating to any rational being that becomes aware of them, but only because we have restricted rationality to beings that happen to share our preferences. It seems to me that this strategy rescues (R1) by trivializing it. The idea behind (R1) is what might be called Aragorn’s Principle: “Good and ill have not changed since yesteryear; nor are they one thing among Elves and Dwarves and another among Men” (Tolkien, 1995, III.ii, p. 428). The idea is that morality is supposed to be binding on all talking, thinking and acting beings because of their capacity for thought, talk and action, not that it is only binding on beings that qualify as rational in some artificially restricted sense of the word. Thus the SRT is important to modern moral philosophy for (at least!) three reasons: (1) it provides a premise for the Motivation Argument which seems to support non-cognitivism; (2) it poses a problem for philosophers who want to found ethics either on reasons for action or objectively prescriptive facts and (3) it supports a premise of the AFQ which, in turn, supports the error theory. 0.4.5

Matters arising

Essays 6–10 deal with matters arising from the Slavery of Reason Thesis. Constantine Sandis (Essay 6) observes that motivation theorists are divided into two factions, those who think that we are motivated by reasons and that those reasons are partly constituted by beliefs and desires and those who think we are motivated by reasons that those reasons can be constituted by pure beliefs. In Sandis’ opinion, Hume denied the assumption on which the two factions are agreed, that we are motivated by reasons. Sandis also develops a cognitivist interpretation of the Motivation Argument according to which it proves what Hume says he is trying to prove, namely that moral distinctions are not derived from reason. Hume has sometimes been read as an instrumentalist about practical reason. Irrespective of our ends what we rationally ought to do is to perform those acts that will achieve our ends. Others have argued that Hume is not even an instrumentalist since he does not really believe that there is anything we rationally ought to do (Millgram, 1995 and Korsgaard, 1986). This view is widely thought to be absurd, so absurd indeed, that it is difficult to believe that Hume himself believed it. Stephen Finlay (Essay 7) argues both that this is Hume’s view and that Hume’s view is correct. There is no plausible norm of practical reason commanding us to do what (we think) will achieve our ends that meets the minimum requirement for being

20

Introduction

a norm, namely that we might fail to meet it. Thus you cannot argue from the plausibility of such a norm, to other more sophisticated norms of practical reason that might be employed to underwrite morality. In an elegant little paper, Kent Hurtig (Essay 8) argues against a latter-day Humean, the late Bernard Williams. Williams is an internalist about practical reason. He thinks that X cannot be a reason for Γ to ϕ unless X relates to something in Γ’s “subjective motivational set”. But unlike Hume himself, who thought of the end-setting members of a motivational set as desires or DTADs, Williams allows that they might be beliefs. But this, Hurtig argues, is a mistake. An internalist about reasons cannot allow their subjective motivational sets to be contaminated by beliefs on pain of paradox. Thus an internalist about reasons must be a “non-cognitivist” about the subjective motivational set: its membership must be confined to mental items that do not rise to the dignity of beliefs. If you want to be a Humean about motivation you have to go the whole “non-cognitivist” hog. Essays 9 and 10 focus on the work of Christine Korsgaard whose ideas loom large in these debates. Though a dedicated Kantian, Korsgaard retains an obvious fondness for Hume, perhaps because she was educated by Rawls. (We can tell what a Rawlsian education was like since his lectures are now available. See Rawls, 2000.) In Essay 9, Herlinde Pauer-Studer develops a complex and subtle critique of Korsgaard’s critique of Hume. The key point, I take it, is this. Korsgaard ascribes to Hume the principle that we have a reason to do X if X will achieve our desires. Then if I want to achieve Y, and it is necessary to do X to achieve Y, it follows that I have a reason to do X. But this is implausible if either my ends Y or my means X are wicked or self-destructive. Pauer-Studer, following Broome, objects that this argument rests on a confusion. For the following inference is not valid: (1) I have a reason to make it the case that (if I want Y and X is a necessary means to Y, that I do X). (2) I want Y. (3) X is a necessary means to Y. Therefore (4) I have a reason to make it the case that I do X. The conditional does not “detach”, governed as it is by an intentional operator. In much the same way it is fallacious to infer “Necessarily I will sin tomorrow” from “Necessarily if God knows I will sin tomorrow I will sin tomorrow” and “God knows that I will sin tomorrow” – a point made by Aquinas long ago. In so far as Humeans can talk about reasons for action, being a reason for someone to do something is actually a four-place relation, since a fact Δ constitutes a reason for an agent Γ to ϕ given ends Ψ if a belief

Charles R. Pigden

21

in the fact Δ would motivate Γ to ϕ in virtue of Γ’s agency and Γ’s ends Ψ. But if Δ constitutes a reason for Γ to ϕ given ends Ψ, it does not follow that if Γ has ends Ψ, then Δ constitutes a reason for Γ to ϕ PERIOD. We cannot transform a four-place relation into a three-place relation by adopting set of desires. According to Korsgaard her Kantian ethic is better than Hume’s because it meets certain requirements that Hume’s fails to meet. (1) An adequate metaethic (adequate at the level of sociology) must explain motivational normativity, the fact that “men are often governed by their duties, and are deterred from some actions by the opinion of injustice, and impelled to others by that of obligation.” But (2) it must also explain why moral behavior is justified, why it is rational to do the right thing. Call this justificatory normativity. The basis of Korsgaard’s complaint is that Hume succeeds in explaining (1) but not (2). We tend to do what we think is the right thing because we are motivated by our moral sentiments. But that does not explain why we are right to do so. Hume thinks that on reflection we tend to approve of the practice of doing what we tend to approve of, but for Korsgaard this kind of reflective endorsement is not enough. In Korsgaard’s view, moral behavior is justified because each of us shares a common practical identity – that of a rational human being – and that therefore we have reason to treat each other with the dignity due to rational agents. Luke Russell (Essay 10) objects that in arguing for this thesis Korsgaard relies on the kind of premise that she scorned in the case of Hume. She points to various kinds of practical necessity to block the justificatory regress – that is, the endless process of “Why should I?” questions – despite her earlier suggestion that practical necessity is no substitute for positive reasons. She thinks, for instance, that it is arbitrary to halt the reflective process along with Hume when we reach our reflectively endorsed desires. Why isn’t it equally arbitrary to stop at our practical identity? There seems to me a further problem if you think that the same theory has to explain both motivational and justificatory normativity. Suppose that Korsgaard’s argument is a success and her complicated theory explains why it is rational to do the right thing. Since nobody thought of this until she arrived on the scene, this theory cannot explain why existing people do what they believe to be the right thing. I certainly do not do what I believe to be right, for Korsgaard’s reasons, since, insofar as I understand her theory, I think that it is false. Thus a novel theory which meets requirement (2) is unlikely to meet requirement (1). You can, of course, solve the problem by positing two kinds of normativity with different explanations. But that leaves us with the uncomfortable consequence that when we do the right thing we do not in general do it for the reasons that we should. It may be rational to do the right thing, but most of the people who do it, don’t do it because it is rational.

22

Introduction

0.5 The Motivation Argument The Motivation Argument poses a puzzle for interpreters of Hume. The textual evidence plus the reactions of contemporaries strongly suggest that Hume was what is nowadays known as a moral realist, with a response-dependent view of the moral properties. He thought (or seemed to think) that moral judgments are in today’s jargon truth-apt, that they are capable of truth and falsity, and that some of them are true. Morality is determined by the reactions of suitably qualified human observers who are presumed to share a moral “sense,” though for Hume, as opposed to Hutcheson, the moral sense, is, in part, a socially constructed affair, the product of a process of social evolution. Thus whether or not a trait is a virtue is “a plain matter of fact” though a fact to be determined by consulting the feelings of suitable spectators (or by trying to approximate those spectators and consulting one’s own feelings). But it is hard to read the Motivation Argument as anything other than an argument for non-cognitivism. This is one reason why Hume was widely regarded as a proto-emotivist (and hence a proto-non-cognitivist) in the middle years of the twentieth century, an interpretation that even today is far from dead, especially in the minds of those with a merely undergraduate acquaintance with Hume. (See Hudson, 1969, Ayer, 1980 and Flew, 1986, for examples of this kind of thing. For a contrasting view see Cohon (1997a).) Here are four formulations of the Motivation Argument, lifted directly from Hume: Motivation Argument A (1) Morals have an influence on the actions and affections. [Premise.] (2) Reason alone, as we have already proved, can never have any such influence [Premise: the SRT.] (3) Morals … cannot be derived from reason. (T, 3.1.1.6/457) Motivation Argument B (1) Morals excite passions, and produce or prevent actions. [Premise.] (2) Reason of itself is utterly impotent in this particular [Premise: the SRT.] (3) The rules of morality, therefore, are not conclusions of our reason. (T, 3.1.1.6/457) Motivation Argument C (1) The merit and demerit of actions frequently contradict, and sometimes control our natural propensities. [Premise.] (2) But reason has no such influence. [Premise: an exaggerated version of the SRT.] (3) Moral distinctions, therefore, are not the offspring of reason. (T, 3.1.1.10/458–9)

Charles R. Pigden

23

Motivation Argument D (1) The distinction betwixt moral good and evil … has an influence upon our actions. [Premise.] (2) Of which reason alone is incapable [Premise: SRT.] (3) It is impossible that the distinction betwixt moral good and evil can be made by reason. (T, 3.1.1.17/462–3) There are several points to note here. (I) When Hume talks of “morals” or the “distinction betwixt moral good and evil” or “the merit and demerit of actions” having an influence on our actions and affections, what he means is that our opinions or beliefs about such things often have an influence. It is “confirmed by common experience … that men are often governed by their duties, and are deterred from some actions by the opinion of injustice, and impelled to others by that of obligation”(T, 3.1.1.5/457). (II) Although Hume thinks that our moral opinions often influence our conduct there is no suggestion that they are necessarily motivating or that they motivate in every case (absent special circumstances). The possibility of indifference or amoralism is not excluded. Indeed, as he subsequently makes plain, Hume thinks that there are “sensible knaves” who “feel no reluctance” at “the thoughts of villainy or baseness” and are therefore not “deterred from some actions by the opinion of injustice.” (III) The Motivation Argument does not add up to an argument for noncognitivism unless we interpret “reason” as our belief-forming faculty and the SRT as the thesis that beliefs as such cannot motivate without the aid of a pre-existing passion. If reason is our belief-forming faculty and our moral “beliefs and opinions” are not the offspring of reason, then they are not genuine beliefs or opinions. But if we allow the possibility of beliefs or opinions that are not derived from reason, then the argument only goes to show that our moral opinions are not derived from reason, which is compatible with their being genuine beliefs based on feeling or sentiment. (IV) Three out of the four formulations of the Motivation Argument are invalid, and version C only approximates validity by exaggerating the SRT. In each case premise (1) says that our moral beliefs have an influence on our actions and affections, a fact which is indeed “confirmed by common experience.” But except in the case of C, all that premise (2) says is that reason alone has no such influence – that is, that reason-based beliefs cannot motivate without the aid of a pre-existing passion – which is all that the SRT purports to prove. But if all that (2) says is that reason-based beliefs cannot motivate by themselves, this opens up the possibility that moral beliefs are in fact derived from reason but do their motivating business with the aid of a pre-existing passion such as a desire to do the right thing. Whatever exactly Hume was trying to prove, he does not manage to prove it.

24

Introduction

Thus anyone who wants to extract a decent argument for non-cognitivism from Hume has a good deal of repair work to do. Particularly influential in this connection is the work of Michael Smith for whom a revised version of the Motivation Argument constitutes “the Moral Problem,” the central issue in meta-ethics (Smith, 1994). For Smith thinks that it is possible to extract a powerful argument for non-cognitivism from Hume, and that this is a problem since the premises appear to be true and the conclusion false. This is his argument: (1) If someone judges that it is right that she φs then, ceteris paribus, [and without the aid of any pre-existing desire] she is motivated to φ. (2) An agent can only be motivated to act by a belief when she has a desire for an end to which the belief points out the means, where belief and desire are, in Hume’s terms, distinct existences. Therefore: (3) If someone judges that it is right that she φs then, her [psychological] “judgment” that it is right that she φs does not constitute a belief. (4) The best explanation of (3) is [insert non-cognitivist theory of choice]. (Smith, 1994, p. 12, slightly doctored. Note that up to step (4), the argument is deductive.) Several of the essays in this collection (specifically Essays 1–5) are devoted to “matters arising,” including matters arising from Smith’s formulations. Was the Motivation Argument intended as an argument for non-cognitivism (or something like it)? Is it possible to extract a decent argument for noncognitivism from Hume’s text and if so, does it ultimately succeed? If Hume was not arguing for non-cognitivism, what was he arguing for and how is his argument supposed to work? Does the argument succeed and if not, can it be patched up? Since Hume can be construed as arguing either for noncognitivism or for a response-dependent theory of ethics (both of which continue to be popular among philosophers), the Motivation Argument takes us to the heart of contemporary meta-ethics. Though famous as the contemporary standard-bearer for the error theory, Richard Joyce (Essay 1) does not argue directly for the error theory in this essay, either as a meta-ethic or as an interpretation of Hume. Instead he challenges the Motivation Argument as developed by Hume’s non-cognitivist disciples. Suppose we reinterpret premise (1) as the claim that (absent specific circumstances) it is necessary and a priori that moral beliefs excite passions, and produce or prevent actions. Call this motivational internalism. Does it follow that the role of moral judgements (conceived as linguistic expressions) must be to express some kind of conative state? Does some kind of non-cognitivism or expressivism follow? Not so, says Joyce, since expressivism does not imply motivational internalism (thus blocking an inference to the best explanation) and internalism does not imply expressivism.

Charles R. Pigden

25

According to Norva Lo (Essay 2) Hume construes moral properties as response-dependent properties, geared to a shared moral sense. His conceptual analysis makes an action or character virtuous/vicious if and only if all human beings who meet certain conditions (including impartiality and adequate information) are disposed, to feel the sentiment of approbation/ disapprobation towards it. Thus Hume cannot have been arguing for noncognitivism on pain of inconsistency. Indeed, she thinks that Hume was not really an internalist about motivation, that is she thinks that it is possible, according to Hume’s theory, to subscribe to a moral belief while remaining altogether unmoved. Indeed, it is theoretically possible to acquire moral knowledge without experiencing the moral sentiments (and hence without being in any morally salient conative state). (Consider a Martian anthropologist who has managed to anatomize the human moral sense.) What then is the purpose of the Motivation Argument? It is meant to disprove moral rationalism, the thesis that moral knowledge is typically the product of some cognitive or discursive process. And Lo provides three reformulations of the Motivation Argument designed to prove precisely this. Charles Pigden agrees that Hume conceives the moral properties as akin to secondary qualities. To say that a trait is a virtue is to say that it would excite the moral sense of a suitably qualified human observer (impartial, dispassionate and devoid of the delusive glosses of superstition and false religion), just as to say that a flower is violet is to say that it would excite the visual sense of a suitably qualified human observer (someone with functioning eyesight but without rose-tinted spectacles). Thus the point of the Motivation Argument cannot be to prove non-cognitivism, since Hume was not a non-cognitivist. Furthermore, the argument for non-cognitivism that has been extracted from Hume’s text is, in Pigden’s view, a failure. Where he parts company from Lo is in thinking that it is impossible to convert the Motivation Argument into a successful argument against moral rationalism and for sentimentalism. Plausible, yes, but successful, no, since it is formally respectable but based on false and question-begging premises. Michael Smith (Essay 4) gallantly goes bat for a non-cognitivist, or at least, an expressivist interpretation of Hume, and a rational reconstruction of the Motivation Argument as designed to prove non-cognitivism. He freely admits that he may be reading more into Hume than is really there, and maybe more than Hume could have conceived of (specifically he suggests a concern with what is practically rational which may have been alien to Hume). But it seems to me that in defending an expressivist interpretation of Hume, Smith is retreating several paces. The expressivism that he is worrying about now is a degenerate descendant of the non-cognitivism that preoccupied him in the nineties. It is okay to say that we have moral beliefs and that these beliefs are true (or false). It is just that in the normative case (but not otherwise) the beliefs are somehow constituted by desires. Even the Motivation Argument that Smith defends in Essay 4 is a feeble abductive

26

Introduction

variant of the deductive argument that he tried to defuse in The Moral Problem. As for the claim that Hume cannot have been arguing for noncognitivism since he was not a non-cognitivist, Smith has two counters. (1) According to Lo, a trait is a virtue for Hume if and only if it meets a certain factual condition. But if “X if and only if Y” is in some sense acceptable, and if “Y” abbreviates a factual statement, then “X” must be factual too. Smith objects that if this principle were correct a non-cognitivist could not coherently suppose that an action A is right if and only if A maximizes utility, since for him “A maximizes utility” is factual whereas “A is right” is not. But I think that Smith misconstrues Lo’s argument. Perhaps it is sometimes possible to accept “X if and only if Y” while interpreting “Y” as factual and “X” as non-cognitive. But this is hardly possible if the biconditional “X if and only if Y” is intended as a definition or a conceptual analysis of “X.” For a definition or a conceptual analysis can only be correct if “X” and “Y” are in some sense synonymous, which they could not be if one were noncognitive and the other factual. And when Hume actually “defines” virtue as “whatever mental action or quality gives to a spectator the pleasing sentiment of approbation” (EPM, App.1.10/289) it does indeed seem that conceptual analysis (or what we would nowadays call conceptual analysis) is what he has in mind. (2) Smith claims that on Pigden’s showing, Hume comes out as a non-cognitivist in the end since moral beliefs, like Pigden’s belief that Zena is desirable, are constituted by desires. But that is not the thesis. The claim is rather that moral beliefs are typically caused and justified by feelings of approval and disapproval not that they are constituted by such feelings, since they are beliefs about what a certain kind of spectator would feel. To believe that Zena is desirable is not necessarily to desire her, even if such beliefs are typically caused by feelings of sexual desire, and even if Hume himself identifies some beliefs with impressions. In Essay 5 (based on his book Value, Reality and Desire) Graham Oddie takes up the idea that to be good, or to be valuable, is to be worthy of desire and that the evidence for value-judgments consists in our desires and aversions. Such evidence is defeasible, of course, since our desires may be due to some pathological condition. Still, to experience something as valuable is to feel the relevant desires. Hence if I know that something is valuable (at least if my knowledge has been derived from experience) I will in fact be motivated to pursue or promote it. Thus moral knowledge (and indeed moral belief) has a tendency to “excite passions, and produce or prevent actions.” Oddie seems to subscribe to a theory rather like Hume’s (at least as analyzed by Lo and Pigden) but this is not quite correct. For Oddie, unlike Hume, does not cash out worthiness to be desired in terms of the dispositions of actual human agents (whether idealized or not). Nonetheless, the fact that something is desired is evidence of its desirability, and it is because we form value-judgments on the basis of desires that value-judgments – or rather the desires on which they are based – tend to motivate. Thus Oddie’s theory is

Charles R. Pigden

27

able to reconcile the premises of the Motivation Argument – that morals “excite passions, and produce or prevent actions” and that “reason itself is utterly impotent in this particular” – with a realist, indeed, a non-naturalist, conception of value.

0.6 Hume as a virtue theorist In the nineteenth century and in the early part of the twentieth, Hume was widely regarded as an ancestor of utilitarianism. He was censured in his own day (and by one of his best friends) for reducing human goodness to “that for which we commend a chest of drawers,” since he regarded the one, like the other, as a matter of being conducive to useful or agreable effects (Smith, 1976 [1759], p. 188). Bentham, more enthusiastically, claimed Hume as an inspiration. “But, after all retrenchments, there will still remains enough [in Book 3 of the Treatise] to have laid mankind under indelible obligations. That the foundations of all virtue are laid in utility, is there demonstrated, after a few exceptions made, with the strongest force of evidence” (Bentham, 1988 [1776], p. 51n). Sidgwick concurred, though with a characteristic caveat: “When Hume presented utilitarianism as a mode of explaining current morality, it was seen or suspected to have a partially destructive tendency” (Sidgwick, 1907, p. 86). But though Hume was indeed a precursor of utilitarianism, he was not really a utilitarian himself. Utilitarianism is a theory of the right and the good, the right being what we ought to do and the good being what we ought to bring about, and it shares the nineteenth and early twentieth century obsession with the “thin” moral concepts: rightness, duty, goodness and badness. These are not the concepts that exercised Hume. It was the virtues in general and some specific virtues in particular, such as justice, benevolence and chastity that interested him. But from about 1850 to about 1950 the big questions in ethics could be summed up in the chapter headings of W. D. Ross’s The Right and the Good (1930): “The Meaning of ‘Right,’” “What makes Right Acts Right?,” “The Meaning of ‘Good,’” “The Nature of Goodness,” etc. with the proviso that as the twentieth century wore on, the semantic status of claims involving these concepts – were they factual statements about non-natural properties or expressions of emotion? – began to be a major preoocupation. But in the 1950s, beginning with the work of Philippa Foot and G. E. M. Anscombe, a reaction set in and the virtues returned to the fore. (See the essays collected in Foot, 1978 and 2002, and in Anscombe, 1981.) The focus was less on what we ought to do and more on what sort of people we ought to be, less on the nature of predicative goodness (a concept regarded with increasing suspicion) and more on what it is to be a good human being. Although the original inspiration of this movement was Aristotelian, and although Foot and Anscombe themselves display a marked hostility to Hume, later virtue ethicists began to see Hume as a kindred spirit. After all, he was preoccupied

28

Introduction

with the virtues, rather than the “thin” moral concepts, and his section headings – “Of vice and virtue,” “Of Virtue and Vice in General,” “Justice, whether a natural or an artificial virtue?,” “Of chastity and modesty” – look much more like paper titles by Foot than chapters headings by Ross. Hume also shares the virtue ethicists’ obsession with human character and human nature, and their interest in social institutions such as promising. And for some virtue ethicists the determinedly modernist and secular cast of his thought was an added attraction. Hume shows that you can put the virtues at the center of the moral life without succumbing to the creeping Thomism that characterizes Anscombe and Foot. The last four essays in this collection deal with Hume as a virtue theorist. “What sort of a virtue theorist is Hume?” asks Christine Swanton (Essay 11). Well, to begin with, he is a virtue theorist who defines a virtue as a trait which arouses approbation in suitably qualified spectators. But this still leaves a wide arena for empirical investigation. What traits do we approve of and why? Do they have anything interesting in common apart from the fact that we are disposed to approve of them? In Swanton’s view, Hume has a complex and pluralistic theory of the virtues according to which we are disposed to approve of traits because of a wide variety of features related to human flourishing and bound up with the exigencies of human life. Traits excite approbation because they are expressive of the bonds of a (non-pathological) love, because they are the objects of a proper pride, because they express joy, because they display a justice-based respect for persons and their property and because they manifest a proper appreciation of status. Many of these traits are productive of human welfare broadly conceived (so that Hume’s utilitarian disciples were not wholly wrong). Nonetheless, for Hume there is a lot more to welfare than pleasure and the absence of pain. For Annette Baier (Essay 12) all this is a bit too pat. She is not sure that Hume is the gung-ho normative virtue theorist that Swanton takes him to be. Sometimes, like Locke, he seems to be using “virtue” in a sociological sense to refer to what is considered as such in this society or in that. The martial virtues of the Scythians, parading their towels made from human hides, are a far cry from the polite virtues of an eighteenth century Scottish gentleman such as Hume himself. Sometimes Hume hints that the crosscultural virtues – qualities that would be approved by any non-deluded and impartial human being – may be in short supply. He often seems to be more interested in anatomizing what are regarded as virtues than in arguing that a particular set of virtues is the real cross-cultural thing. Sometimes he is a cataloging virtue theorist, sometimes an offensive virtue theorist, out to raise puritan hackles. But though not devoid of normative concerns, he is often less of a moral legislator and more of a moral sociologist than Swanton seems to think. Swanton gives us Hume without the ambiguities. Baier wants to put the ambiguities back in.

Charles R. Pigden

29

Swanton replies (Essay 13) by clarifying the concept of a normative virtue ethics, distinguishing Aristotelian virtue ethics in particular from normative virtue ethics in general, and emphasizing that it is possible to be a virtue ethicist without being unduly tainted with Aristotelianism. Which means, of course, that it is not inappropriate to construe an anti-scholastic like Hume as a normative virtue-ethicist. Though Thomas Jefferson notoriously detested David Hume (“this degenerate son of science, this traitor to his fellow-men”) this was primarily on account of his history, not because of his philosophy (Jefferson, 1999, pp. 262, 274, 283–5, 383–4). But if Rosalind Hursthouse is right, (Essay 14) Jefferson would have detested Hume even more if he had properly understood the third book of the Treatise. “We hold these truths to be self-evident, That all men are created equal, that they are endowed by their creator with certain unalienable rights; that among these are life, liberty & the pursuit of happiness, that to secure these rights governments are instituted among men.” Thus for Jefferson and the drafting committee of the Declaration of Independence, rights are primary and virtues secondary. Justice, for instance, consists, in part, in giving each person their due, that is in respecting their pre-existing rights. But according to Hume – or, at least, according to Hume as reconstructed by Hursthouse – this gets things the wrong way round. Justice does not consist in respecting pre-existing rights – rather rights are created by the conventions of justice which, for Hume, is an artificial virtue that has evolved because it serves the interests of the members of society. And this applies as much to human rights as to the rather mundane property rights, that preoccupy Hume in Book 3 of the Treatise. It is not quite that that talk of human rights is “nonsense on stilts” (in Bentham’s famous phrase). It is just that it is a mistake to suppose that human rights were “there” prior to the evolution of the artificial virtue of treating people as if they had them. Not for nothing, it seems, was Hume the buddy of Edmund Burke and the inspiration of Jeremy Bentham, those two notorious critics of the rights of man and of the citizen! (Waldron, 1987.) Thus Hume has something interesting to say about reason, motivation, morals and virtue and maybe even human rights. Now, read on …

1 Expressivism, Motivation Internalism, and Hume Richard Joyce

David Hume is often taken to be a moral expressivist (Flew, 1963; Ayer, 1980, pp. 84–5; Price, 1988, p. 6; Snare, 1991; Harman, 1996, p. 97). He is, moreover, often taken to have presented in the Treatise one of the strongest arguments for moral expressivism: the so-called Motivation Argument. As a metaethicist, I am interested in whether expressivism is true, and thus interested in whether the argument that people think they find in Hume is a sound one. Not being a Hume scholar (but merely a devoted fan), I am less interested in whether Hume really was an expressivist or whether he really did present an argument in its favour. Hume’s metaethical views are very difficult to nail down, and by a careful selection of quotes one can present him as advocating expressivism, or cognitivist subjectivism, or moral scepticism, or a dispositional theory, or an ideal observer theory, or even utilitarianism. It is entirely possible that Hume’s position is indeterminate when considered against these terms of modern moral philosophy; it is also entirely possible that he was hopelessly confused (much as it pains me to admit it). However, I doubt very much that Hume should be interpreted as an expressivist in any straightforward manner, and therefore I am doubtful that he should be interpreted as arguing in its favour. Most of this essay does not discuss Hume directly at all: I critically discuss the Motivation Argument and I advocate a certain positive metaethical view – one that mixes elements of traditional expressivism with elements of cognitivism. This position is neutral between moral realism and radical moral scepticism. I close by wondering – very briefly – whether Hume might have held such a view. Given my reservations about the determinacy of Hume’s metaethical outlook, the case is not pressed with any vigour, but because it is an interpretation of Hume that has not, so far as I know, been articulated before, it may be of interest to note that it seems to be consistent with much of what he says – at least as much as any other precise interpretation.

30

Richard Joyce

31

1.1 Expressivism and motivation internalism Let me start by clarifying terminology. Noncognitivism is the metaethical view according to which public moral judgements do not express beliefs (i.e., are not assertions) in spite of the fact that they are typically formed in the indicative mood. Thus defined, noncognitivism is a view of what moral judgements are not – leaving open space for many different forms of noncognitivism claiming what moral judgements are. One positive form, prescriptivism, holds that moral judgements are really commands.1 My focus in this essay is on another form, expressivism, which holds that moral judgements function to express desires, emotions, pro-/con-attitudes, or (in Simon Blackburn’s words) ‘a stance, or conative state or pressure on choice and action’ (1993, p. 168). I treat ‘expressivism’ and ‘emotivism’, as they appear in metaethical discussions, as synonyms. Why might one be tempted by noncognitivism (and expressivism in particular)? First, noncognitivism sidesteps a number of thorny metaethical puzzles that face the cognitivist. The cognitivist thinks that when we make a public moral judgement, like ‘That act of stealing was wrong’, we are asserting that the act of stealing in question instantiates a certain property: wrongness. But queries arise: What kind of property is wrongness? How does it relate to the natural properties instantiated by the action? How do we have epistemic access to the property? How do we confirm whether something does or does not instantiate the property? The difficulty of answering such questions may lead one to reject the presupposition that prompted them: one might deny that in making a moral judgement we are engaging in the assignment of properties at all. Such a rejection, roughly speaking, is the noncognitivist proposal. Second, the noncognitivist might claim the advantage of more readily accounting for certain aspects of moral disagreement – for example, its vehemence and intractability (see Stevenson, 1963, Essays 1 and 2). The third traditional consideration in favour of noncognitivism is the subject of our attention: that noncognitivism does a better job than its rival of explaining the apparent motivational efficacy of moral judgement (see Smith, 1994, Chapters 1–2). Those who advocate this third argument for noncognitivism often look to Hume for a precedent, finding solace especially in the following passage: Since morals, therefore, have an influence on the actions and affections, it follows, that they cannot be deriv’d from reason; and that because reason alone, as we have already prov’d, can never have any such influence. Morals excite passions, and produce or prevent actions. Reason of itself is utterly impotent in this particular. The rules of morality, therefore, are not conclusions of our reason. (T, 3.1.1.6/457)

32 Expressivism, Motivation Internalism, and Hume

It is not unusual to find Hume’s premise that ‘morals excite passions’ formulated as motivation internalism: Simple Motivational Internalism (simple-MI): It is necessary and a priori that anyone who judges that she is morally required to ␾ will be (defeasibly) motivated to comply. My object here is not to assess the truth of simple-MI, but to investigate its logical relation to metaethical expressivism. There are broadly two ways in which one might use simple-MI in favour of expressivism. One might straightforwardly think that simple-MI entails expressivism, presumably with the supplement of some unobjectionable additional premise(s). Alternatively, one might think that simple-MI is a desideratum that any metaethical theory must strive to satisfy, and thus if expressivism were to entail simple-MI this would count very much in expressivism’s favour. Let us examine both these implications in turn. Does expressivism imply simple motivation internalism? No, it does not. Expressivism, as I have stated it, is a thesis about what kind of mental states are expressed by moral judgements. It is vital to note that the notion of expression that is relevant here is non-causal. One can express a mental state while not having that state, and perhaps having never had that state. Consider a promise, which is a kind of speech act by which we express intentions or commitments. If I, in unexceptional circumstances, say to you ‘I promise to be at the party tonight’, then I have expressed an intention to be at the party tonight. But my promise may nevertheless be insincere, in the sense that I have no intention of coming to the party, and have never had any intention to come. Since that intention appears nowhere in my mental repertoire, it cannot be the cause of my promise utterance. Now consider the following thesis: Simple Promising Internalism: It is necessary and a priori that anyone who promises to ␾ (thereby expressing the intention to ␾) has the intention to ␾. It is clear that simple promising internalism does not follow from the thesis that promises express intentions. The phenomenon of insincerity is sufficient to demonstrate this. If, then, we construe metaethical expressivism as a thesis about what kind of speech act moral judgements are – which is natural if we read it as the denial that moral judgements are assertions, since assertion is a category of speech act (see Austin, 1962; Searle, 1969) – then the phenomenon of insincerity is entirely sufficient to show that simple-MI does not follow.

Richard Joyce

33

The reader might be wondering precisely what kind of relation is denoted by ‘expression’ in this context if it is not a casual one. This is something I discuss later. One might also object that I have construed either expressivism or motivation internalism (or both) incorrectly. They are, I admit, both theses for which there is disagreement as to their correct formulation. I will consider variants in due course, but first let me consider the reverse implication with these simple formulations. Does simple motivation internalism imply expressivism? No, it does not. From the fact that there is a necessary (and a priori) connection between a kind of mental state and a kind of speech act, it does not follow that the speech act expresses that mental state. Let us consider promises again, and consider what criteria must be satisfied in order for X to succeed in making a promise (albeit possibly an insincere one) to Y. The most complete answer to this question comes from John Searle (1969, pp. 57–61), who painstakingly delineates the conditions that must obtain if S is to promise that p to H via uttering T. I shall not rehearse all Searle’s items, but just focus on a couple. First: H would prefer S’s doing A to his not doing A,2 and S believes H would prefer his doing A to his not doing A. Second: It is not obvious to both S and H that S will do A in the normal course of events. Both these criteria make essential reference to the parties to the promise having certain beliefs. These connections are necessary and a priori: it is not possible that any person could succeed in making a promise to another person without their having these beliefs. Yet we would hardly say that the act of promising functions to express the belief that the promisee would prefer that the promised action be performed to its not being performed – rather, the mental state expressed by the promise is as we originally stated: an intention or commitment. This suffices to show that the occurrence of a type of speech act may entail that the speaker has a certain kind of mental state, though the speech act doesn’t function to express that state. A related confusion has cropped up in some quarters over what metaethical conclusions might be drawn from certain recent empirical results that show the important role that emotions play in moral judgement (see Greene et al., 2001; Greene & Haidt, 2002; Moll et al., 2002; Haidt, 2001). Though these scientists’ conclusions are not uncontroversial, let us take them at their word when they assert that ‘recent evidence suggests that moral judgment is more a matter of emotion and affective intuition than deliberative reasoning’ (Greene & Haidt, 2002, p. 517). This conclusion is often referred to as ‘emotivism’ (Haidt, 2001, p. 816; Greene et al., 2001, p. 2107; Greene et al., 2004, p. 397). Anthropologist Daniel Fessler claims that ‘emotivist perspectives on moral reasoning hold that emotional reactions precede propositional reasoning’ (Fessler et al., 2003, p. 31). Let us be a little bolder, and interpret this kind of emotivism as the claim that all moral judgements are caused by emotional arousal.

34 Expressivism, Motivation Internalism, and Hume

Clearly, this use of ‘emotivism’ among empirical scientists is very different from the metaethicist’s usage, for whom it is usually a synonym of ‘expressivism’.3 The terms ‘emotivism’ and ‘expressivism’ in the metaethical tradition do not denote a thesis about the causal origins of moral judgement; they denote (as we have seen) a thesis about what kind of mental state is expressed by public moral judgements. It might be best if we distinguish ‘psychological emotivism’ (the kind advocated by Jon Haidt, for example) from ‘metaethical emotivism’ (advocated by A. J. Ayer and Simon Blackburn, for example).4 The crucial point is to note the logical independence of the two: even if the evidence were to demonstrate that every single moral judgement is caused by emotional arousal (i.e., demonstrate that psychological emotivism is true), this wouldn’t imply anything about the function of moral language. Consider, for example, any kind of metaethical theory according to which moral utterances are veiled reports about one’s own mental states. According to this kind of cognitivist subjectivism, ‘X is morally wrong’ means ‘I feel disapproval towards X.’5 The latter is something that may be asserted, yet in the typical case it will have been prompted by emotional arousal in the speaker. Such a theory would be consistent with psychological emotivism but inconsistent with metaethical emotivism. On at least one occasion, Hume sounds like he endorses some such view: ‘[W]hen you pronounce any action or character to be vicious, you mean nothing, but that from the constitution of your nature you have a feeling or sentiment of blame from the contemplation of it’ (T, 3.1.1.23/469). I doubt that this particular simplistic subjectivist interpretation of Hume ultimately withstands scrutiny, but I have little doubt that Hume should nevertheless be interpreted as some kind of psychological emotivist: he clearly and emphatically thinks that moral judgements have their origin not in the faculty of reason but in sentiment.6 My point is that his advocacy of psychological emotivism does not commit him to the metaethical variety; there is very little evidence that he advocated, or even had much awareness of, metaethical emotivism/expressivism/noncognitivism. Many of his emotivist-sounding moments (for example, ‘Morality … is more properly felt than judg’d of’ (T 3.1.2.1/470)) may be smoothly interpreted as advocating psychological emotivism rather than metaethical. Indeed, in the earlier-quoted passage from the Treatise from which the Motivation Argument is drawn, what we in fact have, I would claim, is an argument for psychological emotivism, not metaethical emotivism. In sum, there is clearly a significant difference between motivation internalism and psychological emotivism. One asserts a necessary connection between moral judgement and conative states whereas the other asserts a causal connection between the two. But advocates of either must be wary of the same potential pitfall: of assuming that their thesis implies or provides support for the thesis that metaethicists have called ‘emotivism’ or ‘expressivism’. I have argued that expressivism does not imply motivation

Richard Joyce

35

internalism, and nor does motivation internalism imply expressivism. The latter denial raises serious problems for an argument that some people have found in Hume: the Motivation Argument. Rather than claiming that Hume argued poorly, however, I urge the conclusion that it is a mistake to foist this argument for expressivism on him in the first place.7 Some will object to all of the preceding on the grounds that I have misconstrued motivation internalism or misconstrued expressivism (or misconstrued both). So it is to variations on these theses that I now turn.

1.2 Variants of expressivism and motivation internalism A prominent variation on motivation internalism is the following, due (inter alios) to Michael Smith (1994): Normative Motivation Internalism: It is necessary and a priori that anyone who judges that she is morally required to ␾ will be (defeasibly) motivated to comply, or she is irrational. I mention this variant simply because it is well known, but in fact there is little to say about it here. The additional clause on the end, though important in other philosophical contexts, does not affect any of the arguments that I have already deployed to show the logical independence of expressivism and simple motivation internalism. This is less obviously so of the following variant: Sincerity Motivation Internalism (sincerity-MI): It is necessary and a priori that anyone who sincerely judges that she is morally required to ␾ will be (defeasibly) motivated to comply. (One may choose to add the suffix ‘… or she is irrational’; it doesn’t matter to anything that follows.) The thesis is not always worded in just this way, but the term ‘sincerely’ is often included in statements of motivation internalism (see Hare, 1999, Chapter 8; Timmons, 1999, p. 53; Svavarsdóttir, 2006, p. 186; Shafer-Landau, 2005, p. 142). Sincerity-MI does not imply expressivism. Proof: Sincerity-MI is implied by simple-MI, therefore if sincerity-MI were to imply expressivism, then so too would simple-MI imply expressivism. But we have already seen that simple-MI does not imply expressivism, therefore nor does sincerity-MI. Thus, construing MI as sincerity-MI provides no succour for the (alleged) Humean Motivation Argument for expressivism. But does expressivism imply sincerity-MI? One might be tempted to think so. Suppose for the sake of argument that expressivism is true: that when one (in ordinary circumstances) utters ‘Stealing is morally wrong’ (say),

36 Expressivism, Motivation Internalism, and Hume

one thereby expresses some conative (that is, motivation-implicating) state. One might think that from this follows something about what it is for such a judgement to be sincere. One might think that any conation-expressing utterance of ‘Stealing is morally wrong’ is sincere if and only if the speaker actually has that conative state at the time of utterance. This would be an instance of a tempting general principle of speech-act sincerity, which I shall name after its advocate, John Searle (1969): Searle-Sincerity: S’s utterance U (at time t) is sincere iff U expresses mental state M, and S has M (at t). With Searle-Sincerity as an additional premise, it does appear that expressivism implies sincerity-MI. The problem is that Searle-Sincerity, plausible as it may appear at a glance, is false. On an earlier occasion (Joyce, 2002), I offered some counterexamples to Searle-Sincerity. I imagined someone saying ‘Thanks!’ as he left a dinner party in a distracted and hurried way, and claimed that though we might admit that at the moment of thanking he was feeling no gratitude whatsoever, nevertheless we wouldn’t ordinarily call his utterance ‘insincere’. A second counterexample along the same lines concerned an act of passing moral judgement. Michael Ridge (2006) criticises these counterexamples on the grounds that they fail to take into account the fact that although the speaker may not have gratitude (say) as an occurrent emotion, he nonetheless may count as having that mental state (at the time of utterance) dispositionally. I harbour misgivings about dispositional mental states (especially emotional ones), but let us not pause to consider them now, for the main point is that Ridge nevertheless agrees with me that Searle-Sincerity is inadequate, and supplies counterexamples of his own that revolve around self-delusion. A person may believe himself to have mental state M when in fact he does not. If there is a speech act that expresses M, and the person performs that speech act, then it seems natural to say (Ridge argues) that the speech act is sincere, even though the speaker lacks the mental state that it expresses. Ridge presents an alternative general thesis of speech-act sincerity: Ridge-Sincerity: S’s utterance U (at time t) is sincere iff S believes that U expresses mental state M, and S believes that she has M (at t).8 I am concerned that Ridge’s version of speech-act sincerity is also problematic, in that it presupposes that ordinary speakers have beliefs about a kind of expression-relation holding between utterances and mental states – but this, I suspect, is far too recherché a belief to require of ordinary speakers in order that they may be granted speech-act sincerity (even if we allow

Richard Joyce

37

that the belief may be implicit, dispositional, and non-conscious). We have already seen the confusions that may entrap the unwary concerning causal versus conventional notions of expression; and if even analytic philosophers stumble over this, what hope should we have that ordinary speakers’ beliefs are in order? Even children of a tender age have the capacity to make assertions, ask questions, bark commands (and so forth) – and to do so in a sincere manner – yet surely they have no beliefs about what kind of mental states various utterances express. I once had an argument with a well-known philosopher (who shall remain nameless) who declared that metaethicists have no idea what they are talking about when they wonder about what mental states moral judgements express. He claimed that it was as if metaethicists had just assumed the existence of some mysterious relation holding between moral utterances and mental states, then given it a name – ‘expression’ (though it may as well have been ‘floog’) – and have then expended endless energy arguing in circles about this baffling relation. This philosopher presumably (or at least conceivably) did not believe that any of his utterances expressed any mental states; he was sufficiently sceptical of the whole notion that he just withheld assent to such thoughts. Do we want to claim that this philosopher, despite himself, ‘implicitly’ had such beliefs? I wouldn’t wish to maintain this; it seems a rejoinder of desperation. I might allow that his speech acts, despite his beliefs, did express various mental states, but I see little plausibility in the claim that, despite himself, he believed this fact. And yet, for all this, I am quite certain that this person was capable of making sincere assertions, sincere promises, sincere apologies, and so forth. Here I am not going to argue for an alternative general theory of speechact sincerity, since I question the assumption that a general account is forthcoming or even particularly desirable. Perhaps what must be added to a promise to ensure its sincerity differs from what must be added to an assertion to ensure its sincerity, while both differ from congratulations, apologies, entreaties, thankings, and so on. If there is anything that unites these cases, in my opinion, it will revolve around the fact that insincere speech acts are ones by which the speaker attempts knowingly to mislead his/her audience – and such audience-directed intentions are not mentioned, nor entailed, by either Searle-Sincerity or Ridge-Sincerity. But I shall not develop this thought on this occasion, for the point that matters to our present purposes does not require it. For our present purposes I can even embrace Ridge-Sincerity. The point is that expressivism promises to imply sincerityMI only with Searle-Sincerity as a bridging premise, but Searle-Sincerity has been refuted, and there is no reason to assume that whatever general thesis of speech-act sincerity replaces it (if there even is one) will also act as a bridge from one thesis to the other. Of course, if we have any account whatsoever of what a sincere moral judgement consists of, it will follow trivially that some kind of vaguely

38 Expressivism, Motivation Internalism, and Hume

motivation-internalism-ish thesis will be implied by expressivism. Consider the generalised argumentative format: Premise 1: Moral judgements express conative state C (regarding the subject of the judgement). Premise 2: Sincere moral judgements have quality Q (vis-à-vis the mental state expressed). Therefore: If S sincerely judges that she is morally required to ␾, then her judgement has quality Q vis-à-vis C (regarding her ␾ing). I call the conclusion ‘motivation-internalism-ish’ on the grounds that it asserts a relation between moral judgement and a kind of motivational state (C). (And if the two premises are necessary and a priori then the relation asserted in the conclusion will also be so.) But in fact this conclusion is derived trivially, and has no useful role to play in the metaethical dialectic. Recall that the expressivist hope has been that some version of motivation internalism might have independent attractions, thus boosting the case for expressivism either by implying expressivism or by being revealed to be a desideratum that expressivism satisfies better than its rivals. But the kind of trivial MI-ish conclusion just mentioned can play neither role. Another obvious problem with any version of motivation internalism that restricts itself to sincere moral judgements is that by encompassing only a proper subset of moral judgements, it fails to tell us anything about moral judgement simpliciter. By comparison, I could tell you something true about all moral judgements made on a Saturday – say, that they are ‘weekend moral judgments’ – but this obviously would tell us nothing about what we are interested in as metaethicists: namely, what a moral judgement is, what its necessary features are, and so on. Similarly, the fact that sincerity-MI reveals a connection between sincere moral judgements and motivational states doesn’t imply any necessary connection between motivation and moral judgements simpliciter, since the motivational aspect may be smuggled in within the concept of sincerity. (If I am correct that the most promising account of speech-act sincerity will make reference to the speaker’s intentions not to deceive, then sincerity will automatically bring motivation along for free, since these kinds of intentions are motivation-engaging states.) Any version of internalism restricted to sincere moral judgements is compatible with the falsity of expressivism concerning moral judgements simpliciter. I have lately been discussing variations on the thesis of motivation internalism, but it may also be objected that I have misconstrued the thesis of expressivism. One might, in particular, complain about my characterization of expressivism as a metaethical theory about speech acts; one might instead insist that expressivism is a theory about mental states: not about what kind of mental state moral judgements express, but about what kind of mental

Richard Joyce

39

state moral judgements are. On such a view the applicability of the sincere/ insincere distinction retreats and, indeed, the whole troublesome expression relation conveniently evaporates. This mentalistic construal of expressivism is unconventional. If we go back to the roots of noncognitivism in the early twentieth century, we see pretty clearly that what is under discussion is the nature of moral language. In their influential 1923 book The Meaning of Meaning, C. K. Ogden and I. A. Richards speak of a use of the word ‘good’ which is ‘purely emotive’, and ‘[w]hen so used the word stands for nothing whatsoever, and has no symbolic function’ (p. 125). A. J. Ayer’s noncognitivism was motivated by the question of how moral utterances might be meaningful statements ([1936] 1971).9 Rudolf Carnap’s noncognitivism was presented as the claim that a ‘value statement’ is not ‘really an assertive proposition’, but is, rather, ‘a command in misleading grammatical form’ (1935, pp. 24–5). Charles Stevenson spoke of ethical judgements as having ‘quasi-imperative force’ which may be ‘intensified by your tone of voice’ (1937, p. 19). At no point in these classic works in the emergence of noncognitivism is it hinted that ‘moral judgment’ might be used primarily to denote a mental state. But perhaps that is all a misguided historical idiosyncrasy, and perhaps we would do better now to treat moral judgements as a species of mental state. In this case, expressivism will be the theory that moral judgements are not beliefs, but rather some kind of conative state (to be specified). One obvious problem with such a decision is that it opens the possibility that moral judgements (qua mental states) might be conative while moral judgements (qua linguistic entities) might be assertoric. This is more or less the same possibility as was noted earlier, when I pointed out that metaethical cognitivism (interpreted in the orthodox manner) is compatible with either simple-MI or psychological emotivism. Of course, this observation doesn’t count as evidence against mentalistic expressivism, but rather indicates how confusing this way of characterising theories might become. In the present context, the important thing to note is that construing expressivism mentalistically would nullify the possibility of any argumentatively interesting relation holding between expressivism and motivation internalism. If expressivism is the theory that moral judgements are episodes of conative state C (where ‘C’ denotes something that is by stipulation necessarily motivation-engaging), then expressivism is essentially equivalent to the thesis of motivation internalism, which states that moral judgements necessarily engage motivation. The connections appear to be so trivial that arguing for either thesis by means of first establishing the other ceases to be a feasible dialectical strategy. This has particular relevance to the so-called Motivation Argument that is drawn from Hume’s Treatise. From the premise that ‘morals excite passions’ (that is, putatively, that moral judgements necessarily engage motivations) one can certainly derive mentalistic expressivism – but only trivially: the conclusion essentially is the premise.10, 11

40 Expressivism, Motivation Internalism, and Hume

One can read the preceding as presenting a dilemma to any expressivist tempted to employ the Humean Motivation Argument. Either expressivism is construed mentalistically (as a theory of what kind of mental states moral judgements are), in which case the argument is valid but question begging, or it is construed linguistically (as a theory of what kind of mental states moral judgements express), in which case the argument is unsound. I have made clear my preference for construing expressivism in the latter fashion (thus impaling the advocate of the Motivation Argument on the second horn), and, in the course of discussion, I made much of the fact that the relevant expression relation must not be understood in a causal manner. I should like now to say more in a positive vein about how that expression relation ought to be understood. This discussion takes us well away from Hume – which is hardly surprising given my contention that thinking of Hume in the guise of a modern expressivist is a serious distortion – but I will close with some brief thoughts applying what we have learned to Hume.

1.3 Expressing mental states When seeking explication of the sense in which types of speech acts express types of mental states, it is useful to start with Moore’s Paradox. G. E. Moore (1942, p. 543)12 noted the oddity of someone claiming: (1)

I went to the pictures last Tuesday. But I don’t believe that I did.

It’s called a ‘paradox’ because although it is not a logical contradiction (for it is perfectly possible that I went to the pictures last Tuesday while I don’t believe that I did), to state the whole is to void the speech act of the first part, leaving the listener confused as to what should be assumed about the speaker’s attitude towards his having been to the pictures last Tuesday. It makes (to quote J. L. Austin) ‘a peculiar kind of nonsense’ ([1970] 1990, p. 112). Moore presents the paradox using the category of assertion, but it seems we can find exactly the same phenomenon with other species of speech act: (2) (3)

I apologize for having lied to you. But I have absolutely no regret about having lied to you. Thank you for the present. But I have no gratitude towards you for giving me the present.

One can imagine someone publicly uttering the first component of any of (1), (2), and (3) while uttering the second component sotto voce – in which case the public speech act (assertion, apology, and thanking, respectively) would simply be insincere. There is nothing very noteworthy about that.

Richard Joyce

41

What is strange is when the second component is uttered out loud along with the first – since a speech act that wears its insincerity on its sleeve is apt to cause confusion.13 It is, to quote Austin again, ‘a statement that fails to get by’ (1971, p. 18). Note that I am not merely saying that (1) –(3) ‘sound odd to my ear’, but am trying to locate a particular kind of oddity, which promises to help us understand the relevant notion of expressing a mental state. David Copp introduces a useful phrase to denote this kind of expression relation: ‘Moore-expression’ (correspondence 2003; see his 2001, p. 10). It is illuminating, I think, to consider these matters when reflecting on pejorative language, such as racial slurs. (4)

Aaron’s a kike. But I have no contempt towards him or people of his ethnicity or religion.14

I am inclined to think that (4) is a manifestation of the same phenomenon as the rest, though Copp would disagree. Though he would concur that calling someone a ‘kike’ certainly in some sense expresses contempt, and that someone who uttered (4) should expect to be challenged to explain, he judges that it doesn’t Moore-express contempt, since the implication can be cancelled. (Copp’s example concerns calling a canine a ‘cur’ as opposed to a ‘mongrel dog’; I am assuming that it is a fair analogy.) One could add further explanatory comments to (4) so as to assuage the audience’s confusion in the following manner: ‘Aaron is a kike; but I have no contempt towards Aaron or people of his ethnicity or religion; it’s just that for the moment I’ve forgotten the usual non-derogatory term for such persons.’ Thus, Copp thinks, the contempt expressed by ‘kike’ may be cancelled. Perhaps he is right, though I confess to feeling uneasy about having to rely on an example of someone forgetting a word or learning a language in order to illustrate the context of cancellation. Is there anything that is non-cancellable if we admit such odd contexts? Consider the case of the Reverend William Spooner, who famously concluded one of his sermons with the addendum: ‘In the sermon I have just preached, whenever I said “Aristotle” I meant to say “St. Paul”.’ Suppose the sermon had included the assertion of the sentence ‘Aristotle was born in Tarsus’ (when in fact it was St. Paul who was born there), and, when queried on this by a bewildered student, Spooner had responded (coming momentarily to his senses, but oblivious of what he had just been saying) ‘No, of course Aristotle wasn’t born in Tarsus; he was born in Stagira!’ Thus we have what appears to be a flat contradiction asserted – ‘Aristotle was born in Tarsus and it is not the case that Aristotle was born in Tarsus’ (the first conjunct uttered as part of the confused sermon, and the second conjunct, being an answer to a question, not part of the sermon) – but one made intelligible by the addendum admitting a linguistic confusion. This possibility hardly undermines the fact that ‘Aristotle was born in Tarsus’ is semantically

42 Expressivism, Motivation Internalism, and Hume

implied by ‘Aristotle was born in Tarsus’, and that this implication is as noncancellable as they come! Recall that Paul Grice confessed to the unreliability of the cancellability test when speakers are talking ‘in a loose or relaxed way’ (Grice, 1989, p. 44), and I suspect that imagining a person who has an imperfect grasp of the language, or who is forced to employ unconventional words because she cannot remember the correct ones, renders the test null. We will return to the question of the status of (4) shortly, but for now apply these thoughts to a moral utterance: (5)

Hitler was evil. But I do not believe that he was evil.

This seems a straightforward instance of Moore’s Paradox, which strongly suggests that an ordinary freestanding utterance of ‘Hitler was evil’ is an assertion. If an utterance of the form ‘But I don’t have mental state M’ can nullify the speech act that the preceding comment would otherwise perform, then it is natural to assume (ceteris paribus) that the preceding comment functions to express M. So does noncognitivism stand refuted? Not by any means, for it seems that we can observe the same phenomenon, or at least a very similar one, if in (5) we substitute for ‘belief’ something conative. Which conative state we choose, and how we opt to describe it, may make a significant difference here, and there are various options. Copp (following Allan Gibbard) prefers to see the expressed conative state as acceptance of, or subscription to, a moral standard (Copp, 2001, p. 30). According to this stipulative use of ‘subscription’, if a person actually does take this attitude, ‘she is in a state of mind that, if effective, constrains and guides her planning so that she is motivated to some degree [to comply]’ (Copp 2001: 30). Since no proposal more plausible suggests itself, let us use the following wording: (6)

Hitler was evil. But I subscribe to no normative standard that condemns him or his actions.

I’m cautiously tempted to treat (6) the same as the others, to hold that all six of the numbered sentence pairs so far considered are Moore-paradoxical, meaning that in each the relation between the speech act of the first sentence and the mental state mentioned in the second sentence is one of Moore-expression. If this is true of (5) and (6), then the conclusion to be drawn is that an ordinary freestanding utterance of ‘Hitler was evil’ expresses both a belief and a conative attitude. This would speak in support of a metaethical view that mixes aspects of traditional noncognitivism with components of traditional cognitivism – a view I favour.15 Copp, however, though agreeing that (1), (2), (3), and (5) reveal instances of Moore-expression, thinks that (6) should go with (4) – as not Mooreparadoxical – for the reason mentioned earlier: that the strangeness of

Richard Joyce

43

(6) may be assuaged via additional commentary, and thus the customary relation between the utterance and the mental state in question may be cancelled. He would find nothing unintelligible about an amoralist who says ‘Hitler was evil; but I subscribe to no normative standard that condemns him or his actions. I just don’t go in for morality; I believe in it all right, but I think it’s a manipulative cultural invention that is best avoided.’ I have two responses to Copp’s claim here. The first is to deny that the amoralist’s declaration is intelligible in the relevant way, and thus maintain grounds for holding that (6) is Moore-paradoxical. The second is to accept that (6) is not Moore-paradoxical, but explore whether there is another kind of expression involved – a kind that, while not being Moore-expression, is nevertheless robust enough to underwrite the noncognitivist element of moral judgements. Either avenue is sufficiently promising that the prospects for the mixed metaethical view to which I have just adverted seem encouraging. Discussion of the former response calls for a section to itself, after which I will turn to the latter response.

1.4

Amoralist cancellation

There is, of course, a sense in which all of the numbered sentence pairs (1)–(6) are intelligible. They are grammatical and non-contradictory sentences. But so too is any Moore–paradoxical sentence pair (and many nonsensical sentences besides). Yet I maintain that they are unintelligible in the sense that someone hearing any such pair (someone who has not been primed in some special way, that is) would be unsure about what speech act has been performed by the first component.16 Considering (6): Copp is correct that any perplexity that may be aroused may also be assuaged with further commentary – the kind of ‘but-I-don’t-go-in-for-morality’ comment that was mentioned. We are familiar with this kind of an amoralist; after all, didn’t Plato’s Thrasymachus proclaim something along similar lines?17 Yet care needs to be taken in our treatment of these pieces of additional commentary that render intelligible preceding speech that would otherwise be confusing. There’s a sense in which any verbal nonsense can be rendered intelligible with the addition of ‘... and what I just uttered was a great example of verbal nonsense’. The crucial matter is whether the additional intelligibilityinstating commentary leaves intact the apparent speech act performed by the first sentence. For example, suppose that A utters after a dinner party ‘Well, the plates were nice’; we might presume that it is being conversationally implicated that the food was unpleasant. But this implicatum may be smoothly cancelled, if A were to add: ‘Of course, I don’t mean to imply that the food was unpleasant; it was nice too.’ Whatever wisps of strangeness might remain hanging over this pair of utterances will be dissipated if A explains: ‘I was just so taken by the plates that for a moment I wasn’t thinking about the food.’ The important thing to notice is that at the end of all this explaining we are

44 Expressivism, Motivation Internalism, and Hume

content that A’s first sentence was indeed what we thought it was: an assertion that the plates were nice. By comparison, if B says ‘The cat is on the mat, but I don’t believe it’, then goes on to add ‘and that’s a good example of confusing language’, then although listeners may be comfortable with the total exchange, they will not know whether B has asserted that the cat is on the mat. The crucial question, then, is not whether the Thrasymachian amoralist who renders intelligible (6) by explaining that he ‘just doesn’t go in for morality’ is someone we can make sense of (I concede that he is); the question is whether after the intelligibility-reinstating amoralist explanation the audience is confident that the speaker has made a genuine moral judgement. And about this I think there is substantial doubt. Before discussing the amoralist further, it will be useful to remind ourselves of how fluid and scrappy linguistic conventions can be. Let me draw attention to four general points. First, a solid linguistic convention may be quite easily overridden by another. There is little doubt that the term ‘slut’ functions as a pejorative in English. Yet by introducing the overarching convention of joking – which may be achieved in a second by a shift in tone or a twitch of an eyebrow – one might in a playful manner say to a close female friend ‘Oh, you’re such a slut’ with all offensiveness nullified. Yet even in these circumstances ‘slut’ continues to be a contempt-expressing term, for that, after all, is what makes the comment funny. Second, many terms that function to express attitudes as well as beliefs will also have purely belief-expressing uses as well. The word ‘queer’ in the sense of unusual and peculiar remains neutral even if ‘queer’ in reference to homosexuality can be used as a term of derision. The word ‘bastard’ for a long while could be used descriptively to mean illegitimate offspring, even when it could also be employed as a term of insult. Similarly, most, if not all, of the terms centrally associated with moral judgement also can be used non-morally. Possibly the most fundamental term of moral appraisal is ‘ought’. The ‘ought’ that appears in ‘Mary morally ought to refrain from stealing’ may express the speaker’s subscription to norms that condemn stealing, but nobody is claiming this of the weather forecaster’s utterance of ‘It ought to rain tomorrow.’ Similarly, to acknowledge that someone is ‘a good assassin’ is not to express any kind of endorsing attitude, whereas to claim that someone is ‘a good person’ is. A third point to note about such conventions is that they can change very quickly. The terms ‘idiot’, ‘moron’, and ‘cretin’ were once respectable scientific labels; the term that largely replaced them – ‘mentally retarded’ – is at present the subject of controversy. There will be transitional times when one should not claim with confidence either that there is or is not a linguistic convention according to which the term expresses an attitude. A fourth point is that linguistic subgroups can create linguistic subconventions. Within certain gay circles, referring to one’s gay friends as

Richard Joyce

45

‘queer’ may be neutral, though it may be highly insulting for an outsider to select that term. Much the same could be said about the use of the word ‘nigger’ among some African American subgroups. In some circles, to call something ‘bad’ (in a certain tone of voice, perhaps) is a way of praising it, and in surfing lingo, to call a wave ‘wicked’ is to express admiration for its qualities. Such conventions may be sequentially embraced and overridden within a single conversation – or, indeed, within a single sentence: ‘Which of you bastards called this bastard a bastard?’18 What I hope these observations call attention to is the fact that although we might be able to imagine someone intelligibly advocating the amoralist line, it doesn’t show that there is not actually an entrenched convention according to which the use of moral terms expresses subscription to a norm. What we need to ask ourselves is whether any such imaginative act involves us thinking of aberrant subgroups, or people speaking in a joking, playful manner, or the speaker using something like a sarcastic or ironic tone of voice, or using a moral term in a non-moral manner, or introducing a new convention by example, or a world with slightly different linguistic conventions than we actually do have, or so on. When Milton’s Satan says ‘Evil, be thou my good’ the natural reading is that Satan is doing something tricky with language. A careful analysis of his comment would take too long here; it’s enough to note that although we know exactly what Satan is trying to communicate, we also recognise that the surface construction is paradoxical (and this, of course, is what gives the line its poetic power). Thus, that Satan’s comment should be intelligible doesn’t reveal that there is not actually a linguistic convention according to which to call something ‘evil’ is to express one’s subscription to a standard that condemns it. It is exactly this convention that Milton has exploited in a clever and mischievous way. I believe the same thing can be said quite generally of the amoralist’s apparent cancellation of the conative component of a moral judgement: ‘Hitler was evil; but I subscribe to no normative standard that condemns him or his actions. I just don’t go in for morality; I believe in it all right, but I think it’s a manipulative cultural invention that is best avoided.’ Although a person would be intelligible if she said this, it seems to me that we would be left in serious doubt as to whether she has really judged Hitler to be evil. And the reason for this indecision, I think, is precisely that the careful explanation that the speaker offers of her position reveals that a degree of stipulative usage is being introduced. She is explicitly suspending a convention that is in place in regular language. But the fact that one can do this, and do it with ease, hardly shows that there is not actually such a convention, any more than the fact that I can say ‘For the next few minutes I will use the word ‘cat’ to stand for dogs’ (or ‘Cats, be thou my dogs’) reveals that there is some doubt concerning whether in English ‘cat’ denotes cats.19

46 Expressivism, Motivation Internalism, and Hume

1.5

Frege-expression

I have just argued in favour of treating the relation between moral judgement and certain conative states as an instance of Moore-expression. The issue of where the line should be drawn between Moore-paradoxical and non-Moore– paradoxical utterances is difficult to settle, since the rules for how we should restrict the contexts in which cancellation may or may not be possible are undecided. But suppose that Copp is correct that (4) and (6) are not examples of Moore-paradox. Does it follow that there is no expressivist component to pejorative slurs and moral judgements? Copp doesn’t think so, and nor do I. Drawing inspiration from Frege’s views on ‘colouring’, Copp claims that pejorative terms and moral terms Frege-express mental states. Frege wrote that two words might have the very same sense and reference, and yet one might lend the utterance a ‘colouring’ (Färbung) that the other does not, such that choosing to use one word rather than the other (for example, ‘kike’ rather than ‘Jewish person’) might be ‘unsuitable, as if a song with a sad subject were to be sung in a lively fashion’ ([1892] 1997, p. 167). Frege’s own example involves the word ‘cur’: he writes that ‘whilst the word “dog” is neutral as between having pleasant or unpleasant associations, the word “cur” certainly has unpleasant rather than pleasant associations and puts us in mind of a dog with a somewhat unkempt appearance’ ([1897] 1997, pp. 240–1). When such colouring becomes an entrenched custom in the linguistic community – as is the case with words like ‘kike’ or ‘slut’ – then we can, according to Copp, consider the relation between the utterance and the proposition that the speaker has the attitude in question (for example, contempt) to be a variety of conventional implicature. Note, though, that Copp thinks that this expressiveness is cancellable, and in this he diverges from Grice, for whom, apparently, non-cancellability is a feature by which conventional implicatures are to be distinguished from conversational implicatures. Though cancellable, these ‘colourings’ are a type of conventional rather than conversational implicature (for Copp) because they are detachable20 and because to employ a coloured term while lacking the attitude in question would be a misuse of the term. At the risk of sounding evasive, I prefer to sidestep the Gricean framework, if only because it strikes me as sufficiently unclear and contested that one only courts controversy in trying to apply it to new domains. I have already expressed my misgivings about the cancellability of colourings, but even if I agreed with Copp on this point, it seems to me imprudent to employ the term ‘conventional implicature’ – a term of art partially defined by reference to non-cancellability – to categorise the phenomenon.21 Nevertheless, I am strongly inclined to agree with Copp (and Frege) on the general point that colourings are, or at least can be, a matter of linguistic convention. This may be a vague claim, but it is good enough for my present purposes. The contemptuous attitude of someone who uses ‘kike’ rather than ‘Jewish person’ is not merely an expectation that interlocutors will

Richard Joyce

47

have formed on the basis of past observation. In teaching the word ‘kike’ to a novice language-user, it would be intolerably negligent to refrain from mentioning the term’s evaluative baggage. Indeed, someone who didn’t know that ‘kike’ was a contempt-expressing term could legitimately be said not to understand the term properly at all, even if able competently to apply it to all and only Jewish people. Any such ignorant person would not require any tutoring concerning what it takes to be Jewish, but is in need of linguistic instruction. Although the comparison threatens to be misunderstood if taken too far, I agree with Copp that moral language is in important respects like pejorative language. More precisely: the way that moral judgements express conative attitudes is very similar to, if not the same as, the way that pejorative terms express attitudes. The manner in which (6) fails to get by may or may not be precisely the same as the manner in which (5) fails to get by, but it is close enough as to make no difference to the general conclusion that I am trying to reach: that moral judgements express, as a matter of entrenched linguistic convention, both beliefs and conative attitudes. The traditional debate between the cognitivist and the noncognitivist has not taken into account such nuances as the distinction between Moore-expression and Frege-expression – indeed, has been scandalously casual about what it means to say that such-and-such judgements express mental state so-and-so – and thus, were we ultimately to conclude that moral judgements Moore-express beliefs and Frege-express attitudes, this could not be construed as a victory for either party. My main point is that a modicum of reflection on the issue reveals the traditional metaethical debate between the cognitivist and the noncognitivist to rest on a false dichotomy.

1.6 Hume: Expressivist, cognitivist, and sceptic? Earlier I claimed that there is very little evidence that Hume advocated expressivism. This was not entirely true; the point I was trying to press is that the places where Hume has traditionally been read as promoting expressivism (or noncognitivism more generally) should not be construed that way. For the real hints of expressivism in Hume, one must look to where he discusses evaluative language. As a preliminary, we should remind ourselves that Hume did have at his disposal a remarkably forward-looking account of how indicative language can be used in non-assertoric ways. His sophisticated discussion of promising foreshadows J. L. Austin’s: [T]here is a certain form of words … by which we bind ourselves to the performance of any action. This form of words constitutes what we call a promise. … When a man says he promises any thing, he in effect expresses a resolution of performing it. (T, 3.2.5.10/522)

48 Expressivism, Motivation Internalism, and Hume

Regarding evaluative language, Hume is (occasionally) clear that there are entrenched conventions that associate conative states with certain words: Every tongue possesses one set of words which are taken in a good sense, and another in the opposite. (EPM, 1.10/174) [W]hen [someone] bestows on any man the epithets of vicious or odious or depraved, he … expresses sentiments, in which, he expects, all his audience are to concur with him. (EPM, 9.6/272) [T]here are certain terms in every language which import blame, and others praise; and all men who use the same tongue must agree in their application of them. … This great unanimity is usually ascribed to the influence of plain reason, which … maintains similar sentiments in all men. … But we must also allow, that some part of the seeming harmony in morals may be accounted for from the very nature of language. The word virtue, with its equivalent in every tongue, implies praise, as that of vice does blame; and no one, without the most obvious and grossest impropriety, could affix reproach to a term, which in general acceptation is understood in a good sense; or bestow applause, where the idiom requires disapprobation. (Of the Standard of Taste [1757] 1996: 134–5) It is in such passages – few and far between as they are – that we find Hume the expressivist. But these comments are presented in a way that makes clear that Hume considers them as peripheral to any of his central arguments; he evidently does not, on these occasions, take himself to be putting forward any weighty and controversial metaethical thesis in need of argumentative support. In other words, to the extent that Hume is an expressivist, it is not something he thinks worthy of making a song and dance about; he hardly notices that he is taking (what we would now classify as) a metaethical stance. More importantly, note that these expressivist musings have no obvious role to play in the moral thesis that Hume is really obsessed with: that morals are the product a sentimental faculty rather than a rational faculty (a thesis I earlier called ‘psychological emotivism’). In particular, these passages are very far indeed – both logically and textually – from the muchtouted Motivation Argument for expressivism. What I hope to have shown in the previous two sections is that even if Hume does have expressivist leanings, this does not exclude his also robustly endorsing a cognitivist metaethical view. Moral cognitivism comes in both realist and sceptic flavours, and each of these possibilities is compatible with expressivism. Copp, for example, articulates and advocates a position

Richard Joyce

49

he calls ‘realist-expressivism’. He thinks that the truth conditions for the belief expressed by a moral judgement like ‘Cursing is wrong’ concern cursing being prohibited by a ‘relevantly justified or authoritative moral standard or norm’ (2001, p. 27). What it takes for a moral standard to be appropriately ‘authoritative’ is, in the first instance, left open. Copp’s own view is a society-centred theory: that a standard is authoritative just in case ‘its currency in the social code of the relevant society would best contribute to the society’s ability to meet its needs – including its needs for physical continuity, internal harmony and cooperative interaction, and peaceful and cooperative relations with its neighbours’ (2001, p. 28). Since, we may assume, such justification is sometimes forthcoming, in Copp’s view moral judgements will turn out sometimes to be true. He combines this realism with the thesis that moral language Frege-expresses conative states; hence: realist-expressivism.22 I should like to draw attention to another branch of the tree of metaethics: error theoretic expressivism. There are different ways that one might argue for this position, but it is convenient to use Copp’s view as a point of departure. Suppose he is correct that the cognitive element of a moral claim refers to something like the ‘relevantly justified or authoritative moral standard or norm’. One may, nevertheless, think that Copp’s preferred explication of justification is too relativistic or too anthropocentric to capture the kind of practical authority we demand of a moral theory. For the kind of familiar reasons outlined by John Mackie, for example, one might think that inherent in moral discourse is a commitment to a kind of institutiontranscendent practical categoricity that is in fact not satisfied by anything in the world (Mackie, 1977; see also Joyce, 2001).23 Thus one might agree with Copp concerning how to understand the expressivist element of moral discourse, and also agree with his views concerning the truth conditions (broadly construed) of moral judgements, while holding that these truth conditions are never satisfied. Hence: error theoretic expressivism. Might Hume be a realist-expressivist or an error theoretic expressivist? I wouldn’t want to press either case with any confidence, but I will nevertheless close with a brief exploration of Hume’s commitment to the moral error theory. Note that I do not claim for a moment that Hume thought of himself as an error theorist but just expressed himself poorly, nor even that he would have embraced the view had it been articulated to him. But there are certainly threads in Hume’s moral philosophy that lean in that direction.24 In looking for evidence against this interpretation, one might bring forth any of a number of Hume’s critical comments aimed at moral scepticism. But it should be remembered that the kind of sceptic whom Hume has in mind is invariably the Pyrrhonic sceptic: someone who thinks that we cannot know whether claims of a certain kind are true or false and therefore ought to withhold passing judgement on the matter. The error theorist, by

50 Expressivism, Motivation Internalism, and Hume

contrast, is no Pyrrhonic sceptic, but (in classical terms) should be classified as a negative dogmatist (or nihilist). Similarly, many of Hume’s comments apparently targeting moral nihilists (for example, ‘those who have denied the reality of moral distinctions’ (EPM, 1.2/169)) on more careful examination seem to be admonishing those who would pretend indifference, who would claim not to care whether a person was honest or a thief. But this is also something that a moral error theorist may distance himself from. The moral error theorist may be as opposed to tax fraud, as sickened by paedophilia, as horrified by genocide, as anyone else. Error theoretic moral scepticism implies nothing about how tolerant its advocates will be. Hume is opposed not merely to Pyrrhonic ataraxic indifference towards morality, but is averse to any suggestion that philosophising should lead us to give up the practice of making moral judgements. I can discern no hint of moral eliminativism in his writings. The widespread assumption that eliminativism is the natural consequence of a moral error theory may have something to do with a reluctance to press the error theoretic interpretation of Hume. This assumption, however, is flawed. There may be pragmatic reasons for maintaining moral thought and moral language even once moral scepticism has been embraced (see Joyce, 2001; Kalderon, 2005). Or it may be that the human mind is simply unable to give up these practices, even when philosophical considerations have led one to see the flaws. One can find allusions to the former in Hume’s writings, and the latter is something of a recurring theme. In his essay ‘The Sceptic’, he notes ‘that famous doctrine’ that colours exist not in nature but only in the eye, then poses the rhetorical question: ‘[If this were so,] would dyers or painters ever be less regarded or esteemed?’ He goes on to ask ‘why should a like discovery in moral philosophy make any alteration?’ ([1742] 1996: 354). The closing section of Book 1 of the Treatise waxes lyrical about how simply living life will drive all sceptical musings from one’s mind: I dine, I play a game of back-gammon, I converse, and am merry with my friends; and when after three or four hour’s amusement, I wou’d return to these speculations, they appear so cold, and strain’d, and ridiculous, that I cannot find it in my heart to enter into them any farther. (T, 1.4.7.9/269) In his abstract for the Treatise (in which he writes of himself in the third person), Hume sums up this view: Our author insists upon several other sceptical topics; and on the whole concludes, that we assent to our faculties, and employ our reason only because we cannot help it. Philosophy would render us entirely Pyrrhonian were not nature too strong for it. (T, Abs. 27/657)

Richard Joyce

51

Of anyone who would profess indifference to moral distinctions, Hume counsels that the best response is ‘to leave him to himself’, trusting that ‘it is probable he will, at last, of himself, from mere weariness, come over to the side of common sense and reason’ (EPM, 1.2/170). In other words, even if one came to espouse a moral error theory, ‘nature herself’ would eventually drive that philosophical allegiance from one’s mind. And Hume evidently thinks that this would be no bad thing. Of those ‘honest gentlemen of England’ – who ‘being always employ’d in domestic affairs, or amusing themselves in common recreations, have carried their thoughts very little beyond those objects’ – Hume tells us that ‘they do well to keep themselves in their present situation’ (T, 1.4.7.14/272). Yet even if these down-to-earth folk are acknowledged to be well off, Hume is not attempting to dissuade anyone from engaging in philosophical speculations: he simply thinks that one either will or will not, according to temperament and mood, and that to the extent that one will it is likely that one’s efforts will be temporary (that is, until someone calls out ‘Anyone for backgammon?’). Yet none of this is at odds with the possible truth of an error theory. Regarding causal relations and the continued existence of external objects, Hume is explicit that experience ‘leads us into errors’ (T, 1.4.7.6/267) (for he thinks that the two beliefs are jointly affirmed contraries), but even here – where we seem to have Hume clearly endorsing some kind of error theory – he thinks it remains an open question the extent to which he should ‘torture my brain with subtilities’ (T, 1.4.7.10/270), the extent to which he should ‘yield to these illusions’ (T, 1.4.7.6/267). Nature may ineluctably reassert herself against the awareness that one has fallen into error, but the errors are no less errors for that. Might things stand similarly for morality, in Hume’s eyes? He certainly never claims outright that morality doesn’t exist. He is more likely to say something along the lines of ‘[m]oral properties exist not in bodies but merely in the senses’. It is worth noting that such claims remain consistent with a moral error theory. Telling someone ‘The pink elephants exist only in your mind’ is in fact a way of saying that the pink elephants do not exist at all. At one point in the Treatise Hume declares that sounds and smells ‘really exist no where’ (T, 1.3.14.25/167), and it is reasonable to think that he will say the same of colour and causation (the section in question concerns ‘necessary connexion’). This bald claim of non-existence comes immediately after he has spoken of the mind’s ‘great propensity to spread itself on external objects’ – a thesis that I will here refer to as ‘projectivism’. The other well-known projectivist passage from Hume is in EPM, where he claims that ‘taste’ (as opposed to reason) ‘has a productive faculty, and gilding and staining all natural objects with the colours, borrowed from internal sentiment, raises in a manner a new creation’ (EPM, App. 1.21/294). Here he mentions ‘beauty and deformity, vice and virtue’ as the products of sentimental projection. The correlation of these two projectivist passages

52 Expressivism, Motivation Internalism, and Hume

suggests that what goes for one (that is, that sounds and smells ‘really exist no where’) should go for the other (that is, that virtue and vice ‘really exist no where’). That moral qualities should receive the same treatment as colour, sound, smell, heat, and cold is reaffirmed elsewhere in the Treatise (3.1.1.21/469) and in ‘The Sceptic’. Hume is, moreover, explicit that the folk do indeed think of colours, smells, sounds, and heat as objective qualities of objects. In a letter to Hugh Blair of 1762 he is dismissive of the idea that the folk might not be objectivists; evidently, ‘the Vulgar’ (as Hume refers to them) are taken in by their own projectivist tendencies. ‘Philosophy scarce ever advances a greater Paradox in the Eyes of the People, than when it affirms that Snow is neither cold nor white: Fire hot nor red’ ([1762] 1986, p. 416). On the assumption that what is said here will carry over for moral qualities as well, then the folk are generally fooled by their moral projectivist tendencies: they are unaware that their moral judgements are the product of sentiments being projected onto the world; they both experience the world as morally ‘coloured’ and believe it to be. But if someone believes something to be the case, then it is natural to assume that her utterances on the matter will be assertions. Thus, I claim, Hume’s moral projectivism is a form of psychological emotivism that leads naturally to (but I would not go so far as to say implies) moral cognitivism – though a cognitivism that remains compatible with an expressivist component.25 But are the assertions in question true? If snow is not white – something that Hume seemingly endorses – and someone asserts the sentence ‘Snow is white’, then the very natural conclusion to draw is that she has simply asserted something false. Similarly, if Fred’s character does not have the quality of virtuousness, and someone asserts the sentence ‘Fred is virtuous’, then the natural conclusion is that she has asserted something false. One way to avoid this error theoretic interpretation of Hume’s metaethical commitments is to problematise the analogy that allows us to draw conclusions about his implicit moral views from what he somewhat more explicitly claims about sensory modalities. I have nothing to say here on that score, except to reaffirm that Hume draws the analogy sufficiently frequently that it is safe to assume that he thinks there are illuminating similarities among these topics. Another way is to leave the analogy intact but to deny the error theoretic construal of both analogues. Perhaps when Hume denies that snow is white what he means is something like ‘Snow – considered in itself, restricting ourselves to the consideration of only its intrinsic qualities – is not white’. But perhaps he also thinks that it is not obligatory to understand the truth conditions of ‘Snow is white’ in this manner. If we allow the possibility of whiteness being some relational, subject-implicating property, then perhaps Hume will consider the sentence true, after all. The problem is that he has adamantly affirmed that the general folk think of the whiteness of snow in the former objectivist fashion. The crux of the issue, then,

Richard Joyce

53

is whether the weight of dominant folk belief on the matter is sufficient to determine how the word ‘white’ must be understood (or, if you prefer, how the identity conditions of the concept whiteness must be construed). Understood one way it leads to an error theory, understood in another way it leads to a success theory. Hume had no resources for addressing this question, and nor have we. This leads to an impasse in establishing whether Hume’s occasional expressivist tendencies are mixed with a cognitivism that is committed to success or a cognitivism that is committed to scepticism. The contemporary debate between the moral error theorist and the moral success theorist seems locked up at exactly the same point: a problematic (or ‘queer’) quality of morality is brought forth, regarding which some will argue that this quality is an essential aspect of the moral conceptual framework (such that any normative system stripped of this problematic element would no longer deserve the name ‘morality’), whereas others will argue that extirpation of the flawed element would amount merely to a benign revision and demystification of morality. Lacking an accepted methodology for deciding such disputes, the modern metaethical debate is at a disappointing stalemate.26 It is also possible that on occasions there is no fact of the matter about whether a given discourse (for example, morality) is committed to some putatively queer property, leading to the conclusion that the dispute between the moral error theorist and her many detractors may in fact be fundamentally undecidable.27 If this is correct, then it is tempting to suppose that any interpretation of Hume that leaves him sitting on the fence over this matter is a charitable one.28

Notes 1. See, for example, Carnap (1935); Stevenson (1937). 2. ‘A’ denotes a future act that proposition p predicates of S. 3. See Joyce (2008) for further discussion of this disparity between psychologists’ and metaethicists’ use of ‘emotivism’. 4. Páll Árdal (1966) once distinguished ‘emotionism’ from ‘emotivism’, in a way that maps closely to the contrast between what I am calling ‘psychological emotivism’ and ‘metaethical emotivism’ (though perhaps it is even closer to the distinction that I will make shortly, between the mentalistic construal of expressivism and the metaethical construal of expressivism). I quite like Árdal’s terminology, but it never took off. Prinz (2007) has recently reintroduced the word ‘emotionism’ for a somewhat different thesis. 5. Charles Stevenson (1937; 1963) held a view of this sort, though he maintained that in addition the moral judgement includes a command. 6. John Bricke writes of the subjectivist-sounding Treatise passage (T, 3.1.1.23/469): ‘The autobiographical rendering of evaluative sentences being so utterly implausible, it is fortunate that there is no reason whatever to think that Hume here means by “meaning” what, when concerned with language, we now mean’ (1996, p. 162). (See also Ayer, 1980, p. 84.) Nicholas Sturgeon (2008, p. 514) interprets Hume as a subjectivist, but not the speaker-oriented kind mentioned here. After quoting the subjectivist-sounding passage, Sturgeon notes that Hume

54 Expressivism, Motivation Internalism, and Hume

7.

8. 9.

10.

11.

12.

13.

14.

subsequently ‘modifies this view to make the truth of one’s ascription of virtue or vice depend, not on one’s actual feelings, but on the feelings one would have under the right conditions, whether or not one now is (or even could be) in those conditions’. I am not sure whether Sturgeon’s Hume ultimately counts as a psychological emotivist. This is my own fault, since I have characterised psychological emotivism only as carefully as is necessary to reveal its logical independence from metaethical emotivism – but have left it intentionally indeterminate in several respects. Rachel Cohen concurs that Hume’s Motivation Argument ‘is irrelevant to noncognitivism’ (1996, p. 261); however, her positive interpretation differs from mine: she maintains that the argument concerns the nature of moral properties. For the sake of brevity I have stripped Ridge’s thesis of a few details that don’t matter on this occasion. For the full account see Ridge 2006, p. 501. Ayer thought that all meaningful statements must be either analytic or empirically verifiable. Given that moral utterances appear to be neither, Ayer was forced to claim that they are not meaningful statements. But rather than concluding that moral judgements are meaningless, Ayer’s preferred conclusion is that they are not statements, but are, rather, ways of evincing ones emotions and issuing commands. It is possible that one might construe MI and mentalistic expressivism such that they have a different modal and/or epistemological status. MI, recall, is presented as a necessary and a priori thesis; perhaps mentalistic expressivism need not be. It is difficult, however, to see how this would create the possibility of a viable argumentative strategy from one to the other. From ‘It is actually the case that X’ we cannot conclude ‘It is necessarily the case that X’. The reverse implication does hold, of course, but then the question is on what grounds we could establish the necessity claim as the antecedent. If we had any such grounds, then the consequent would hardly be in doubt. Frank Snare interprets the argument from motivation as aiming to establish mentalistic expressivism (he calls it ‘emotivism’). He concludes that the influence of the argument is due to philosophers having ‘been so completely convinced of the conclusion that they did not realise that the conclusion itself provides much of the reason for believing the premises’ (Snare, 1975, p. 9). I am not sure whether Moore mentioned the paradox in print on any earlier occasion. Wittgenstein reports Moore having mentioned it in a lecture, probably from before World War 1. (Norman Malcolm mentions that Wittgenstein opined that this was the only work of Moore’s that had ever ‘greatly impressed him’ (Malcolm, 1958, p. 66).) Compare the phenomenon of sarcasm: If Fred looks out at the pelting English rain and sighs ‘another glorious day in paradise’ in tones of sarcasm clear for all to hear, then he has not asserted that it is a glorious day in paradise. But sarcasm is such an entrenched convention that his audience will not be confused – they will know that he did not assert it, and did not intend to assert it, and Fred will know that they know. Thus sarcasm is not a case of insincerity ‘worn on the sleeve’, since it is not a case of insincerity at all. A lot depends on tone of voice here. It does with (1)–(3) as well, since uttering any of these in tones of sarcasm or in a clearly joking way completely alters the speech acts involved. Let me stipulate that we are to imagine these uttered in serious tones. (Insert the word ‘stinking’ before ‘kike’ if it helps to reinforce this point.) I apologise to anyone who finds the term offensive even in the context of mentioning the word as an example of an extremely offensive word. On another

Richard Joyce

15.

16.

17.

18.

19.

55

occasion of making this argument, I employed the less provocative term ‘kraut’ – but the very choice of a word whose offensiveness is less deeply ingrained undermined the cogency of the argument. Such a mixed view has been maintained by C. L. Stevenson, R. M. Hare, and P. H. Nowell-Smith, among others. If one defines noncognitivism simply as the denial of cognitivism, then the two theories are, of course, contradictories. Similarly, if one defines cognitivism as the theory that moral judgements express only beliefs, then any ‘mixed theory’ will be excluded. However, if we think of expressivism as a positive proposal (about what moral judgements do express), and we drop the ‘only’ clause in both expressivism and cognitivism, then moral judgements may be two things: They may be assertions and ways of expressing conative attitudes. Cf. John MacFarlane (2005, p. 334), who writes: ‘Imagine someone saying: “I concede that what I asserted wasn’t true, but I stand by what I said anyway.” We would have a very difficult time taking such a person seriously as an asserter. If she continued to manifest this kind of indifference to established truth, we would stop regarding the noises coming out of her mouth as assertions.’ Note how ‘amoralism’ is a term of art here. In the vernacular, ‘amoralist’ often denotes someone who rejects morality altogether, who doesn’t believe in it at all. In recent philosophical debates, by contrast, it denotes someone who makes genuine moral judgements but lacks any motivation to comply (and the topic of the debate is whether the amoralist is even a possibility). I am using the term in a third way: to denote someone who tries to cancel the motivation-implicating aspect of a moral judgement. The debate here is not whether such agents exist, but whether they succeed in making moral judgements. This sentence was reportedly uttered by the Australian cricket captain during the ‘bodyline series’ of 1932–3. The English captain came to the Australian dressing room to complain about one of his players having been called a ‘bastard’ during play. Bill Woodfull, the Aussie captain, turned to his team and uttered the memorable line. (I owe this example to the late David Lewis (correspondence 2000).) There is a well-known interpretation of the amoralist from R. M. Hare (1952, pp. 124–6, 167 ff.), according to which the amoralist’s statement is not literally a moral judgement at all, but rather is best read as having quotation marks round the term ‘morally ought’. My view is not unlike this, though it is important to bear in mind that Hare’s amoralist utters something like ‘For me to steal would be wrong’ while having no motivation to refrain from stealing, while my amoralist says, ‘For me to steal would be wrong, though I subscribe to no normative framework that condemns stealing.’ Given the careful verbal qualification that the latter offers, it seems to me quite plausible that something rather like quotation marks are being imposed. Copp objects that Hare’s view fails to accommodate the possibility of moralists and amoralists entering into moral debate (Copp, 2001, p. 13). If the amoralist says ‘Liberalism is a great evil’ and the moralist responds ‘No, liberalism is morally defensible’, but in fact the former statement is equivalent to something like ‘It is considered hereabouts that liberalism is a great evil’, then there is no real disagreement. But I don’t find this the reductio that Copp seems to think it to be. Perhaps any intuition we have that there can be genuine moral debate with amoralists just stems from the fact that the way they speak (making the quotation marks tacit) is apt to encourage us to forget that they are not really making moral judgements at all. If I were to hear someone claim that liberalism is a great evil then I would want to protest; but if I were then to discover that this person had earlier asserted ‘Evil, be thou my good’, then I should become quite

56 Expressivism, Motivation Internalism, and Hume

20.

21.

22.

23.

24.

25.

26. 27.

28.

confused as to what she thought about liberalism, and thus not at all confident that I should disagree. For Grice, the implicatum p is detachable from an utterance iff there are ways of saying the same thing that do not implicate p (see Grice 1989, p. 39). What Copp has in mind in saying that colourings are detachable is that instead of ‘Aaron is a kike’ one could say ‘Aaron is Jewish’, and the latter, though saying the same thing (having the same sense and reference?), lacks the implicatum that the speaker has contempt. Copp himself admits that ‘nothing [in my argument] turns on whether coloring is an example of conventional implicature or simply a phenomenon that is similar to conventional implicature’ (2001, p. 23). I actually harbour some reservations that Copp’s view quite deserves the label ‘realism’, but the fact that it is a version of moral cognitivist ‘success theory’ is enough to underwrite the distinction I am highlighting. For my views on how to characterise moral (anti)realism, see Joyce, 2007a. Indeed, Mackie’s general definition of ‘good’ is not a million miles away from the cognitivist element of morality articulated by Copp. Mackie defines ‘good’ as ‘such as to satisfy requirements (etc.) of the kind in question’ (1977, pp. 55–6). With many non-moral uses of ‘good’ Mackie thinks the predicate is satisfied. But in moral contexts, he thinks, the pertinent requirements are those that are ‘simply there, in the nature of things, without being the requirements of any person or body of persons, even God’ (p. 59). It is Mackie’s conviction that there are no such ‘intrinsic requirements’ that lead to his moral scepticism. Mackie (1980) argues along similar lines to me for the error theoretic interpretation of Hume. David Gauthier (1992) also toys with this interpretation of Hume’s account of the artificial virtues, though doesn’t firmly endorse it. In saying this I am flying in the face of recent tradition, which tends to lump projectivism together with noncognitivism. I have argued elsewhere that this is at best optional and at worst a mistake. See Joyce, 2006 (Chapter 4) and Joyce (2009). In the latter, I delineate different species of projectivism. I discuss this impasse in Joyce, 2006 (Chapter 6), 2007a, and 2007b. David Lewis writes: ‘Strictly speaking, Mackie is right: genuine values would have to meet an impossible condition, so it is an error to think there are any. Loosely speaking, the name may go to a claimant that deserves it imperfectly. … What to make of the situation is mainly a matter of temperament’ (1989, pp. 136–7). Early portions of this chapter closely follow Joyce, 2002. A youthful version of this essay once went by the name of ‘Noncognitivism, Motivation, and Assertion’, and it was helped along by feedback from David Lewis and Simon Kirchin. I thank David Copp for the very useful discussion.

2 Is Hume Inconsistent? – Motivation and Morals Norva Y. S. Lo

2.1 Hume’s Motivation Argument Morality, Hume argues in his Treatise of Human Nature, cannot be ‘deriv’d’ from reason alone.1 Among his many arguments2 for this anti-rationalist claim is what might be called the ‘Motivation Argument’.3 He writes: Those who affirm that virtue is nothing but a conformity to reason; that there are eternal fitnesses and unfitnesses of things, which are the same to every rational being that considers them; that the immutable measures of right and wrong impose an obligation, not only on human creatures, but also on the Deity himself: All these systems concur in the opinion, that morality, like truth, is discern’d merely by ideas, and by their juxta-position and comparison. In order, therefore, to judge of these systems, we need only consider, whether it be possible, from reason alone, to distinguish betwixt moral good and evil, or whether there must concur some other principles to enable us to make that distinction […] [M]orality is […] supposed to influence our passions and actions […] And this is confirm’d by common experience, which informs us, that men are often govern’d by their duties, and are deter’d from some actions by the opinion of injustice, and impell’d to others by that of obligation. Since morals, therefore, have an influence on the actions and affections, it follows, that they cannot be deriv’d from reason; and that because reason alone […] can never have any such influence. Morals excite passions, and produce or prevent actions. Reason of itself is utterly impotent in this particular. The rules of morality, therefore, are not conclusions of our reason. […] An active principle can never be founded on an inactive; and if reason be inactive in itself, it must remain so in all its shapes and appearances […] (T, 3.1.1.4–7/456–7. Cf. EPM, 1.7–9/172) 57

58 Is Hume Inconsistent? – Motivation and Morals

Accordingly, Hume’s Motivation Argument can be formalized as follows: H1 Morality excites passions and motivates actions. H2 Reason alone cannot excite passions or motivate actions. H3 That which influences passions and actions cannot be derived merely from something which by itself cannot have such an influence. H4 Morality cannot be derived from reason alone. [from H1, H2 and H3] Hume’s argument is clearly valid. It is unclear, however, how we should read the argument in contemporary metaethical terms, which express finer distinctions apparently not discussed or made by Hume himself. In the passages where Hume articulates his Motivation Argument, he uses a variety of terms interchangeably with the term ‘morality’ – including ‘opinion of injustice, and […] that of obligation’, ‘rules of morality’, ‘moral good and evil’,4 ‘merit and demerit’5 and ‘moral distinctions’.6 Hence, the major premise and the conclusion of Hume’s Motivation Argument, both containing the ambiguous term ‘morality’, are open to multiple interpretations. When Hume says that ‘morality is […] supposed to influence our passions and actions’, he can be interpreted as supposing (a) that judgements of moral good and evil have such an influence on us,7 and/or (b) that knowledge of moral good and evil has such an influence8 and/or (c) that the properties of moral good and evil themselves have such an influence9 – regardless of whether or not we make any judgement, or have any knowledge, about them. Given the ambiguity in Hume’s terminology, all three readings seem plausible. The three different readings of Hume’s motivation argument are formulated below. The First Internalist Reading H1a

Judgments of moral good and evil excite passions and motivate actions – that is, if we judge that ␾-ing is virtuous/vicious, then (automatically) we feel a favourable/unfavourable sentiment towards ␾-ing, and are motivated (though not necessarily overridingly) to pursue/avoid ␾-ing. H2a Reasoning is a process which cannot by itself excite passions or motivate actions. H3a That which excites passions and motivates actions cannot be derived merely by processes which cannot by themselves excite passions or motivate actions. H4a Judgments of moral good and evil cannot be derived merely by reasoning. [from H1a, H2a and H3a] The Second Internalist Reading H1b

Knowledge of moral good and evil excites passions and motivates actions – that is, if we know that ␾-ing is virtuous/vicious, then we

Norva Y. S. Lo

59

feel a favourable/unfavourable sentiment towards ␾-ing, and are motivated to pursue/avoid ␾-ing. H2b Reasoning is a process which cannot by itself excite passions or motivate actions. [same as H2a] H3b That which excites passions and motivates actions cannot be derived merely by processes which cannot by themselves excite passions or motivate actions. [same as H3a] H4b Knowledge of moral good and evil cannot be derived merely by reasoning. [from H1b, H2b and H3b] The Third Internalist Reading H1c Properties of moral good and evil excite passions and motivate actions – that is, if ␾-ing has the property of being virtuous/vicious, then we feel a favourable/unfavourable sentiment towards ␾-ing, and are motivated to pursue/avoid ␾-ing. H2c Relations that are distinguishable10 simply by reasoning (for example, mathematical relations) cannot by themselves excite passions or motivate actions. H3c That which excites passions and motivates actions is not reducible to things that cannot by themselves excite passions or motivate actions. H4c Properties of moral good and evil are not reducible to relations that are distinguishable simply by reasoning. [from H1c, H2c and H3c]11 These readings of Hume’s Motivation Argument are not uncommon. Different versions of H1 have been taken to express different versions of internalism. The core idea of internalism is that morality is practical in that it has some necessary and internal connection to motivation. H1a expresses the version of internalism according to which there is an internal connection between one’s making a moral judgement and one’s being motivated to act accordingly; whereas according to the version of internalism represented by H1b, not all judgements about morals motivate, only the sound ones. H1c, however, expresses yet another form of internalism, the idea of which is that we are somehow attracted by the good and repelled by the evil. And so it is moral properties themselves that motivate us to act accordingly. Given these internalist interpretations of the major premise of his Motivation Argument, Hume is now very often considered as a precursor of internalism.12 Hume’s Motivation Argument, as we have seen, is devised to disprove the general rationalist thesis that morality is derived from reason. The above three readings of Hume’s anti-rationalist argument, however, show that the general rationalist thesis can be divided into two more specific claims – one about the epistemology of morals, another about the ‘essence’13

60 Is Hume Inconsistent? – Motivation and Morals

of morals. The first two versions of Hume’s Motivation Argument both reject the rationalist epistemological claim that rational processes alone suffice to generate moral beliefs and knowledge, whereas the third version of the argument rejects the rationalist’s analysis of good as essentially reducible to the ‘conformity to reason’ or ‘truth’ on the one hand, and evil as reducible to the ‘contrariety to it’ or ‘falsehood’ on the other, which are relations14 distinguishable by mere reasoning (See T, 3.1.1.4/456, 9–15/458–61). If Hume is right in rejecting the rationalist account of morality, then what are Hume’s own answers to these two questions: First, are moral properties reducible to properties or relations of some other kind? Second, by what means, other than reasoning, do we arrive at judgements, and acquire knowledge, about moral good and evil? I shall answer these two questions in Sections 2.2 and 2.3, respectively, by putting together Hume’s own metaethical analysis of morals. In Section 2.4, I shall examine the above three internalist readings of Hume’s Motivation Argument in the light of his own analysis of morals, and argue that all of them are, in one way or another, inconsistent with the analysis. In Section 2.5, I shall consider a possible defence of the internalist readings of Hume’s Motivation Argument, which appeals to his view on cognitive but non-rational processes, such as observation and intuition. But I shall argue that the internalist readings of Hume are indefensible, and that Hume is not an internalist. Finally, I offer three non-internalist readings of Hume’s argument, which are consistent with his analysis of morals and are also effective devices to disprove the kind of moral rationalism that he rejects.

2.2 Hume’s analysis of morals Consider the following two passages where Hume analyses the moral properties of vice and virtue, the first from the Treatise of Human Nature, the second from the Enquiry Concerning the Principles of Morals. [W]hen you pronounce any action or character to be vicious, you mean nothing, but that from the constitution of your nature you have a feeling or sentiment of blame from the contemplation of it. (T, 3.1.1.26/468–9, emphasis added)15 The hypothesis which we embrace is plain. It maintains, that morality is determined by sentiment. It defines virtue to be whatever mental action or quality gives to a spectator the pleasing sentiment of approbation; and vice the contrary. (EPM, approx. 1.10/173–5, first emphasis added, the rest original) Accordingly, Hume’s conceptual analysis, or definition, of ‘virtue’ and ‘vice’ can be reformulated as follows: an action or character is virtuous/vicious if and only if a ‘spectator’ feels the sentiment of approbation/disapprobation16

Norva Y. S. Lo

61

towards it. It is very important, however, to notice that what Hume calls a ‘spectator’ or ‘judicious spectator’ (T, 3.3.1.14/580–1) is not just anyone. Rather, Hume’s spectator is a person who must meet a set of conditions. The first condition that Hume’s spectator must satisfy is: C1 The condition of having full knowledge and awareness of all the circumstances and relations relevant to the case under evaluation. As Hume puts it: All the circumstances of the case are supposed to be laid before us, ere we can fix any sentence of blame or approbation. If any material circumstances be yet unknown or doubtful, we must first employ our enquiry or intellectual faculties to assure us of it; and must suspend for a time all moral decision or sentiment. […] In moral decisions, all the circumstances and relations must be previously known; and the mind, from the contemplation of the whole, feels some new impression of affection or disgust, esteem or contempt, approbation or blame. (EPM, approx. 1.11/175, emphasis added) This doctrine will become still more evident, if we compare moral beauty with natural, to which, in many particulars, it bears so near a resemblance. […] in all decisions of taste or external beauty, all the relations are before-hand obvious to the eye; and we thence proceed to feel a sentiment of complacency or disgust, according to the nature of the object, and disposition of our organs. (EPM, approx. 1.13, emphases added. Cf. T, 3.1.2.3/471.) Furthermore, in making moral decisions, Hume’s ‘judicious spectator’ must adopt what he calls the ‘steady and general points of view’. He writes: [E]very particular man has a peculiar position with regard to others; and ’tis impossible we cou’d ever converse together on any reasonable terms, were each of us to consider characters and persons, only as they appear from his peculiar point of view. In order, therefore, to prevent those continual contradictions, and arrive at a more stable judgment of things, we fix on some steady and general points of view; and always, in our thoughts, place ourselves in them, whatever may be our present situation. (T, 3.3.1.15/581–2, emphases original) In general, all sentiments of blame or praise are variable, according to our situation of nearness or remoteness, with regard to the person blam’d or prais’d, and according to the present disposition of our mind. But these variations we regard not in our general decisions, but still apply the terms

62 Is Hume Inconsistent? – Motivation and Morals

expressive of our liking and dislike, in the same manner, as if we remain’d in one point of view. Experience soon teaches us this method of correcting our sentiments, or at least, correcting our language, where the sentiments are more stubborn and inalterable. […] Such corrections are common with regard to all the senses. (T, 3.3.1.16/582 emphasis added. Cf. T, 3.3.1.21.) We blame equally a bad action, which we read of in history, with one perform’d in our neighbourhood the other day: The meaning of which is, that we know from reflection, that the former action wou’d excite as strong sentiments of disapprobation as the latter, were it plac’d in the same position. (T, 3.3.1.18/583–4, emphases added) A case of ‘correcting our sentiments, or at least, correcting our language’, which Hume considers to be ‘common with regard to all the senses’ is that of making colour judgements under less than ideal conditions. A red object, for instance, may appear to have different colours under different circumstances – orangey at sunset, brownish on a dark night, and purplish at dawn. But an experienced spectator judges it to be red because she knows that if the same object were to be placed under normal daylight (the ideal condition for perceiving colours), then it would look red (Cf. T, 3.3.3.2/602–3 and Hume, 1760, p. 140). Similarly, in the case of making moral judgements, we might be in less than ideal positions so that our moral sentiments might not reliably indicate the presence of moral properties. For instance, as Hume points out, our sentiments are naturally stronger when the interests of ourselves or those who are close to us are at stake. So we may have a stronger feeling of disapproval towards the same evil behaviour or character when it causes harm to ourselves and our friends than when it causes harm to people who are complete strangers to us. But, according to Hume, in making sound moral judgements, we judge not according to our actual moral sentiments. Rather, we judge according to the moral sentiments that we belive we would have if we were to adopt the ‘steady and general points of view’ (the ideal perspective for perceiving moral properties). But what exactly is it for someone to adopt the so-called steady and general points of view? According to Hume, these points of view are ‘general’ and ‘steady’ in that people adopting such perspectives would overlook their own particular situations and interests, but focus instead on what is universal and common to all human beings, so that people adopting such perspectives could expect each other to converge in their moral sentiments and judgements. He writes: ’Tis […] from the influence of characters and qualities, upon those who have an intercourse with any person, that we blame or praise him. We consider not the persons, affected by the qualities, be they our

Norva Y. S. Lo

63

acquaintance or strangers, countrymen or foreigners. Nay, we over-look our own interest in those general judgments; and blame not a man for opposing us in any of our pretensions, when his own interest is particularly concern’d. We make allowance for a certain degree of selfishness in men; because we know it to be inseparable from human nature, and inherent in our frame and constitution. By this reflection we correct those sentiments of blame, which so naturally arise upon any opposition. (T, 3.3.1.17/582–3) ’Tis only when a character is consider’d in general, without reference to our particular interest, that it causes such a feeling or sentiment, as denominates it morally good or evil. (T, 3.1.2.4) In short, to adopt the ‘steady and general points of view’ is to meet all of the following conditions: C2 The condition of taking into account of facts about human nature.17 C3 The condition of overlooking one’s personal interests. C4 The condition of overlooking one’s personal relations to the parties involved in the case under evaluation. Given the above discussion on the conditions that Hume’s ‘judicious spectator’ must satisfy, his analysis of virtue and vice can be expounded as follows: H

An action18 or character is virtuous/vicious if and only if all human beings are disposed, under the conditions C1, C2, C3 and C4, to feel the sentiment of approbation/disapprobation towards it.19

Now, according to Hume, it is via sympathy that we form the sentiments of approbation and disapprobation towards social virtues and vices (for example, justice and injustice), respectively. Sympathy, in Hume’s view, is a natural psychological mechanism of the human mind, by which the mind’s thoughts of the sentiments of others (for example, my idea of another person’s pleasure) can be converted into its own sentiments of the same kind (for example, my own feeling of pleasure).20 For Hume, the psychological mechanism of sympathy causally explains our experience of moral sentiments, just as for a medical scientist, certain physiological mechanisms causally explain our experience of various bodily sensations (for example, the sensations of colour, smell, sound and heat). The reason not to include the proper functioning of sympathy as one of the ideal conditions for having reliable moral sentiments is as follows: (a) the operation of sympathy, in Hume’s view, is based on two more basic psychological principles – namely,

64 Is Hume Inconsistent? – Motivation and Morals

the association of ideas and impressions21 and the associated ideas and impressions sharing a similar degree of vivacity,22 and (b) the proper functioning of these two psychological principles (and thus the proper functioning of sympathy) is, for Hume, part of the normal conditions of a human being, and (c) in general, dispositional analyses23 (of which H is one) implicitly assume that the spectator satisfies the conditions that are normal for its kind. For instance, a dispositional analysis of colour often implicitly assumes the normal functioning of those organs or bodily parts of the observer that are causally responsible for her optical experiences. Likewise, a Humean dispositional analysis of moral qualities implicitly assumes the normal functioning of sympathy in the ‘judicious spectator’. It should be noted that H is a humanist as well as a universalist metaethical analysis of morality. It is humanist in that it understands value and disvalue as essentially anchored on the psychological dispositions of human beings, not that of God or any other superhuman beings. And it is universalist in that it implies that something is valuable/disvaluable only if all human beings are disposed under ideal conditions to approve/disapprove of it. On his universal humanism, Hume writes: The notion of morals implies some sentiment common to all mankind, which recommends the same object to general approbation, and makes every man, or most men, agree in the same opinion or decision concerning it. (EPM, 9.5/272–2, emphasis added) When a man […] bestows on any man the epithets of vicious or odious or depraved, he then […] expresses sentiments, in which he expects all his audience are to concur with him. (EPM, 9.6/272–3, last emphasis added, the rest original) [T]he sentiments, which arise from humanity, are not only the same in all human creatures, and produce the same approbation or censure; but they also comprehend all human creatures; nor is there any one whose conduct or character is not, by their means, an object, to every one, of censure or approbation. (EPM, 9.7/273)24 It is perhaps his optimism in the uniformity of human nature (which is, after all, an empirical matter) that has led Hume to consider universality as a conceptual component of moral properties. Those Humeans who lack such an optimism (which is common for enlightenment thinkers) might want to hold a neutralized variation of Hume’s analysis of morals, which says: Something is a relative/universal value (or disvalue) if and only if some/all human beings are disposed, under C1, C2, C3 and C4, to feel the

Norva Y. S. Lo

65

sentiment of approbation (or disapprobation) towards it. As we have seen, H is concerned only with identifying moral properties – for example, whether an action or character is or is not obligatory or virtuous. But even if we could identify all the values and obligations there are, we would still have the problem of how to balance them if they happened to be in conflict with each other. So, in the case of comparing25 moral properties, Hume holds the following parallel analysis: H*

An action or character is more/less virtuous (or vicious) than another if and only if all human beings are disposed, under the conditions C1, C2, C3, C4, to feel a stronger/weaker sentiment of approbation (or disapprobation) towards one than towards another.

According to Hume, C1, C2, C3 and C4 are conditions under which our sentiments of approbation and disapprobation will reliably mark virtue and vice, respectively. The reference to these favourable conditions for reliable moral sentiments is crucial to Hume’s analysis of morals. For that is what gives his otherwise subjective metaethics a certain degree of objectivism. Let me explain. Given H, a moral judgement is a judgement about the nature of the object under moral evaluation, as much as it is a judgement about the moral psychological dispositions of our mind – or, more precisely, it is, in Hume’s words, about the ‘relation between the object and the organs or faculties of the mind’.26 An action or character is virtuous/vicious just in case it is naturally fitted to excite a favourable/unfavourable sentiment in us according to the dispositions and structures of our mind. Hence, the truth or falsity of a moral judgement is objective in the sense that it is not an arbitrary matter, but determined by the characteristic of the object under evaluation as well as by the facts of human evaluative psychologies. Furthermore, according to H, to say that something is virtuous is equivalent to saying that if we are/were in the hypothetical conditions C1, C2, C3 and C4, then we will/would feel the sentiment of approbation towards it. This allows the possibility that an action is virtuous without us actually feeling such a sentiment towards it. For it might be that we have not actually satisfied all those conditions favourable for having reliable moral sentiments. The lack of a positive sentiment from us towards the action does not imply that it is not virtuous. In other words, H allows the possibility that moral properties, such as virtue and vice, good and evil, can exist (that is, be instantiated) in the absence of valuers’ matching evaluative responses. Or, more generally speaking, things can be (dis)valuable without being (dis)valued. In this sense, moral properties are said to be ontologically objective. But in another sense, moral properties are subjective under Hume’s account. As we have seen, H is a conceptual analysis of virtue and vice in terms of the psychological dispositions of human subjects. If H is true, then the concepts ‘virtue’ and ‘vice’ cannot be correctly analysed without employing the concept ‘subject’.

66 Is Hume Inconsistent? – Motivation and Morals

In other words, there is a necessary conceptual connection between moral properties and human subjects. In this sense, then, moral properties are said to be conceptually subjective. In short, given Hume’s analysis, facts about morals are nothing but empirical facts about the psychological dispositions of human beings to feel moral sentiments under certain hypothetical conditions. Moral properties are objective as well as subjective, but in ways compatible with, and indeed complementary to, each other. This is an advantage of Hume’s theory, which can accommodate and account for the insights from both the objectivist and the subjectivist camps.

2.3 Hume’s epistemology of morals Hume’s metaethics, as we have just seen, reduces facts of morals to the empirical facts of human psychology. This is an important part of Hume’s wide-ranging naturalism, which presupposes nothing other than the things studied and investigated by the empirical sciences (broadly construed as to include psychology and various other social sciences). Hence, given H, the Humean canonical method of acquiring knowledge about morals is not mysterious, but empirical and potentially scientific.27 For example, in order to decide whether someone’s killing of another person is a vicious act, we need to go through the following procedures: (i) we need to get ourselves into condition C1, that is, we need to gather all the information about the circumstances under which the action is performed and the relations between the actor and the other people involved, such as what the motive behind the action is, whether it was an accident, or whether it is a case of self-defence, what personal histories and relations the killer and the victim have, and so on. And (ii) we need to get ourselves into condition C2, that is, we take into account the facts of human nature. For instance, in cases where the killer has a history of suffering from long-term physical and psychological abuses, we pay attention to the fact that it is part of the human condition that people sometimes lose sensitivity and sanity in a traumatic environment. Furthermore, (iii) we need to get ourselves into condition C3, that is, we try to overlook our self-interests. For instance, in cases where the victim is our benefactor so that we suffer a material loss due to her death, we pretend as hard as possible that our interests are not at stake. And (iv) we need to get ourselves into condition C4, that is, we try to overlook our personal relations. For instance, in cases where the killer is a friend of ours, we pretend as hard as possible that the killer was a stranger instead. Now, suppose we have got ourselves into all these conditions, C1, C2, C3 and C4, which are favourable for having reliable moral sentiments. Then, while we are still under these condition, (v) we proceed to feel what sentiments we have towards the action under evaluation. The better we get ourselves into C1, C2, C3 and C4, the better indicators of moral properties our

Norva Y. S. Lo

67

subsequent moral sentiments will be. If we could get ourselves completely into those conditions, then, according to H, our subsequent moral sentiments would be infallible indicators: our positive sentiments of approbation, delight and esteem would indicate virtue; whereas our negative sentiments would indicate vice. It is unlikely that we will completely satisfy those conditions ideal for having reliable sentiments. So we approximate the test and get less than ideal but approximate results. We also try to improve the test by trying harder to meet those ideal conditions. This is how we make progress in our knowledge about morals. Hume’s metaethical account of morals, like many other dispositional theories of values nowadays, ‘makes a place for truth, and in principle for certain knowledge, and in practice for less-than-certain knowledge about value. But also it makes a place for ignorance and error, for hesitant opinion and modesty, for trying to learn more and hoping to succeed’ (Lewis, 1989, p. 123). Now, let us consider again the five processes that we need to go through in order to identify or compare the moral properties of actions or characters. Processes (i) and (ii) are both cognitive, in that they aim at delivering a complete true description of the object under moral evaluation, namely: a certain action performed by (or a certain character of) a being of a certain nature. Processes (iii) and (iv), however, are partly imaginative, in that they require us to actively entertain thoughts that we know to be not completely true: we pretend that our private interests are not at stake, and that people involved are strangers to us. Finally, process (v) is purely emotive and subjective, in that we simply let ourselves experience the subjective properties of the sentiments that we have towards the object under moral evaluation: at this point, no judgement is made, and no descriptive thought is entertained. It is a process of forming emotions, not beliefs or make-beliefs. According to Hume, the first four steps in our inquiry of morals are merely preparatory – they prepare us for the reliability of our subsequent sentiments as indicators of moral properties. It is not until we engage ourselves in the final, emotive process, in which we feel the sentiment of approbation or disapprobation, that we can tell whether the object under evaluation is good or evil. On the necessity of experiencing moral sentiments in our making judgements about morals, Hume writes: Take any action allow’d to be vicious. […] The vice entirely escapes you, as long as you consider the object. You never can find it, till you turn your reflection into your own breast, and find a sentiment of disapprobation, which arises in you, towards this action. Here is a matter of fact; but ’tis the object of feeling, not of reason. It lies in yourself, not in the object. (T, 3.1.1.26/468–9, first emphasis original, all others added)28 If any material circumstances be yet unknown or doubtful, we must first employ our enquiry or intellectual faculties to assure us of it; and must suspend for a time all moral decision or sentiment. […] But after every

68 Is Hume Inconsistent? – Motivation and Morals

circumstance, every relation is known, the understanding has no further room to operate. […] The approbation or blame, which then ensues, cannot be the work of judgment, but of the heart; and is not a speculative proposition or affirmation, but an active feeling or sentiment. In the disquisitions of understanding, from known circumstances and relations, we infer some new and unknown. In moral decisions, all the circumstances and relations must be previously known; and the mind, from the contemplation of the whole, feels some new impression of affection or disgust, esteem or contempt, approbation or blame. (EPM, approx. 1.11/175, emphasis added)29 In short, under Hume’s canonical method, which is warranted by H and H*, the final non-rational process of feeling moral sentiments is a necessary step in acquiring knowledge, or making approximate judgements, about moral good and evil, virtue and vice. In short, if we only employ Hume’s canonical method in our moral inquiry, then it is true to say that judgements and knowledge of moral good and evil cannot be derived merely by reasoning (that is, H4a and H4b).

2.4 The split between judgement and sentiment Reasoning, on Hume’s view, is essentially a process of making inferences – to reason is simply to infer something unknown from something already known or presupposed.30 According to Hume, there are only ‘two kinds’31 of reasoning, namely: ‘demonstrative’ (that is, deductive) reasoning and ‘probable’ (that is, inductive) reasoning (T, 1.3.1/455; 1.3.2–6/73–92). By demonstrative reasoning, we infer previously unknown ‘abstract relations of ideas’ (that is, conceptual relations: T, 2.3.3.2/413). By probable reasoning, we infer previously unknown ‘real existence or the relations of objects’ (that is, cause or effect, or causal relation between them: T, 1.3.2.2/73–4). Now, consider the following argument, which is a combination of demonstrative and probable reasonings. P1

P2 P3

For all the people whom I have observed from all walks of life, if (i) they believe that an action of type ␾ has been performed by a person under the circumstances of type S, (ii) they take into account facts of human nature, (iii) their interests are not at stake, and (iv) none of their acquaintances are involved, then (v) they feel the sentiment of disapprobation towards the action. [empirical observation] My friend’s attack on my enemy is an action of type ␾ performed under the circumstance of type S. [empirical observation] For all human beings, they are disposed, under the Humean ideal conditions C1, C2, C3, C4, to feel the sentiment of disapprobation

Norva Y. S. Lo

P4

P5

69

towards my friend’s attack on my enemy. [from P1 and P2, by probable reasoning] An action or character is virtuous/vicious if and only if all human beings are disposed, under the conditions C1, C2, C3 and C4, to feel the sentiment of approbation/disapprobation towards it. [H, Hume’s conceptual analysis of morals] My friend’s attack on my enemy is vicious. [from P3 and P4, by demonstrative reasoning]

This argument is inductively as well as deductively valid. It (its premise P4) endorses Hume’s analysis of morals (that is, H), and delivers a moral conclusion.32 Hume himself never discussed acquiring moral knowledge by arguments similar to the one above. Nonetheless, his analysis of morals allows this way of gaining moral knowledge. Now, this is a bad news for the internalist readings of Hume. For, contrary to the first and second internalist readings of the conclusion of his Motivation Argument (that is, H4a and H4b), the above argument derives a moral conclusion purely by probable and demonstrative reasonings. Furthermore, contrary to those internalist readings of the major premise of Hume’s Motivation Argument (that is, H1a and H1b), the above argument shows that it is possible for someone to arrive at a moral conclusion without feeling any matching moral sentiment, and thus without being motivated to behave morally. For example, suppose by running the above argument, I arrive at the negative moral conclusion on my friend’s attack on my enemy. Despite my unfavourable judgement, it is possible that I might not feel any negative moral sentiment towards my friend’s behaviour. This is because I might not be in the ideal conditions C3 and C4, and so I might not be impartial, when I contemplate upon what my friend has done. Worse still, I might even delight in my friend’s behaviour due to my dislike of the victim. Here, we have a case of the split between judgement and sentiment – rationally one might judge something to be vicious, but emotionally one might feel no unfavourable moral sentiment towards it. Moreover, that such a divide is possible is a logical consequence of Hume’s analysis of morals, which reduces facts about good and evil to empirical facts about human psychological dispositions, which can be studied from the detached perspective of an onlooker. In short, if H is true, then H1a and H1b are both false. H1c is also false under H. For given H, if something is virtuous/vicious, then we are disposed under conditions C1, C2, C3 and C4 to feel the sentiment of approbation/ disapprobation towards it. If we are more or less under those conditions, then we will more or less feel those favourable/unfavourable moral sentiments towards it. Unfortunately, there is no guarantee that we will satisfy all those conditions, and thus no guarantee that we will feel the corresponding moral sentiments or be motivated to act accordingly. Thus, moral properties themselves have no more reliable influence on our passions and actions than our judgements or knowledge of them.

70 Is Hume Inconsistent? – Motivation and Morals

How exactly does the split between judgement and sentiment come about? It has nothing to do with whether the judgement is correct. For a split between moral knowledge and moral sentiment is just as likely. Some might argue that such a split is due to the person not really endorsing the judgement. In other words, someone does not really judge something to be virtuous/vicious unless she feels a positive/negative sentiment towards that thing. But this way of thinking about the problem is not helpful. For isn’t a so-called real moral judgement (or a real endorsement of it) simply the judgement accompanied by a matching sentiment? If a real moral judgement is simply that, then it becomes a trivial truism to say that the divide between judgement and sentiment is due to a lack of real endorsement – there is a split between my judgement and my sentiment because my judgement is not accompanied by a suitable sentiment! Rather, the occasional split between our moral judgements and sentiments is to be explained in terms of how we arrive at those judgements. Consider again the above case of my friend’s attack on my enemy. Why do I have a negative judgement of my friend and yet fail to feel a corresponding negative sentiment towards him? The answer is: because in making such a moral judgement, I did not use the Humean canonical method, which requires the experience of moral sentiments as an essential step – that is, because I have arrived at the judgement by completely non-affective means. That is equivalent to the case of a purely rational being, incapable of any feelings, arriving at the same moral conclusion by running the same argument above. This being, a perfect observer and moral judge though it might be, could never feel the moral sentiment that is required to motivate a corresponding action from a human being. The Humean canonical method of making moral judgement and acquiring moral knowledge, though certified by Hume’s conceptual analysis of morals, is not the only method available. There is the participant’s way of finding out facts of morals; but there is also the onlooker’s way – both compatible with H. In short, there are two conclusions to be drawn: (1) The major premises in the three internalist readings of Hume’s Motivation Argument (namely: H1a, H1b and H1c) are all incompatible with his own analysis of morals (that is, H). Neither do we always form emotions, nor are we always motivated to act, according to our judgements, knowledge, or the facts, of moral properties. (2) The anti-rationalist conclusions in the three internalist readings of Hume’s argument (namely: H4a, H4b and H4c) also turn out to be all false under Hume’s analysis of morals. For Hume’s analysis warrants the onlooker’s method of making judgements, and acquiring knowledge, about good and evil. And this method, as we have seen, demonstrates that moral judgements and knowledge can indeed be derived (in the sense of ‘inferred’)33 simply by reasoning. For similar reasons, moral properties also turn out to be reducible to relations that are distinguishable by reasoning – namely, the empirical relations between different objects under evaluation

Norva Y. S. Lo

71

and our different psychological dispositions to react emotionally towards them, which are distinguishable by probable reasoning.

2.5 Reasoning and other cognitive processes It can be argued, however, that when Hume concludes that morality cannot be ‘deriv’d’ from reason alone, what he means is that our judgements or knowledge about morals cannot be the product of mere reasoning.34 Why not? Because even if we employ the onlooker’s method to acquire moral knowledge or make approximate moral judgements, which involves no exercise of the emotions, we still need to do more than just reasoning. Namely: we need to observe. For example, both premises P1 and P2 in the onlooker’s argument are the results of empirical observation, without which we would not arrive at the conclusion of the argument. Observation, however, does not belong to the province of reason. For it involves no inference making, and is in Hume’s view ‘a mere passive admission of the impressions thro’ the organs of the sensation’ (T, 1.3.2.2/73). Whether we use the onlooker’s method or the Humean canonical method (that is, the participant’s method) in our moral inquiry, the non-rational procedure of observation is indispensable. Hence, those who want to defend the internalist readings of Hume’s Motivation Argument can argue that it is after all correct for a Humean to hold H4a and H4b, the anti-rationalist conclusion that moral judgements and knowledge cannot be ‘deriv’d’ by (that is, the product of) mere reasoning. Similarly, the defenders of the internalist readings might argue that it is correct for a Humean to hold H4c, the anti-rationalist conclusion that moral properties cannot be reduced to relations that are distinguishable by mere reasoning. For non-rational observation is needed before we can distinguish the different empirical relations (between objects and our dispositions to react emotionally towards them) which, according to H, constitute different moral properties. I do not think that the above defence can save the internalist readings of Hume’s Motivation Argument. First of all, we have seen that a core problem with the three internalist versions of Hume’s argument is that their major premises (that is, H1a, H1b and H1c) are all inconsistent with Hume’s own analysis of morals (that is, H). But the above defence of the internalist interpretations does not address this problem at all. Second, it follows from the above defence that strictly speaking, reasoning cannot by itself produce knowledge of any kind at all. For even a priori knowledge, such as that of logic and mathematics, still requires the exercise of non-inferential ‘intuition’ (for example, the intuition of the principle of excluded middle, and the intuition that 2 is less than 5), which, according to Hume, is not reasoning because it involves no inference making but merely ‘discover[y] at first sight’ (T, 1.3.1.2/70). Thus, the above defence can save (the three versions of) the conclusion of Hume’s Motivation Argument only

72 Is Hume Inconsistent? – Motivation and Morals

at the cost of making them almost trivially true – reasoning by itself could not produce knowledge or sound judgements about morals simply because reasoning by itself could not produce any knowledge or judgement whatsoever; likewise, moral properties cannot be reduced to relations that are distinguishable by mere reasoning simply because no relations whatsoever can be distinguished by mere reasoning. If the anti-rationalist conclusion of Hume’s Motivation Argument were to be defended in this way, then the point about motivation would be completely lost – all the three premises in the Motivation Argument would become redundant. So, if we are to defend Hume’s whole argument, not just its conclusion considered in isolation, then we’d better keep the focus on the relation between motivation and morality, and the difference between passion and reason in terms of their relation or non-relation to motivation. Third, when Hume refers to the rational process of reasoning in the course of giving his Motivation Argument, his points about reasoning seem to apply to all cognitive processes, understood as the ‘discovery of truth or falsehood’, of which reasoning is just one particular case. For example, he writes: Reason is the discovery of truth or falsehood. Truth or falsehood consists in an agreement or disagreement either to the real relations of ideas, or to real existence and matter of fact. Whatever, therefore, is not susceptible of this agreement or disagreement, is incapable of being true or false, and can never be an object of our reason. Now ’tis evident our passions, volitions, and actions, are not susceptible of any such agreement or disagreement; being original facts and realities, compleat in themselves, and implying no reference to other passions, volitions, and actions. Tis impossible, therefore, they can be pronounced either true or false, and be either contrary or conformable to reason. (T, 3.1.1.9/458, emphases added) This argument […] proves […] that as reason can never immediately prevent or produce any action by contradicting or approving of it, it cannot be the source of moral good and evil, which are found to have that influence. (T, 3.1.1.10/458) It is clear that this argument from Hume can be readily extended to cover cognition in general, which includes the non-rational processes of observation and intuition. The extended argument will look exactly the same as the original one save that the term ‘reason’ is uniformly replaced by ‘cognition’. For example, instead of saying ‘reason is the discovery of truth or falsehood,’ the extended argument says ‘cognition is the discovery of truth or falsehood.’ And instead of saying ‘reason can never immediately prevent

Norva Y. S. Lo

73

or produce any action by contradicting or approving of it,’ the extended argument says ‘cognition can never immediately prevent or produce any action …’, and so on, and so on. Fourth, Hume himself needs to extend his point on the inertness of reason to that of cognition. For he argues that ‘since vice and virtue are not discoverable merely by reason, … it must be by means of some … sentiment they occasion, that we are able to mark the difference betwixt them’ (T, 3.1.2.1/470, emphases added). But Hume himself recognizes operations of our mind, such as observation and intuition, which are neither the rational process of reasoning nor the emotive process of experiencing sentiments. So, in order not to commit the fallacy of false alternatives when he argues from ‘not … reason’ to ‘sentiment’, he needs an extra premise which says that those other operations, such as observation and intuition, cannot by themselves distinguish good and evil. But how can he justify this extra premise? – obviously, by extending his Motivation Argument to cover cognition in general, not just reasoning. For the reasons given above, when Hume says ‘vice and virtue are not discoverable merely by reason,’ I read him as meaning, or at least implying, that moral properties are not discoverable merely by cognitive processes. Accordingly, the best we can do for the three internalist readings of Hume’s Motivation Argument is to modify them in the following manner: The First Modified Internalist Reading H1a Judgments of moral good and evil excite passions and motivate actions. H2a* Cognitive processes (for example, reasoning, observation and intuition) cannot by themselves excite passions or motivate actions. H3a* That which excites passions and motivates actions cannot be merely the product of processes which cannot by themselves excite passions or motivate actions. H4a* Judgments of moral good and evil cannot be merely the product of cognitive processes. [from H1a, H2a* and H3a*] The Second Modified Internalist Reading H1b Knowledge of moral good and evil excites passions and motivates actions. H2b* Cognitive processes cannot by themselves excite passions or motivate actions. [same as H2a*] H3b* That which excites passions and motivates actions cannot be merely the product of processes which cannot by themselves excite passions or motivate actions. [same as H3a*] H4b* Knowledge of moral good and evil cannot be merely the product of cognitive processes. [from H1b, H2b* and H3b*]

74 Is Hume Inconsistent? – Motivation and Morals

The Third Modified Internalist Reading H1c Properties of moral good and evil excite passions and motivate actions. H2c* Relations that are distinguishable simply by cognitive processes (for example, mathematical relations and causal relations) cannot by themselves excite passions or motivate actions. H3c That which excites passions and motivates actions is not reducible to things that cannot by themselves excite passions or motivate actions. H4c* Properties of moral good and evil are not reducible to relations that are distinguishable simply by cognitive processes. [from H1c, H2c* and H3c] Note that the above modifications keep the anti-rationalist aspects of Hume’s Motivation Argument. This is because reasoning is one kind of cognitive process, and therefore the modified conclusions, H4a*, H4b* and H4c* (which express the idea that morality cannot be derived from cognition alone), entail the original anti-rationalist conclusions, H4a, H4b and H4c (which express the idea that morality cannot be derived from reason alone). This is, I believe, the best way that the internalist readings of Hume’s Motivation Argument can take into account his view on cognitive but nonrational mental processes. Unfortunately, however, the core problem with the three internalist readings of Hume’s Motivation Argument still remains. The major premises in the modified versions are still the same as before (that is, H1a, H1b and H1c). They all assert some necessary connection between morality and motivation. But, as we have seen, that is inconsistent with Hume’s own conceptual analysis of morality (that is, H), which allows the possibilities that people can make judgements and have knowledge about morals without automatically experiencing any matching sentiments and thus without being motivated to act accordingly, and similarly that things can have the property of being morally good or evil without people reacting appropriately towards them. Moreover, contrary to the conclusions in the three modified internalist readings (that is, H4a, H4b and H4c), people can indeed make judgements and acquire knowledge about morals by purely cognitive means, as the onlooker’s method (which is warranted by H) has demonstrated, and therefore the properties of good and evil are indeed reducible to relations distinguishable purely by cognition. In short, if Hume’s analysis of morals is correct, then all three modified internalist readings of Hume’s Motivation Argument (like the unmodified ones) are unsound. If Hume is to be consistent, then he had better not hold any necessary connection between morality and motivation. Contrary to the prevailing view, I do not think that Hume holds any such internalist connection. After all, he himself says: ‘’Tis one thing to know virtue, and another to conform

Norva Y. S. Lo

75

the will to it’ (T, 3.1.1.22/465–6). And, although he argues that we should judge morality by reflecting on how we would feel under the conditions ideal for having reliable moral sentiments, he also acknowledges that the sentiments that we actually feel in our actual situations can be too ‘stubborn and inalterable’ to match our considered judgements, so that the considerations of our ideal sentiments under the ideal conditions may merely help in ‘correcting our language’, but not correcting our subjective sentiments.35,36 ‘The passions’, he says, ‘do not always follow our corrections’ of moral language (T, 3.3.1.21/585). Clearly, Hume is aware of the possible divide between morals and motivation. For these reasons, we should be suspicious of any internalist reading of Hume. Then, what are the best readings of Hume’s Motivation Argument? I shall conclude this article with the following three non-internalist interpretations, which, I believe, are not far from what Hume had in mind. The First Non-Internalist Reading H1d Some (but not all) moral judgements and pieces of moral knowledge excite passions and motivate actions. H2d Cognitive processes cannot by themselves excite passions or motivate actions. [same as H2a*] H3d That which excites passions and motivates actions cannot be purely the product of processes which cannot by themselves excite passions or motivate actions. [same as H3a*] H4d Some moral judgements and pieces of moral knowledge (namely those that excite passions and motivate actions) cannot be purely the product of cognitive processes. [from H1d, H2d, H3d and H4d] H5 Some moral judgements and pieces of moral knowledge cannot be purely the product of reasoning (which is a cognitive process). [from H4d] H6 The epistemological claim of moral rationalism is that moral knowledge is purely the product of reasoning (which is a cognitive process). H7 The epistemological claim of moral rationalism is false. [from H5 and H6] The Second Non-Internalist Reading H1e The internalist claim of moral rationalism is that moral knowledge always excites passions and motivates actions. H2e Cognitive processes (for example, reasoning) cannot by themselves excite passions or motivate actions. [same as H2a*] H3e That which excites passions and motivates actions cannot be purely the product of processes which cannot by themselves excite passions or motivate actions. [same as H3a*]

76 Is Hume Inconsistent? – Motivation and Morals

H4e

H6

H8

The internalist claim of moral rationalism requires that moral knowledge cannot be purely the product of cognitive processes (for example, reasoning). [from H1e, H2e and H3e] The epistemological claim of moral rationalism is that moral knowledge is purely the product of reasoning (which is a cognitive process). Moral rationalism is an inconsistent position in that either its internalist claim or its epistemological claim is false. [from H4e and H6]

The Third Non-Internalist Reading H1e The internalist claim of moral rationalism is that moral knowledge always excites passions and motivates actions. H2e Cognitive processes (for example, reasoning) cannot by themselves excite passions or motivate actions. [same as H2a*] H3e That which excites passions and motivates actions cannot be purely the product of processes which cannot by themselves excite passions or motivate actions. [same as H3a*] H4e The internalist claim of moral rationalism requires that moral knowledge cannot be purely the product of cognitive processes (for example, reasoning). [from H1e, H2e and H3e] H9 The analysis of the essence of morals given by moral rationalism reduces moral good and evil to the conformity to reason and the contrariety to it, respectively, which can be known purely by reasoning. H10 Moral rationalism is an inconsistent position in that either its internalist claim or its analysis of the essence of morals is false. [from H4e and H9] Consider the three non-internalist reading of Hume in turn. The first one shows that Hume’s Motivation Argument can validly reject moral rationalism without its major premise (that is, H1) being given an internalist reading. The second and third readings both show that it is moral rationalism which endorses internalism, and that Hume’s Motivation Argument validly rejects moral rationalism by showing that its internalism is inconsistent with its other theoretical commitments. In conclusion, Hume is not an internalist, and he is not inconsistent. It is the internalist readings of Hume that are inconsistent with his analysis of morals.

Notes 1. See T, 3.1.1/455, which is entitled ‘Moral distinctions not deriv’d from reason’. 2. See T, 3.1.1/455, and the editors’ summary of Hume’s various arguments in Norton and Norton (2000, pp. 176–8).

Norva Y. S. Lo

77

3. According to Sayre-McCord (1997), Rachel Cohon and David Owen use the term ‘Motivation Argument’ in their ‘Hume on Representation, Reason and Motivation’ which was delivered at the University of Nottingham at the 1996 meeting of the International Hume Society. 4. See T, 3.1.1.10/458, where Hume writes: ‘as reason can never immediately prevent or produce any action by contradicting or approving of it, it cannot be the source of moral good and evil, which are found to have that influence’ (emphasis added). 5. See T, 3.1.1.10/458, where Hume writes: ‘The merit and demerit of actions frequently contradict, and sometimes controul our natural propensities. But reason has no such influence. Moral distinctions, therefore, are not the offspring of reason’ (emphasis added). 6. See n. 5 above. Also see T, 3.1.1.16/462–3, where Hume writes: ‘’tis impossible, that the distinction betwixt moral good and evil, can be made by reason; since that distinction has an influence upon our actions, of which reason alone is incapable’ (emphasis added). 7. Hume (T, 3.1.1.2/456) uses the phrase ‘those judgments, by which we distinguish moral good and evil’ (emphasis added). 8. Hume (T, 3.1.1.22/465–6) discusses knowledge of virtue and vice and its influence on the will. 9. Hume (T, 3.1.1.4/456–7) speaks of ‘measures of right and wrong’ as ‘impos[ing] an obligation’ on rational beings. 10. See T, 3.1.1.19/463–4, n. 69, where Hume attributes to the rationalists who assert that ‘morality is demonstrable’ an implicit argument ‘that morality lies in the relations, and that the relations are distinguishable by reason’ (emphasis added). 11. See T, 3.1.1.4/456, 19/463, n. 69, 21–3/464–6, where Hume rejects the rationalist claim that moral properties consist of relations that can be inferred by reasoning. See T, 3.1.1.22–3/465–6, which can be interpreted as supportive of the third reading of Hume’s Motivation Argument. 12. In the Routledge Encyclopedia of Philosophy, Wallace (1998, pp. 522–3) says, ‘David Hume is presumably giving voice to an internalist point of view when he makes central to his theory the idea that “morals excite passions, and produce or prevent actions.”’ For various internalist readings of Hume, see Harrison (1976), Foot (1978), Mackie (1980), Brown (1988), Coleman (1992), Dimock (1992), Radcliffe (1996), Bricke (1996), Sayre-McCord (1997) and Liu (2003). 13. For Hume’s discussion of the rationalist thesis concerning the ‘essence’ of morality, see T, 3.1.1.4/456–7, 13/460, 15/461, n. 68, 19/463–4, 25/467–8. 14. ‘Truth or falsehood’, according to T, 3.1.1.9/458, ‘consists in an agreement or disagreement either to the real relations of ideas, or to real existence and matter of fact’. Both of these, like ‘conformity to reason’ and ‘contrariety to it’, are regarded by Hume (T, 3.1.1.18–26/463–9) as relations distinguishable by reason. 15. Also see T, 3.3.1.3/574–5, where Hume writes: ‘we … may pronounce any quality of the mind virtuous, which causes love or pride; and any one vicious, which causes hatred or humility’. On his view, pride/humility is a certain pleasant/ painful sentiment that we feel towards ourselves when we experience or contemplate upon some agreeable/disagreeable qualities of ourselves. That is why Hume writes: ‘everything related to us, which produces pleasures or pain, produces likewise pride or humility’ (T, 2.1.6.3/291, emphasis removed; see also T, 2.1.7.8/297–8), and ‘[t]o approve of a character is to feel an original delight upon its appearance. To disapprove of it is to be sensible of an uneasiness. The pain and

78 Is Hume Inconsistent? – Motivation and Morals

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

pleasure, therefore, being the primary causes of vice and virtue, must also be the causes of all their effects, and consequently of pride and humility, which are the unavoidable attendants of this distinction’ (T, 2.1.7.5/296). Hume’s treatment of love and hatred is similar to his treatment of pride and humility. The difference is only that while pride and humility are directed to oneself, love and hatred are feelings directed to some other person (see T, 2.2.1/275). According to Hume (T, 3.1.2.2–6/470–3, 11/475), the sentiment of approbation is a ‘pleasing sentiment’, and disapprobation the contrary. Furthermore, Hume (T, 1.3.10.2/118–9; 2.1.1.4/276–7; 2.1.10.8/314; 2.3.3.3/414; 2.3.9.1/438; 3.3.1.2/574, and EPM, approx. 1.21/141) maintains that we naturally desire what pleases us and are repelled by what brings us pain. It follows that the moral sentiments can excite desires and thus to some extent motivate moral behaviour (Cf. EPM, approx. 1.3/170). For example, Hume considers the following features as ingredients of human nature: regard to self-interest or, what Hume sometimes calls, ‘selfishness’ (T, 3.2.2.16/494), ‘limited generosity’ (T, 3.2.2.16/494), a tendency to ‘overlook’ the value of things, ‘to which we have been long accustom’d’ (T, 2.1.6.3/291). According to Hume (T, 3.2.1.2–8/477–9), an action is virtuous/vicious only if it is the product of a virtuous/vicious ‘motive or principle’. Hence, in the case of evaluating an action, condition C1 requires that the spectator is aware of the motive or principle, from which the action was produced. For other examples of realist readings of Hume, see Firth (1952), Wiggins (1987), Sayre-McCord (1994). The realist readings of Hume are not currently dominant. It will take another occasion to defend that reading that I advocate in this essay against its various rivals. For instances of different anti-realist readings of Hume, see Hunter (1962), Flew (1963), Mackie (1980), Ayer (1980) and Blackburn (1984 and 1998a). For Hume’s account of sympathy, see T, 1.1.1.1–7/1–4; 1.1.2–3/7–9; 1.1.4.1– 2/10–11; 1.3.6.13/92–3; 1.3.8.2/98–9; 1.3.10.3/119–20; 1.3.14.28/168–9; 1.3.15.1, 5/173; 1.4.6.4; 2.1.4.2/252–3; 2.1.11.4–8/317–20; 2.2.4.7/354; 3.3.1.7/575. According to this Humean psychological principle, it is a natural propensity of the mind to associate the impression or idea of any thing with the idea of any other thing, where the two things are related to each other by causation, resemblance, and/or contiguity in time and place. See T, 1.3.6.13/92–3. Cf. T, 1.1.4.1–2/10–11, 2.1.4.2/283. This Humean psychological principle says that it is a natural propensity of the mind to communicate a share of the force and vivacity of any present impression to such ideas as are associated with it. See T, 1.3.8.2/98–9. The general form is dispositional analyses of value which is as follows: X is valuable/disvaluable if and only if valuers are disposed, under certain conditions, to have a favourable/unfavourable response towards X. See EPM (5.46/231–2) where Hume understands ‘humanity’ as ‘a concern for others’ regardless of our personal interests and relations. Also see EPM (2.5/177–8) where he takes ‘humanity’ to be a virtue that ‘proceeds from a tender sympathy with others, and a generous concern for our kind and species’. Cf. EPM, 5.18/220. Hume (T, 3.3.1.21/585) talks of the ‘degrees of vice and virtue’. Also see, for example, T, 3.1.1.13/460; 3.3.4.10/611, where Hume takes virtue and vice as admitting of degrees. Likewise, Hume (EPM, approx. 1.11/175) talks about some moral duty and obligation as ‘superior’. See Hume (1741) for a discussion on cultivating our

Norva Y. S. Lo

26. 27.

28.

29. 30. 31.

32.

33. 34.

35. 36.

79

abilities (which Hume calls ‘taste’) to make finer distinctions and comparisons of moral and aesthetic properties. Hume (1760, p. 136). Also see n. 19 above. See EPM, approx. 1.10/173, where after putting forward his conceptual analysis of morals, he says: ‘We then proceed to examine a plain matter of fact, to wit, what actions have this influence [on the sentiments of the judicious “spectator”]: We endeavour to extract some general observations with regard to these sentiments’ (emphasis added). Also see Ardal, 1966. And see Lo, 2001 for a discussion of the operation of Humean sympathy in the production of moral sentiments in the case of the virtue of justice and the vice of injustice. Also see EPM, approx. 1.12/13. See T, 1.3.1.2/70; 1.3.2.2/73; 1.3.3.6/81. Also see Owen, 1999, ch. 5. See T, 3.1.1.18/463, where Hume writes: ‘the operations of human understanding divide themselves into two kinds, the comparing of ideas, and the inferring of matter of fact; […] nor is there any third operation of the understanding.’ The inference from P3 and P4 to P5 is an example of a ‘deduction’ from ‘is’ to ‘ought’, about which Hume (T, 3.1.1.27/469) says ‘’tis necessary that it shou’d be observ’d and explain’d; and at the same time that a reason shou’d be given, for what seems altogether inconceivable, how this new relation can be a deduction from others, which are entirely different from it’ (emphases added). The Humean explanation for the deduction from ‘is’ to ‘ought’ is that Hume’s conceptual analysis of morals, H, is a bridging principle linking ‘ought’ (the LHS of H) to ‘is’ (the RHS of H). For a debate on whether Hume holds a gap between ‘is’ and ‘ought’, see Hudson, 1969. See, for example, T, 3.1.1.6/457, where he appears to use the term ‘deriv’d’ in the sense of ‘inferred’. See, for example, T, 3.1.1.24/466; 3.1.2.6/473; 3.2.1.6/478, 9/479, 17/483; 3.2.2.9/489, 18/495, 20/496, 23/498; 3.2.3.7/506; 3.2.6.9–10/531–3, where Hume appears to use the term ‘deriv’d’ in the sense of ‘the product of’. See n. 20 above. For Hume’s own examples of ‘correcting our language, where the sentiments are more stubborn and inalterable’, see T, 3.3.1.16/582, 18/583, 21/585.

3 If Not Non-Cognitivism, Then What? Charles R. Pigden

3.1

Introduction

According to Michael Smith, the big issue in metaethics (what he calls “The Moral Problem”) is how to accept the seemingly true premises of Hume’s Motivation Argument while fending off its non-cognitivist conclusion, a conclusion that Smith takes to be both repugnant and false (Smith, 1994, The Moral Problem, henceforward MP). This presupposes two claims: (A) that Hume was arguing for non-cognitivism; and (B) that the Motivation Argument (as construed by non-cognitivists) is a good one, so good indeed that it can only be evaded by some very fancy footwork. I shall argue that both claims are false. Hume was not arguing for non-cognitivism since he was not a non-cognitivist. For Hume, moral properties are akin to secondary qualities, a view he derived from his sometime hero Francis Hutcheson. “Vice and virtue, therefore, may be compar’d to sounds, colours, heat and cold, which, according to modern philosophy, are not qualities in objects, but perceptions in the mind” (T, 3.1.1.26/469). We don’t attribute to Hume a non-cognitive theory of colors. Why then should we attribute to him a non-cognitivist theory of vice and virtue? Thus the non-cognitivist argument that has been extracted from Hume’s work is not the argument that he intended. But whether he intended it or not, the non-cognitivist argument is a failure. So too is the argument he actually advanced. Hume fails to show what he intended to show, that our moral distinctions are derived from a moral sense. Taking my cue from Michael Smith, I shall try to extract a decent argument for non-cognitivism from the text of the Treatise. I argue that the premises are false and that the whole thing rests on a petitio principi. I then re-jig the argument so as to support the conclusion that Hume actually believed (namely that an action is virtuous if it would excite the approbation of a suitably qualified spectator). This argument too rests on false premises and a begged question. Thus the Motivation Argument fails both as an argument for non-cognitivism and as an argument for Hume’s favored 80

Charles R. Pigden

81

thesis, that moral distinctions are not derived from reason. So far as the Motivation Argument is concerned, cognitivists and rationalists can both rest easy.

3.2 Hume – grandfather of non-cognitivism? Hume’s posthumous reputation as the grandfather of non-cognitivism (which for me is primarily the semantic thesis that moral judgments are not full-bloodedly true or false) rests principally on three short passages in his Treatise: the Slavery of Reason Argument (T, 2.3.3/413–18), the Motivation Argument (T, 3.1.1.5–10/456–8) and the No-Ought-From-Is observation (T, 3.1.1.27/469–70). Having thrashed No-Ought-From-Is to death elsewhere, I shall simply remark that on my interpretation, it provides no support for non-cognitivism whatsoever (see Pigden, 1989, and Pigden, 2009b and the Introduction to this volume, §3). Now the Slavery of Reason Thesis (“Reason is, and ought only to be the slave of the passions”) does not furnish a direct argument for noncognitivism. Rather it supplies one of the premises for the Motivation Argument, on which non-cognitivists chiefly rely. The “magnetism” of moral judgments, the power they possess to motivate the true believer, provides the other. According to Michael Smith, the central problem in meta-ethics is an apparently inconsistent triad of propositions, one of which is the negation of non-cognitivism (MP, p. 12). But, in fact, the inconsistent triad can easily be recast as an argument for non-cognitivism, using the other two members of the triad as premises.

The Motivation Argument: Non-Cognitivist Version A (1)

(2)

If someone judges that it is right that she φs then, ceteris paribus, [and without the aid of any pre-existing desire] she is motivated to φ. An agent can only be motivated to act by a belief when she has a desire for an end to which the belief points out the means, where belief and desire are, in Hume’s terms, distinct existences. Therefore:

(3)

(4)

If someone judges that it is right that she φs then, her [psychological] “judgment” that it is right that she φs does not constitute a belief. The best explanation of (3) is [insert non-cognitivist theory of choice].

82

If Not Non-Cognitivism, Then What?

Premise (1) is a variant of Hume’s thesis that it is “confirm’d by common experience” that “men are often govern’d by their duties, and are deter’d from some actions by the opinion of injustice, and impell’d to others by that of obligation.” Premise (2) is the Slavery of Reason Thesis as it is usually understood. When it comes to motivation, reason – that is, the faculty which forms beliefs – is “utterly impotent in this particular,” since beliefs by themselves cannot motivate. Thus we have a psychological argument for a semantic conclusion. Moral “beliefs” motivate in a way that genuine, product-of-reason beliefs do not. Hence moral “beliefs” are not genuine beliefs and the judgments that express them are not genuine propositions. Moral judgments have a different role, which is to express emotions or prescribe conduct (yadda, yadda, yadda).

3.3 Actions, affections and the influence of reason This Smithian formulation of the argument is still a little too far from Hume’s text. For a start, Hume tends to talk about virtue rather than rightness and in this context is concerned with moral beliefs and moral judgments, generally not just judgments that it is right to φ. Moreover, it is clearly Hume’s opinion that unlike the products of reason, moral “beliefs” can excite passions as well as producing or preventing actions (T, 3.1.1.6/457). That is, moral beliefs can induce new desires by themselves, which may or may not be the immediate causes of actions, whereas product-of-reason beliefs cannot. This is an important point. Hume’s proof would break down if product-ofreason beliefs could excite new desires by themselves, even if those desires were at one or two removes from action. Even if action-producing wants, the immediate causes of action, were always caused either by moral “beliefs” or beliefs-plus-desires, reason would lose its impotence if those desires were themselves solely the products of product-of-reason beliefs. On Smith’s reading, the second premise of the Motivation Argument goes something like this:

(2*) An action can only be motivated by a belief if that belief is conjoined with a desire to which the belief points out the means.

Motivating reasons – the psychological states that explain an agent’s actions – are “pairs comprising desires and means-end beliefs.” But the Smithian thesis (2*) is quite compatible with the idea that the desire in such a pair should itself be motivated by a belief but not a desire, thus:

Charles R. Pigden

83

A (product-of-reason) belief (for example, that φ will avert my utter ruin)

Desire for φ

The (product-of-reason) belief that ψ-ing conduces to φ

The desire to ψ

The act of ψ-ing Given (2*) a (product-of-reason) belief cannot be the parthenogenetic mother of an action, but it can be the parthenogenetic mother of a desire, and hence the parthenogenetic grandmother of an action. For in the above diagram, the causal antecedents of the action are all (product-of-reason) beliefs even though one of them causes the action through the intermediary of a desire of which it is the sole psychological cause. But if this could happen, there would be no essential distinction between the moral “beliefs” which can produce actions without the aid of pre-existing desires and the (product-of-reason) beliefs which cannot, except that the moral “beliefs” would be able to do it directly whereas the (product-of-reason) beliefs would be condemned to doing it at one remove. Thus Hume and (I think) Smith require something stronger than (2*): (2**) An action or a new desire can only be motivated by a belief if that belief is conjoined with a desire to which the belief points out the means. One final point. What it means to say that moral beliefs motivate “alone” or “by themselves” is either that moral “beliefs” cause new desires, (whether or not these desires result in action) or that they in some sense register new desires so that subscribing to a moral “belief” involves adopting a new desire. And when we say that genuine or product-of-reason beliefs alone cannot motivate, what this means is that they cannot cause new desires,

84

If Not Non-Cognitivism, Then What?

except desires for the means to ends set by pre-existing desires and that they themselves are not (even partly) constituted by desires.

3.4 Hume and reason The non-cognitivist interpretation of Hume depends upon a particular reading of reason as our belief-forming faculty. By Book II of the Treatise, Hume has adopted a relatively inclusive conception of reason which not only incorporates deduction but a substantial dollop of custom, habit and imagination, enough of the sensitive part of our natures to enable us to form beliefs about relations of objects including causal connections. But if the non-cognitivist argument is to work we must identify reason with the belief-forming faculty. For unless we assume that all genuine beliefs are product-of-reason beliefs, the Motivation Argument will prove, at best, that moral “beliefs” are not product-of-reason beliefs, thus allowing for the possibility that they are genuine beliefs, but not the products of reason. Thus the non-cognitivist must identify reason with the belief-forming faculty and genuine beliefs with product-of-reason beliefs.

3.5

The non-cognitivist argument restated

We are now in a position to restate the Motivation Argument: The Motivation Argument: Non-Cognitivist Version B (1’)

(2’)

(3’) (4’)

(5)

Moral “beliefs” alone often motivate (either by causing actions or by giving rise to new desires). [“Men are often governed by their duties, and are deterred from some actions by the opinion of injustice, and impelled to others by that of obligation.” “Morals excite passions, and produce or prevent actions.”] Genuine beliefs alone cannot motivate (either by causing actions or by giving rise to new desires). [“Reason alone can never be a motive to any action of the will.” “Reason of itself is utterly impotent in this particular.] Therefore Moral “beliefs” are not genuine beliefs. The best explanation of (3’) is [insert non-cognitivist theory of choice]. Therefore probably [Insert non-cognitivist theory of choice].

This argument has to be understood in the light of Section 3.3 above. Thus when we say that moral beliefs alone often motivate, what this means is not just that they often move us to act without the aid of a pre-existing desire but also that they often give rise to new desires without the aid of any

Charles R. Pigden

85

pre-existing desire. When we say that no genuine belief alone can motivate, what we mean is not just that genuine beliefs by themselves cannot cause actions but that genuine beliefs cannot cause new desires except desires for the means to an end that is already desired. Arguments for Premises (1’) and (2’) can easily be extracted from Hume, but before we get on to that, the argument has to be amended yet again. For, as it stands, the argument does not quite work. Premise (1’) says that moral “beliefs” by themselves often motivate. It does not say that they always do so. And this was pretty clearly Hume’s opinion. “And this is confirmed by common experience, which informs us, that men are often governed by their duties, and are deterred from some actions by the opinion of injustice, and impelled to others by that of obligation [my italics].” Hume is sensibly agnostic about the more extravagant claim that moral “beliefs” alone always motivate or even that they always motivate other things being equal.1 Premise (2’) says that no genuine belief can motivate by itself. But (1’) and (2’) taken together do not imply (3’). For the possibility remains that some moral “beliefs” are genuine beliefs, namely those which do not motivate or those which manage to motivate but only with the aid of a pre-existing desire. The way to deal with this difficulty is not to beef up Premise (1’) – that would be taking too much of a departure from Hume as well as too much of an epistemic risk – but to convert the whole argument into an inference to the best explanation. Moral “beliefs” often do something that genuine beliefs can never do – they give rise to new desires without the aid of a pre-existing desire. And the best explanation of this is that they are fundamentally different kinds of items. This gives us: The Motivation Argument: Non-Cognitivist Version C (1’)

Moral “beliefs” alone often motivate. [That is, they often give rise to both actions and new desires without the aid of pre-existing desires.] (2’) Genuine beliefs alone cannot motivate. [“Reason alone can never be a motive to any action of the will.”] (3a) The best explanation of (1’) and (2’) is that moral “beliefs” are not genuine beliefs. Therefore probably (3’) Moral “beliefs” are not genuine beliefs. (4’) The best explanation of (3’) is [insert non-cognitivist theory of choice]. Therefore probably (5) [Insert non-cognitivist theory of choice].

Thus the argument consists of two inferences to the best explanation, one mounted on top of the other. It is not a deductive argument, but it qualifies as a cogent argument if (1’) and (2’) are true and if the best explanation of this really is that moral “beliefs” are not genuine beliefs.

86

If Not Non-Cognitivism, Then What?

3.6 The Slavery of Reason Argument Premise (2’) is the Slavery of Reason Thesis construed as a non-cognitivist must construe it. Can an argument for this claim be extracted from Hume? Indeed it can. Reason or the understanding “exerts itself after two different ways, as it judges from demonstration or probability; as it regards the abstract relations of our ideas, or the relations of [external] objects.” Since (on the non-cognitivist reading) reason is our belief-forming faculty, this means that all genuine beliefs fall into one of two classes: beliefs about relations of ideas (or analytic beliefs) or beliefs about matters of fact or existence (or synthetic beliefs). “I believe it scarce will be asserted, that the first species of reasoning alone is ever the cause of any action.” Analytic, or relations-of-ideas beliefs, cannot give rise to new desires, except in so far as they suggest beliefs about matters of fact. “A merchant is desirous of knowing the sum total of his accounts with any person: Why? But that he may learn what sum will have the same effects in paying his debt, and going to market, as all the particular articles taken together. Abstract or demonstrative reasoning, therefore, never influences any of our actions, but only as it directs our judgment concerning causes and effects.” This “leads us to the second operation of the understanding.” Beliefs about matters of fact (and more specifically causal connections) do indeed have an influence on our conduct. But it is clear that they cannot do it alone. “It can never in the least concern us to know, that such objects are causes, and such others effects, if both the causes and effects be indifferent to us.” If the merchant has ceased to care about his business and is content to allow it to go to rack and ruin, then the fact that a payment of so many pounds sterling is causally necessary to appease his creditors will be a matter of indifference to him. He must want to pay his bills (or escape the consequences of bankruptcy) before this belief can influence his conduct or even induce a new desire. Reason, therefore, is the slave of the passions, by which we mean that genuine beliefs – the products of reason – cannot motivate by themselves, but require the assistance of a want (T, 2.3.3.2–3/413–14). More formally: Slavery of Reason Argument Mark 1: Non-Cognitivist Version (i)

(ii)

All genuine beliefs are either analytic (to do with the relations of ideas) or synthetic (to do with causal relations). [“The understanding exerts itself after two different ways, as it judges from demonstration or probability; as it regards the abstract relations of our ideas, or those relations of objects, of which experience only gives us information.”] Analytic beliefs cannot motivate except in so far as they lead to synthetic beliefs. [“Abstract or demonstrative reasoning, therefore,

Charles R. Pigden

(iii)

(iv)

(v)

87

never influences any of our actions, but only as it directs our judgment concerning causes and effects.”] Synthetic beliefs cannot motivate if the causes and effects believed in are indifferent to us. [“It can never in the least concern us to know, that such objects are causes, and such others effects, if both the causes and effects be indifferent to us.”] Hence synthetic beliefs cannot motivate unless they point the way to realize some pre-existing desire (from iii). [“’Tis evident in this case that the impulse arises not from reason, but is only directed by it.”] Genuine beliefs alone cannot motivate (from (i), (ii) & (iv)). [“Reason alone can never be a motive to any action of the will.” “Reason is, and ought only to be the slave of the passions, and can never pretend to any other office than to serve and obey them.”]

And (v) is, of course, the second premise of the Non-Cognitivist Argument C.

3.7 Motivation and morals What about Premise (1’)? Does Hume have an argument? Yes. “If morality had naturally no influence on human passions and actions,” he says, “’twere in vain to take such pains to inculcate it; and nothing would be more fruitless than that multitude of rules and precepts, with which all moralists abound.” (T, 3.1.1.5/457.) In other words, our cultural practices presuppose that moral opinions have an influence on our passions and actions, a presupposition that is “confirmed by common experience.” Unfortunately this does not prove Hume’s point, namely that our moral opinions “go beyond the calm and indolent judgments of the understanding.” If the information that smoking kills had naturally no influence on human passions and actions, ’twere in vain to print it on cigarette packets and if “Danger: High Voltage!” had naturally no such influence, nothing would be more fruitless than the multitude of such signs affixed to high-voltage equipment. It is, moreover, confirmed by common experience, that men are often governed by their belief that high-voltage equipment can be dangerous, and are sometimes deterred from smoking by the opinion that it can kill. Yet Hume himself would have to agree that the opinions induced are nothing but the calm and indolent judgments of the understanding, and only have an influence because of our desire not to die. Take away that desire and the beliefs in question would soon become perfectly inert. It might be the same with our moral beliefs. When we bring up children, we don’t just teach them the difference between right and wrong – we teach them to desire the right and abhor the wrong. What Hume needs to prove is not just the innocuous thesis that moral

88

If Not Non-Cognitivism, Then What?

beliefs motivate – something that nobody is likely to disagree with – but belief-internalism, the thesis that they do it “alone.” We must do our duty solely because we believe it to be our duty and not because we have a de dicto desire to do it. And for that thesis Hume has no argument.

3.8 Michael Smith to the rescue? But where Hume fails, perhaps his latter-day disciple can come to the rescue. In my externalist opinion, the reason people do what they think it is right to do is that on the whole they want to do the right thing, where this qualifies as a de dicto rather than a de re desire. That is they want to do whatever the right thing is, not the particular things that they believe to be right. What has Smith got against this desire (or rather these desires, since the desire to avoid the wrong is probably more important than the desire to do right)? After all, by positing this desire the externalist can meet a desideratum that Smith regards as important. He can explain why a change in motivation follows reliably in the wake of a change in moral judgment, at least in the good and strong-willed person, since a strong-willed and conscientious person (conscientiousness being a component of goodness) just is someone who is dominated by his desire to do the right thing. And he can explain why it is that if I want you to give to charity there is some point in persuading you that it would be right for you to do so. Since you want to do the right thing and I have persuaded you that the right thing to do is to give to Oxfam, there is a good chance that you will actually cough up. Smith admits all this, but at the last moment he balks. Why? His reasons seem to be these: (I)

“Commonsense tells us” that if good people judge it to be right to be honest or to care for their children, friends and fellows, they care non-derivatively for these things. That is, it is not the case, that they care for these things because (a) it is right to do so, and (b) they desire to do right. The concern is de re rather than de dicto. (MP, p. 75)

(II)

The externalist supplies the good person with “one thought too many.” The good husband should save his wife because she is his wife, not because she is his wife and husbands ought to save their wives. Even if (as some benighted persons suppose) we should be motivated by an impartial concern for others, this should be a de re concern not a concern derived from a de dicto desire to do the right thing. Someone with that type of concern would not be really virtuous, even by the debased standards of utilitarians. (MP, pp. 75–6)

Charles R. Pigden

89

(III) Externalism enshrines a form of moral fetishism, and indeed elevates a moral fetish into the one and only virtue. (MP, p. 75) The first thing to notice about these arguments is that Smith has lost his grip on what it is that we are trying to explain. The question is not why good people do what they think they ought to do but why people in general tend to do so. There are many highly conscientious people who are very far from good – Nazis and Inquisitors, for example. Indeed the phenomenon we are trying to explain, the phenomenon alluded to in Hume’s thesis and Smith’s own thesis (2) is not goodness at all but conscientiousness, the widespread tendency to live up (or down) to ones precepts. That said, an adequate metaethic must not misrepresent the motives of the virtuous either, but I don’t think the externalist has to do so. Let us start with (I). Is the externalist committed to the view that the concerns of the virtuous agent are either de dicto or derivative? No, and this for two reasons. The first reason is that our beliefs about what is right, good, etc. are largely dependent on our de re concerns. If I wish to persuade you that you ought to give to Oxfam, I may (if you are a rationalistic philosopher) appeal to your reason, but I am much more likely to tug at your heart-strings by trying to arouse a de re concern for individual starving children. There may be people whose moral beliefs cut across their inclinations, but with most people this is not so – rather their inclinations, humane or otherwise, help determine the content of their moral beliefs. Thus a de dicto desire to do whatever you believe is the right thing is modified in its operation by the de re concerns that influence your beliefs. The second reason is that a derived desire or concern may acquire a certain autonomy and hence cease to be derivative. I originally got into Amnesty International because in the true style of a de dicto moral fetishist, I wanted a good cause to which I could devote my energies. But, having internalized the Amnesty ethos, I now have an autonomous and de re concern for political prisoners, human rights and refugees. This means that the good agent (in something like Smith’s sense) may be less conscientious than most, since when she changes her moral beliefs (by switching from the libertarians to the social democrats) she may have so many de re concerns dating from the days of yore that it takes her a while for her motivations to catch up. It is Smith, who seems to think otherwise, who misrepresents the motivations of the virtuous. (II) Does the externalist credit the virtuous with one thought too many? Frankly, if we are to do the right thing, we need all the thoughts we can get. Suppose as an Amnesty member, I get two Urgent Actions on successive nights. The first is for a Peruvian philosopher who has recently been arrested for her outspoken criticisms of the Fujimori regime. She is a socialist with

90

If Not Non-Cognitivism, Then What?

impeccable liberal and democratic credentials and has written an article in Analysis which I agree with. In every way she is congenial, my heart goes out to her and I sit down to write the letter with a will. The second is for a member of a grisly Muslim sect who has been arrested on suspicion in Egypt. His group believes in clitoridectomies, cutting off the hands of thieves and thinks that democracy and women’s rights are the wiles of Satan designed to tempt God’s people. So far as my de re, non-derivative inclinations are concerned, I would be happy to let him rot. But I reflect that even scumbags like him have rights, that as an Amnesty member, I am not supposed to play favorites. Then, out of a desire to do the right thing, I write the letter anyway. One thought too many? I’d say that even with the second thought, I had only just enough thoughts to get by. But the real problem is with (III). Smith is, of course, quite wrong to suppose that the externalist elevates a fetish (the desire to do the right thing) to the status of the one and only virtue. Conscientiousness is indeed a component of virtue and this requires a fetish in the form of a desire to do the right thing. But virtue requires other things as well, such as sensible views about what the right thing is, and perhaps a dollop of de re concerns. But what Smith really objects to is not the non-existent claim that a fetish forms the whole of virtue but the idea that a fetish forms any part of virtue whatsoever. In so far as conscientiousness is a part of virtue, and in so far as the desire to do the right thing is the basis of conscientiousness, it is indeed a consequence of externalism that a fetish forms a part of virtue. And it is this that Smith finds offensive. Offensive? Why? Well, why does Smith call the desire to do the right thing a “fetish” in the first place? The idea, I think, is this. When we were very young our Mummies and Daddies taught us right from wrong. (That is, they taught us that some things were right and others wrong, though if they were enlightened they usually hinted that these categories were subject to revision.) And they taught us to desire the right and shun the wrong by praising us when we were good and punishing us when we were bad. Yet this infantile compulsion, this fetish foisted upon us when we were too young to do anything about it, has dominated our lives ever since! This silly little fixation, this survival of the kindergarten, is a crucial component of the noblest human characters! The idea is absurd, degrading – in short, it does not bear thinking of! Well, it may be that this idea is offensive and degrading, but that does not give us the slightest reason to suppose that it false.

3.9 A question begged So Premise (1’) is in trouble. But there is worse to come. If we combine the two arguments together – so that the Slavery of Reason Argument provides the second premise for the Motivation Argument C – it soon becomes apparent that the argument as a whole is a flagrantly question-begging affair.

Charles R. Pigden

91

The point is to prove that moral “beliefs” are not genuine beliefs, and thence to argue for some kind of non-cognitivism. But Premise (i) of the Slavery of Reason Argument implicitly assumes that moral “beliefs” are not genuine beliefs, since it divides genuine beliefs into two categories from which moral “beliefs” are excluded. The idea is that moral “beliefs” are not genuine beliefs since they possess a power which genuine beliefs do not, the ability to influence our actions and affections without the aid of a pre-existing desire. But our chief reason for supposing that genuine beliefs are utterly impotent in this particular is that moral ‘beliefs’ (which enjoy the psychological equivalent of Viagra) have been left off the list of genuine beliefs! This is undeniably sharp practice as moral “beliefs” are generally regarded as bona fide beliefs. Hence an argument is required to prove that they are otherwise. The non-cognitivists can hardly disagree since they endeavor to supply such an argument. Unfortunately it assumes what it sets out to prove.

3.10 Desires and dispositions But wait – there’s more! The Slavery of Reason Argument suffers from a hitherto unregarded defect. Crucial to the argument is the transition from (iii) Synthetic beliefs cannot motivate if the causes and effects believed in are alike indifferent to us. to (iv) Hence synthetic beliefs cannot motivate unless they point the way to realize some pre-existing desire. Thesis (iii) seems to me true, but thesis (iv) does not follow. Hume and his followers implicitly assume that if the causes and effects we believe in are not indifferent to us, this must be because they relate to some pre-existing desire. But this assumption is false. Like Michael Smith, I take desires to be propositional attitudes (of which we are not necessarily conscious) involving some representation of the desired state of affairs. They are distinguished from beliefs by their different “directions of fit.” Desires are characterized by their causal roles. Thus a desire for φ is a state which grounds a range of dispositions such as the disposition to seek or promote φ under conditions C, the disposition to desire ψ given the belief that ␺ promotes φ and the disposition to intend φ if φ consists in my performing an action and the performance of that action does not conflict with any of my other desires. But desires are not the only motivational states that interact with beliefs to produce new desires. Crucial to our psychology are what I call dispositions to acquire desires (DTADs for short). These are like desires in some respects, but although they may be

92

If Not Non-Cognitivism, Then What?

attitudes, they are not propositional attitudes, since they are lacking in propositional content. Examples of such dispositions are a taste for chocolate or an amorous disposition. If I like chocolate this does not mean that I want to keep up a certain daily intake of the stuff (since my liking for chocolate takes no account of averages). But when I see and smell chocolate I tend to acquire a desire for it. If I am of an amorous disposition, this does not mean that I have some propositional project – say, to seduce as many women as possible. What it means is that I am unusually prone to sexual desire. Such dispositions are not propositional attitudes (and hence cannot be identified with desires) because it is impossible to specify the proposition to which they are the attitude. If I am of an amorous disposition, there will be states of affairs that I desire to bring about – usually states of sexual congress with this or that person – but my amorous disposition is not my desire for those states but the tendency to acquire those desires. The stimulus that activates such dispositions can be a perception but can also be a newly acquired belief. And this brings us back to the move from (iii) to (iv). Suppose we are sitting on the sofa watching the Simpsons on TV. Suppose too that you inform me of a causal connection, namely that there is beer in the fridge (that is, that if I go to the fridge I can get some beer). Now it is true that if both the cause – going to the fridge – and the effect – getting the beer – were both indifferent to me, this new belief would leave me as calm and indolent as it found me. But if I am not indifferent, this does not mean that I have to have a pre-existing desire for beer. I may simply be such – on this occasion, at any rate – that if I believe there is beer in the fridge, I will acquire a desire for that beer. That is, I may have a (perhaps temporary) disposition to acquire a desire for beer, if informed that there is a cold beer in the offing. Not to be indifferent is not – or need not be – to have a pre-existing desire, though it is to be in some motivational state or other. What underlies the move from (iii) to (iv) in the Slavery of Reason Argument is the thought that it is causally possible for two people to share the same [relevant] beliefs but nevertheless to act differently. If this can happen, it must be because of some causal factor which makes the difference, specifically a difference in their motivational states. Why did Dmitri hang around after that appalling talk? Because he believed that the departmental chairman would be serving free drinks in his office. Why didn’t Dunya hang around given that she too believed in the likelihood of free drinks? Because she was badly hung-over from a vodkasodden evening the night before, and did not want any more alcohol. Since they share the same (relevant) beliefs, the difference in their behaviors must be explained by something else (in this case a desire for free drinks which is present in one and absent in the other). But although we must posit some difference in people’s motivational states to explain why shared beliefs can produce different behaviors in different people, the motivational state that makes the difference need not always be a desire. Sometimes a disposition to acquire a desire will do the trick.2

Charles R. Pigden

93

3.11 Russell, Hobbes and abstract or demonstrative reasoning According to Hume “abstract or demonstrative reasoning never influences any of our actions, but only as it directs our judgment concerning causes and effects.” This claim is false. Consider the following: “At the age of eleven I began Euclid, with my brother as my tutor. This was one of the great events of my life as dazzling as first love. I had not imagined anything so delicious in the world. … From that moment, until Whitehead and I finished Principia Mathematica, when I was thirty-eight, mathematics was my chief interest and my chief source of happiness.” The author of course, is Bertrand Russell and the passage comes from his Autobiography (Russell, 2000, pp. 30–1). Ray Monk, who recounts this episode, together with some details from Frank Russell’s diary, remarks that Russell’s reactions to Euclid seem a little extreme. “‘Dazzling’ and ‘delicious’ are not words one would naturally associate with learning geometry,” he says, but then goes on to point out that Russell’s reactions have an interesting precedent (Monk, 1996, pp. 25–6). When Thomas Hobbes was 40 he chanced upon a copy of Euclid’s Elements which was lying open in a gentleman’s library. It was open at the proof of the famous Pythagorean theorem. “‘By God’, says he, ‘this is impossible!’ So he reads the Demonstration of it, which referred him back to another, which he also read. Et sic deinceps that at last he was demonstratively convinced of its truth. This made him fall in love with geometry” (see Hobbes, 1994, p. lxvii. The story comes from Aubrey’s Brief Lives). Though Russell and Hobbes both valued the geometric information they acquired, what really entranced them was the idea of demonstration. It was clearly the logic that inspired them, the fact that this followed from that. This was abstract and demonstrative reasoning at its most abstract and demonstrative. In both cases the abstract and demonstrative beliefs led to new desires. Being in love, they both sought to know the beloved better and embarked on a course of study. Russell wanted to demonstrate not only the inferences but the axioms while Hobbes acquired a desire to square the circle and to establish a deductive science of politics. But these desires were not desires for the means toward things they already wanted. Their logical and geometric beliefs did not “canalize” pre-existing wants nor did they “influence their actions” only by directing their judgment concerning causes and effects. Thus Hume is doubly mistaken, both about the inertness of reason and about the peculiar inertness of demonstrative reason. Of course, these abstract and demonstrative beliefs did not act quite alone. We must postulate a rather peculiar motivational state, an unusual disposition to acquire new desires, since not everybody reacts to Euclid as Russell and Hobbes did. But we do have two clear instances refuting (2’) and its Smithian variant (2**). This is all the more ironic, because Hobbes anticipated Hume while Russell explicitly agreed with him.3 They both thought that (reason-derived)

94

If Not Non-Cognitivism, Then What?

beliefs could not motivate without the aid of pre-existing desires though in each case their own most formative experiences proved them wrong.

3.12 Thomas Reid and the Vulcan valediction The Humean must suppose that we have a settled desire for long-term survival and satisfaction if he is to escape refutation. Although I do enjoy myself once I am away, I prefer to work rather than go on holiday. Yet I sometimes yield to family and friends and take a break for the sake of my mental and physical health. The belief that a holiday would be good for me prompts me to act. Indeed, it prompts me to overcome my strong preference for hanging around in my study. In order to reconcile this behavior with the theory, the Humean must postulate a “calm desire,” specifically a desire for my long-term welfare. The realization that a holiday would be good for me together with this calm but powerful passion overcomes my reluctance and prompts me to pack my bags and book the tickets. We might call this desire the Vulcan desire, since it corresponds to the Vulcan valediction “Live long and prosper!” Thomas Reid however prefers to talk of a desire for our good on the whole. Now one of Reid’s great achievements was to realize that the Vulcan desire, so necessary to the Slavery of Reason Thesis, is at the same time incompatible with it, at least it is construed as thesis that beliefs cannot give rise to new desires without the aid of a pre-existing desire (as opposed to a DTAD). Why? Well, to begin with it is a desire that presupposes a belief, or rather a family of beliefs. Before you can desire your long-term survival and satisfaction, you must conceive of yourself as a temporally extended being, endowed with desires and susceptible to pleasure and pain. And since the desire presupposes these beliefs, and since we are not born with these beliefs, the desire itself cannot be one we are born with. But neither is it a desire for the means toward something we already want. With one or two exceptions, we do not want to live long and prosper for the sake of something else. It just seems that once we acquire the beliefs, we tend to acquire the desire. Or as Reid puts it, “as soon as we have the conception of what is good or ill for us on the whole, we are led by our constitution to seek the good and avoid the ill.” This desire is the product of reason since it presupposes certain beliefs but it is not arrived at via a means/end calculation. It therefore refutes the Slavery of Reason Thesis in its no-new-desirewithout-a-pre-existing-desire variant (see Reid, 1843, 3.3.2–3, pp. 208–17). And this is all the more damaging because without this desire this version of the Slavery of Reason Thesis would be obviously bankrupt. Whether the disposition to acquire this desire is itself a component of rationality (as Reid seems to suggest) is rather more moot. But this does not affect the point at issue. There is a family of beliefs that motivate by themselves in the relevant sense – that is, they generate a desire which is not for the means toward something we already want. Yet this desire is one

Charles R. Pigden

95

of the most important we are possessed of. Hence, again, the transition from (iii) to (iv) is fallacious and (2’) is false.

3.13 A dead dog? The Motivation Argument (in its Non-Cognitivist Variant) is a two-tiered inference to the best explanation. But inferences to the best explanation fail if the facts to be explained are not genuine facts. Premise (1’) is dubious at best. It is obvious that moral beliefs motivate, less so that they do so without the aid of pre-existing desires. Our parents take considerable pains not only to teach us right from wrong but also to generate a hunger and thirst after righteousness. Sometimes they seem to succeed. But even if (1’) is true that is not enough to save the inference to the best explanation. For the argument to work, there must be a contrast between moral “beliefs” and genuine beliefs. And the claim that genuine beliefs cannot motivate without the aid of a pre-existing desire – the non-cognitivist variant of the Slavery of Reason Thesis – has been shown to be false. The argument for the Slavery of Reason Thesis has been shown to be fallacious, since the key move – from (iii) that synthetic beliefs cannot motivate if the causes and effects believed in are indifferent to us, to (iv) that synthetic beliefs cannot motivate unless they point the way to realize some pre-existing desire – ignores the possibility of dispositions to acquire desires. The examples of Hobbes and Russell have shown that abstract and demonstrative reasoning can influence our actions and desires even when it does not direct our judgment concerning causes and effects. Finally the entire argument has proved to be flagrantly question-begging, since the first premise of the Slavery of Reason Argument implicitly assumes that moral “beliefs” are not genuine beliefs, as it divides genuine beliefs into two categories from which moral “beliefs” are excluded. So far from being a mighty dragon in search of a slayer, the Motivation Argument for non-cognitivism is a dead dog in search of a decent burial. The only consolation for Humeans is that it is not a dead dog that can be laid at David’s door.

3.14 If not non-cognitivism, then what? The non-cognitivist interpretation of Hume faces a further problem. When it comes to what is nowadays known as “meta-ethics,” Hume was regarded both in his own day and for the next 200 years (roughly 1740–1940) as a dissident disciple of Frances Hutcheson, and Hutcheson was a moral realist in the Sayre-McCord sense. He believed both that moral judgments are “truth-apt,” true-or-false, and that some such judgments are (literally) true; and true, what’s more, with respect to their distinctively moral contents. He was, therefore, neither a non-cognitivist nor an error-theorist. But he was not a moral realist in the strong sense of the word since he did not believe the moral

96

If Not Non-Cognitivism, Then What?

facts to be mind-independent. Moral properties for Hutcheson were akin to secondary qualities, and thus in today’s jargon, “response-dependent.” To say that an action is right or virtuous is to say (very roughly) that it tends to arouse the approbation of a suitably qualified human spectator, just as to say that a rose is red is to say that it has a tendency to trigger sensations of a certain kind in a suitably qualified human observer. Hume agreed with him on this, a point which was generally acknowledged. And since Hutcheson was not a non-cognitivist, neither was Hume. To suppose otherwise is to suppose that virtually all Hume’s contemporaries, many of them quite astute, got him quite horribly wrong, leaving it to the likes of A. J. Ayer (not a man noted for his historical scholarship) to get him right (see Ayer, 1992; Dunn et al., 1992, p. 226). Indeed, it follows on this interpretation not only that other people got Hume wrong, but that Hume misunderstood himself. Here is Hume in a paragraph deleted from later editions of the EHU: “A late Philosopher [note: Mr Hutcheson] has taught us, by the most convincing Arguments that Morality is nothing in the abstract Nature of things but is entirely relative to the Sentiment or mental Taste of each particular Being [that is each kind of being] in the same Manner as the Distinction between sweet and bitter, hot and cold, arise from the particular feeling of each Sense or Organ. Moral perception, therefore, ought not to be class’d with the Operations of the Understanding but with the Tastes or Sentiments.” (Quoted in Kemp Smith, 1941, p. 19. See also Beauchamp’s critical edition of the ECU, p. 232.) Moral properties, as Hutcheson has taught us, are relative to a mental taste, they are secondary qualities like hot and cold. Though hot and cold may be in the mind of the perceiver, the expressions, “This fire is hot,” “This hand is cold,” “This rose is red,” are not non-cognitive expressions. But if rightness is like redness, why should we suppose that truthaptness gives out when we move from “This rose is red” to “This action is right”? If Hume is not a non-cognitivist about the one (as surely he is not), why should we suppose that he is a non-cognitivist about the other?

3.15 Motivation, reason and slavery Perhaps because is hard to construe the Motivation Argument as an argument for anything except non-cognitivism. It rests on a contrast between morals which motivate and reason which does not (or which can only motivate with the aid of a pre-existing passion). And if reason is construed as our belief-forming faculty, this strongly suggests that moral “beliefs” – the opinions of injustice that deter some and the opinions of obligation that impel others – are not genuine beliefs. But suppose that reason is not our belief-forming faculty but only one of the mechanisms by which we acquire beliefs. Then the conclusion of the Slavery of Reason Argument is not that genuine beliefs do not motivate by themselves but that beliefs derived from reason do not motivate themselves. In other words what we get is this:

Charles R. Pigden

97

Slavery of Reason Argument Mark 2: Sentimentalist Version (i) Beliefs derived from reason are either analytic (to do with the relations of ideas) or synthetic (to do with causal relations). (ii) Analytic beliefs cannot motivate except in so far as they lead to synthetic beliefs. (iii) Synthetic beliefs cannot motivate if the causes and effects believed in are indifferent to us. (iv) Hence synthetic beliefs cannot motivate unless they point the way to realize some pre-existing desire (from iii). (v) Beliefs derived from reason cannot motivate by themselves (from (i), (ii) & (iv)). This new interpretation of the Slavery of Reason Argument demands a new interpretation of the Motivation Argument since the conclusion of the one is a premise of the other. We retain Premise (1’) – that moral beliefs alone often motivate – but modify the second premise once again. This is no longer thesis (2’) – that genuine beliefs cannot motivate by themselves. It now becomes (2”) – that beliefs derived from reason cannot motivate by themselves. This leaves open the possibility that there are some beliefs not derived from reason which can motivate without the aid of a pre-existing passion. What sort of beliefs might they be? Perhaps beliefs based on impressions. Hume certainly believed in such beliefs. “To believe is in this case to feel an immediate impression of the senses, or a repetition of that impression in the memory” (T, 1.3.5.8/86). Thus my belief that there is a red book in front of me (in fact my rebound copy of the Treatise) is not much more than a vivid impression of a red book. A little later Hume gives a different definition: “An opinion, therefore, or belief, may be most accurately defined [as] a lively idea related to or associated with a present impression” (T, 1.3.7.5/96). Here the belief is not an impression but an idea associated with an impression.4 But if some beliefs are impressions and others are ideas associated with impressions, not all our beliefs are the products of an inferential process.5 Obviously the kind of belief that consists in “feel[ing] an immediate impression of the senses” is not the kind of thing that needs to be inferred. If my belief that there is a red book in front of me simply consists of my impression of a red book then (if there is in fact a red book in front of me) all I need to do to acquire the relevant belief is to open my eyes and see the book. And if some beliefs are “lively ideas related to or associated with a present impressions” then some of these beliefs can be acquired without the aid of inference. I open my eyes and form the belief that there is a computer screen in front of me. I tuck into my breakfast of toast and mussels and form the belief that it is delicious. Or I contemplate my wife Zena in her new red dress and form the belief that she is sexy. In the first case it is my visual impressions that do the trick, in the second my oral impressions and in the third case … well, it’s a complicated business, but you get the general idea. Now in all these

98

If Not Non-Cognitivism, Then What?

cases inference is minimal to nonexistent. I do not deduce or infer that my dinner is delicious or my wife sexy. I feel them to be so in consequence of my sensations. An alien anthropologist might infer that somebody is (humanly) sexy by consulting a checklist or a set of INUS conditions and reasoning her way to an evaluation, but that is not how human beings typically do it. Or at least, that is not how Hume thinks that human beings typically do it. Now, Hume’s conception of reason undergoes a metamorphosis in the course of the Treatise. It starts off as deduction (and perhaps the perception of analytic truths) and gradually expands to include substantial dollops of custom, habit and imagination, those “sensitive parts of our natures” which make up for the defects in our purely cognitive capacities. But “reason” still carries connotations of discursus, of running around from the premises of an argument to the conclusion, though, by 3.1.1, reason runs with the aid of custom, habit and imagination. In other words Hume tends to conceive of reason as our inferential organ. Thus part of what Hume is denying in 3.1.1 is that the basic principles of morals can be inferred or deduced from anything else. Moral distinctions, he insists, are not derived from reason and are not “discovered … by a deduction of reason” (T, 3.1.1.7/457). They are not “conclusions of our reason” (T, 3.1.1.7/457) and are not “matters of fact, whose existence we can infer by reason” (T, 3.1.1.27/468). Furthermore, Hume seems to think that when reason does its running around, it runs from one idea (or set of ideas) to another. The topic of “our present enquiry concerning morals,” he says, is “whether it is by means of our ideas or impressions we distinguish betwixt vice and virtue” since “those who affirm that virtue is nothing but a conformity to reason … concur in the opinion, that morality, like truth, is discerned merely by ideas, and by their juxtaposition and comparison” (T, 3.1.1.4–5/456–7). Thus Hume tends to think of reason as an inferential capacity that operates on ideas. That leaves plenty of room for beliefs not derived from reason, namely those based on impressions (or identical with impressions) which are not the products of inference. Moreover the idea that such beliefs are motivating has got a lot going for it. It is at least plausible to suppose that my belief that my breakfast is delicious will motivate me to go on eating, while Dunya’s belief that Dmitri is sexy might motivate her to give him the eye. So the conclusion of the Slavery of Reason Argument should not be (as is commonly assumed) that genuine beliefs do not motivate by themselves but that beliefs derived from reason do not motivate by themselves, leaving the way open for beliefs not derived from reason (but based on impressions) that can sometimes motivate by themselves or, at least, that can motivate without the aid of a pre-existing desire. But if we modify the Slavery of Reason Argument we must modify the Motivation Argument too, since the one supplies one of the premises of the other. If we replace (2’) with (2’’) what we get is the following: Motivation Argument: Sentimentalist Version Mark 1 (1’) Moral beliefs alone often motivate.

Charles R. Pigden

99

(2’’) Beliefs derived from reason cannot motivate by themselves. (3a’) The best explanation of (1’) and (2’’) is that moral “beliefs” are not derived from reason. Therefore probably (3’’) Moral beliefs are not derived from reason. [The rules of morality, therefore, are not conclusions of our reason.] (3’’) is precious close to the claim that Hume says he is arguing for in 3.1.1, namely that “moral distinctions are not derived from reason.” It neither entails nor suggests non-cognitivism, and is quite consistent with the view that moral properties are akin to secondary qualities. Furthermore if we add a premise that Hume seems to have believed, we get the core thesis of 3.1.2. Here is the premise: (4’’) All our beliefs are either based on ideas (reason/thinking) or on impressions (sensation/feeling). According to Hume, all our perceptions, that is all our mental contents, can be divided into these two categories, impressions and ideas. Impressions are “sensations, passions, and emotions, as they make their first appearance in the soul,” whereas ideas are the “faint images of these in thinking and reasoning” (T, 1.1.1.1/1). The difference between impressions and ideas is the “difference betwixt feeling and thinking.” But if all our mental contents are ideas or impressions, then all our beliefs that are not based upon (or identical with) impressions must be arrived at by reason, either by intuitive perception as with an analytic truth or by some inferential process. Thus the chief question with respect to our moral beliefs is whether it is by means of our ideas or impressions we distinguish betwixt vice and virtue, and pronounce an action blamable or praiseworthy – that is whether our moral beliefs are derived from an inferential process or based upon feeling or sentiment. And given (3’’) and (4’’), the answer is clear: (5’) Moral beliefs, therefore, are based on impressions [specifically the sentiments of approbation and disapprobation]. And this is precisely the conclusion that Hume announces at the outset of 3.1.2, “that since vice and virtue are not discoverable merely by reason, or the comparison of ideas, it must be by means of some impression or sentiment they occasion. … Our decisions concerning moral rectitude and depravity are evidently perceptions; and as all perceptions are either impressions or ideas, the exclusion of the one is a convincing argument for the other. Morality therefore, is more properly felt than judged of” (T, 3.3.2.1/470). This suggests (though it does not entail) that moral properties are akin to secondary

100

If Not Non-Cognitivism, Then What?

qualities. For Hume, the belief that an action is right is caused in much the same way as the belief that the book is red. I contemplate the book and, if my sense of sight is not disturbed by disco-lights, blue-tinted spectacles or other distorting influences, my visual impressions enable me to see that it is red. For its redness simply consists in its tendency to produce such impressions in suitably qualified spectators, where being suitably qualified involves such things as not wearing blue-tinted spectacles. I contemplate an action and, if my moral sense is not distorted by partial passions or false opinions (such as “the delusive glosses of superstition and false religion,” EPM, 9.1.3/270), my impression of approbation enables me to feel that it is right. For its rightness simply consists in its tendency to arouse a sentiment of approbation in suitably qualified spectators, where being suitably qualified includes not having relevantly false opinions (such as the delusive glosses) and not being subject to partial passions. Thus the conclusion we have arrived at is very close to the conclusion that Hume says he is arguing for, namely that “vice and virtue … may be compared to sounds, colours, heat, and cold,” that is to secondary qualities. We have managed to recast both the Slavery of Reason and the Motivation Arguments so that they support the quasi-Hutchesonian doctrine that Hume probably believed rather than the non-cognitivism that has been read into him by his Twentieth Century admirers.

3.16

An objection

But isn’t the position I attribute to Hume rather an odd one? The thesis is that beliefs based on impressions sometimes motivate without the aid of a pre-existing desire or passion, but that beliefs based on ideas or reasoning cannot. But the Slavery of Reason Arguments in 2.3.3 seem to be pretty general. The conclusion they suggest (as opposed to the conclusion that they actually support) is that beliefs cannot motivate without the aid of a preexisting desire or passion, not that beliefs derived from reason cannot motivate without the aid of a pre-existing desire or passion. To put the point another way, when people become persuaded that beliefs derived from reason cannot motivate by themselves, the reason that they are inclined to think this is that beliefs derived from reason are beliefs not that beliefs derived from reason are derived from reason. The provenance of the beliefs does not enter into it. Indeed, it is perhaps worth pointing out that Hume’s predecessors – such as Hobbes and Hutcheson – seem to have subscribed to the general thesis that beliefs cannot motivate by themselves, not the restricted thesis that beliefs derived from reason cannot motivate by themselves. And so too do latter-day Humeans such as Michael Smith. Why (if I am right) does Hume differentiate between reason-based and impression-based beliefs when his predecessors and disciples do not? Let’s get the thesis clear. I have attributed to Hume the view that some impression-based beliefs influence our actions and affections without the

Charles R. Pigden

101

aid of a pre-existing desire. The claim is not that they always do so or that they do so necessarily. What I am going to suggest is that when impressions prompt a belief which does cause a new desire or passion, it is not really the belief that is doing the business, but the painful or pleasurable impressions we feel in formulating the belief. Suppose I bite into my toast and mussels and form the belief that they are delicious. This prompts the desire to take another bite. Conversely, if I bit into them and formed the belief that the mussels were off this would prompt the desire to spit them out. If Dunya contemplates Dmitri in his new tuxedo and forms the belief that he is sexy this might prompt an amorous advance. Conversely, if her impressions led her to form the belief that he had a bad case of BO, this might well “oppose or retard the impulse of passion.” But in all four cases, it is not really the belief that is prompting the desire but the impressions that we feel when formulating the belief. It is the pleasant taste of the mussels that prompts me to eat more, not the belief that they taste pleasant. It is the unpleasant taste that prompts me to spit the mussels out, not the belief that they taste unpleasant. It is the sexual feelings that Dmitri arouses in Dunya that cause her desire to seduce him, not the belief that he is sexy (though this belief might have such an influence if conjoined with some desires such as a Merteuil-like project of seducing as many attractive men as possible). It is Dmitri’s passion-killing body odor that causes her to back off, not the belief that he has it. Hume’s thesis, I suggest, is that our moral motivations are sometimes due to the same kind of process. Morals have an influence on the actions and affections and they can exercise this influence without the aid of pre-existing desires. But it is not strictly speaking our beliefs that have this influence, but the sentiments we feel when we form those beliefs. It is the approval I feel when I contemplate some generous action that moves me to emulate the doer, not the approval-based belief that he was doing a good deed. The claim that we are “governed by our duties” is correct but the thesis that we are “deter’d from some actions by the opinion of injustice” is only true as a first approximation. It is not, strictly speaking, the opinions of injustice and obligation that deter or impel us but the sentiments we feel when we formulate those opinions. When it comes to morality, we are moved by our impressions not the copies of those impressions that figure in our beliefs.

3.17 The Motivation Argument reformulated But if this is correct, the Motivation Argument needs to be modified yet again. The second Premise (2’’) – the conclusion of the Slavery of Reason Argument – can be left as it stands, but the first premise needs to be reformulated. Morals may have “an influence on the actions and affections,” but when morality exerts its influence perhaps it is not our moral beliefs that do the motivating so much as the impressions on which they are based.

102

If Not Non-Cognitivism, Then What?

But there is a difficulty here. For Hume construes some beliefs as impressions, since “to believe is [in some cases] to feel an immediate impression of the senses” (T,1.3.5.8/86). Perhaps he is confounding moral beliefs with immediate impressions of the moral sense in the Motivation Argument (a confusion suggested by some of his rhetoric)? It seems to me that we can get around this problem with the aid of a studied ambiguity. I define “moral considerations” as the mental contents that move us when we are motivated by morality. Perhaps they are impressions (namely the sentiments of approbation and disapprobation), perhaps they are beliefs based upon such impressions or perhaps they are beliefs confounded with impressions. Never mind, we simply leave it open. Then the first premise of the Motivation Argument becomes the thesis that moral considerations by themselves sometimes motivate. This gives us: Motivation Argument: Sentimentalist Version Mark 2 (1’’) Moral considerations alone often motivate. (2’’) Beliefs derived from reason cannot motivate by themselves. (3a’’) The best explanation of (1’’) and (2’’) is that moral considerations are not beliefs derived from reason. Therefore probably (3’’’) Moral considerations are not beliefs derived from reason. (3b) Moral beliefs either are, or are based upon, moral considerations. (3c) Moral beliefs are not beliefs derived from reason. [Moral beliefs are not beliefs derived from reason if they are identical with moral considerations and they are not beliefs derived from reason if they are based upon items – namely moral considerations –– that are not beliefs derived from reason.] (4’’) All our beliefs are either based on ideas (reason/thinking) or on impressions (sensation/feeling). (5’) Moral beliefs, therefore, are based on impressions [specifically the sentiments of approbation and disapprobation]. Once again this gives us the conclusion that Hume says he is arguing for, namely that moral distinctions are not deriv’d from reason (T, 3.1.1.1/455) but that vice and virtue are discoverable by means of some impression or sentiment that they occasion (T, 3.1.2.1/470). But the argument is no longer committed to the claim that either moral beliefs or beliefs-not-derived-from-reason can motivate without the aid of a pre-existing desire. When morals excite passions, or produce and prevent actions, maybe it is moral beliefs that do the motivating, maybe it is the impressions or sentiments on which they are based or maybe it is the sentiments which Hume has oddly confounded with beliefs. The argument is agnostic between these three alternatives.

Charles R. Pigden

103

3.18 An argument undone In formulating (what I take to be) Hume’s argument, I have reversed the historical sequence. The non-cognitivist argument was derived from Hume’s via a series of misunderstandings. I have derived Hume’s argument from its non-cognitivist descendant via a process of rational reconstruction. Unfortunately Hume’s argument shares many of the defects of the noncognitivist argument from which it is derived. Like the non-cognitivist argument, it is formally respectable, though not formally valid, since the move from (3a’’) to (3’’’) is an inference to the best explanation. However an inference to the best explanation is only as good as the “facts” that the conclusion purports to explain. And Premises (1’’) and (2’’), the supposed facts that subconclusion (3’’’) purports to explain, are not good enough to sustain the inference. Premise (1’’) – that moral considerations alone often motivate – is a less specific variant of Premise (1’) – that moral beliefs alone often motivate. And as we have seen (§§7–8), when moral considerations (the mental contents that move us when we are motivated by morality) motivate, they often do it with the aid of a hunger and thirst after de dicto righteousness. Most of us want to do the right thing (or at least to avoid actions that are conspicuously wrong), a desire carefully fostered by conscientious parents. But if Premise (1’’) is in a bad way, Premise (2’’) is worse. To begin with it is derived from the Slavery of Reason Argument, which is just as questionbegging in its sentimentalist incarnation as was its non-cognitivist variant. Hume arrives at sub-conclusion (v) – that beliefs derived from reason cannot motivate by themselves – by assuming, without argument, that moral considerations are not beliefs-derived-from-reason. But the whole point of the two arguments taken together – the Slavery of Reason Argument which provides one of the premises for the Motivation Argument – was to prove precisely that! Yet the Slavery of Reason Argument only works (in so far as it works at all) by dividing beliefs-derived-from-reason into two categories, relations of ideas and matters of fact, thereby excluding moral beliefs as not derived from reason. Premise (ii) – that abstract or demonstrative reasoning never influences any of our actions, but only as it directs our judgment concerning causes and effects – is falsified by the examples of Russell and Hobbes who were mightily motivated by abstract or demonstrative reasoning even though that reasoning had no direct causal consequences. The move from (iii) – synthetic beliefs cannot motivate if the causes and effects believed in are indifferent to us – to (iv) – synthetic beliefs cannot motivate unless they point the way to realize some pre-existing desire – is fallacious since it ignores the possibility of DTADs. Indeed, once we recognize the reality of DTADs, it becomes apparent that Premise (2’’) is simply false. Beliefs of all sorts can motivate without the aid of a pre-existing desire so long as they mesh with a DTAD, a disposition to acquire desires, such as curiosity, amorousness or a fondness for chocolate. (The Slavery of Reason thesis can

104

If Not Non-Cognitivism, Then What?

perhaps be saved but only if it is reformulated as the claim that no belief can give rise to a new desire or action without the aid of a pre-existing desire or DTAD.) As we have noted, the Motivation Argument, in its sentimentalist as in its non-cognitivist variants, is best construed as an inference to the best explanation. But one of the “facts” that the inference purports to explain is distinctly dubious while the other – Premise (2’’) – is no fact at all. And there is no reason to accept the explanation of a non-fact. The Motivation Argument in its sentimentalist version is a complete write-off. We began with Michael Smith’s thesis that Hume’s Motivation Argument mounts a powerful case for non-cognitivism. But it is not really an argument for non-cognitivism and in so far as it can be converted into one, it is not at all powerful. But if it was not an argument for non-cognitivism what was it an argument for? The thesis that moral distinctions are not derived from reason, but that vice and virtue are analogous to “sounds, colours, heat and cold,” and hence more properly felt than judged of. But so far from proving Hume’s point the Motivation Argument has proved to be, in his own words, “utterly impotent in this particular.”

Notes 1. Smith, unwisely in my opinion, is rather less cautious, at least about judgments about what is right for the agent to do. See §2 and MP, pp. 60–91. 2. For more on DTAD’s and the deficiencies of Hume’s argument, see Pigden, 2007. 3. For Hobbes “the thoughts are to desires as scouts and spies, to range abroad, and find the way to the things desired” which strongly suggests that reason as the thinking faculty is the servant, if not the slave, of the passions (Leviathan, Chapter viii. 16). As Darwall puts it “Reason [for Hobbes] can recommend no conduct or end directly or intrinsically. Its practical function is purely instrumental, to work out the means or ‘way to the things desired’” (Darwall, 1995, p. 59; see also p. 60). Thus Hobbes’s psychology does not allow for thoughts which can generate fundamentally new desires such as his own thoughts on the subject of geometry. According to Russell, “‘Reason’ has a perfectly clear and precise meaning. It signifies the choice of the right means to an end that you wish to achieve. It has nothing whatever to do with the choice of ends” (Russell, 1999, p. 170). Thus for Russell, reason cannot produce beliefs which lead to desires for new ends, such as his own beliefs about the demonstrations of Euclid. 4. The oddities of Hume’s theory of belief – and hence the inadequacy of my analysis – were forcefully pointed out to me by Annette Baier. 5. Hume’s theory of belief is, of course, utterly silly and a big step backwards from that of the “scholastick headpieces” he so arrogantly decries. For a critical discussion see Bennett, 2001, Chapters 32–3, though he is by no means critical enough.

4 The Motivation Argument for Non-Cognitivism Michael Smith

[M]orality is … supposed to influence our passions and actions. … And this is confirm’d by common experience, which informs us, that men are often govern’d by their duties, and are deter’d from some actions by the opinion of injustice, and impell’d to others by that of obligation. Since morals, therefore, have an influence on the actions and affections, it follows, that they cannot be deriv’d from reason; and that because reason alone … can never have any such influence. Morals excite passions, and produce or prevent actions. Reason of itself is utterly impotent in this particular. The rules of morality, therefore, are not conclusions of our reason. … An active principle can never be founded on an inactive; and if reason be inactive in itself, it must remain so in all its shapes and appearances. (T, 3.1.1.5–7/457) This passage from Hume’s Treatise is often cited as the inspiration for the Motivation Argument for non-cognitivism, an argument that can be stated more explicitly as follows: (1)

(2) (3)

So: (4)

Actions are the product of agents’ desires about the ways things are to be and their beliefs about what they can do to make things those ways. People generally act in accordance with their moral views. The fact that people generally act in accordance with their moral views would be hard to explain if having a moral view were simply a matter of having beliefs about the ways things are, easy to explain if having a moral view were a matter of having desires about the ways things are to be. Having a moral view is a matter of having desires. 105

106 The Motivation Argument for Non-Cognitivism

Much needs to be said to clarify both this argument and its relationship to Hume’s. The basic ideas should, however, be clear enough. Premise (1) is the standard Humean view about the psychological springs of action (T, 2.3.3/413–15); premise (2) expresses a commitment to internalism, a doctrine Hume seems to endorse explicitly in the passage quoted; and premise (3) expresses skepticism about the prospects for explaining the acquisition of the desires that we need to posit in order to explain people’s actions in accordance with their moral views unless having a moral view is a matter of having desires, skepticism of the sort that Hume expresses not only in the passage quoted, but also and more famously in the “is-ought” passage (T, 3.1.1.27/469–70). However, whether Hume would recognize and accept the Motivation Argument is a matter of considerable debate, as too is the cogency of the argument itself. My aim in what follows is to address both these issues. More specifically, after some preliminary clarifications (Section 4.1), I will focus on the reasons provided by the authors of two essays in this volume – Norva Lo in her “Is Hume Inconsistent? – Motivation and Morals” (hereafter IHI) and Charles Pigden in his “If Not Non-Cognitivism, Then What?” (hereafter INNC) – for supposing that the argument, at least as stated, isn’t one that Hume could accept (Section 4.2). I will then consider Pigden’s reasons for thinking that the argument is, in any case, a dismal failure (Section 4.3). Though Pigden has some independent criticisms of arguments that I have put forward, I will ignore these in what follows. The focus is to be on the Motivation Argument itself and its relationship to Hume’s.

4.1

Clarifications

Let me begin by identifying what seem to me to be the crucial features of the Motivation Argument. It is important that we do this because whether the argument is one that Hume would recognize or accept depends on the specific claims it makes. The most important question to address in this connection is what kind of internalism figures as a premise in the argument and whether Hume really is an internalist in that sense. Of course, Hume does say that “men are often govern’d by their duties, and are deter’d from some actions by the opinion of injustice,” and he also says that “morals … have an influence on the actions and affections.” But as Norva Lo points out, there is equivocation here. He begins by saying that men are motivated by “their duties,” and then immediately switches tack, telling us they are motivated by “the opinion of injustice.” When he goes on to say that “morals … have an influence on the actions and affections” (my emphasis), the “morals” in question thus seem to wobble between moral views (“opinions” about duties) and moral facts (“duties” themselves). In the Motivation Argument as formulated, by contrast, (2) posits a quite specific link between moral views – that is, opinions about duties, not duties themselves – and motivations.

Michael Smith

107

Ascribing the Motivation Argument to Hume thus requires us to interpret him as being what Steve Darwall calls a “judgment” internalist, as opposed to an “existence” internalist (Darwall, 1983, 54–5)). But in so interpreting him we focus on one of the things he says at the expense of the other. Another source of difficulty in interpreting both Hume’s argument and the Motivation Argument is that, insofar Hume formulates internalism at all, he formulates it as an empirical claim. To repeat, he tells us that “morality is … supposed to influence our passions,” something which is “confirm’d by common experience, which informs us, that men are often govern’d … by the opinion of injustice.” Premise (2) of the Motivation Argument is similarly empirical in nature. It tells us that people generally act in accordance with their moral views. But why only “often” or generally? Why not always? And, much more importantly, what underlying conceptual truth is this empirical truth supposed to convey in both these arguments? As I understand it, the “generally” in (2) – Hume’s “often” – is there to acknowledge the fact that, from time to time, people do fail to act in accordance with their moral views, as for example when they suffer from weakness of will, or compulsion, or some similar form of practical irrationality (Smith, 1988). This too, we might say, is “confirm’d by common experience.” What (2) says is thus that, though people are not always practically rational, they are practically rational generally speaking, and, moreover, that whenever they are practically rational, they act in accordance with their moral views. This is the underlying conceptual truth (2) is supposed to convey. But if this is right then the empirical content of (2) turns out to be irrelevant. The same may be true with regard to the empirical content of Hume’s argument. Can we restate the Motivation Argument in such a way as to remove the irrelevant empirical content? We can if we replace (2) with: (2*) People act in accordance with their moral views, insofar as they are practically rational. and replace (3) with: (3*) The fact that people act in accordance with their moral views, insofar as they are practically rational, would be hard to explain if having a moral view were simply a matter of having beliefs about the ways things are, easy to explain if having a moral view were a matter of having desires about the ways things are to be. The non-cognitivist conclusion, (4), follows just as readily from (2*) and (3*), if it followed from (2) and (3). The general question with which we began was whether Hume is properly interpreted as putting forward the Motivation Argument. The difficulty facing that interpretation should now be readily apparent. We would have to suppose that, according Hume, it is a conceptual truth that people are motivated in accordance with their moral views, insofar as they are practically rational. The trouble is, however, that we find no mention of the fact that the connection between our moral views and our actions can be

108 The Motivation Argument for Non-Cognitivism

defeated by practical irrationality when Hume says things that make him sound like a judgment internalist. Ascribing the Motivation Argument to him – supposing that he accepts (2*) – thus requires us to go beyond the text. Moreover, and much more worryingly, as Elijah Millgram (1995) and Christine Korsgaard (1986) have pointed out, it isn’t at all clear that Hume even has the resources to make sense of the idea of someone’s being practically rational. For insofar as he has a conception of norms of rationality at all, these norms simply aren’t the sorts of thing that could govern the desires that motivate us. Ascribing the Motivation Argument to him may thus require us not just to go beyond the text, but to ascribe to him concepts that he is incapable of making sense of. There are therefore serious difficulties associated with interpreting Hume as putting forward the Motivation Argument in the passage quoted at the outset. But there is, of course, something weaker that we might say about the relationship between what Hume says and the Motivation Argument. We might say that the Motivation Argument is a rational reconstruction of Hume’s argument. To be sure, in coming up with this rational reconstruction we may have to focus on some of the things he says as opposed to others, and we may have to ascribe to him concepts that he doesn’t have the resources to make sense of, but our doing so will be forgivable if it allows us to make the best overall sense of what he says. My own view is that this is indeed the case: the Motivation Argument is the best rational reconstruction of the argument Hume puts forward in the passage quoted at the outset. Though it would be nice if we could make a stronger claim about the relationship between the two arguments, doing so would invite too much by way of controversy. According to Norva Lo and Charles Pidgen, however, even this weaker claim about the relationship between the Motivation Argument and Hume’s own goes too far. They tell us that the Motivation Argument couldn’t possibly be a rational reconstruction of anything Hume says because he wouldn’t have dreamt of arguing for non-cognitivism. As they see things, Hume was plainly a cognitivist, not a non-cognitivist. In what follows I will begin by saying a little about this interpretative issue. To anticipate, it seems to me that both Lo’s and Pigden’s reasons for insisting that Hume is a cognitivist are inconclusive. Though Hume does say things that suggest a cognitivist interpretation, on closer examination it turns out that what he has to say is consistent with non-cognitivism (Section 4.2). Assuming that Hume is a non-cognitivist, I will then turn to consider briefly the plausibility of the Motivation Argument itself. The argument is abductive, rather than deductive, so if someone could come up with a plausible explanation of the connection between moral views and action without supposing that having a moral view is a matter of having desires, it would lose its appeal altogether. This is Charles Pigden’s strategy. However it seems to me that what Pigden manages to explain is something less than what needs explaining. If I were forced to choose between Pigden’s

Michael Smith

109

cognitivist explanation of the connection between having a moral view and motivation and the Humean non-cognitivist explanation, I would choose the Humean explanation (Section 4.3).

4.2 Is Hume a cognitivist or a non-cognitivist? As I said, Norva Lo and Charles Pigden insist that Hume is a cognitivist, rather than a non-cognitivist. They therefore think that he wouldn’t dream of putting forward the Motivation Argument. Before we consider the arguments they provide for this conclusion, however, it is important that we first settle what it means to say that someone is a cognitivist rather than a non-cognitivist. The names “cognitivism” and “non-cognitivism” suggest, somewhat misleadingly, that cognitivsts hold that having a moral view is a matter of having a set of moral beliefs and that non-cognitivists hold that having a moral view is not a matter of having a set of moral beliefs. This cannot be right, however, as most non-cognitivists have been quite happy to admit that there are moral beliefs (see, for example, Blackburn, 1998b). So what exactly is the difference between cognitivism and non-cognitivism? It seems to me that non-cognitivists are best understood as subscribing to an explanatory thesis (Dreier, 2005). They hold that we can explain what it is for a subject to have moral beliefs in terms of his non-belief states – typically, his desires – and that we cannot tell a similar explanatory story about what it is for a subject to have non-moral beliefs. The second, negative point is important, otherwise there would be no rationale for the “non-” in non-cognitivism. Non-cognitivists are thus best understood as insisting not that there are no moral beliefs, but rather that beliefs fall into two quite distinct categories. On the one hand, there are moral beliefs (or perhaps better, normative beliefs more generally). These are constituted by desires. On the other hand, there are non-moral beliefs (or, perhaps better, non-normative beliefs more generally). These beliefs are not constituted by desires. Cognitivists deny that we can explain an agent’s moral beliefs in terms of constitution by his desires. There is thus no easy move from the premise that someone thinks that there are moral beliefs to the conclusion that she is a cognitivist. This will be important in what follows. With this understanding of the terms “cognitivism” and “non-cognitivism” firmly before us, let’s now turn to consider Lo’s and Pigden’s reasons for thinking that Hume is a cognitivist, rather than a non-cognitivist. Pigden makes a number of different points. One is that, up until recently, Hume was regarded as a “dissident disciple” of Frances Hutcheson, himself a cognitivist. When it comes to what is nowadays known as “meta-ethics”, Hume was regarded both in his own day and for the next two hundred years (roughly

110 The Motivation Argument for Non-Cognitivism

1740–1940) as a dissident disciple of Frances Hutcheson, and Hutcheson was a moral realist in the Sayre-McCord sense. He believed both that moral judgments are “truth-apt”, true-or-false, and that some such judgments are (literally) true; and true, what’s more, with respect to their distinctively moral contents. He was, therefore, neither a non-cognitivist nor an error-theorist. But he was not a moral realist in the strong sense of the word since he did not believe the moral facts to be mindindependent. Moral properties for Hutcheson were akin to secondary qualities, and thus in today’s jargon, “response-dependent”. To say that an action is right or virtuous is to say (very roughly) that it tends to arouse the approbation of a suitably qualified human spectator, just as to say that a rose is red is to say that it has a tendency to trigger sensations of a certain kind in a suitably qualified human observer. Hume agreed with him on this, a point which was generally acknowledged. And since Hutcheson was not a non-cognitivist, neither was Hume. (INNC, § 14) I will have more to say about the relationship between Hume’s views and the view Hutcheson held presently, for Norva Lo also reminds us of this aspect of Hume’s thinking. In fact, however, by Pigden’s own account of the matter, Hume’s analysis of moral belief is far more complicated than this would suggest. He tells us that Hume distinguished between two kinds of beliefs. On the one hand, there are beliefs that we can arrive at by inference from other things we believe – Pigden calls these “product-of-reason beliefs” – and, on the other, there are beliefs that we aren’t able to arrive at in this way. Perceptual beliefs are supposed to be a prime example of the latter kind, as Pigden tells us that Hume held perceptual beliefs to be constituted by the very impressions that are caused in us by the objects that we perceive. Perceptual beliefs thus turn out not to be product-of-reason beliefs because impressions are not something that we arrive at by inference. We typically enjoy them simply by having them caused in us by the objects of perception. This distinction between two kinds of beliefs casts a different light on Hume’s claim that our moral views are not the product of reason, says Pigden, because it opens up the possibility that he wasn’t so much concerned to deny that our moral views are a matter of belief, as that they are beliefs that we can arrive at by inference: The topic of “our present enquiry concerning morals”, [Hume] says, is “whether it is by means of our ideas or impressions we distinguish betwixt vice and virtue” since “those who affirm that virtue is nothing but a conformity to reason … concur in the opinion, that morality, like truth, is discerned merely by ideas, and by their juxtaposition and comparison” (T, 3.1.1.4–5/456–7). Thus Hume tends to think of reason as an inferential

Michael Smith

111

capacity that operates on ideas. That leaves plenty of room for beliefs not derived from reason, namely those based on impressions (or identical with impressions) which are not the products of inference. (INNC, § 15) Pigden’s suggestion is thus that Hume holds that moral beliefs, and indeed normative beliefs more generally, are not product-of-reason beliefs but are rather identical with “impressions,” much like perceptual beliefs. Pigden’s suggestion has some initial appeal because it suggests a diagnosis of why so many have been tempted to interpret Hume as a non-cognitivist. Hume famously holds that our passions are not the product of reason either (something Pigden thinks is manifestly false, by the way, but more on this presently). We therefore get the non-cognitivist interpretation of Hume if we further assume, incorrectly according to Pigden, that Hume thought that passions are the only psychological states that are not the product of reason: “Moral beliefs are not products of reason. Passions are not products of reason. So moral beliefs are passions.” The mistake is to overlook the other possibility, namely, that Hume thought moral beliefs were identical with impressions, which are also psychological states that are not the products of reason. Initially appealing though this suggestion is, however, when you think it through more fully it seems to me that it collapses into the view that Hume is a non-cognitivist. Pigden’s idea, to repeat, is that perceptual beliefs and normative beliefs are on all fours insofar as having such beliefs is simply a matter of having relevant “impressions”: I open my eyes and form the belief that there is a computer screen in front of me. I tuck into my breakfast of toast and mussels and form the belief that it is delicious. Or I contemplate my wife Zena in her new red dress and form the belief that she is sexy. In the first case it is my visual impressions that do the trick, in the second my oral impressions and in the third case … well, it’s a complicated business, but you get the general idea. Now in all these cases inference is minimal to non-existent. I do not deduce or infer that my dinner is delicious or my wife sexy. I feel them to be so in consequence of my sensations. (INNC, § 15) But Pigden forgets to mention that when we see Zena in her new red dress there are in fact two quite different “sensations” at play. There is a sensation much like that we have when we open our eyes and form the belief that there is a computer screen in front of us: we open our eyes and have a visual image – an “impression” – of Zena in her new red dress. Unsurprisingly, since it is a perceptual belief, this belief is indeed on all fours with other perceptual beliefs. And there is also the more “complicated business” Pigden

112 The Motivation Argument for Non-Cognitivism

tactfully passes over, namely, the attraction we feel towards Zena when we have this perceptual belief, that is, when we enjoy that “impression” of her. But attraction is, of course, a matter of desire. So let’s now put the whole story together. According to Pigden’s own analysis, Hume holds that the perceptual belief that Zena, standing there in her new red dress, is sexy isn’t a productof-reason belief, but is rather constituted by two “sensations”: the visual image – the “impression” – of Zena in her new red dress and the attraction we feel towards her. But, as is I hope now obvious, this amounts to the suggestion that Hume is a non-cognitivist, not a cognitivist. For, according to this suggestion, Hume holds that our perceptual belief that Zena is sexy is constituted, inter alia, by a desire, namely, the attraction we feel towards her. Indeed, there seem to be two quite different beliefs here, the perceptual belief that Zena is sexy and the normative belief that she is sexy, with the perceptual belief being constituted by the “impression” and the normative belief being constituted by the desire. Since, as explained at the beginning of this section, the signature view of a non-cognitivist is that normative beliefs are constituted by desires, it follows that Pigden’s more complicated interpretation of Hume’s account of moral belief entails a non-cognitivist interpretation, not a cognitivist interpretation. Let’s now return to Pigden’s original idea. He tells us that, as a follower of Francis Hutcheson, Hume held that moral features are, in today’s jargon, “response-dependent.” Norva Lo reminds us of this feature of Hume’s thinking as well, though she concentrates on what’s distinctive of Hume’s own version of a response-dependent view. As she points out, Hume holds that an action or character is virtuous or vicious if and only if human beings would approve or disapprove of the action or character if they satisfied the following condition: C1 The condition of having full knowledge of all the circumstances and relations relevant to the case under evaluation. (IHI, §2) From what Hume calls the general point of view, a point of view Lo summarizes is as follows: In short, to adopt the “steady and general points of view” is to meet all of the following conditions: C2. The condition of taking into account facts about human nature. C3. The condition of overlooking one’s personal interests. C4. The condition of overlooking one’s personal relations to the parties involved in the case under evaluation. (IHI, section 2)

Michael Smith

113

Thus she concludes: Hume’s analysis of virtue and vice can be further expounded as follows: (H) An action or character is virtuous/vicious if and only if all human beings are disposed, under the conditions C1, C2, C3 and C4, to feel the sentiment of approbation/disapprobation towards it. (IHI, § 2) Since (H) is a biconditional, and since Lo thinks that the right hand side of this biconditional is a statement of fact – the statement that all human beings are disposed, when they have relevant knowledge and adopt the general point of view, to feel certain sentiments of approbation or disapprobation – she thinks it follows that the left hand side of the biconditional – the claim that an action or character is virtuous or vicious – must be a statement of fact too. Hume must therefore really believe that morality is a matter of fact and that our moral views are beliefs about these matters of fact; he is a cognitivist, not a non-cognitivist, or so Lo and Pigden contend. But though this line of thought also has some initial appeal, it seems to me that it faces a fairly serious objection, namely, that the form of argument involved is plainly invalid. In order to see that it is invalid, it suffices that we ask ourselves whether a non-cognitivist could be a utilitarian. The answer is presumably that yes, a non-cognitivist could believe that (say) some particular action is forbidden if and only if that action fails to maximize happiness and minimize suffering. But look again at what a non-cognitivist who is a utilitarian has to believe. He has to believe a biconditional, the right hand side of which is a statement of fact. Assuming now that Lo’s and Pigden’s argument in the case of Hume on the general point of view is valid, it would follow by parity of reasoning that the left hand side of this biconditional – the claim that some particular action is forbidden – is a statement of fact too. From the fact that a non-cognitivist is a utilitarian, it would therefore follow that he is a cognitivist. Reductio? I think not. So where did we go wrong? We went wrong in supposing that the mere acceptance of a biconditional, where the right hand side is a statement of fact, means that the person who accepts that biconditional thinks that the left hand side is a statement of fact too. This is the argument form implicit in Lo’s explanation of why Hume is a cognitivist given what he says about the general point of view. But if the person who accepts the biconditional is a non-cognitivist, then there is an obvious alternative. He might think that the left hand side of the biconditional – the claim that some particular act is forbidden, say – is, inter alia, an endorsement of certain standards, and he might think that the right hand side – the claim that action fails to maximize happiness and minimize suffering, say – tells us what those standards are. The biconditional might, in other words, be an implicit endorsement or recommendation of certain standards.

114 The Motivation Argument for Non-Cognitivism

With this alternative understanding of what a non-cognitivist might have in mind when he accepts a biconditional linking a moral claim to a nonmoral claim, let’s return to consider Hume on the general point of view. To be sure, Hume does accept (H), a biconditional linking statements of virtue and vice to statements about the sentiments of approbation and disapprobation that human beings feel towards certain actions and characters when they adopt the general point of view. But should we suppose that, when he says this, he is really doing what Pigden calls “what is nowadays known as ‘meta-ethics’”? Is Hume doing meta-ethics when he tells us about the general point of view, or is he, inter alia, making a substantive normative proposal, much like a non-cognitivist who advocates utilitarianism? The crucial question, as we have seen, is whether Hume is implicitly endorsing the deliverances of the general point of view. In his account of Hume on the general point of view, Geoffrey SayreMcCord argues that Hume has both explanatory and justificatory aims (SayreMcCord, 1994, p. 203). He not only wants to explain what we are doing when we make moral judgments, but he also wants to justify the fact that we make moral judgments in the way that we do, as opposed to in some other way. As Sayre-McCord puts it: [A]ccording to Hume, “[t]he only point of view in which our sentiments concur with those of others is when we consider the tendency of any passion to the advantage or harm of those who have any immediate connexion or intercourse with the person possess’d of it” (T, 3.3.2.17/ 602–3). And because we have compelling reason to seek a common ground with our fellows, we have reason to adopt that particular point of view as setting the standard for our moral judgment. According to Hume, “reason requires such an impartial conduct” because reason informs us that we will be unable to achieve a suitable consensus in evaluation without establishing a shared standard at least for our judgments, and sentiment makes attractive the end of achieving such a consensus. Reason here plays a role as “a general calm determination of the passions, founded on some distant view or reflection.” (T, 3.3.1.18/583) (Sayre-McCord, 1994, p. 219) Sayre-McCord’s idea, in other words, is that Hume is indeed endorsing the deliverances of the general point of view to us. He endorses these deliverances on the grounds that they represent a consensus, something “we have compelling reason to seek.” Sayre-McCord’s discussion of the justificatory aim of Hume’s discussion of the general point of view suggests an alternative understanding of (H). Perhaps we should understand (H) in much the same way as we understand the biconditional accepted by a non-cognitivist utilitarian. In other words, Hume may hold that the left hand side of (H) – the claim that certain

Michael Smith

115

characters are virtues and others are vices – is implicitly an endorsement of certain standards, and that the right hand side – the claim that all human beings are disposed to approve or disapprove of those characters when they have relevant knowledge and adopt the general point of view – tells us what those standards are. The standards, to repeat, are standards of achieving a consensus among our fellows. If this is the right way to understand (H) then it follows that we cannot move immediately from the premise that Hume accepts that biconditional to the conclusion that he is a cognitivist. Given that Hume is endorsing the convergent deliverances of the general point of view on the grounds of the consensus they represent, that move is itself licensed only if Hume is a cognitivist, rather than a non-cognitivist, about the attitude of endorsement. Yet whether or not Hume is a cognitivist or a non-cognitivist about the attitudes of endorsement is much the same as the question with which we began, which is whether he is a cognitivist or a non-cognitivist about claims about virtue and vice. The detour via Hume on the general point of view thus hasn’t helped us answer that question. It has merely relocated it.

4.3 Is the Motivation Argument successful? So far we have been concerned with a purely interpretative question. Is the Motivation Argument appropriately thought of as a rational reconstruction of Hume’s argument quoted at the outset? Now let’s consider the argument itself. Is the Motivation Argument for non-cognitivism successful? The question splits into several parts. The first is whether we can explain why people are motivated in accordance with their moral views, insofar as they are practically rational, if we suppose that having a moral view is a matter of having certain desires; the second is whether we can explain this if we suppose that having a moral view is a matter of having certain beliefs; and then, if both of these questions are answered positively, the third concerns the relative merits of the two explanations: which is better? The discussion of Pigden’s and Lo’s reasons for thinking that Hume is a cognitivist has made it clear what kind of non-cognitivism Hume would have to embrace, if he were indeed a non-cognitivist. So let’s assume that he is a non-cognitivist and begin by asking the first question with respect to non-cognitivism of that kind. If Hume is a non-cognitivist, then he is the kind who holds that a moral judgment is an endorsement of the desires that we would have under certain conditions. Hume himself, for example, holds that a moral judgment is an endorsement of the attitudes of approbation and disapprobation that we would have if we were to adopt the general point of view. This reflects his substantive commitment to consensus. Those with different substantive commitments may endorse attitudes of approbation and disapprobation under rather different conditions. Assuming that Hume is such a non-cognitivist,

116 The Motivation Argument for Non-Cognitivism

it follows that, in his view, having a moral view is a matter of having certain second-order desires: having a moral view is a matter of having an attitude of approval toward those attitudes of approbation and disapprobation that we would have if we were in the relevant conditions. Note that non-cognitivism of this kind would seem to provide Hume with an explanation of why people are motivated in accordance with their moral views, insofar as they are practically rational – or, at least, it would if we made the further assumption that, for Hume, a practically rational agent is one who has and exercises the capacity to realize the contents of his desires, given his beliefs. For any agent who has such a second-order desire and who has and exercises this capacity would have the attitudes of approbation and disapprobation that he believes he would have if he were in the relevant conditions. Indeed, a failure to exercise this capacity would seem to constitute a kind of weakness – there would be a mismatch between the agent’s desires about the sentiments of approbation and disapprobation he is to have and his sentiments themselves, a mismatch he has the capacity to ensure does not occur – and a lack of this capacity would constitute a kind of compulsion – a mismatch he has no capacity to correct. Non-cognitivism of the kind Hume is committed to, assuming that he is indeed a non-cognitivist, thus makes it possible to give familiar Humean explanations of the various kinds of practical irrationality that undermine motivation in accordance with moral views (see, for example, Frankfurt, 1971; Bigelow, et al. 1988; Bigelow, et al. 1990). I said earlier that interpreting Hume as a judgment internalist requires us to ascribe to him concepts he may not have resources to make sense of. It should now be clear why I said this. As I mentioned earlier, some theorists insist that Hume denies that there are any norms of rationality governing the ways in which our desires relate to our beliefs. If they are right about this then, in particular, it follows that Hume can make no sense of an agent’s being practically rational in the sense of having and exercising the capacity to realize the contents of his desires, given his beliefs. Interpreting Hume as accepting judgment internalism of the kind we have been discussing, by contrast, requires us to ascribe to him precisely this view about what it is to be practically rational. This is why I have been careful to say that the interpretation of Hume as a non-cognitivist who can explain judgment internalism is a rational reconstruction of his views. But, having said this, note that the rational reconstruction doesn’t require us to ascribe to Hume views that are a million miles away from those that are usually associated with him. It is, in essence, simply to see him as a familiar kind of Humean instrumentalist about practical rationality (Hubin, 2001). So much for the non-cognitivist’s explanation of the fact that people are motivated in accordance with their moral views, insofar as they are practically rational. Can a cognitivist come up with an alternative explanation? Hume famously thinks not. As he says in the passage quoted at the outset, “[m]orals excite passions, and produce or prevent actions. Reason of itself

Michael Smith

117

is utterly impotent in this particular.” This is where Pigden thinks the cognitivist should take issue with Hume. He explains what Hume has in mind by his talk of the impotence of reason in the following terms. When we say that no genuine belief alone can motivate, what we mean is not just that genuine beliefs by themselves cannot cause actions but that genuine beliefs cannot cause new desires except desires for the means to an end that is already desired. (INNC, § 5) The crucial point is thus that, according to Pigden, Hume is making a claim that is wholly and solely about causation. This is not, I think, the most plausible interpretation of what Hume has in mind when he says that reason is “utterly impotent,” given that he thought that anything could cause anything at least in principle, but let’s put that to one side for the moment. We will return to consider what else he might have in mind at the end. Pigden finds himself baffled that Hume should think that beliefs don’t cause new desires, except desires for the means to an end that is already desired (from here on I will take this exception as read), because he thinks that there are lots of beliefs that do this. Consider the following: “At the age of eleven I began Euclid, with my brother as my tutor. This was one of the great events of my life as dazzling as first love. I had not imagined anything so delicious in the world. … From that moment, until Whitehead and I finished Principia Mathematica, when I was thirty-eight, mathematics was my chief interest and my chief source of happiness.” The author of course, is Bertrand Russell and the passage comes from his Autobiography … Ray Monk, who recounts this episode … goes on to point out that Russell’s reactions have an interesting precedent. When Thomas Hobbes was forty he chanced upon a copy of Euclid’s Elements which was lying open in a gentleman’s library. It was open at the proof of the famous Pythagorean theorem. “‘By God’, says he, ‘this is impossible!’ So he reads the Demonstration of it, which referred him back to another, which he also read. Et sic deinceps that at last he was demonstratively convinced of it truth. This made him in love with geometry.” … In both cases the abstract and demonstrative beliefs led to new desires. Being in love, they both sought to know the beloved better and embarked on a course of study. Russell wanted to demonstrate not only the inferences but the axioms while Hobbes acquired a desire to square the circle and to establish a deductive science of politics. But these desires were not desires for the means towards things they already wanted. … Thus Hume is doubly mistaken, both about the inertness of reason and about the peculiar inertness of demonstrative reason. Of course, these abstract and demonstrative beliefs did not act quite alone.

118 The Motivation Argument for Non-Cognitivism

We must postulate a rather peculiar motivational state, an unusual disposition to acquire new desires, since not everybody reacts to Euclid as Russell and Hobbes did. But we do have two clear instances refuting [the claim that genuine beliefs cannot motivate by themselves]. This is all the more ironic, because Hobbes anticipated Hume while Russell explicitly agreed with him. They both thought that (reason-derived) beliefs could not motivate without the aid of preexisting desires though in each case their own most formative experiences proved them wrong. (INNC, § 11) Pigden also mentions another striking example. Thomas Reid claimed that almost all of us have the disposition to acquire the desire for our own good on the whole upon grasp of the fact that we are persisting creatures with a good of our own: “as Reid puts it, ‘as soon as we have the conception of what is good or ill for us on the whole, we are led by our constitution to seek the good and avoid the ill’” (INNC, § 12). The fact that there are such dispositions provides the cognitivist with a response to the Motivation Argument, or so Pigden insists. The Motivation Argument assumes that beliefs do not cause new desires. But the phenomenon of beliefs causing new desires is utterly ubiquitous, as the examples he mentions make clear. The underlying mechanism is something that Pigden describes as an “unusual disposition to acquire new desires” upon acquisition of a relevant belief. But this disposition is unusual only in the sense that philosophers don’t often talk about it. It is not unusual in the sense of being uncommon. To repeat, there are many beliefs that give rise to new desires, and, since this is so, Pigden thinks that it should come as no surprise that moral beliefs might be among them. The Motivation Argument, which assumes that moral beliefs lack such causal powers, is thus a dismal failure, or so he concludes. What should we make of this? Pigden is surely right that there are all sorts of dispositions to acquire new desires upon the acquisition of beliefs. An example I’ve talked about myself in an earlier work is best understood in exactly these terms (Smith, 1991). Consider the way in which many people are repulsed by spiders upon first encountering them. For such people, the mere perception of a spider seems to cause all sorts of new desires: the desire to be far removed from spiders, the desire to kill those that are nearby if they can’t get away from them, the desire never to look at them, and so on. Such people thus seem to have the disposition to acquire these new desires, which are all about spiders, immediately upon the acquisition of perceptual beliefs about them, beliefs which are often the first they have ever formed about spiders. It therefore seems to me that Pigden is absolutely right that there are such dispositions; he is right that beliefs can cause new desires. The question is, however, whether the fact this is so suffices to take the sting out of the Motivation Argument, and the answer is that it does not.

Michael Smith

119

As I said at the outset, what needs to be explained is why people are motivated in accordance with their moral views insofar as they are practically rational. This is the crucial judgment internalist thesis that plays a role in the Motivation Argument. But the mere fact that people have the disposition to acquire new desires when they acquire certain moral beliefs would be insufficient to underwrite the truth of this thesis. Imagine, for example, that Russell’s disposition malfunctioned and that, as a result, he didn’t fall in love with the idea of constructing proofs after having grasped Euclid’s demonstrations, while Hobbes’s didn’t. That wouldn’t suffice to make Russell practically irrational and Hobbes practically rational because, absent some suitable background desire, there is simply no rational relation between the acquisition of such geometrical beliefs and the desire to construct proofs. Or again, imagine someone who lacks the disposition to be repulsed by spiders upon the acquisition of perceptual beliefs about them. Such a person isn’t for that reason appropriately thought to be practically irrational. Again, this is because there is no rational relation between someone’s mere acquisition of the belief that she is confronted by a spider and her desiring to run away, or kill it, or the like, at least without a suitable background desire. For much the same reason, if the disposition to acquire desires that accord with our moral views were on all fours with these other dispositions, those in whom that disposition functioned properly would not appropriately be deemed to be practically rational. Of course, matters would be very different if Pigden were able to provide some reason for supposing not just there is such a disposition, but that this disposition is partially constitutive of being rational (in essence, this was my strategy in Chapter 5 of The Moral Problem (1994)). This is, however, precisely what Pigden does not do. The disposition he claims to have identified is meant to be brutally causal, something his examples make clear. Though a disposition of the kind Pigden identifies would suffice to explain how moral beliefs could cause new desires, it is thus totally incapable of explaining the distinctive practical rationality of this connection. Yet the practical rationality of the connection is what the cognitivist needs to explain, if he is to have a plausible response to the Motivation Argument. This raises an interesting question. Why doesn’t Pigden realize that what a cognitivist needs to explain is the practical rationality of the connection between moral views and motivations? The explanation lies, I think, in his implausible interpretation of Hume on the impotence of reason. As noted earlier, when Hume says that reason is “utterly impotent,” Pigden thinks he is making a claim that is wholly and solely about the causal powers of belief: beliefs don’t cause new desires. As a result, he thinks we can refute Hume’s claim if we find examples of beliefs that do cause new desires. But an obvious alternative interpretation of Hume on the impotence of reason now suggests itself. Hume’s claim is that, if we restrict our attention to beliefs that are not constituted by desires, there are no beliefs that make it rational

120 The Motivation Argument for Non-Cognitivism

to have new desires. Hume’s claim is thus not wholly and solely about the causal powers of belief, but is rather a claim about the causal and rationalizing powers of belief. According to this interpretation of Hume on the impotence of reason, what Hume is saying is that no beliefs that are not constituted by desires stand to new desires in the same sort of relation that beliefs stand to the beliefs with which they are inferentially connected, or that desires for ends and beliefs about means stand to the desires for the means which they produce. This is the interpretation of Hume on the impotence of reason that lies behind the Motivation Argument. In Hume’s view, though beliefs that aren’t constituted by desires do not stand in a causal and rationalizing relation to new desires, moral beliefs do stand in such a causal and rationalizing relation to new desires. He therefore urges us to draw the obvious conclusion, which is that moral beliefs are constituted by desires. If what Pigden says on behalf of cognitivism is the best that can be said by way of a response, is it any wonder that this argument for non-cognitivism has seemed so convincing?

5 Experiences of Value Graham Oddie*

Hume, in the Treatise of Human Nature rejects the thesis that morality is the province of reason – that it is the job of reason to reveal the moral truth, and then, via the will, to direct action, moderating and overriding the unruly influence of the passions. Hume announces – famously and shockingly – that [r]eason is, and ought only to be the slave of the passions, and can never pretend to any other office than to serve and obey them. (T, 2.3.3.4/415) He concedes that ‘this opinion may appear somewhat extraordinary’, so that ‘it may not be improper to confirm it by some other considerations’ (T, 2.3.3.4/415). The considerations which Hume adduces in favour of his thesis are rich and suggestive. Indeed they have suggested to different interpreters various versions of non-cognitivism, expressivism, subjectivism, projectivism and the response dependence of the moral – in short, a variety of extreme to moderate anti-realisms about morality and value. I am sympathetic to a Humean moral phenomenology and epistemology of value, but I also favour a robust value realism. Here, I want to sketch a compatibility argument – an account of our knowledge of value that sits happily with Hume’s phenomenology, but which also sits happily with a realist account of the metaphysics of value. By a realist theory I mean one that claims, firstly, that, value judgements have genuine propositional content, and that endorsing a value judgement is affirming the truth of the associated proposition. (This is cognitivism.1) Secondly, that the existential presuppositions of value judgements are satisfied, that there really are value properties and relations. (That is to say, nihilism, or an error theory of value judgements, is rejected.) And thirdly, that value truths do not reduce to any more basic non-evaluative truths. (That is, idealism about value in particular, and naturalism in general, are false.) I won’t argue for these theses here2 but show rather that they are in fact compatible with many of Hume’s central insights about the phenomenology of value, and that in fact 121

122

Experiences of Value

Hume’s insights can be put to service to help meet the epistemic challenge to value realism.

5.1 Reason and the passions One powerful strand in Hume’s thought is that pure reason alone could not discover truths about morality in particular or about value in general. Absent specific value data a pure ratiocinator would not alight on any truths about value – or at least no truths other than those that involve necessary relations between value concepts. (For example, if something is good it is not bad; that if X is better than Y and Y is at least as good as Z, then X is better than Z and so on.) With this line of the argument I am in broad agreement. The rational core of Hume’s thesis, however hard it is to articulate it, is that you cannot derive an interesting evaluative judgement (for example, an ‘ought’) from a purely non-evaluative judgement (an ‘is’), no matter how rich and interesting the latter may be. (The difficulty, of course, lies in characterizing the evaluative and non-evaluative judgements and dealing appropriately with mixed and hybrid judgements. See Oddie and Demetriou (2007) for a fuller discussion.) Hume argues, further, that the direct passions (‘the impressions, which arise immediately from good or evil, from pain or pleasure’), like desires and aversions, cannot logically conflict with ‘truth or reason’, because these passions are ‘original existences’ which do not represent the world as being a certain way. A passion is an original existence, or, if you will, modification of existence, and contains not any representative quality, which renders it a copy of any other existence or modification. When I am angry, I am actually possessed with the passion, and in that emotion have no more a reference to any other object, than when I am thirsty, or sick, or more than five foot high. ’Tis impossible, therefore, that this passion can be opposed by, or be contradictory to truth and reason; since this contradiction consists in the disagreement of ideas, consider’d as copies, with those objects, which they represent. (T, 2.3.3.5/415) The central premise of this argument (that the passions do not represent anything), as well as one of the conclusions Hume draws from this (that the passions cannot be ‘contradictory to truth’), I reject. However, the premise is not needed to secure Hume’s main thesis that unaided reason cannot secure interesting truths about value, and the conclusion that the passions cannot be ‘contradictory to truth’ goes beyond what Hume is entitled to. Hume himself concedes that it is not universally true of the passions that they don’t represent the world as being in a certain state. He recognizes

Graham Oddie

123

that the indirect passions (those that ‘proceed from the same principles, but by the conjunction of other qualities’), like envy or pride, do indeed represent the world as being a certain way. If you are envious of your neighbour’s fancy new car then you must represent to yourself that your neighbour has a fancy new car. You cannot rationally combine your envy of your neighbour’s new car with the belief that he does not have a car. Thus your envy is misplaced if he has no car, and irrational if you don’t believe he has a car. So the idea of an indirect passion like envy is something like this: you first notice that your neighbour has a fancy new car, and you yourself would like just such a car, and you are averse to his having that kind of car while you lack such. The conjunction of this particular belief with these particular desires and aversions is what constitutes envy. If we look at Hume’s examples the argument for the non-representative status of the passions seems even less persuasive: When I am angry, I am actually possessed with the passion, and in that emotion have no more a reference to any other object, than when I am thirsty, or sick, or more than five foot high. (T, 2.3.3.5/415) The state of being five foot high is an anomaly in Hume’s list – it quite obviously does not contain a reference to anything else. But two of the others do contain, within themselves, not only a representation of some other state but also a representation of the value of a state. If I am angry I am angry about something and part of the anger is that that thing presents itself to me as unjust. If I am thirsty, it seems to me that my body needs water, and that it would be good to drink some. My being sick, like my being five foot tall, may have no reference to any state other than being sick, but being sick does contain a reference to a standard of health, and feeling sick involves, at least, it’s seeming to me that my body is in a bad way. So, passions can be and are representative. Further, they can represent not just non-evaluative states, but evaluations of the states they represent. In this essay I develop the Humean idea that our knowledge of value is based on experience – specifically, on value experiences or appearances of value – and further, that these basic experiences of value are not just theoretical postulates required by a realist empiricist theory of value, but are states with which we are all intimately familiar: they are, in fact, Hume’s direct passions, desires and aversions. Desires and aversions represent to us the values of the states that are their objects. This idea is quite compatible with numerous Humean themes: that beliefs alone are incapable of motivating; that reason alone is not sufficient to reveal to us substantive value truths; that evaluative judgements are, in some sense, necessarily connected with motivation. All of this, however, is

124

Experiences of Value

also compatible with a realist account of value and of our epistemic access to the world of value.

5.2 The puzzling asymmetry One powerful argument that Hume levels against rationalism – an idea that has now enjoyed a long and fruitful history – appeals to the idea that mere beliefs alone cannot motivate. Since evaluations, such as moral judgements, do motivate, evaluation cannot be a matter of mere belief, and hence cannot be a matter of pure reason. ’[T]is impossible that the distinction between moral good and evil can be made by reason; since the distinction has an influence upon our actions, of which reason alone is incapable. (T, 3.1.1.16/462) I want to illustrate the apparent motivational force of value judgements by means of a puzzling asymmetry between first person and third person attributions of such. Consider the fact that millions of people suffer from debilitating conditions – like severe malnutrition – that would take only a few dollars to remedy. You know that. And as you cast your eye over my assets, you might notice that I have a surplus which I am not sending to charities. In that case it wouldn’t sound odd for you to endorse the following pair of judgements: The best thing he could do with his surplus income would be to donate it to charity. But he has absolutely no desire to do that. But I have to admit it sounds a bit odd for me to say: The best thing I could do with my surplus income would be to donate it to charity. But I have absolutely no desire to do that. Now a cognitivist about evaluation holds that value judgements have propositional content. And it is natural for the cognitivist to go on to hold that value judgements share standard features with other judgements: that a typical affirmation or endorsement of a value judgement, for example, is just like the typical affirmation of any other judgement. Typically, to sincerely affirm a judgement with propositional content is to express one’s belief in that proposition. Thus, typically, to sincerely affirm a value judgement is to express one’s belief in a proposition about value. So, if cognitivism is right there is a perfectly good proposition which serves as the common propositional content of both your value judgement and mine, and both our affirmations are simply affirmations of the truth of that proposition. So why should my affirmation of that

Graham Oddie

125

proposition sound odd, whereas your affirmation of the very same proposition sound not at all odd? Call this the puzzling asymmetry.

5.3 Internalism Suppose that to endorse a value judgement necessarily involves being moved to behave or respond appropriately in the light of it – at least to some extent – and being moved to behave involves having appropriate desires. To summarize: Judgement Internalism Necessarily, one who endorses a value judgement possesses the corresponding desire to act or respond appropriately. Hume, in the passage quoted in the previous section, along with many others, might plausibly be construed as a judgement internalist. Given judgement internalism, the oddness of my affirmation of the pair of judgements is immediately laid bare. That I endorse the value judgement necessitates my possessing an appropriate corresponding desire – a desire to donate my surplus income to charity. So by expressing my endorsement I thereby commit myself to the existence of that desire. But straight away I go on to deny the existence of any such desire. So the oddness of my utterance consists in a kind of contradiction – between what I imply by affirming the value judgement (viz. that I have a certain desire) and what I imply by affirming the non-value judgement (viz. that I lack that very desire). However, your utterance does not commit you to any kind of contradiction. By endorsing the very same value judgement you affirm your desire that I donate my surplus income to charity. Your affirmation of that is quite compatible with your subsequent denial that I lack the right desires. Internalism thus provides an apparently neat explanation of the puzzling asymmetry. Internalism flows rather naturally out of a non-cognitivist analysis of value judgements. It’s a platitude that one can express attitudes other than belief by uttering a declarative sentence. If I say to someone – Cheney, say – ‘I hate you!’ then I don’t merely express my belief in the proposition that I hate Cheney (as someone else might do by saying ‘He hates Cheney.’). Rather, what I express is my hatred for Cheney. So an apparently declarative affirmation can be used to express an attitude like hatred. When I exclaim ‘Pistachio ice-cream is the best!’ maybe what I am expressing is not a belief about value at all, but rather a desire, a desire for pistachio ice-cream. Suppose that, quite generally, endorsing a value judgement, that P is good, is just a way of expressing the desire for P. Then internalism flows naturally out of that brand of non-cognitivism and the non-cognitivist can thus happily appropriate the internalist explanation of the puzzling asymmetry. Internalism and non-cognitivism dovetail nicely.

126

Experiences of Value

5.4

Independence

The cognitivist is also free to explain the puzzling asymmetry by embracing internalism. But internalism does not flow freely from cognitivism as it does from its rival. It is an extra commitment, and it may be an embarrassing commitment at that. True propositions about value would be queer. They would be queer, according to this line of reasoning, because grasping or believing them would violate the Humean doctrine that beliefs alone cannot motivate us. The basic idea is that there are two broad classes of mental states, beliefs and desires, and these are logically independent states: Independence For any belief, the possession of that belief is logically compatible with the possession of any given set of desires. Belief and desire suffer from the standard state-content ambiguity which afflicts all our talk about the mental. By belief one might mean the propositional content of a belief – the proposition believed. (As in: it was our belief that McCain would win the election easily.) Alternatively, one might mean the mental state of believing that propositional content. (As in: it was your belief that McCain would win the election easily that made you neglect your civic duty to vote.) Similarly, by desire one might mean the content of a desire – what is desired – or else one might mean the mental state of desiring that thing. In the content-sense, beliefs are obviously logically dependent on other beliefs, because beliefs (in this sense) just are propositions, which, of course, stand in various relations of logical dependence. But in the state-sense, beliefs seem just as obviously logically independent of each other: the having of one belief does not, by itself, entail the having of any other belief, even if the contents of those two beliefs are logically related. Now, given that belief states are logically independent of one another, how unlikely that any given belief state should entail a desire state. In the statesense, then, beliefs and desires are obviously logically independent. Being in some belief state does not entail that one is in any particular desire state.

5.5 The Queerness Argument Here is a recent spelling-out of the queerness argument, offered by Michael Smith, and it is apposite here because it is clearly informed by neoHumeanism. [T]he idea of a [value] judgment thus looks like it may well be incoherent, for what is required to make sense of such a judgment is a queer sort of fact about the universe: a fact whose recognition necessarily impacts upon our desires. But the standard picture tells us that there are no such facts. (1991, p. 402)

Graham Oddie

127

So in a nutshell the argument is this: Internalism tells us that the endorsement of a particular value judgement is necessarily accompanied by the appropriate desire; Cognitivism tells us that the endorsement or affirmation of a value judgement just is the expression of a belief; and Independence tells us that the mere having of a belief does not entail the having of any desire at all. So the triad consisting of internalism, cognitivism and independence threatens a contradiction – something has to go. Consequently, if the realist wants to embrace internalism she will have to reject the independence of belief and desire – and that’s queer, or at least, deeply un-Humean.

5.6 Independence examined The cognitivist might begin with an examination of the independence thesis. Is there anything decisive to be said either in favour of it or against it? Smith cites two arguments in favour of the thesis. First, there is the notorious ‘direction-of-fit’ argument. Beliefs purport to represent the world as it is, and a belief is faulty to the extent that its content fails to represent the world accurately. Briefly, beliefs are required to fit the world, and they can be justly criticized if they fail to fit the world – if they are false. Desires, on the other hand, do not purport to represent the world, and are not usually said to be true or false. The content of a desire may be true or false, just as the content of a belief is either true or false. But it is clear that a desire cannot be criticized merely by pointing out that the content of the desire is not the case. In that sense, the content of a desire does not have to fit the world, or fit the world now, for it to be a perfectly defensible desire. Rather, the world is in some sense ‘defective’ for not matching the content of one’s desires. Beliefs have to fit the world, but the world has to fit desires. There are thus two classes of mental states, characterized by different ‘directions of fit’ and (perhaps because they are orthogonal dimensions) beliefs and desires are independent (see Anscombe (1957), Smith (1987), and Humberstone (1992)). The direction-of-fit metaphor may be unnecessary here. Smith spells out the essential contrast between beliefs and desires directly in terms of rational criticisability. Since our beliefs purport to represent the world they are subject to rational criticism: specifically they are assessable in terms of truth and falsehood according to whether or not they succeed in representing the world to be the way it really is. Desires are unlike beliefs in that they do not even purport to represent the world as it is. They are therefore not assessable in terms of truth and falsehood. Indeed, according to the standard picture they are at bottom not subject to any rational criticism at all. (Smith, 1991, p. 400)

128

Experiences of Value

Or, as Hume puts it: ’Tis not contrary to reason to prefer the destruction of the whole world to the scratching of my finger. ’Tis not contrary to reason for me to choose my total ruin, to prevent the least uneasiness of an Indian or person wholly unknown to me. (T, 2.3.3.6/416) However, the claim that desires are not subject to any rational criticism is rather implausible, at least without hefty qualifications, and in fact Hume concedes most of these (at least with respect to the indirect passions). Firstly, a combination of desires may be logically inconsistent in the sense that they cannot all be jointly realized. (I want to be the richest man in the world, but I don’t want everyone to be poorer than me.) Such combinations of desires are clearly rationally criticisable. Secondly, even if not logically incompatible, desires may not be jointly realizable given my beliefs. (I want to go to the concert and I also want to go to the lecture, but I believe they are being held at the same time.) Admittedly it is the total package of belief and desire that is here irrational and criticisable. But if the belief component is rationally impeccable, it is desires that must suffer reason’s censure. Thirdly, desires may also be based on irrational belief. (I desire to leap out of a tenth floor window because I have a more basic desire to arrive safely on the groundfloor, and a belief that jumping out of tenth-floor windows is an efficient way of achieving my basic desire. The derived desire is rationally criticisable because the belief that mediates between the basic and the derived desire is crazy.) Smith (like Hume) acknowledges this: ‘desires are subject to rational criticism, but only insofar as they are based on beliefs that are subject to rational criticism’ (Smith, 1991, p. 401). And there are other objections to the claim that desires are not rationally criticisable.3 The argument for independence from state-content ambiguity (stated in section 5.4) is more promising but turns out, somewhat surprisingly, to be fallacious. There are, in fact, purely logical reasons why belief states cannot be totally logically independent of one another as demanded, and for the same reason belief states are not totally logically independent of desire states. The reason is that there are vastly more propositions than there are possible situations, and as such there are too many propositional objects of belief and desire for there to be enough room in logical space to combine any old belief with any old desire set. There are just not enough possibilities to underwrite the logical possibility of all the different combinations of belief and desire required for the independence thesis to be true (see Oddie (2001)). This rather abstract refutation of independence is based on the Cantorian fact that a power-set is always of greater cardinality than the set itself. If propositions are classes of possible worlds then the set of propositions has a higher cardinality than the set of possible worlds. So there are more

Graham Oddie

129

propositions than there are possible worlds, and hence more belief-desire pairs than there are possible worlds. If propositions are more fine-grained than sets of possible worlds, then there are even more propositions. Concrete proposition-pairs that are not logically possible to embrace as belief and desire can be directly constructed (see Appendix 2 of Oddie (2005)). Here is a direct and down-to-earth example against independence involving particular beliefs and desires. Suppose that to be in pain is to experience a very unpleasant sensation, and suppose the unpleasantness of a sensation consists (at least in part) of the strong desire to be rid of the sensation. So, to be in pain is ipso facto to desire that the painful sensation cease. Suppose further that one cannot be mistaken in one’s pain-beliefs – that is to say, of necessity, if one believes one is in pain then one is in pain. It follows that a certain kind of belief (the belief that I am experiencing pain) necessitates a desire (the desire for the painful sensation I am experiencing to stop). So a typical pain-belief necessitates a desire – the desire for a certain sensation to cease. The weakness in this refutation is, of course, the claim that pain beliefs are infallible. While it would certainly be an odd kind of cognitive defect to believe one is in pain without being in pain, it is not an unimaginable defect. People can certainly have false beliefs which are just as odd. Still, the Cantorian argument outlined above, while admittedly somewhat abstract, is sufficient to show that the quite general independence thesis fails.

5.7 Desire-judgement gaps Still, even when one grants these or other limitations on the general independence thesis, a friend of the queerness argument could always retreat to a more local and plausible principle, one which would do the job. Isn’t it just downright implausible that having the particular belief that P is good necessitates having the desire for P? Are we not familiar with lots of cases in which that claim is obviously false? Isn’t denying judgement internalism a much more plausible response for the cognitivist? Does judging something to be valuable entail that one is motivated to pursue it? Clearly judging that something is good does not always go hand in hand with an overriding motivation to pursue that good. Something judged to be good may be just one such among many competing goods, and so may be traded off against other goods. Pursuing something judged to be good, even judged to be the best, may be rather risky, and unlikely to succeed. Pursuing what you judge to be best may run a high risk of landing you with something you judge not very good at all. (The best can be the enemy of the good.) Further, pursuing what is judged to be the best may conflict drastically with what you take to be in your own interests, which channel your desires in quite a different direction. Finally, one may suffer from akrasia – judging something to be the best, all things considered, but imply lacking the desire to pursue it. So, even if my charitable desires do not

130

Experiences of Value

measure up to my charitable beliefs, I am not so exceptional after all. I have plenty of company among the lazy, the self-interested, the risk-aversive and the akratic. But the phenomenon is more widespread than this gallery of rogues might suggest. Quite generally, we do not think it particularly exceptional to find gaps between our value judgements and our desires. Gaps come in different sizes. A gap may be small or it may be large. In the case of desire and value judgement, the gap will be large if the value attributed is large while the corresponding desire is small. The larger the gap, the more out of kilter one’s desires are with one’s value judgements. Suppose we grant the possibility of a gap here, and that it can vary in size. By continuity, there seems nothing to block the logical possibility that one have a maximal degree of belief that something is good and yet have the absolutely minimal corresponding desire. If this is right then the cognitivist should eschew any version of internalism which entails that a maximal belief-desire gap is not even possible. Cognitivism is well equipped to account for these gaps – of the various ways in which one’s value judgements can come apart from one’s desires – even without a generalized thesis of independence of belief and desire. By contrast, simple versions of non-cognitivism, like expressivism, seem badly placed to save the phenomena. If a value judgement is simply the expression of a desire, rather than of a genuine belief, then it is a mystery how one can sincerely make such a judgement and fail to have the appropriate desire. In such cases the expressivist will have to deny that the judgement is sincere, or heartfelt, or something like that. If one makes a show of endorsing the judgement, it may be an attempt to deceive some audience into believing that one has desires that one in fact lacks, perhaps an audience consisting of oneself. But as a general rule, that seems far-fetched. The phenomena at issue here, the gaps between one’s desires and one’s judgements of value, are just too familiar to all of us. The upshot is that a strong Humean thesis of belief-desire independence seems to sit more happily with cognitivism about evaluative judgements and a rejection of judgement internalism than it does with non-cognitivism.

5.8 The merit connection If a cognitivist eschews judgement internalism, what can he then say about the puzzling asymmetry? What is the connection between value and desire that explains why it is odd for me to affirm the following judgement even when the propositions expressed are true? P is really very good, but I don’t desire P at all. Here’s a possibility. Many value theorists with a realist bent have suggested variants of a thesis about the relation between value and desire that I will dub the merit connection. Briefly, the thesis is that the good is what merits desire, or some other closely related motivating state. Given that something is good it is appropriate, or fitting, to desire it. Variations on this theme can be stated

Graham Oddie

131

using closely related attitudes like approval and love. This thesis, broadly construed, can be found in Aristotle, Franz Brentano, Alexius Meinong, Max Scheler, C. D. Broad, and Roderick Chisholm, and more recently Kevin Mulligan.4 C. D. Broad stated the merit connection in this way: I’m not sure that ‘X is good’ could not be defined as meaning that X is such that it would be a fitting object of desire to any mind which had an adequate idea of its non-ethical characteristics. (Broad, 1930, p. 283) The right-hand side might be necessary and sufficient for the left, and if so we can treat the right-hand side as a definition of good. But even if we do treat it as a definition, it is clearly not a definition of good in non-evaluative terms. The term fitting – like the terms merit, worthy of and appropriate – is clearly evaluative. And as evaluative terms go, fitting seems a lot less basic than good. It would be much more natural to define what it takes for a desire to be fitting, or appropriate, in terms of the apparently more fundamental notion of goodness. But we can put to one side the issue of definition and reduction, and simply note that the thesis posits a necessary connection between value and desire. We can state the thesis in a way that allows for degrees of goodness: The merit connection P is good just to the degree that P merits being desired (or: just to the degree that it is appropriate for P to be desired) by anyone with an adequate idea of P’s non-value characteristics. Given the merit connection, and given that it is a matter of necessity, the following also holds of necessity: P is very good if and only if P is worthy of being strongly desired by anyone with an adequate idea of its non-value characteristics. But then someone who utters P is very good, but I don’t desire P at all. is obliged to accept: P is worthy of being strongly desired by anyone with an adequate idea of its non-value characteristics, but I don’t desire P at all. There is a tension here, but is the tension more obvious here than in the original? Perhaps it’s a little more obvious. Both clauses now make explicit claims

132

Experiences of Value

about desire, but note that the first clause isn’t a claim that the second clause repudiates. The first clause is about the value of a certain desire, and the second clause about the non-existence of that desire. The tension between acknowledging the value of desiring P while also acknowledging the actual lack of desire for P is rather like the tension between acknowledging the value of P while also acknowledging my actual lack of desire for P. In both cases, I acknowledge that I am not responding appropriately to value, but the oddness of that acknowledgement has not really been explained.

5.9 The value experience thesis There is another thesis about the connection between value and desire which explains the asymmetry somewhat more naturally. Here I state the thesis with a brief defence, and show how it explains the asymmetry. Hume is clearly right that we experience a rich array of values. We experience the world as replete with goods and evils: pleasure and pain, joy and misery, kindness and callousness, graciousness and greed, the beauty of Britten’s Billy Budd and the banality of Britney’s Blackout. The value of some of these – like intense bodily pleasures, or the unexpected kindness of a stranger – forces itself upon us. Their value lies on or near the very surface of appearance. The value of others – like the beauty of Billy Budd or of forgiving one’s enemies – may be not so easily discerned. They may lie some distance from the surface, to be discovered only through repeated exposure and the refining of specific sensibilities. These experiences are surely connected with reasonable beliefs about value, just as our experiences of material objects are connected with reasonable beliefs about those. Consider the normal relation between experience and belief. When I have a visual experience of a bright red rose, say, the rose presents itself to me as bright red. It seems or appears bright red to me in that sense of seems which is tied to perceptual presentation rather than to belief. Normally, of course, the visual experience of a bright red rose – that is to say, the rose’s appearing bright red to me – gives me a reason to believe that the rose really is bright red. But it is a defeasible reason in the sense that any number of conditions might crop up which indicate that, despite its appearance, the rose is not really red after all. Conditions may not be ideal for seeing, or my visual apparatus might be in an impaired condition, or I might be on drugs, whatever. If I discover any of these defeaters I might well believe that the rose is not bright red after all, or not nearly as red as it seems, despite the appearances and despite the fact that the rose persists in appearing bright red to me. If the rose can seem to me (is experienced by me as) bright red, even though I believe it not to be bright red, then there must be seemings which do not entail believing. I can also have the belief, of course, without the seeming. P’s seeming to me to be the case (in this sense) and my believing P to be the case are thus logically independent. That is what I mean by the non-doxastic sense of seems.

Graham Oddie

133

Grant, then, that we do have experiences of value – value appearances, or value-seemings. If there are genuine experiences of value they could stand to values as ordinary perceptual experiences stand to the objects of perceptual experience. An experience of the goodness of P, say, would be the state of P’s seeming (appearing, presenting itself as) good where this seeming is an experiential, non-doxastic take on the value of P. Given an appearance of the goodness of P then, by analogy with the perceptual case, it would give me a reason to believe that P is good. But, again by analogy, it would be a defeasible reason, in the sense that the totality of information available to me might indicate that P is not good, despite its seeming to me to be good. Either conditions may not be favourable for my experiencing P’s value, or my ability to experience value might be impaired in some way. If I uncover any such defeaters, I might well come to the conclusion that P is not good after all, or not nearly as good as it seems – despite both the initial appearances, and despite the fact that P may persist in seeming good to me. What exactly are these appearances of the good? We could of course leave things here, and be a bit mystified about the nature of these appearances. Alternatively we could rummage around in the collection of familiar mental states, and see whether any of them fit the characterization. There may be more than one – for example, approvals, emotions, feelings, or maybe combinations of these with other states – but consider desires. When I desire that P, P has a certain magnetic appeal for me. It presents itself to me as something needing to be pursued, or promoted, or embraced. Now the good is that which needs to be pursued, or promoted, or embraced. So, my desire that P involves P’s seeming good (seeming to be worth pursuing, promoting, embracing). So the desire that P looks as though it just is the experience of P as good. The desire that P (which, on this proposal would be the experience of P as good) is clearly not the belief that P is good (just as the experience of the rose as red is not the belief that the rose is red). Nor does my desiring P – my experiencing P as good – entail that I have the belief that P is good, just as my experiencing the rose as bright red does not entail that I believe it to be bright red. The belief that the rose is bright red is something which one could have without experiencing the rose as bright red, and something which one might not have even if one were having the experience of the rose as bright red. Similarly, the belief that P is good is something which one could have without experiencing P as good, and something which one might not have even if one were experiencing P as good. Let’s gather these two theses (that there are experiences of value, which can serve as reasons for evaluative beliefs, and that desires are such experiences of value) into the following conjecture: The value experience thesis The desire that P is P’s seeming or appearing good to me (P’s being experienced as good by me).

134

Experiences of Value

If the value experience thesis is right, experiences of value are not exotic states which some otherwise mysterious faculty of ‘value intuition’ delivers. They are mundane states with which we are totally familiar – desires and aversions. Experiences of value are, in short, Hume’s direct passions. I have said desires are experiences of value. It does not follow that experiences of value are all and one desires. Are they? I want to leave that open here. Maybe all experiences of value are either desires or contain a desiderative component. Alternatively, however, desire may be just one way of experiencing value, as visual experiences are just one way of experiencing shapes. (One might have a tactile experience of a shape or even an auditory one – witness the bats and the dolphins.) It is quite possible that value experience is a determinable of which desire is a determinate. I myself think that desires stand to the goodness of states of affairs as visual experiences stand to colours. Visual experiences are the standard and paradigmatic way of experiencing colours of objects (although one might dream up other ways of doing so). Likewise, desires are the way we paradigmatically experience the value of states of affairs. The value of other categories of object (persons, or works of art for example) might demand a different determinate realization of value experience. It may be that the way one experiences the value of a person is through loving; or of a work of art, through appreciating. Here I will leave this possibility open, as well as the possibility that these also reduce to congeries of desires. The value experience thesis is developed and defended in depth in Value, Reality and Desire (Oddie (2005)). My main purpose here, though, is to put the conjecture to work to explain the puzzling asymmetry. The point is really quite a simple one. While the belief that P is good does not necessitate the desire that P, there is clearly a connection between the belief and the desire, one which typically makes it a bit odd to espouse the one without having the other. The oddness can be explicated as an instance of what I will call the shadow of Moore’s paradox.

5.10 Moore’s paradox and its shadow We start with the well-known paradox, discovered by G.E. Moore, which bears a superficial similarity to the asymmetry problem. Consider the following schema: Q is true, but he doesn’t believe that Q. There are many instances of this schema that someone could truly utter while pointing at me (viz. most of the true propositions can be substituted for Q). Further it is entirely unproblematic for you to point at me and affirm of some particular proposition that it is true, but that he doesn’t believe it. However, it is highly problematic for me to affirm: Q is true, but I don’t believe that Q ,

Graham Oddie

135

even while I can happily concede that almost all true propositions are such that I don’t believe them. A pretty good explanation for this asymmetry involves the notion of conversational implicature. When I assert Q , I not only affirm Q itself, but I thereby express the proposition that I believe Q. Although in affirming Q I do not explicitly say that I believe Q, I do thereby express the fact that I believe Q. It is not an implication of what I affirm that I believe Q, but it is an implication of my affirming Q. By affirming Q I conversationally imply that I believe that Q because the proposition that X affirms Q sincerely does entail that X believes Q. So in my original affirmation what I say implies that I don’t believe Q and conversationally implies that I do believe Q. That’s why it is odd. Let’s now turn to a related phenomenon, which as far as I know has not been discussed. It is the shadow of Moore’s paradox. You can assert, unproblematically, the following of me: Q is true although Q doesn’t seem true to him. However it does sound a bit odd for me to say: Q is true although Q doesn’t seem true to me. Now this utterance isn’t as obviously paradoxical as the Moorean assertion. But in typical circumstances there is something a little bit odd about it. This oddness can also be explained by the notion of conversational implicature. When I assert, under typical conditions, that Q is true, there is a conversational implicature that I believe that Q is true (Moore’s insight), but there also seems to be some sort of presumption that I have evidence for Q (typically I do not assert things for which I have no evidence) and one standard source of such evidence is Q’s seeming true to me. Of course, even if Q seems to me to be true, this is still only a defeasible reason for believing Q to be true. I can see a rose as bright red even though it is not (I might be looking at a white rose through red-tinted glasses). And I might not see it as bright red even though it is (I might have blue-tinted glasses on). In the latter case I might well affirm an instance of the odd-sounding schema above. However, for the most part, I base my beliefs about the visual properties of things on the way they present themselves to me in experience, and so, absent odd conditions, an assertion like ‘the rose is bright red but doesn’t seem at all bright red to me’ is a little bit jarring. The asymmetry between the first and third person assertions in this case is not nearly as pronounced as it is in Moore’s paradox. We might call it the shadow of Moore’s paradox. Substitute P is good for Q in the schema, and abbreviate P is good is true to P is good: P is good but P doesn’t seem good to him.

136

Experiences of Value

That’s a claim you might assert unproblematically of me. But it sounds a bit odd for me to say: P is good, but P doesn’t seem good to me. When I assert that P is good, there is the standard conversational implicature that I believe P is good. But if P doesn’t seem good to me then I don’t have a, perhaps the, standard reason for believing that it is good. So much simply lies in Moore’s shadow. Now let’s add the value experience thesis to this. My desire that P just is P’s seeming good to me. So the above is equivalent to: P is good but I don’t desire that P. Whatever oddness attaches to this can be explained in terms of the oddness of the proposition above which (according to the value experience thesis) is necessarily equivalent to it, namely: P is good, but it doesn’t seem good to me. The utterance is an acknowledgement that my experience of goodness (how good things seem to me) and my evaluative beliefs (how good I believe things to be) have come apart. That’s not impossible, of course. Indeed, it is not so very unusual. But acknowledging a belief that something is the case, while also acknowledging things don’t seem that way to me, does require a bit of explaining. The table summarizes the explanation of the puzzling asymmetry. Not odd Moore’s paradox Moore’s shadow Substitute P is good for Q Substitute X desires that P for P seems good to X.

Odd

Q, but he doesn’t believe Q. Q, but I don’t believe Q. Q, but it doesn’t seem that Q, but it doesn’t seem that Q Q to him. to me. P is good, but P doesn’t P is good, but P doesn’t seem seem good to him. good to me. P is good, but he doesn’t desire P.

P is good, but I don’t desire P.

The value experience thesis thus has at least one attractive consequence – it enables the cognitivist to explain the puzzling asymmetry without violating independence. If desires are experiences of the value of states of affairs, then it follows immediately, even on the standard Humean account, that experiences of value are intrinsically motivating. The experience of P as good would necessarily motivate one to promote, pursue or embrace P, because it would simply be the desire that P, and desires are intrinsically motivating. It is not the evaluative belief that P is good that is intrinsically motivating, and so the independence of belief and desire is not threatened.

Graham Oddie

137

However, there is an appropriate internal connection between a desire and an evaluative judgement. The desire that P is the experience of P’s being good. An experience should not be confused with a belief, not even with the belief which the experience would, under normal circumstances, give one some reason to adopt. But if desires are the experiences which ground our beliefs about the good, then that would also explain why value judgements – which clearly involve beliefs about the good – do not typically float entirely freely of motivation. So by means of the value experience thesis we can vindicate the essential insight driving internalism. Suppose I know that P is good and I know so by direct acquaintance with the goodness of P. That is to say, P is genuinely good, I believe P is good, and I believe so because I directly experience P as good. My direct experience of P’s goodness is a reason (and, let’s suppose, a good one) for my believing that P is good. Then, given the value experience thesis, my experiencing P as good is my desire that P. If I know the goodness of P in this way, by direct perceptual acquaintance, then ipso facto I am necessarily motivated to pursue or preserve P. So the value experience thesis explains the puzzling asymmetry, allows for the appropriate valuedesire gaps, and also explains the internalist insight that certain kinds of knowledge of value would be intrinsically – that is, necessarily – motivating.

5.11 The perspectival nature of experience The merit connection, recall, says that the good is what merits desire. If this were right, then presumably if I were desiring properly or adequately or ideally, I would desire things just to the degree that they are good. But given the value experience thesis, that means that ideally I should experience value-reality as it is in itself. Anything less than an isomorphism between value and desire would be a defect in a responder to value. Call this the isomorphic response thesis: that for one to be experiencing value appropriately one’s desires must be perfectly isomorphic to the structure of value. That seems very far-fetched. Something has to give. Consider two cases. Imagine I have a badly fractured limb caused by a skiing accident, part of the femur is protruding through the skin, and the thing is excruciatingly painful. A stranger skiing the same slope just behind me has suffered a similar fracture, and he is now lying in the snow alongside me, suffering what appears to be the same degree of pain. I would like his pain to cease, naturally, but I am even more desirous of the cessation of my own pain. When the first-aid team appears it turns out they have only one shot of morphine left. Since the relief of the stranger’s pain is just as valuable as the relief of my pain, the merit principle demands that I be indifferent as to whose pain is treated. But I am not. Or imagine I am in that unhappy situation in which my daughter is drowning a hundred yards from where some other girl is drowning, some

138

Experiences of Value

stranger’s daughter. As is the norm in such cases, I can’t save both. I certainly would like to be able to save both, but I cannot do so, and not surprisingly I have a stronger desire to save my daughter while the other girl drowns, rather than to save the other girl while my daughter drowns. Now I know that, in all probability, my daughter’s life is no more (or less) valuable than the life of the stranger’s daughter, or that my happiness is no more (or less) valuable than the stranger’s. But even though I know that, I cannot desire to save the stranger’s daughter as much as I desire to save my own daughter. We face a dilemma here. Given the isomorphic response thesis, I should be indifferent between saving my daughter and saving the stranger’s daughter. If that’s what value demands of desire then it is a defect in me, as a valuer and desirer, that I am not completely indifferent about who I save. And that seems wrong. However, if it is not a defect then it is hard to see how my desires give me any reason at all for corresponding value beliefs. If it is quite in order for me to want to save my daughter (or to want my pain to cease) more than I want to save the stranger’s daughter (or to want the stranger’s pain to cease), even though the one is not more valuable than the other, how can my desires be construed as any kind of evidence of value? In neither case does it strike me as at all plausible that I should be indifferent, that my desires (or the stranger’s) should track the actual values. Rather, it is perfectly in order for me to desire the cessation of my pain more than I desire the cessation of the stranger’s pain. And it is also perfectly in order for me to desire to save my daughter somewhat more than I desire to save the stranger’s daughter. In fact I would be regarded as something of a monster if I had to toss a coin to decide which of them to save. And it isn’t clear how I could have a pain just as bad as the stranger’s and yet have identical desires concerning the cessation of both. If this is right then there is something very wrong with the merit connection, at least as I have stated it. Suppose desires are experiences of value, just as sensory perceptions are experiences of the material world. Suppose I look into the heavens at the setting sun and the rising moon. The sun is much bigger than the moon, of course, but the moon looks the bigger of the two. That’s because the sun is much further away. Are my experiences here defective? Would my perceptual apparatus be better if it always yielded an appearance of the sun as vastly bigger than the moon, no matter which position I took up with respect to them? That doesn’t seem right. I am having a perfectly apt experience of the two celestial bodies given that I am located on earth. That is exactly how one would expect those two bodies to look from earth to a human being whose perceptual apparatus is in perfect working order. Clearly you cannot be expected to perceive things exactly as they are no matter what your situation is with respect to them. Nor should you believe that things are exactly as you perceive them to be no matter how you are situated with respect to them. The relation between experience and the world and one’s beliefs is more complex and interesting than that.

Graham Oddie

139

If desires are experiences of value, and valuers are differently situated with respect to value, then we should expect something analogous to the phenomenon of perspective. And the two cases illustrate this. While my pain and the stranger’s pain are equally bad, I am clearly situated very differently with respect to my pain than I am with respect to the stranger’s. In fact, part of what makes pain bad is surely the fact that its bearer badly wants it to stop, and that is clearly not under any direct control. (That’s why we need morphine.) I cannot be expected to detach myself from that desire and be indifferent between the cessation of my pain and the cessation of the stranger’s pain. So, it is perfectly in order for me to want my pain to stop more than I want the stranger’s pain to stop. What about the drowning girls? Just as with pain, I am very differently situated with respect to my daughter and her survival than I am with respect to the stranger’s daughter and her survival. I love my daughter, fiercely, in a way that completely precludes me from being indifferent between her death and the death of a stranger’s daughter. And it would be absurd to demand that my love for my daughter not inform the strength of my desire for her to go on living. It is not that I don’t care about the stranger’s pain, or the death of his daughter. I am not indifferent to either of them. I certainly believe them to be as bad, respectively, as my own pain, and the death of my own daughter, and in addition I do desire their well-being. But given that I care about myself and my daughter more than I do about the stranger and his daughter, I cannot be indifferent as to where good and evil flow. Consider size and perception of size again. An observer, even an ideal one, has to take up a wide range of different observation posts in the world, and his perceptual apparatus should be appropriately sensitive to variations in perspective. Even to an ideal observer in my current position, the sun should appear smaller than the moon, even if to an ideal observer equidistant from both, but within viewing range, they should look just the reverse. The moon should look bigger than the sun from the vantage point of earth and it is no defect of the observer that it appears that way to him. So, even though the facts about size are observer-independent, experiences of size are appropriately observer-dependent, since they depend on the situation, point of view, or perspective of the observer. Analogously, even though facts about value are valuer-neutral, experiences of value can and should be valuer-relative, since they depend on exactly where the valuer is situated with respect to the objects of desire. In general we do not require that a perceiver’s experiences be isomorphic to reality for him to be perceiving appropriately. Analogously we should not require that a valuer’s desires be isomorphic to value for her to be responding appropriately. Now, if desire has a perspectival component – as the two examples suggest – then the merit connection, as stated, has to go. So we need not demand that a person, adequately responding to value, desire things exactly to the degree that they are good.

140

Experiences of Value

Interestingly Hume himself noted the heavily perspectival nature of value experiences in the passages quoted at the start of this essay. ’Tis as little contrary to reason to prefer even my own acknowledg’d lesser good to my greater, and have a more ardent affection for the former than the latter. A trivial good may, from certain circumstances, produce a desire superior to what arises from the greatest and most valuable enjoyment; nor is there anything more extraordinary in this, than in mechanics to see one pound weight raise up a hundred by advantage of its situation. (T, 2.3.3.6/416) Hume is using this to press his case for the thesis that the passions cannot be contradictory with the truth. But that does not quite follow. The moon looks larger than the sun, and the content of that experience does indeed contradict the truth. What makes it, nevertheless, a perfectly appropriate experience to have is the fact that the sun is much further away. So it is entirely ‘reasonable’ or ‘fitting’ to experience things that way, even if the experience does in fact contradict the truth of the matter. If desires are experiences of value, and experiences of value not only can but should be perspectival, then this goes a long way in explaining away what is otherwise a deep problem for realism about value. We can embrace a world of agent-neutral value without condemning as defective or perverse the persistently agent-relative nature of desire. But also, the perspectival nature of value experiences no more undermines their claim to yield a glimpse of value, than the perspectival nature of ordinary perception undermines its claim to yield data about the material world.

5.12

Further work

The value experience thesis has done some good work here, dissolving some puzzles and explaining the attractions of various Humean themes. It can do more. It can explain the attractions of the Meinongian thesis that emotions provide us with epistemic access to value, just to the extent that they do. Most emotions have a desiderative component, and thus incorporate value data. But it can also explain why emotions are a tricky source of value data. It can explain the role of Frankfurtian higher-order desiring in our attempts to get a firmer grip on value. It can explain the value of having accurate experiences of value. It also raises some problems which have to be tackled. Is value idealism – the idea that value supervenes on value experiences, on desires – an option? How do we refine our desires in extending and improving our value beliefs? What can we say about perverse desires which presumably lead us away from the truth about value rather than reveal it? What can we say about the phenomenon of disappointment, of finding desire-satisfaction to be bad rather than good, and so on? To do all this, however, one would need to write a book.5

Graham Oddie

141

Notes * This essay draws heavily on Chapter 2 and parts of Chapters 3 and 8 of my 2005 book Value, Reality and Desire (Oxford, Oxford University Press). I wish to thank the editor for granting me permission to reprint material from the book here. 1. On this construal cognitivism is the conjunction of two distinct theses that can be pried apart: (1) that value judgements have propositional content, and (2) that endorsing a value judgement is a matter of affirming the truth of the associated proposition. Mark Kalderon (2005) has suggested that a fictionalist is a non-cognitivist who accepts the first of these but rejects the second. For the fictionalist endorsement is some attitude to the value proposition other than belief. See Demetriou and Oddie (2007) for an analysis and discussion. 2. See Oddie (2005) for an extended argument. 3. Clearly, desire sets can embody incoherencies the elimination of which reason demands – see Chapters 4 and 5 of Oddie (2005) for an extended discussion. 4. Aristotle (1962), Brentano (1969), Meinong (1972), Scheler (1973), Broad (1930), Chisholm (1986), Mulligan (1998). A recent defence of this kind of view can be found in Johnston (2001). 5. And the book is, of course, Value Reality and Desire (Oddie (2005)).

6 Hume and the Debate on ‘Motivating Reasons’ Constantine Sandis*

It is a heuristic maxim that the truth lies not in one of the two disputed views but in some third possibility which has not yet been thought of, which we can only discover by rejecting something assumed as obvious by both disputants. (Ramsey, 1931, pp. 115–6)

Introduction This essay offers a new interpretation of Hume’s account of motivation before relating it to certain disputes in modern moral psychology. The essay is divided into three parts. First, I lay down some general distinctions in the so-called theory of motivation (TOM), introducing two related but distinct ongoing debates. Next, in the middle and largest parts of the essay, I focus on what Hume has to say on these matters, concluding that the standard map of available positions leaves no space for his view, as it rejects an assumption shared by all concerned. Finally, I demonstrate how the disputes most central to the debates we began with evaporate once we follow Hume in rejecting this shared assumption, taking this to count in his favour.

6.1 The motivational map The term ‘motivation’ is most widely used to refer to either: (a) Whatever it is that motivates us (often misleadingly termed our ‘motives’ or ‘motivators’). or (b) Our being motivated (by (a)). Accordingly, TOM is prima facie a central part of what used to be called ‘philosophical psychology’ whose principle focus is the provision of a satisfactory account of (b) in terms of (a). While this project is more closely 142

Constantine Sandis

143

tied to moral psychology than to the philosophy of cognitive science which currently passes as ‘philosophy of psychology’ it is not concerned with the normative question of what ought to motivate us to act, but only with what possibly could and/or actually does. There are numerous debates within TOM but I shall here restrict myself to just two, both concerning the nature of (a). The first of these is between Humeanism and Anti-Humeanism (cf. Smith, 1994), the second between what has been called Psychologism and Non-Psychologism (cf. Dancy, 2000). What I shall argue will, for the most part, be of mere passing interest to the causalism/anti-causalism controversy and the disputes between various forms of internalism and externalism. The definitions that follow are fairly standard: Humeanism: The view that we are motivated by reasons for acting that are constituted by our desires and beliefs. Proponents include Mele (1992), Smith (1994) and Lenman (1996). Anti-Humeanism: The view that we are motivated by reasons for acting that need only be constituted by beliefs. For example Nagel (1970), Foot (1972b), McDowell (1978 &1979) and McNaughton (1988). Pure Anti-Humeanism: The view that we are motivated by reasons for acting that are constituted purely by beliefs. Proponents include Parfit (1997) and Dancy (1993 & 2000). Psychologism: The view that we are motivated by reasons for acting that are psychologically real, and therefore, of an entirely different ontological category to normative reasons which are conceived of as truths, facts, states of affairs, or propositions. Proponents include Nagel (1970), and Mele (1992), Smith (1994) and Brink (1997). Non-Psychologism: We are motivated by reasons for acting that are not psychologically real but, rather, of the same ontological category as normative reasons. Proponents include Dancy (1995 & 2000) and Collins (1997); hints of this view may also be found in Williams (1980) and Nagel (1997). The non-psychologism about normative reasons plugged into this view is defended by Broome (1997), Raz (1986), Scanlon (1999), Dancy (2000) and Quinn (1993). Throughout these disagreements all disputants share a common assumption, namely the view that the things that motivate us are (at times by definition) reasons for which we act. Hence TOM is thought to be a theory of motivating reasons for action. We might characterise this assumption as follows: The Common View (CV): TOM is a theory about the reasons we act for.

144 Hume and the Debate on ‘Motivating Reasons’

I shall eventually conclude that Hume cannot partake in either of the two aforementioned debates within TOM precisely because he rejects CV. Still, I begin by trying to place him there.

6.2 Hume on reason and influence If Hume is to at all resemble a Humean, he will insist that desires are a constitutive part of what motivates us. Belief alone, so the official Humean line goes, is inert. Here are the passages most commonly quoted in defence of this understanding of Hume (pace most modern interpreters Hume only ever capitalises the word ‘reason’ when it starts a new sentence): [R]eason alone can never be a motive to any action of the will (T, 2.3.3.1/413); Abstract or demonstrative reasoning, therefore, never influences any of our actions, but only as it directs our judgment concerning causes and effects (T, 2.3.3.2/414); impulse arises not from reason but is only directed by it (T, 2.3.3.3/414); reason alone can never produce any action, or give rise to volition … the same faculty is as incapable of preventing volition (T, 2.3.3.4/414 5); reason has no influence on our passions (T, 3.1.1.7/457); I have prov’d, that reason is perfectly inert, and can never either prevent or produce any action of affection (T, 3.1.1.8/458); Reason is wholly inactive (T, 3.1.1.10/458); The action may cause a judgment, or may be obliquely caused by one, when the judgment concurs with a passion (T, 3.1.1.11/459). There is much tension between these varied formulations of the first premise of his Influence Argument which seem to range from the claim that reason is wholly inactive to the suggestion that it can (albeit obliquely) cause action. Critics are divided over the issue of whether or not Hume is contradicting himself (cf. Botros, 2006, Chapter 1 for various exegetical positions, including her own arguments in favour of the ‘contradiction’ view; in what follows I shall retain neutrality here by keeping the term ‘alone’ in brackets). They all agree, however, that by ‘reason’ Hume means the same thing as ‘belief’ or ‘judgment’ (cf. Mackie, 1980, p. 53 for a typical example of this near-universal assumption). Yet how true is this? Beliefs, for Hume, are lively ideas, which in turn are copies of impressions: An opinion or belief is nothing but a strong and lively idea deriv’d from a present impression related to it. (T, 1.3.8.16/119; cf. T, 1.1.1.1/1–2) In the section ‘Of the Influence of Belief’ Hume thereby assigns the same effects to them. [T]he ideas of those objects, which we believe either are or will be existent, produce in a lesser degree the same effect with those impressions,

Constantine Sandis

145

which are immediately present to the senses and perception. The effect, then, of belief is to raise up a simple idea to an equality with our impressions, and bestow on it a like influence on the passions. (T, 1.3.10.3/119) The context makes it clear that by ‘like influence’ Hume means ‘brings about the same effect to a lesser degree’, the degree in question being proportionate to the degree to which the idea in question is fainter to the impression it is a copy of, beliefs being the most lively of all ideas. Moreover, while Hume contrasts reason with sentiment (for example in EPM, 1.3/134) he clearly identifies belief with a kind of sentiment: Belief is nothing but a peculiar feeling … or sentiment. … ’Tis felt rather than conceived, and approaches the impression, from which it is deriv’d, in its force and influence … it is something felt by the mind which distinguishes the ideas of judgments from the ideas of the imagination … and renders them the governing principles of all our actions. (T, App. 3/624; cf. Korsgaard, 1997a, p. 24) There is nothing non-cognitivist in this construal of belief. On the contrary, Hume clearly states that beliefs are capable of being true or false (T, 3.1.1.12/459, quoted further below). More importantly, we have not yet seen anything which suggests that Hume thought that no belief could motivate, whether alone or otherwise. Indeed, I shall later be claiming that this is precisely what Hume takes moral beliefs to be capable of doing. Hume’s claim that reason cannot motivate (alone), it would begin to appear, cannot be as straightforward as the thesis that beliefs cannot motivate, let alone the Humean variety which specifies desires or pro-attitudes of some related kind as the missing ingredient (cf. Snare 1991:55). We might therefore do better to understand it as the (sometimes doubly) qualified claim that (only) those beliefs derived from reason (alone) cannot motivate (alone). So understood, it states that there is a subset of beliefs – viz. those derived from reason (alone) – whose motivational power is at best constrained and at worst nil. Yet even this interpretation may be restricting the role of belief further than Hume intended, for two related reasons. The first is that it remains neutral on the issue of whether Hume takes all action to be produced by passions, or whether he allows that, at least sometimes, beliefs may be said to move us to action by exciting a passion that is not derived from reason (alone). In the latter kind of case, desires need not be anything more than enablers of action (cf. Dancy 2000:127ff.). We should, after all, take care not to conflate the claim that we cannot only be moved to act unless we already have certain passions with the claim that the passions in question are always part of our motivation.

146 Hume and the Debate on ‘Motivating Reasons’

Plato’s Socrates warns against the basic fallacy that such a conflation would involve: τὸ γὰρ µὴ διελέσθαι οἷόν τ᾽ εἶναι ὅτι ἄλλο µέν τί ἐστι τὸ αἴτιον τῷ ὄντι, ἄλλο δὲ ἐκεῖνο ἄνευ οὗ τὸ αἴτιον οὐκ ἄν ποτ᾽ εἴη αἴτιον

(Phaedo, 99b, translated by R. Hackforth as ‘fancy not being able to distinguish between the cause of a thing and that without which the cause would not be a cause!’ and by D. Gallop as ‘fancy being unable to distinguish between two things: the reason proper and that without which the reason could never be a reason!’; cf. Dancy 2004: 45) Is Hume guilty of such a fallacy? The answer partly depends on how we disambiguate his phrase ‘concurs with a passion’. A clue is offered in the following passage: [T]he impulse arises not from reason, but is only directed by it. ’Tis from the prospect of pain or pleasure that the aversion or propensity arises towards any object. (T, 2.3.3.3/414) The impulse in question may not arise from reason yet this does not preclude the idea that once it is there, reason may move us to take various means as ways of satisfying it. Korsgaard describes such cases as ones where Hume makes it clear that ‘we can be moved by a judgement about a probable passion’ (Korsgaard 2009: 64, n.6). To this extent, even beliefs derived from reason alone may be said to motivate us (alone), though Hume would not himself have put it this way, not least because he takes this to be a case of directing action rather than producing it, which renders the causation in question oblique. This brings us to the second reason for being cautious in our interpretation which is that Hume does not employ modern psychological terms such as ‘motivation’, ‘motivating’ and ‘motivate(s)’. He only talks of ‘motives’ which he introduces as things that ‘produce’ or ‘influence’ action. The motivation of action, however, is a different phenomenon from its production (cf. Sandis, 2008a). Indeed, we are frequently motivated to perform actions that never take place. Strictly speaking, then, Hume’s view is closer to the claim that only beliefs derived from reason (alone) cannot (alone) produce an action, a thesis which is relatively weak given his presumption that actions may be caused by anything which regularly precedes them so long as it is true that ‘if the first object had not been, the second never had existed’ (EHU, 8.2.4/76; cf. T, 1.3.14.21/166). So what is it for a belief to be derived from reason (alone)? To answer this we must return to our initial question which was ‘what does Hume mean by “reason?”’ Here is what ‘Hume’s Fork’ tells us:

Constantine Sandis

147

Reason is the discovery of truth or falsehood. Truth or falsehood consists in an agreement or disagreement either to the real relations of ideas, or to real existence and matter of fact. Whatever, therefore, is not susceptible of this agreement or disagreement, is incapable of being true or false, and can never be an object of our reason. (T, 3.1.1.9/458; cf. T, 2.3.2/413) So portrayed, reason is concerned with beliefs and not desires. Hume rejects the Aristotelian account of emotion as a cognitive phenomenon, hence his claim that ‘passions can be contrary to reason only so far as they are accompany’d with some judgment or opinion’ which swiftly leads him to the infamous remark about it not being contrary to reason ‘to prefer the destruction of the whole world to the scratching of my finger’ (T, 2.3.3.6./416). Yet this is not to say that the objects of reason (i.e. beliefs) are inert but only that a subset of them, viz. those beliefs that are themselves reached through reason (alone), do not influence action (alone). In sum, beliefs for Hume are all sentiments, some of which are derived from reason alone and some from reason combined with some other sentiment(s). This helps to explain those instances of the doubly qualified claim that it is only beliefs derived from reason alone that cannot influence alone. Mutatis mutandis, it could also account for the reasoning of Influence Argument: Morals excite passions and produce or prevent actions. Reason of itself is utterly impotent in this matter. The rules of morality therefore, are not conclusions of our reason. (T, 3.1.1.6/457, my emphasis; cf. EPM, 1.5–9/135–7) This passage is standardly interpreted as an argument for moral non-cognitivism, the form of non-cognitivism usually attributed to Hume being one which asserts that moral judgements are expressions of our desires (cf. Blackburn 1996, p. 180). On such readings, the conclusion of the argument is meant to be that moral judgements are not beliefs but, rather, expressions of desire. We have already seen, however, that Hume allows that beliefs can can excite passions and actions (alone or otherwise). It would make no sense for Hume to consequently suggest that since moral judgements can influence the passions on their own they cannot be cognitive (i.e. beliefs). I propose, instead, to reconstruct the Influence Argument as follows: (P1) Moral beliefs produce action (alone). (P2) Beliefs that are conclusions of reason (alone) do not produce action (alone). (C) Moral beliefs are not derived from reason (alone).

148 Hume and the Debate on ‘Motivating Reasons’

So conceived, the argument aims only to show that since (P1) moral judgements can move us (alone) and (P2) beliefs derived from reason (alone) cannot produce action (alone) we should conclude that (C) moral judgements are not derived from reason (alone), whatever that turns out to mean. But why should this amount to anything more than the claim that it ‘is in vain to pretend, that morality is discovered only by a deduction of reason’? (T, 3.1.1.7/457) since ‘reason … can never be the source of so active a principle as conscience, or a sense of morals’ (T, 3.1.1.10/458)? There is nothing non-cognitivist in these remarks. As Francis Snare (1991: Ch. 2) has pointed out it could, for example, be read as an anti-realist remark and indeed – for all he has said so far – Hume may even be some kind of intuitionist realist (Snare ultimately interprets Hume as a non-cognitivist but rejects the notion that this position is conclusively established by any formulation of the Influence Argument). In his splendid book, Snare also notes that – though valid (and possibly also true) – this syllogism is question-begging. He offers the following ‘sentimental zoology’ by analogy, noting that ‘if we substitute “whale” for “trout” in the parody we get a valid argument with true premises and conclusion, but an argument just as question-begging’ (1991, p. 51): (P1*) No sort of fish interests me. (P2*) Trout do interest me. (C*) Trout are not fish. Unless (P1*) is an a priori truth, we cannot dismiss the possibility that (P2*) is not a counter-example to it. The same logic holds for the preceding syllogism regarding moral beliefs (above) and yet (P1) appears to be an empirical proposition. In fact so does (P2), but if Hume is to reach the conclusion that no moral belief can be cognitive (as opposed to the considerably weaker thesis that all the moral beliefs he happens to have observed – or, at best, that we all happen to have had so far – are non-cognitive) (P2) must be a necessary truth. This would only be plausible if we could somehow conceive of it as a logical remark concerning the limits of reason (perhaps resulting from the narrow scope that Hume restricts it to) rather than as a statement regarding the observed motivational inertness of belief. Either way, there are independent reasons for thinking that the Influence Argument is compatible with cognitivist internalism concerning the relation of moral judgements to the production of action (of which (P1) could even be a weak statement of), It is to these that I now turn. Beliefs for Hume, we have already seen, are ‘felt sentiments’, so it would be prudent to read the Influence Argument in the light of this general ‘sentimentalism’. On this picture to have a belief is not to merely assent to a proposition but to feel that it is true (cf. Owen, 2003). Eo ipso,

Constantine Sandis

149

to have a moral belief would be to feel that something is morally right or wrong. This strengthens the case for an internalist reading of (P1), and could possibly explain why Hume may have taken (P1) to be a universal truth of some kind. Given that Hume’s sentimentalism does not lead him to a non-cognitivist characterisation of beliefs in general, there is no reason to suppose he takes morality to be any different when he claims that it ‘is more properly felt than judg’d of’ (T 3.1.2/470) and that ‘to have a sense of virtue is nothing but to feel satisfaction of a particular kind’ (T 3.1.2/471). Vice and virtue are discoverable ‘by means of some impression or sentiment they occasion’ (T 3.1.2/470), but this is not to say that the resulting beliefs are impressions. And even if they were, their effects would be no different in kind to those of ideas, besides which our moral feelings are said to be ‘so soft and gentle’ that we are ‘apt to confound’ them with ideas (T 3.1.2/470), which only differ from impressions in their vivacity). If not the lack of a non-cognitive component, then what is it that makes beliefs derived from reason (alone), inert? The obvious candidate would be the actual cause or origin of the judgements. But what exactly is it about their origin that makes the beliefs in question inert? It cannot simply be that reason itself just happens to be inert since this would be clearly at odds with Hume’s more general views regarding causation (which, incidentally, is in tension with the Humean theory of motivation in general: [A]ny thing may produce any thing. … Creation, annihilation, motion, reason, volition; all these may arise from one another, or from any other object we can imagine … the constant conjunction of objects determines their causation … ’tis possible for all objects to become causes or effects to each other. (T, 1.3.15.1–2/173, emphasis in the original) Jonathan Dancy, who voices a similar concern (1993, p. 15), qualifies it by noting that this point is not so strong since there are many inconsistencies in Hume. Be that as it may, the name of the game is to offer the most consistent interpretation possible, and nothing in Hume’s argument so far suggests that actions do not (let alone cannot) follow beliefs with appropriate regularity, this being all that Hume’s notion of causation appears to require (cf. T 2.3.1/403ff, Korsgaard 2009:64, & Sandis 2010a and 2010b). Indeed as far as the Treatise is concerned we can assign Hume the view that the inertness in question is attributable to a further fact about judgements derived from reason (alone), namely that their intentional objects are limited to the two alternatives provided by his fork (see p. 147 above). In his later Dissertation on the Passions (5.1-2/161ff.) Hume explicitly distinguishes between a strict and a popular sense of ‘reason’ with the aim of equating the former with the judgement of truth and falsehood (which can only influence to the degree

150 Hume and the Debate on ‘Motivating Reasons’

that they touch some passion or affection) and the latter with the aforementioned ‘calm’ passion (which ‘actuates the will’).

6.3 Hume’s error theories of reasons Any judgement derived from reason alone, can only pertain to the kind of relations described in the fork. These relations, Hume states, tell us nothing about how we ought to act and are consequently incapable of moving us on their own. There is an obvious analogy here with Hume’s remarks concerning the powers of theoretical reason (for example T, 1.3.6/16) which aim to show that we cannot infer from any relation discovered by reason that the future will resemble the past. Judgements about these relations cannot (alone) cause us to make the transition from an impression of constant conjunction to the idea of necessary connection, anymore than they can produce action. Rather we are determined to do so ‘by a certain association and relation of perceptions’ (T 1.3/88-9); no ‘ought’ is derived from reason (alone). To recap: if we have no reason to form a certain belief and/or be moved to action in a certain way, then our beliefs and actions cannot be the result of a judgement derived from reason (alone). In each case there is a missing principle which reason (alone) cannot provide. In the theoretical case, reason cannot provide us a principle that would confirm that the future must necessarily continue to resemble the past. This is why Hume attributes our belief to custom, though he arguably takes it to be in with ‘reason’ in the popular sense given above. In the practical case, Hume’s reason cannot tell us whether or not we ought to satisfy our desires, it can only tell us how to go about doing so. Given the absence of a suitable impression, Hume would have arguably claimed that we cannot have a meaningful idea of a normative reason insofar as this is thought to amount to anything more than a description of our natural propensities and aversions (see further below). But we might also call him a nihilist with regard to the practicality of pure reason: no consideration derived from reason (alone) could (alone) count in favour of an action. Indeed, Christine Korsgaard (1986 & 1997a), Elijah Millgram (1995) and Jean Hampton (1995) have all suggested, for a diversity of reasons too complicated to repeat here, that on Hume’s view the hypothetical imperatives generated by means-end reasoning cannot even provide us with instrumental reasons (consequently Hume either never generates imperatives, or does so as a kind of proto-Kantian or empirical revisionist (see further below). On such understandings of the ‘reason is, and ought only to be the slave of the passions’ passage (T, 2.3.3.4./415), reason ought to be a slave of the passions in the sense that it can only tell you what you should do in order to satisfy your desires. Whether or not you ought to satisfy them it cannot

Constantine Sandis

151

comment on. Practical reasoning thus reduces to means-end reasoning. The alternative, more common, interpretation of Hume is he takes desire to command in the sense of it (alone) being able to tell us what we have reason to do, namely whatever will satisfy them (but see Schroeder 2007 for a more subtle version of the Humeanism theory of reasons). Yet if this interpretation is correct, not only would Hume be making a logical fallacy which he seems to have been sensitive to (that of detaching an ought from the conjunction it applied to in one of the premises and applying it to just one of the conjuncts in the conclusion), he would also be breaking his own law (that you cannot derive an ‘ought’ from an ‘is’). Let us call the view that practical reason cannot transcend the meansend relation Hume’s error theory of normative reasons. This theory maintains that no judgement or desire (or combination of both) could ever tell us (unconditionally) how we ought to act (cf. Mackie, 1977, pp. 18 and 35). It entails that when beliefs produce action, they do not do so by giving us (what we take to be) a good reason for acting. One’s taking something to be a good reason could, of course, have the appropriate influence. The thesis merely states is that the (normative) judgement that X is a good reason for action cannot be derived from reason (alone). Consequently, no judgement can move us to action purely qua any reason-giving capacity it might have because such capacities are an illusion. This leads to what we might call Hume’s error theory of motivating reasons: our actions are never produced by reasons, for there are no such reasons around: influencing reasons do not exist, only influencing sentiments (which may include one’s taking something to be a reason). One might object here that moral on this view moral judgements cannot give us any reason to act either, Hume should not allow that they can influence action any more than judgements derived from reason (alone) can. Moral judgements are practical in their concern: they are judgements about how we ought to act. Hume takes them to be capable of moving us (alone) precisely because they are concerned with ends rather than with means. As anticipated above, neither error theory prevents Hume from subsequently redefining normative reasons in a strongly naturalised light according to which passions might be thought to be defective if they fail to conform to some norm or standard of nature. We might even be said to not be as we ‘should’ be, so long as this means no more than that we do not function in a standard fashion and thereby fall short of what happens to be the natural norm. Indeed Hume arguably constructs such an account of naturalistic ‘reasons’ for acting in (T, 3.1.2–3.2–2/468–84) (cf. Baier 2009 & Sandis 2010a), having first tested this revisionist tack in his account of necessary connexions: Necessity then is the effect of this observation and is nothing but an internal impression of the mind. (T, 1.3.4.18/165)

152 Hume and the Debate on ‘Motivating Reasons’

It is the constant conjunction of objects, along with the determination of the mind, which constitutes a physical necessity. (T, 1.3.4.30/171) He thus becomes a realist about naturalised reasons (for various references to such entities in both his philosophical and his historical work see Baier, 2008), as he is regarding necessity. In both cases, it is not reason that produces our judgements but human nature, in virtue of all the sentiments, habits, instincts, and other dispositions it bestows on us (phenomena upon which Hume will ultimately ground his virtue and consequentialist normative ethics (in a move later mimicked by Mackie and Hare).

6.4 The motivational map revisited Returning to the map introduced in 6.1, Hume most closely resembles the weak anti-Humean since he allows that at least some beliefs can do the motivating ‘on their own’. As for the psychologism/non-psychologism debate, although at times he appears like a psychologist it is difficult to know just what to say about him since it is far from obvious that he makes a state-content distinction. What is obvious however is that in denying that influencing reasons (under a certain conception) exist, he rejects CV (the view that TOM is a theory about the reasons we act for). Interestingly, both the psychologism/non-psychologism and the Humean/ anti-Humean debates arise (largely) because of CV. Non-psychologism takes its cue from two truisms: that it is possible to act for a normative reason (viz, a reason that counts in favour of an action), and that normative reasons are non-psychologistic. From this it infers that what moves us to action cannot be a psychological state (or, for that matter, any other feature of our psychology). Psychologism takes its cue from two different truisms: that we are – at least sometimes – motivated by psychological states and that motivation is causal notion (which is not to say that it need treat action explanation as being causal). If we reject non-psychologism we are faced with two horns of a dilemma: either we cannot ever act for good reasons or the good reasons we act for are (necessarily) psychological phenomena. Conversely, if we reject psychologism we are left with an account that claims that (it is necessarily true that) we are never motivated by either our beliefs or our desires, or indeed the two combined together. What counts in favour of the former view is precisely what counts against the latter, and vice versa. The trouble, then, is that each of the two positions have irresistible selling points (⫹) which we cannot buy into without also committing ourselves to accepting an unwanted consequence (–) of having adopted the position in question. In rejecting either

Constantine Sandis

153

unwanted consequence (as each rival view does) we find ourselves having to deny the obvious. Claims counting for (ⴙ) Claims counting against (ⴚ) Psychologism

T: We are (at least sometimes) motivated by our psychological states.

F: Good reasons (we act for) are not psychologically real.

Non-psychologism

T: Good reasons (we act for) are not psychologically real.

F: We are (at least sometimes) motivated by our psychological states.

This is problematic because it is natural to side with psychologism with regard to motivation and non-psychologism as far as normative reasons are concerned. The motivational map, being drawn by the assumption that we are motivated by the reasons for which we act, has no theoretical space for such a position. By contrast Hume, having rejected CV, is free to do just this. Be that as it may, Hume nonetheless seems committed to the thought that any successful explanation of action will lie within TOM. That is to say, he thinks of action explanation as motive-giving explanation. Moreover, he understands this as a species of causal explanation, introducing the ‘influencing motives of the will’ as the particular causes of actions (T, 2.3.2.8/412 and 2.3.3/413 ff.). So his view on action causation is one and the same with his view on the motives of action. If we reject Hume’s causalism, however, we are free to side with psychologists on motivation, and non-psychologists on reasons for action. A similar conclusion might be reached regarding the Humeanism/anti-Humeanism debate, as illustrated:

Humeanism Anti-Humeanism

Claims counting for (ⴙ)

Claims counting against (ⴚ)

T: We are (at least sometimes) partly motivated by desires. T: Desires are not parts of the reasons we act for.

F: Desires are not parts of reasons we act for. F: We are (at least sometimes) partly motivated by desires.

To paraphrase House M.D.: they’re both right, in the sense that they’ve convinced me that they’re both wrong.

Notes * This essay was finished with help from the project ‘Aspectos Modales del Realismo Materialista’, HUM2007-61108, MCYT-Spanish Government. I would like to thank Jonathan Dancy, Jeppe Høj, Tom Joyce, Michael Smith and Galen Strawson for

154 Hume and the Debate on ‘Motivating Reasons’ their very helpful comments on earlier drafts of this essay. Various versions of it were trialled at the University of Reading (November 2001), SPPA – 5th Mark Powell Conference, University of Stirling (26 January 2002), Hume Studies in Britain II, University of Edinburgh (18–19 September 2002), Hume: Motivation, ‘Is’ and ‘Ought’, University of Otago, New Zealand (19–24 January 2003) and the British Society for Ethical Theory conference, Queens University, Belfast (21–23 July 2003). Many thanks to all those who commented and/or asked questions, particularly Annette Baier, Simon Blackburn, Stephen Finlay, Rosalind Hursthouse, Richard Joyce, Peter Kail, Charles Pigden, Sabine Roeser and Michael Smith (again).

7 Against All Reason? Scepticism about the Instrumental Norm* Stephen Finlay

Introduction A naturalistic project descended from Hume seeks to explain ‘ought’ and normativity as a product of motivational states such as desires and aversions.1 Following Kant, rationalists reject this thesis, holding that ‘ought’ rather expresses a command of reason or intellect independent of desires. On Hume’s view the only genuine form of practical reason is theoretical reason operating in the service of desire, as in calculation of means to ends. Reason at most discovers normative requirements, which exist through the interrelation of subjective desires and objective world. The Humean desiredependence view of the source of normativity is commonly associated with instrumentalism, an influential theory of normative content according to which agents ought always and only to act so as to optimise satisfaction of their own desires. But rationalists (including Thomas Nagel, Jean Hampton, and Christine Korsgaard) have recently argued that proponents of desiredependence are not entitled even to this instrumentalist ‘ought’ (see also Wallace, 2003). Instrumentalism holds that all normativity derives from the instrumental norm: approximately, the principle that one ought to take the means to one’s ends, or (᭙ agents x, ᭙ ends y, ᭙ actions z) ((x wills y & z is the means to y) → x ought to do z).2 The normativity of the instrumental norm, rationalists observe, cannot itself be dependent on any particular desires. Whatever I desire, I ought to acknowledge its normative authority and take the means to my ends.3 Rationalists therefore charge that instrumentalism itself assumes a practical principle with normativity not derivable from any desire, but commanded purely by reason itself.4 This provides leverage against resistance to other alleged rational requirements such as Kant’s Categorical Imperative: once we acknowledge one command of pure reason, we have no principled ground 155

156

Against All Reason?

for scepticism about the possibility of others.5 In any case, once we concede the desire-independent normativity of an instrumental norm of reason, the desire-dependence thesis is compromised. Is the desire-dependence view of normativity therefore hopelessly inconsistent, denying the possibility of rational norms while assuming the rational authority of the instrumental norm? Hume himself, unlike those who claim his inheritance, is said not to accept the instrumental norm (see Hampton, 1995, 1998; Millgram, 1995; Wallace, 2003). While I am unaware that he ever explicitly denies it to be contrary to reason to fail to pursue means to our ends,6 this is clearly entailed by his infamous rejection of practical irrationality per se: Reason is the discovery of truth or falshood. Truth or falshood consists in an agreement or disagreement either to the real relations of ideas, or to real existence and matter of fact. … Now ’tis evident our passions, volitions, and actions, are not susceptible of any such agreement or disagreement. … ’Tis impossible, therefore, they can be pronounced either true or false, and be either contrary or conformable to reason. (T, 3.1.1.9/458) Hume is a radical and consistent champion of the desire-dependence thesis, who denies the existence of any rational norm. This scepticism is claimed to be absurdly radical; the existence of the instrumental norm is a dogma in contemporary ethical theory, and many declare its denial simply unthinkable.7 R. Jay Wallace writes that ‘In the modern era, this form of rationality has been widely viewed as the single unproblematic requirement of practical reason’ (2003). Hampton claims that Hume’s view of reason ‘defies common sense’ (1995, p. 57). To champion the desire-dependence thesis consistently, one must adopt the heresy of Hume, which no right-minded person could. It is therefore apparently against all reason that I seek to defend Hume and the desire-dependence thesis, by arguing that the instrumental norm is a mere phantasm of our confusion. There is no desire-independent normative principle commanded by reason (or anything else) that we ought to take the means to our ends, and all the intuitions and phenomena that seem to lend the idea credence are explicable in other, Humean ways. My Argument will be that no formulation of the alleged norm can satisfy two basic conditions for a genuine normative command of reason. First, there is the possible violation criterion. As Kant saw, for there to be a principle that can tell me what I should do and thereby counsel, guide, or command me, it must be (at least logically) possible for me to act otherwise than as it directs. Second, there is the reasonable expectation criterion. Unlike a putative normative requirement springing from the will of a deity, or from the metaphysical nature of things, a requirement arising from pure reason (hence, as Kant claims, a priori) must be evidently reasonable. I must, in exercising my

Stephen Finlay

157

reason, be able to recognise the legitimacy of such an expectation, and it must be impossible to sustain an intelligible challenge to it. This is implicit already in the view that conformity to the instrumental norm is so selfevidently required that to reject it is absurd.8 Section 7.1 aims to clarify the dictate of the alleged norm. Section 7.2 presents my Argument debunking it. Section 7.3 critically examines alternative rationalist formulations, and Section 7.4 reinterprets the phenomena that have been taken as evidence for the norm.

7.1 Formulating the instrumental norm What does the instrumental norm command, and why should we obey? It is surprisingly difficult to formulate, given its alleged firm foundation in common sense, but it is important that we do so painstakingly, as the rationalist case for its existence trades upon ambiguity and indeterminacy. My method is to start with the simplest, most commonplace formulation, and progressively refine it. We commonly utter statements of the form ‘If you want y, you should do z,’ where doing z is a means to the end y.9 I shall mean by ‘ends’ sets of possible states of affairs, and by a ‘means’ to an end a possible course of action, or necessary part thereof, that may result in a member of that set obtaining. My first approximation is then IN(1): (᭙ agents x, ᭙ ends y, ᭙ actions z) ((x desires y & z is a means to y) → x ought to do z). Three elements need refinement. I will first address the nature of the meansrelation between action and end, then the proper formulation of the consequent imperative (what is it one ought to do if the conditions specified in the antecedent obtain?), and finally the proper relation between agent and end. According to IN(1), if an action is a means to my end, I ought to perform it. But there are many logically possible means to accomplishing some ends that are physically impossible for a given agent. Are there superhuman feats I ought nonetheless to perform? Requiring this for rationality clearly violates the reasonable expectation criterion. Reasonable criticism must be sensitive to agents’ situational abilities, hence the instrumental norm must command only means that are performable by the agent. Epistemic circumstances are also significant. On IN(1) if I will an end to which, unbeknownst to me, ␸-ing is the means, and I fail to ␸, I violate the instrumental norm. But I would more appropriately be called irrational if I were to ␸, not believing that I thereby further my ends.10 The norm must only command me to action that I believe to be performable means to my ends. IN(1) tells me that if some action is a means to my end, I ought to perform it. But there can be many alternative means to attaining one and the same

158

Against All Reason?

end, and it is unreasonable to demand that I perform them all (and often counterproductive to try). The instrumental norm must only direct me to take one sufficient means to my end. One suggestion is that the norm directs me to take the best means, but are we to maximise moral acceptability, aesthetic appeal, likelihood of success, ease of execution, or something else entirely? The most plausible candidate would be efficiency (instrumentally best). But a requirement of efficiency can be accommodated by IN(1) since we never have just one end, so that other ends affect reason’s counsel in regard to means. If a more substantive account of efficiency is offered, we may observe that efficiency is not the only virtue of action, and it would be unreasonable to require that we always act in the most efficient manner. I here presume that I can only be justly accused of instrumental irrationality if I fail to take any means sufficient to my end11 (this is certainly suggested by the standard examples), so we should reject evaluative criteria in favour of modal ones. I must take (what I believe to be) the performable means necessary for attaining my ends. Isn’t it common, however, for an end to have several possible, hence no necessary means? In this case there is a necessary disjunctive means (for example d, where d ⫽ a or b or c, and a, b, and c are the only alternatives); here the instrumental norm commands performance of d. Combining these refinements we get IN(2): (᭙ agents x, ᭙ ends y, ᭙ actions z) ((x desires y & x believes z to be the necessary performable means for achieving y) → x ought to do z). Let us now consider what is commanded if the antecedent conditions are met. First, note that IN(2) commands performance of the action z, hence if the antecedent conditions hold of me and I fail to perform z, I am instrumentally irrational. But this is unreasonable. Suppose I try to perform z, but my efforts go awry: my shoelaces being tied together, I instead trip and fall. I may then be guilty of clumsiness, incompetence, or ignorance, but not of irrationality. Let us say for now that the instrumental norm commands not that I perform, but that I try to perform z. Time introduces further difficulties. I can desire an end without desiring that it obtains at that moment. I may desire it for some future time, or for the non-specific future. The appropriate means to our ends therefore need not be presently performable. I need not choose among presently performable means if I anticipate that another (perhaps preferable) means will become available later. IN(2) does not say when I ought to try to perform z, but it cannot be at any time. If I will the end on Monday, but the means z is the action of ␸-ing on the following Thursday, it would be absurd to counsel me on Monday to attempt to perform z on Monday. The ‘when’ of the imperative may then appear to be the time of the chosen means. But suppose I desire the end on Monday and not thereafter, and the means is

Stephen Finlay

159

an action on the following Thursday, ought I on Thursday to perform the means to an end that I no longer desire? This is an unreasonable expectation: I am at a time t subject to instrumental requirements only in virtue of ends that I still have at t. I provisionally amend the norm as commanding me to try to perform at t an action z if I believe that performing z at t is possible, and necessary for realising an end I desire at t (an unsatisfactory solution, as we shall soon see). There is, further, a problem with the very suggestion that this norm is ‘hypothetically imperative’ in form. With a genuine conditional, If A then B, satisfaction of the antecedent condition A entails the unqualified truth of the consequent. If my end is to give my wife a sapphire ring for our anniversary, and the only performable means is to steal one, then on IN(2) it is unqualifiedly the case that I should steal one. Reason commands that I do so. This can’t be right (see Broome, 2000, pp. 89–90). We often set ends without appreciating what their attainment will require, only later realising that performing the necessary means is unthinkable. The reasonable thing to do is to give up the end. Of course, by surrendering the end we make the antecedent false: don’t we thereby escape the imperative? But in doing so, on IN(2), we are violating the command of reason, which doesn’t give the option of surrendering the end. The reasonable expectation criterion is not satisfied. Some rationalists therefore rightly hold that the instrumental norm counsels us either to perform the means z, or to give up the end y.12 This disjunctive imperative yields our third formulation: IN(3): (᭙ agents x, ᭙ ends y, ᭙ actions z, ᭙ times t) ((x at t desires y & x at t believes z to be the necessary performable means at t for achieving y) → (x ought at t to try to do z or cease desiring y)). This solves the preceding problems. Reason never demands that I take immoral or intolerable means to my ends; rather, that I pursue what I believe to be the necessary means to my end or else give up that end. Instrumental irrationality is the offence of not adopting what one believes to be the necessary means to an end while maintaining that end. This brings us to the most difficult question: the character of the relation between agent and end. So far, taking my cue from an ordinary turn of speech, I have represented the relation as desiring – being desired. Am I indeed subject to hypothetical imperatives of reason as a consequence of my desires? Humeans who admit an instrumental norm often favour this formulation, and even Kant sometimes wrote this way (for example, 1997, p. 18). But I have desires of which I myself disapprove, and believe should not be satisfied. It is unreasonable to expect that I should pursue the means to an end I desire, when I should not desire the end. (Of course IN(3) also gives me the option of ceasing to desire the end, but desires cannot be

160

Against All Reason?

eliminated simply by choice. It would be unreasonable to call me instrumentally irrational because I continue to have such desires.) Perhaps an improved formulation would counsel pursuing means to the ends of only a certain kind of desire: endorsed desires, second-order desires, fundamental valuings, etc. Donald Hubin writes, [A means/ends irrational] agent has ends that he values intrinsically, recognises that certain actions are necessary means to these ends but never feels any Motivation to perform the acts in question. (1996, p. 40) But there is a general problem with formulating the norm in terms of any kind of desire. Suppose I intrinsically value world peace, but believe there is no means to it except the slaughter of half the world’s population. I therefore despair of realising my ideal; must I also cease valuing it? IN(3) says I must, on pain of irrationality. But this is not a reasonable expectation! (It may even be impossible.) It does not follow from my having a desire that I rationally ought to seek its satisfaction.13 As Kant intended, it is when I will an end, not when I desire something, that there are actions I instrumentally ought to perform. The notion of ‘willing an end’ is however significantly vague. We know at least that it contrasts with desiring something, which doesn’t obviously require anything of us. Desires can be idle, unendorsed, and merely wishful, while having an end involves resolve and commitment. This difference needs greater articulation. What is ‘willing’? Rationalists believe that willing is one critical element of action neglected by Hume and those who would reduce deliberation to the interplay of rival desires. Voluntary action requires states and actions of intellect, as well as (perhaps even in place of) conative states, and willing is to be identified with some such intellectual state or action – such as decision, volition, or intention. So which, if any, of these modes of intellectual activity is ‘willing’? When we decide to act on an inclination – to do a desired action or seek a desired end – then its object becomes an object of volition. (Korsgaard, 1997a, p. 234) Here decision plays a pivotal role: should we identify willing with deciding? But to decide (or resolve or commit oneself) to realise an end is to perform a momentary, temporally discrete action. The end that I decide upon and the means I believe necessary for attaining that end may be temporally distant, so deciding upon some end does not always precipitate nor require immediate action. It cannot therefore be merely when I decide upon an end that I will it, and the antecedent of the instrumental norm should not specify the act of deciding as the pivotal relation between agent and end, or

Stephen Finlay

161

the norm will fail to command me after my decision is made. Korsgaard’s point is more plausibly that willing is a product of decision, which is the process of arriving at the intellectual state of willing. We could, then, identify willing as the state of being decided (resolved, determined, committed) upon an end – but here this would be circular and uninformative. The concept of decision does not help to isolate the meaning of ‘willing’. Korsgaard clearly suggests volition as the product of decision – perhaps this is our quarry. (Indeed, etymology favours identification of will with volition.) The concept of volition is that of aiming at literal or figurative motion by an exercise of intellect. To ‘will’ an end would then be to exert causal influence towards self-activity by power of thought – it is to actually aim at, seek, or pursue that end. (Note this is not equivalent to being causally determined to activity by thought, since volition can fail.)14 Clearly rationalists, including Kant, often mean by ‘will’ precisely this. But so interpreted Korsgaard’s claim is false; decision does not always result in volition. Deciding tonight to drive to work tomorrow morning does not generate volition tonight (and could even fail to do so tomorrow). Commitment to my decision requires volition only when morning arrives. Evidently another mental state is at work, one acquired when a decision is made, and able to produce volition when conditions are right. Shouldn’t we call this, also, willing an end? Since rationalists insist that one can will an end while being irrationally unmoved, they cannot mean to identify willing with volition. The product of decision and cause of volition is the mental state of intending to realise the end. Intention is the most plausible candidate for a state to identify with ‘willing’: it incorporates the commitment to the end that makes plausible the claim of normative requirement of the means, and intending an end does not entail present volition towards that end.15 This also resolves a problem regarding what the instrumental norm counsels when the conditions of its antecedent are met. Any formulation that counsels me to perform, attempt to perform, or seek the means to my ends has trouble with temporally dislocated means and ends, as previously discussed. I provisionally suggested we interpret the norm as counselling merely that I attempt to perform any means to my ends that are both presently performable and necessary, but this is incompatible with disjunctive means (performance of which cannot be required at a specific time if the disjuncts specify different times), and entails that the instrumental norm gives no counsel about pursuing our ends until all but one of the possible means have been declined. Those who refuse to put off today what can be done tomorrow are then neither following nor flouting the counsel of instrumental rationality, whereas they appear rather to be following one disjunct of its advice. The dictum, ‘whoever wills the end, wills the means’ invites us rather to consider intention for the imperative as well as the antecedent. My ends require that I intend to perform the means.16 This can be required when

162

Against All Reason?

I will the end on Monday, even though the means are not performable until the following Thursday, and I am not required to pursue on Thursday the means to ends erstwhile abandoned. We can also eliminate talk of attempting. Since intention is a matter of attitude rather than performance, it is not a condition on my intending to ␸ that I successfully ␸. We then get: IN(4): (᭙ agents x, ᭙ ends y, ᭙ actions z, ᭙ times t) ((x intends at t to achieve y & x at t believes z to be the necessary performable means for achieving y) → (x ought at t to intend to do z or cease intending to achieve y)). A noteworthy objection is that a norm commanding only intention and not attempted action fails to proscribe an important form of instrumental irrationality. Suppose I believe the necessary means to my end is to ␸ at t1. On IN(4), to comply with the counsel of reason at t1 it is sufficient that at t1 I intend to ␸ at t1, and not necessary that I attempt or have volition to ␸ at t1. The strange consequence is that failing to pursue what I believe to be the necessary means to my end is not instrumentally irrational. The solution is to recognise the intimate connection between intention and volition. To intend in the present to ␸ in the present is to have volition to ␸ in the present. A gap between intention and volition only opens through temporal distance between intending and what is intended. Therefore, if in the present I do not attempt to ␸, it follows that in the present I do not intend to ␸ in the present.

7.2 The impossibility of instrumental irrationality With IN(4) we have at last met the reasonable expectation criterion. Obviously I instrumentally should intend to do what I believe necessary for achieving the result I intend to bring about – and it is prima facie plausible to say that reason itself demands this of me. Indeed, I believe IN(4) successfully captures what we have in mind when we think of an instrumental norm of reason. To be a genuine normative principle, however, it must also satisfy the possible violation criterion. Instrumental irrationality, as defined by IN(4), is the failure to intend to do what one believes necessary for achieving some end, while intending that very end. Is such irrationality possible? Kant himself seems at times to maintain that the principle, whosoever wills the end wills the necessary means, is an analytic hence necessary truth. Failure to comply would then be impossible – but a normative principle that cannot be transgressed is no normative principle at all, per Kant and the possible violation criterion. This interpretation of Kant is mistaken: elsewhere (for example, 1911, p. 417) he clearly explains that the connection is analytic only insofar as we are rational. Since we are capable of irrationality we can violate the command of reason and fail to pursue the necessary means to our ends.

Stephen Finlay

163

I believe Kant’s principle is, nonetheless, analytic without this qualification (see also von Wright, 1978). Observe that the expression, ‘to intend the end y’, is significantly incomplete. Intention has to do with action: what is intended is always action (or inaction) of some sort. To intend the end y, therefore, is really to intend to act so as to bring the end y about.17 Now observe that the expression, ‘to intend the necessary means z’, is similarly incomplete. Talk of the ‘necessary means’ implicitly invokes the end, so that to intend the necessary means is to intend to act so as to bring the end z about. Intending the end, therefore, simply is intending the necessary means to that end – the two locutions give partial and complementary descriptions of the same intention, so one logically cannot intend an end without intending the necessary means.18 This result is significant, but I’m seeking something more. Suppose I believe that the necessary means to my intended end y is to do z. We have established that my intending the end entails that I intend the necessary means. Doesn’t it therefore follow that I intend to do z? Actually it doesn’t – not straightforwardly, anyhow – because of the opacity (‘intensionality’) of intention. The content of intention, like other propositional attitudes, does not permit the substitution of identicals salva verite. Suppose I aim to spot the kitten on the floor so as to avoid stepping on it in the dark, and believe that turning on the light is the necessary means. I also believe that the action of turning on the light is at the same time the action of startling the fish. I intend to turn on the light – do I also intend to startle the fish? Clearly not, even though I believe the actions to be one and the same. Intending A and believing that A⫽B does not entail intending B. Therefore what follows from my intending the end y is merely that I intend the necessary means to y de dicto, and not that I intend the necessary means de re (that is, to do z). Surely (one says) I rationally ought to intend to do z – but the claim I am defending is rather that I metaphysically must. Isn’t this implausibly to insist upon the impossibility of failing to acquire a new intention on identifying the means? It might be enough for my purposes to point out that we’ve established an analytic requirement of intending the necessary means. If I intend the end y at t, and believe that doing z at t is the necessary means, then (given the identification of volition with present intention for the present) it follows that I will knowingly do z (barring failure) even if I don’t intend to do z.19 But I believe we can, further, show an analytic requirement to intend to do z. To do so, we have to look beneath the principle of the opacity of intention, at its explanatory ground. The opacity of intention is often glossed as a matter of being ‘descriptionsensitive’. We must note, however, that it is not strictly speaking descriptions – that is, sentences – to which opacity is sensitive. My intention to turn on the light is also the intention to illuminate the room by flipping the light switch. The description reports some aspect or element of the action,

164

Against All Reason?

which permits multiple descriptions, and opacity is sensitive to this content of description. Every action can be characterised in indefinitely many ways, according to the impact that action has upon different parts of the world (for example illuminating the room, startling the fish). Intentions can be thought of as agents’ automatic targeting – that is, they involve dispositions for one’s deliberative endeavours to track a certain content. This explains the general opacity of intention: my target is illuminating the room, not startling the fish. As circumstances change I will deliberatively track the goal of illuminating the room, but not the goal of startling the fish. Hence we say that I do not intend to startle the fish, although I do perform that action. An intention, then, will entail some further intention provided the content of the further intention is appropriately connected to the content of the first. The aspect of being a necessary means to an end (such as spotting the kitten) is very different from the aspect figuring in the description of the means itself (turning on the light). In this example, the former characterises the action in terms of its impact on my awareness of a part of my environment, the latter characterises the action in terms of an impact directly on the environment itself. There is no connection of constitutivity here, therefore – the connection is rather causal. The impact of illuminating the room is causally responsible for the impact of my spotting the kitten. To defeat opacity we need more than just a causal connection; illuminating the room is causally responsible for startling the fish, but my intention for the former doesn’t translate to an intention for the latter. However, consider the significance of my belief that turning on the light is the necessary means to my seeing the kitten. A belief identifying an action as a means is not simply an identification of an action with itself under two separate aspects. We would not say that my startling the fish is the necessary means to seeing the kitten, even though my startling the fish is the same action as turning on the light. Rather, a means-identification indicates the aspect of an action which is causally significant as a means. So long as I have this belief in the necessity of going through the event of my turning on the light in order to reach the event of my spotting the kitten, my deliberative efforts to perform the necessary means to spotting the kitten will of necessity track the goal of turning on the light – which simply means (given the identification of intending with tracking) that so long as I have this means-belief, my intention to perform the necessary means is an intention to turn on the light. So long as I believe that the necessary means to my intended end is z, I do indeed intend to do z. The connection between intending the end and intending what I believe to be the necessary means to the end is analytic. On IN(4), if I am not willing what I believe to be the necessary means, then I am not willing the end, hence the possible violation criterion is not met. We have a logical rather than a normative law (per Hume, what else should we expect from pure reason?). Nonetheless, I believe that IN(4) does indeed articulate our

Stephen Finlay

165

intuitive notion of a rationally required instrumental norm. It is precisely when – and only when – I aim at or intend an end that the expectation that I aim at or intend the means appears incontrovertibly reasonable: because it is a requirement with which I cannot fail to comply! There is no such thing as an instrumental norm of reason, because Hume is right: pure reason never commands. The illusion that the idea has substance, even that its denial is absurd, is a conjuring trick, rationalist sleight-of-hand. Rationalists offer a vague formula (‘willing an end’) admitting of differing interpretations. Some satisfy the possible violation criterion, others the reasonable expectation criterion, but none satisfy both. On one interpretation, IN(4), it is analytic that if I will the end, I will what I believe to be the necessary means to the end: our grasp of this yields our impression that the principle derives purely from reason. But the world cannot violate the laws of logic, and so to give the impression that we have a normative principle, rationalism substitutes different interpretations.

7.3 Rationalist responses Rationalists must resist this claim of analyticit y: it is possible to violate the instrumental norm – indeed, it is a common occurrence. There are two strategies I must address: first, one might accept IN(4) as an adequate interpretation of the instrumental norm, but claim that my Arguments have failed to prove that it cannot accommodate possible violation: one can indeed fail to intend what one knows to be the necessary means to one’s intended ends. Second, the rationalist can deny that IN(4) adequately captures the instrumental norm. I shall attend to these objections in turn. Surely, it may be insisted, such irrationality as proscribed by IN(4) is possible. The rationally required means/end connection can be broken by a number of familiar influences, such as ‘rage, passion, depression, distraction, grief, physical or mental illness’ (Korsgaard, 1997b, p. 378). To support the claim, rationalists appeal to a wide variety of alleged examples of instrumental irrationality: Korsgaard offers the case of a woman who wills to overcome her fear of riding roller coasters, but walks away at her moment of truth, overcome by terror (1997b, p. 228–9). I shall instead consider the modern paradigm of instrumental irrationality, the addicted smoker who, despite being resolved to kick the habit, succumbs to addiction and smokes another cigarette. I ought not deny that such things occur, nor that these people deserve criticism. But are they genuine cases of violation of IN(4)? For that to be the case, they must both be aware of the necessity of the means, and continue to will the end. The problem for the rationalist case is that the relevant behaviour calls for interpretation, and there are alternative, arguably more plausible analyses of such weak-willed actions. Is the smoker passing up what he believes to be a necessary means to his end? The means under consideration involves abstaining from

166

Against All Reason?

a cigarette on this particular occasion. But he could give up tomorrow instead of today and still achieve his end – smoking one more cigarette is compatible with his end. It is partly this knowledge, that the means at hand is not necessary, that makes his addiction so hard to escape. The smoker doesn’t violate IN(4), therefore, unless we modify the example. Suppose he resolves never to smoke another cigarette, effective immediately (at t). When later tempted to light up, he knows that doing so is incompatible with attaining his end. I ask: can we conceive of his succumbing to temptation while still intending never to smoke again after t? I find myself incapable of imagining this. However other, plausible diagnoses come readily to mind: in choosing to smoke, he abandons his end, perhaps forming a new one (‘this will be the last, I swear’), but perhaps resigning himself in self-disgust never to escape his addiction. He may try to deceive himself that he hasn’t abandoned his end (‘it’s still before t in the Pacific time zone’), which would be to self-deceptively change his end. He may well continue to wish for the end, but this is a different matter, and involves no violation of IN(4). What about the usual suspects – ‘rage, passion, depression’, and so on? One common effect of both violent passion and depression is the abandonment (even if only temporary) of previously stable ends. When, in Korsgaard’s example, the woman turns in terror away from the roller coaster, she seems in no way still resolved upon riding it (in the present). Another common effect of violent passion is to make us oblivious to things we generally know. Indeed, four out of the five ways Korsgaard suggests these mental perturbances make us irrational involve causing us to be unaware of our reasons for acting (1997b, p. 378n)! But, most plausibly, this is to say that in those circumstances we are unaware that our actions are incompatible with our ends! This would therefore be, at best, a form of theoretical irrationality, but not of practical irrationality, which must be assessed relative to an agent’s epistemic limitations. The case for genuine violations of IN(4) depends upon the possibility of knowingly failing to take necessary means to our ends – and I submit that there is no anecdotal evidence that clearly supports this. The very paradigms of instrumental irrationality fail to demonstrate violations of IN(4). However, this may reasonably be thought to evince rather that IN(4) is not the true instrumental norm. Rationalists offer various distinct formulations of the norm: might any of these meet both criteria? Perhaps I have not yet captured what is meant by ‘willing an end’ in this context: is there a viable interpretation without the same close link to action? In particular, my observations regarding the failure to meet the possible violation criterion have been based on constitutive claims about what it is to have present intention. But suppose what it is to will an end is free of any conditions on an agent’s present state of mind? Some rationalists identify instrumental irrationality with a form of diachronic inconsistency: We continually make [resolutions to act in a certain way at some future time], and sometimes when the time comes for carrying them out, we do

Stephen Finlay

167

in fact act otherwise under the influence of passion or mere habit, without consciously cancelling our previous resolve. This inconsistency of will our practical reason condemns as irrational. (Sidgwick, 1907, p. 37, my emphasis) When I will an end, I must ipso facto will that even on another occasion, even when I am tempted not to, I will stay on the track of that end … So when you will an end, the form of the act of your will is general: you will a kind of law for yourself, a law that applies not only now, but on other possible occasions. (Korsgaard, 1996b, p. 230–1) If what we will is determined by our past resolutions rather than our present mental states, then the possible violation criterion is easily met. We cannot, of course, be rationally beholden to every end we have ever set for ourselves, and Sidgwick acknowledges our entitlement to change our minds: we are normatively bound only where ‘we do not consciously retract our adoption of the end’ (1907, p. 38). Why, however, take past resolutions as normatively binding? Suppose that as a child I resolved never to force my future children to eat vegetables – surely I am not now irrational when I insist that my daughter eat her vegetables, never having consciously disowned that resolution. Korsgaard explains that ‘the claim to generality, to universality, is essential to an act’s being an act of the will’ (1996, p. 232) – to will at t is to commit yourself at t to a law that is to apply to you at all times, so that ‘when I follow a hypothetical imperative, one part of me – say my will at one moment – governs another part of me – say at another moment’ (1996, p. 230).20 This may appear incompatible, however, with our entitlement to change our minds, and hence an unreasonble expectation. Korsgaard’s response is that while we may indeed change our minds, we may not do so all the time: The reason I must follow hypothetical imperatives in general is that if I don’t follow them, if I always allow myself to be derailed by difficulty or dread or dullness, then I never really will an end. The desire to pursue the end and the desires that draw me away from it each hold sway in their turn, but my will is never active. (1996, p. 230) Why does it matter that I never will an end? Korsgaard explains, ‘The distinction between my will and the operation of the desires and impulses in me does not exist, and that means that I, considered as an agent, do not exist’ (1996, p. 230). This claim is extended to normativity generally: ‘The function of the normative principles of the will … is to bring integrity and therefore unity – and therefore, really, existence – to the acting self’ (1996, p. 229).

168

Against All Reason?

She explains this as follows. Strictly speaking, I am my reflective self. In order to see my actions as arising from my self (hence to see myself as an agent), rather than the mere products of various desires and impulses ‘in me,’ I must be able to see myself as something other than my particular, momentary states of mind. But since a ‘self’ is never an immediate object of experience, this is only possible if I can infer some persisting reflective character as the cause of my particular actions. As causation can only be discerned by regular, law-like connections, I must act according to some universal, constant principles (my practical identity, as law), in order to exist at all as a temporally extended, reflective agent. Supposing this to be a true story, there remain several avenues of intelligible challenge to this ‘rational requirement’. First, if non-existence is a consequence merely of never (or only seldom) pursuing means to our ends, it would seem quite reasonable to flout the norm’s counsel on many occasions (like the counsel to brush one’s teeth). Second, why can’t we reasonably opt for non-existence of this kind? Indeed, repudiation of the myth of the ‘self’21 is a central teaching in many philosophical and religious systems, including existentialism and Buddhism. One could appeal to a desire for existence – but this would be to concede the desire-dependence of instrumental normativity! One could appeal to a further rational norm commanding us to maintain our existence – but this would again compromise the rationalist case, as the normative authority of such a principle would be less self-evident than that of the instrumental norm. Korsgaard’s answer to this question is unclear, but she seems to offer the following: it is not that we have reason to obey the instrumental norm because we have reason to maintain our existence. Rather, obeying the instrumental norm (and other normative principles) is constitutive of agency: [T]he reason that [the reflecting self] has to unify itself into an agent who can persist through a series of relevant occasions is not that it has some reason to want or anticipate that it will persist into the future … The reason is rather that the view of itself as active now essentially involves a projection of itself into other possible occasions. (1996, p. 229) This may mean that I am simply not acting if I am not willing ends and pursuing means, hence it is impossible for me to act otherwise. If this is the account (and a number of inconsistencies suggest otherwise), it is an attempt to meet the reasonable expectation criterion by denying the possibility of violation. As Korsgaard acknowledges (1996, p. 231), constitutive rules that do not admit the possibility of deviation cannot be normative. Either violation of the instrumental norm is impossible, or else we are yet to receive an adequate explanation as to why we ought to obey it.

Stephen Finlay

169

The diachronic inconsistency strategy, however, may significantly illuminate that my Argument errs in assuming people to be time-slice agents. It must be observed that we frequently attribute to people, as genuine constituents of their psychology, non-occurrent mental states. I can be said to believe some proposition p even at moments when p is ‘the farthest thing from my mind’, and to intend some end y similarly, providing my dispositions meet certain conditions. It follows that I may concurrently believe and intend, and hence that there may be no occurrent mental presence of something I nonetheless intend. One effect of violent passion is, indeed, to cause us to overlook non-occurrent beliefs and intentions. While the failure to recall non-occurrent beliefs can be chalked up once more to theoretical irrationality, the failure to recall non-occurrent intentions may seem rather a practical failure. Clearly it is possible to fail to act in ways that I believe to be necessary means to my non-occurrent and overlooked ends – and this may even appear a genuine violation of IN(4), avoidance of which is reasonably required. To this problem I offer four responses. First: this too might turn out to be a case of epistemic rather than practical failure. Such lapses in intention (unlike lapses in desire) seem to be failures of memory, which suggests that intentions, if not reducible to beliefs, have at least a significant belief component, which is what fails in these cases. Second: it still seems appropriate to say that I did not intend the end at that moment. Attribution of non-occurrent mental states might be a pragmatic turn of speech that imposes false but useful simplicity on complex psychological phenomena: in other words, I don’t really intend y at such moments, hence I don’t really violate IN(4). Third: any victory the rationalist may win here is hollow, as the paradigms of alleged instrumental irrationality are not manifestations of this particular form of psychological failure. Fourth: supposing both possible violation and reasonable expectation criteria are met, this is not yet sufficient to prove a normative command of reason, and rationalism remains vulnerable (as I shall argue in Section 7.4) to an Argument from an inference to the best explanation. Perhaps, it might be thought, a rationalist can concede the analyticity of the means-end connection, and formulate a genuine instrumental norm of reason consistent with this. On Hampton’s interpretation, the norm rather commands us to will certain ends (those we reflectively judge to be the best), and if we do, then necessarily we will the means, as we cannot will an end without being instrumentally rational: ‘to be instrumentally rational is to will the end that is in some way best or strongest in the circumstances, where willing means “sustaining a commitment to that end through action”’ (1998, p. 163).22 Willing is a ‘norm-responsive psychological state’, so that I am not willing an end unless I am committed to it on the basis of judging it to be best. Instrumentally irrational agents are thus those who ‘fail to will the end that they acknowledge is the best end in the circumstances

170

Against All Reason?

(they want it, but they don’t will it)’ (1998, p. 165). Two problems are evident here. First, this distorts the intuitive notion of an instrumental norm – as a principle that gives normative guidance when we will some end – to the point of unrecognisability. Second, no true Humean will concede the existence of rationally mandated ends, or the self-evidence of this alleged normative principle (which Hampton herself describes as ‘metaphysically occult’ (1998, p. 6) and as having ‘mysterious objective authority’ (1998, p. 166)) – hence her ‘instrumental norm’ offers no leverage from common sense against the thesis of the desire-dependence of normativity. These rationalist attempts to find a satisfactory interpretation of the instrumental norm yield results that are alien and implausible. The notion of willing an end invoked in intuitive appeals to an instrumental norm is adequately captured in IN(4)’s terms of intention, and not in Sidgwick’s talk of past commitments, Korsgaard’s talk of law-like conceptions of practical identity, or Hampton’s talk of commitments to ends we judge to be best. Although I have not examined every possible rationalist account, I believe these failures strongly suggest that the rationalist project can only be fruitless.23

7.4 A humean instrumentalism? There is no instrumental norm of reason. This conclusion will be too radical for many, who will point to familiar phenomena in everyday thought and discourse appearing to show that common sense supports rationalism here, and hence that scepticism must embrace an implausible error theory of practical rationality. Two ordinary practices in particular may appear to furnish proof of this everyday wisdom. First, there is ascription of ‘irrationality’. We make judgements about what agents ‘rationally’ ought to do, and we frequently ascribe ‘irrationality’ to those who act otherwise, furnishing countless concrete examples of irrational action. Second, utterance of hypothetical imperatives is not the preserve of rationalist philosophers, but similarly a familiar part of everyday discourse and judgement. I will now gesture at how these phenomena can be accommodated by a desire-dependence account of normativity. My goal in this section is merely to present an alternative, anti-rationalist point of view. I have argued that even the paradigms of instrumental irrationality fail to violate any plausible command of reason. Must we then decide against ordinary wisdom that irrationality is impossible? No. We may, like Hume, deny merely that irrationality as defined by the rationalist is possible. Rationalism deceives us, by speaking in the language of common sense, into believing that it speaks for common sense. But it misinterprets these practices and utterances. Hume infamously writes: What we commonly understand by passion is a violent and sensible emotion of mind. … By reason we mean affections of the very same kind with

Stephen Finlay

171

the former; but such as operate more calmly, and cause no disorder in the temper. (T, 2.3.8.13/437)24 We need not endorse Hume’s exact translations: perhaps ‘reason’ and ‘rationality’ are ordinarily used to refer to behaviour controlled by stable or persistent rather than calm desires, ‘unreason’ and ‘irrationality’ referring to that behaviour prompted by unstable or fleeting rather than violent passions (particularly where they are in conflict with ‘reason’) – or perhaps these terms are used in a variety of distinct ways. A football player who, overcome with rage, commits a personal foul depriving his team of the victory that is the object of his stable desires is acting ‘irrationally’ (hereafter I use scare-quotes to distinguish the ordinary, as opposed to the rationalists’ use). Indeed, this Humean account can, unlike rationalism, accommodate the rationalists’ own paradigms of irrational behaviour. The addicted smoker acts on the prompting of a temporary nicotine-induced urge, rather than on his stable, reflective desire that he not smoke. Everyday wisdom seems here to vindicate Hume. Rationalists will challenge the claim that stability in reflection (for example) is definitive of ‘rationality’ in desire and action. That a desire survives reflection is rather a consequence of its being such as to be endorsed by dispassionate practical reason. Here I firmly dissent. But what, then, is to be made of the normativity of ‘rational’ criticism, if it simply amounts to a conflict of calm/stable against violent/fleeting passions? The answer is that ‘rational’ criticism is endorsement or censure from the point of view of the ‘rational’ self – that is, from our calm/stable concerns. Judgement of the ‘rationality’ of others’ behaviour involves stepping into their shoes, and viewing their actions from the perspective of their ‘rational’ desires together with their beliefs.25 It guides us insofar as we are motivated by such desires, which is not always the case. The judgement that we ‘ought’ to be so guided is itself made from this point of view. So far, I have assumed with rationalism that ‘ought’ is to be interpreted as expressing a command of reason (as logically consistent thought) – but we need a different interpretation. ‘Ought’ is rather teleological or end-relational.26 Every ‘ought’ presupposes some particular end as framing a normative perspective, and recommends a course of action from that particular point of view as serving that end. Normative judgements do not express commands of dispassionate reason, but endorsement from particular motivated perspectives (hence the frequent conflict of moral and prudential ‘oughts’). There is, then, room for ‘rational’ criticism in the desire-dependence theory of normativity. But this does not extend to criticism of instrumental ‘rationality’ – as I cannot but intend what I believe to be the necessary means to my intended ends. This may sound like rejection of any kind of instrumental normativity whatsoever. But what about the judgements we utter in the form

172

Against All Reason?

of ‘hypothetical imperatives’ – are they not evidence of normativity in instrumental matters? Rationalism misinterprets the function of these judgements, which fall into two categories: propositions of instrumental necessity, and practical directives. First, there are statements of the form, ‘In order to realise y, you must do z’. This ‘must’ is an alethic modal: it reports the instrumental necessity of doing z if it is to be the case that y obtains. It is a constitutive principle, that realising y is impossible without doing z, which (as Korsgaard observes) can be used normatively in case an agent intends to realise y. In this circumstance, such propositions are used as practical directives, which I must now address. It seems that not all apparent that ‘hypothetical imperatives’ can be mere statements of alethic necessity, and that some involve deontic modals instead: particularly those incorporating ‘ought’ or ‘should’ rather than ‘must’.27 These practical directives are not uttered in order to tell people that reason commands they take the means to their ends, but rather to inform them about the means to their ends. Should I say to someone, ‘If you want to see a modern defence of rationalism, you should read Korsgaard’s The Sources of Normativity’, I am aiming to inform her not of a rational requirement operating on facts we both recognise, but that reading this work is an excellent means to that end. There is never any need to prescribe to anybody that they take what they know to be the necessary means to their ends – although reminding or informing about means and prodding to induce lacking resolve or decision are often called for and easily mistaken for the former.28 This instrumental ‘ought’ still needs explanation. Haven’t I denied that failure to take the means to our ends is even possible? Since I endorse the possible violation criterion, in what sense am I entitled to this ‘ought’? All I have denied is that we can fail to will what we believe to be the necessary means to our ends. Ignorance, false belief, and confusion can lead us to neglect required means, and hence there is indeed room for an instrumental ‘ought’. But this cannot be the rationalists’ ‘ought’. First, it is insensitive to the epistemic circumstances of the agent; if the necessary means to my end is to ␸, then, instrumentally speaking, I ought to ␸ – even if unaware of this fact.29 Second, it is not limited to what is directly within my power: it is actually true that I (instrumentally) ought to ␸, not simply that I ought to try or intend to ␸. These features of instrumental normativity fit the model of the endrelational ‘ought’ – a thesis that allows reintroduction of further intuitive features we had to strip from everyday ‘hypothetical imperatives’ when seeking to formulate a satisfactory rational norm. Third, therefore, the antecedent of a practical directive need not specify the willing of an end; it can specify a desire, so that ‘If you want y, you ought to do z’ legitimately recommends some course of action z from the perspective of a desire for y.30 Fourth, the means need not be necessary. We can recommend a means as the best, from the perspective of some desires. We can say ‘Well, if you want y, then you ought to do z … but you may also do u.’ Rationalism cannot

Stephen Finlay

173

account for this ‘may’, since it interprets ‘ought’ as expressing a command of reason, so that not doing y is contrary to reason, hence impermissible. In being insensitive to an agent’s epistemic and physical limitations, judgements of instrumental normativity are not judgements of ‘rational’ criticism. I cannot justly be condemned as ‘irrational’ for failing to do what I instrumentally ought. Instrumental ‘oughts’ and ‘rational oughts’ fall into separate categories of normative propositions. The desire-dependence thesis makes room for instrumental normativity and ‘rational’ criticism, but not for criticism of instrumental ‘rationality’. Given this affirmation of instrumental normativity, can we after all formulate a legitimate instrumental norm – of desire? The rationalists’ response to this proposal is to challenge, ‘which desire?’ Taking the means to my ends is instrumentally required of me whatever my desires, but desires are contingent. Rationalists conclude that the requirement must derive from an authority independent of all desire. But how satisfactory as a solution is appeal to the command of reason? Why would reason command such a thing? Rationalists often maintain that the justification of the instrumental norm is simply brute.31 There is no further explanation why we should take the means to our ends; it is a fact about reason we must simply accept. But what harm could there be in disobeying a gratuitous command without justification or sanction?32 The explanation why I ought to pursue the means really isn’t at all mysterious. Failure to take the means signifies failure to attain the end – this is the sanction and reason. To what does it matter that we fail to attain our ends, regardless of whether they are evil or misguided? Objective reason? But why would it – and how could it – care? If we are seeking that patron that always cares whether or not we attain the object of our desire, whatever the desire, there is one obvious candidate. Failure to satisfy a desire always matters – analytically! – to that desire itself. It is from the perspective of the desire for y that we judge we ought to do z when we believe doing z to be the means to y. To the question, ‘which desire counsels us to pursue the means?’ I answer, whichever desire would have us seek the end. Every desire is instrumentally normative: whatever the object of my desire y might be, I oughty to take some means to satisfy it. We can even abstract a true general principle that looks like an instrumental norm; approximately, IN(5): (᭙ agents x, ᭙ ends y, ᭙ actions z) ((x desires y & z is the means to y) → x oughty to do z). This is compatible with the rationalists’ claim that it is possible there is no desire that counsels taking the means in every circumstance, or, approximately, Possibly, ~᭚ a desire u such that (᭙x, ᭙y, ᭙z) ((x desires y & z is the means to y) → x oughtu to do z).

174

Against All Reason?

We can now see that it does not follow from this, as rationalists claim, that there is a desire-independent (un-subscripted) ‘ought’. But might it nonetheless be true? Rationalists argue for norms of reason over desire-dependent normativity on the grounds that the correctness of taking means to our ends is not internal to that particular motivated perspective. We criticise those who fail to take such means even where we do not share their desires. The desire-independence of instrumental criticism is alleged to prove the desireindependence of instrumental normativity. Instrumental and rational criticism is thereby seen as possible from a perspective devoid of all desires (the ‘view from nowhere’). Observe, however, that when challenged to justify the claim that the subjects of criticism ought to care, rationalists insist that such agents are making a mistake even within the context of their own desires and beliefs. Rather, it is a ‘mistake’ simply because it is a mistake within the context of that perspective. Rationalism overlooks the possibility of imaginatively and conversationally borrowing another’s perspective of concern, and making judgements indexed to it (a phenomenon amenable to Hume’s doctrine of sympathy). Judgements of instrumental normativity involve assessing people’s actions from the perspective of their own motivating desires, just as judgements of ‘rationality’ involve adopting the perspective of others’ ‘rational’ desires plus beliefs. Suppose while watching a crime documentary I declare, ‘The murderer should have used a gun!’ I am not invoking a desire-independent rational norm, but making a normative judgement from the point of view of the murderer’s ends. The crucial rationalist error lies in the inference that because one always ‘ought’ to take the means to one’s ends,33 therefore it must be commanded from outside all desires. Is there an instrumental norm of desire? IN(5) may be true, but it is not normatively authoritative. On the end-relational account offered here, ‘ought’ has only contingent authority. An ‘oughte’ has authority for an agent only insofar as the end e is integrated into the agent’s desires and concerns. IN(5) is therefore a generalization from instrumental propositions, various of which are contingently authoritative for various agents, but does not itself possess any authority for those agents. Its subscript y is a variable ranging over all possible ends, only a few of which are important to any actual agent. I conclude that there is no authoritative instrumental norm at all.

Notes * Since this essay was originally written, in 2000–1, there has been an explosion of scholarly work on instrumental normativity. But while the progress in the dialectic makes some of this essay’s emphases and omissions seem odd, none of the recent literature duplicates my viewpoint or weakens my confidence in it. I owe thanks to Gary Ebbs, Gideon Yaffe, David Copp, Charles Pigden and the participants of the ‘Hume, Is and Ought’ conference at the University of Otago in February 2003, some anonymous referees, and others whom I may have forgotten.

Stephen Finlay

175

1. The existence of Motivation-external ‘oughts’ seems indubitable, but Humeans maintain they are also normatively external. 2. Individual ‘hypothetical imperatives’ are obtained by instantiating y with a particular end e, and z with a particular action a; if my end is e, and a is the means to e, then I ought to do a. 3. A universal desire (perhaps explicable in evolutionary terms) to take the means to our ends is sometimes proposed in explanation. But any such desire, and hence consequent normativity, could only be contingent, while the rational requirement to take the means to our ends clearly is not. 4. ‘However contingent the hypothetical ‘ought’ is on a desire, it is still not the same as a desire; to say, therefore, that its objective normative authority is what moves us to act rationally is to analyse the ‘prescriptive force’ of hypothetical imperatives such that it is identical to the prescriptive force of categorical imperatives. If naturalists reject the plausibility of this analysis for categorical imperatives, they must do so for hypothetical imperatives’ (Hampton, 1998, pp. 162–3). See also Wallace, 2003. 5. Some, such as Korsgaard and Hampton, argue more ambitiously that if we are to formulate a coherent instrumental norm then we must assume the existence of further rational norms. This is a case of their modus ponens being my modus tollens. 6. Hampton (1995) claims he does with his infamous claim, ‘’Tis not contrary to reason … to prefer even my own acknowledg’d lesser good to my greater’, (T, 2.3.3.6/416) interpreting ‘greater good’ as meaning means to the more preferred end, and ‘interest’ as meaning end. This leads her uncharitably to find Hume guilty of direct self-contradiction about whether we can fail to take the means to our ends. 7. For example see Dreier, 1997, pp. 90–4; Korsgaard, 1997a, p. 319; 1997b, p. 215. Nicholas Sturgeon writes of Hume’s scepticism concerning practical reason, ‘His “strict and philosophical” account of reasonableness and unreasonableness … is not worth taking seriously, and properly receives almost no attention at all. … It is a measure of my respect for Hume’s intellect that I find it hard to believe that he took it seriously either’ (quoted by Millgram, 1995, p. 87, from an unpublished paper). 8. Satisfying this criterion is not sufficient for normativity, as the requirement that one act in a way logically or psychologically necessary would certainly qualify as reasonable, and would be invulnerable to intelligible challenge. 9. Here, ‘You should do z’, which I provisionally shall follow rationalists in supposing to report a command of reason, occupies the position of the consequent following the hypothetical antecedent, ‘If you want y’: hence, a ‘hypothetical imperative’. 10. Korsgaard, however, discusses at one point ‘people acting irrationally only because they do not know about the relevant means/end connection’ (1997a, p. 380). 11. This may be mistaken; arguably it is instrumentally irrational to take an inferior means over a superior means without good reason. This indicates an avenue for rationalists to escape the Arguments of this essay, although given space I would extend my case to argue that even this form of behaviour is not possible. Even the more limited result is significant, however; defenders of the principle usually formulate it in terms of necessity (for example Wallace, 2003), some defences of it cannot easily be extended beyond necessary means (for example Setiya, 2007),

176

12.

13.

14.

15.

16. 17.

18.

19. 20.

21.

Against All Reason? and the classic examples of instrumental irrationality on offer all involve failure to take any sufficient means. Korsgaard, 1997a, p. 237; Hill, 1992, p. 24; Broome, 2001, p. 180. Schroeder (2004) notes further options: eliminating the efficacy of the means, or ceasing to believe that the action is a means. I ignore these here, because no one suggests that these are rational courses of action. As Ramon Das has observed to me, I haven’t shown that there isn’t a rational requirement to seek the satisfaction of my desire in the absence of opposing reasons – perhaps IN(3) could be fixed with a ceteris paribus clause. My rationalist opponents, however, will concede my claim. Volition can fail to produce action; a creature unaware of the amputation of a limb may still attempt to move it, and psychological states like terror may immobilise our muscles so that volition is ineffective. These phenomena do not constitute instrumental irrationality, as here the means are willed, and it is unreasonable to demand action where there is physical incapacity. Henry Sidgwick writes, ‘the “hypothetical imperative”… prescribes the fittest means to any end that we may have determined to aim at … .The adoption of an end as paramount … is quite a distinct psychical phenomenon from desire: it is a kind of volition, though it is, of course, specifically different from a volition initiating a particular immediate action. As a species intermediate between the two, we may place resolutions to act in a certain way at some future time’ (1907, p. 37). See also Broome, 2000 and 2001, who concurs with this formulation of the instrumental norm. It is has been objected that an intention (say) to kill Bob is not identical to an intention to act to bring it about that I kill Bob; the former intention need not include a commitment to perform any merely instrumental means. But such an intention is really something quite different: a conditional intention, to kill Bob if a certain kind of situation arises. The unconditional intention to kill Bob involves a commitment to bringing it about that I kill Bob. This much is admitted by Broome, who nonetheless maintains the possibility of such irrationality (without Argument, that I am aware). He writes, ‘When you intend to ␸, a causal process is in train that you believe will result in your ␸-ing. … To form the intention of ␸-ing, you must set such a process in train. If there is no process you can set in train that you believe will result in your ␸-ing, you cannot form the intention’ (2001, p. 187). In a weaker sense I can even be said to do it intentionally. Unlike thrashing in an epileptic fit, which I may do knowingly, my action is still intentional. Korsgaard insists that instrumental irrationality does not require diachronic inconsistency: I can flout hypothetical imperatives while simultaneously willing the end, because I have two distinct parts that can separately determine my actions; ‘one that wills, and one that is capable of resisting my will’ (1996, p. 231). The part that wills is my ‘self’, a reflective agent or ‘mind’ (1996, p. 228), the other consists of the impulses and desires ‘causally effective in or through my body’. I violate the instrumental norm when my mind is resolved upon an end, and my impulses and desires cause me to neglect required means. However, behaviour motivated by desire is voluntary, unlike reflexive behaviour, hence it cannot be external to or in conflict with the active will. Korsgaard’s Argument has a serious flaw; at most, absence of perceived law-like connection between self and action would entail that I cannot recognise that I exist. This evinces that Korsgaard’s transcendent ‘self’ is a non-existent fiction;

Stephen Finlay

22.

23.

24.

25.

26.

27.

28.

29.

30.

177

her normative principles command behaviour contrived to create the appearance of a unified cause. Existentialists condemn this as ‘Bad Faith’, the fabrication of a fictional self standing behind our individual actions. ‘I am trying … to make a conceptual point here about the nature of rational willing – that all and only instrumentally rational agents can be said to will ends, as opposed to merely “wanting” them. … Whereas we may merely “want” an end, but refuse to take what we regard as hateful means to achieving it … if we are instrumentally rational agents then, via our reason, we will commit ourselves to that end – that is, “will” it, insofar as we regard it as the best or the only end in the circumstances – and thus will not become derailed by the temptation for lesser goods’ (Hampton, 1998, p. 164). Joseph Raz writes, ‘The fact that the pursuit of the means is part of what makes one have the ends misled some into thinking that instrumental irrationality is impossible. If one does not pursue what one believes to be the necessary and available means, then one does not have the end. … However, even in [simple] cases it is not failure to pursue the means alone, but also the absence of feelings of conflict … guilt and remorse that establish I do not have the end’ (2005, p. 27). This Argument seems without merit; surely we can feel conflict, guilt, and remorse over ends that we have not willed, but merely desire or believe we ought to have willed. Also, ‘When any of these passions are calm, and cause no disorder in the soul, they are very readily taken for the determinations of reason. … What we call strength of mind, implies the prevalence of the calm passions above the violent’ (T, 2.3.3.8–10/417–18). A similar, albeit non-Humean, view of rational criticism, as observation of what is wrong with an agent’s behaviour from her own point of view, has since been persuasively advanced by Niko Kolodny (2005). Kolodny does not, however, question the possibility of instrumental irrationality. This suggestion can also be found in Foot, 1972a and Mackie, 1977. I argue for this view in my ‘Oughts and Ends’ (forthcoming), and more generally for an endrelational view of normativity in my 2004 and 2006. Elsewhere (forthcoming) I reject this appearance, and argue that the normative ‘must’ is nothing other than the alethic ‘must’ in an end-relational (in order that … ) context, and further that the normative ‘ought’ is nothing other than the so-called predictive ‘ought’ in an end-relational context. But the Argument of this essay doesn’t depend on these radical theses. This is now recognised by a number of philosophers (for example Scanlon, 2003; Raz, 2005), who deny that rational principles are normative in the strict sense of giving guidance. But these philosophers still accept the possibility of instrumental irrationality, viewing the principle as evaluative or as underlying criticism. Kolodny (2005, pp. 554–6) defends the normative status of rational principles. Suppose you ask me how you should accomplish some end. My answer, if cooperative, will not be to recommend you pursue what I take you to believe to be the best means, but rather to recommend to you what I believe to be the best means. The imperative I utter is not constrained by your epistemic situation. Note, however, that the hypothetical form is misleading – hence my rejection of the label ‘hypothetical imperative’. The preceding clause does not actually function as the antecedent of a conditional, but rather indicates the perspective of the ‘should’. (However, ‘If you have paedophiliac desires, then you should stay away from kindergartens’, is (ironically) legitimately described as a hypothetical imperative).

178

Against All Reason?

31. Hampton takes this stance, Korsgaard resists it. See also Beardman, 2007. 32. Similar scepticism about the importance of rationality is pushed by Kolodny (2005). 33. There is always a sense of ‘ought’ upon which this is true, but there might not be a sense on which it is always true.

8 Why Internalists about Reasons should be Humeans about Motivation Kent Hurtig

Introduction In ‘Internal and External Reasons’ Bernard Williams suggests that internalism about reasons is compatible with the idea that beliefs can motivate; that is, that internalism is compatible with cognitivism about motivation. However, contrary to Williams’ suggestion, I will argue that internalists about reasons are in fact committed to Humeanism about motivation; that is, that internalists about reasons cannot allow that beliefs can motivate in their own right. Since I believe that Humeanism about motivation is false, I think internalism about reasons implies something false, and that therefore internalism about reasons should be rejected. In this essay I shall say nothing about why I think Humeanism is false; rather, my aim will be to establish that internalism about reasons is compatible only with (or makes sense only on) a Humean theory of motivation. To do this, I will try to show that cognitive motivational internalism (CMI) is, at best, an unstable position.

8.1 Humeanism and internalism Although there are different versions of internalism and ‘the internal interpretation’ of statements like ‘A has a reason to ␾’, Williams’ thought that [A]ny model for the internal interpretation must display a relativity of the reasons statement to the agent’s subjective motivational set… (1981, p. 102) seems to be central to all versions of internalism. Externalists deny that reasons are relative in this way. According to one influential version of internalism about reasons for action which I will call ‘motivational internalism’: MI: A has reason to ␾ if and only if A is, or would be, motivated to ␾ after having deliberated soundly on all relevant facts. 179

180 How about: Why Reasons Internalists should be Humeans

This model says nothing about what it takes for an agent to be motivated, and as such it seems neutral between non-cognitivist and cognitivist theories of motivation. For motivational internalism (MI) to be plausible, however, and as Williams of course recognises, the ‘reason generating’ power of an agent’s motivations must be subject to what I will call the ‘no reasons from false beliefs’ proviso: NRFFB: ‘A member of [an agent’s subjective motivational set] S, D will not give A a reason for ␾-ing if either the existence of D is dependent on false belief, or if A’s belief in the relevance of ␾-ing to the satisfaction of D is false.’ (Williams, 1981, p. 103) In other words, if an agent’s motivation to ␾ is dependent on a false belief, then, ceteris paribus, it is not true of that agent that he has a reason to ␾. This, I take it, is common ground between internalists and externalists. Although Williams’ internalism has traditionally been viewed as a paradigmatically ‘desire-based’ theory of reasons, in a much overlooked passage in ‘Internal and External Reasons’, Williams says: Does believing that a particular consideration is a reason to act in a particular way provide, or indeed constitute, a motivation to act? … Let us grant that it does – this claim indeed seems plausible, so long at least as the connexion between such beliefs and the disposition to act is not tightened to that unnecessary degree which excludes akrasia. The claim is in fact so plausible, that this agent, with this belief, appears to be one about whom, now, an internal reason statement could truly be made: he is one with an appropriate motivation in his [subjective motivational state] S. (Williams, 1981, p. 107) It seems Williams is here offering us a cognitivist version of MI which we may call cognitive motivational internalism (or CMI for short). CMI is certainly on the face of it incompatible with the standard non-cognitive (Humean) theory of motivation. According to this theory: H: Beliefs are, by themselves, incapable of motivating an agent; only when an agent’s beliefs are combined with some suitable desire will he or she be motivated to act. Cognitivists, on the other hand, reject the idea that desires play this central role in motivation. According to cognitivism, at least some beliefs are capable of motivating in their own right. Normative and evaluative beliefs certainly seem to have this motivational dimension to them. While some cognitivists agree that the presence of a desire is indeed necessary for motivation,

Kent Hurtig

181

the necessity involved need not be a causal one. As Nagel puts it: ‘That I have the appropriate desire simply follows from the fact that [my beliefs] motivate me’ (1970, p. 29). So even if desires will be present whenever an agent is motivated, these desires need not play the role Hume assigned to them. On this understanding of the distinction between cognitivism and non-cognitivism, Williams’ internalism certainly seems compatible with cognitivism.

8.2 Beliefs and motivation According to CMI, then, A’s belief that [p] is a member of A’s subjective motivational set if A’s belief that [p] has motivational force.1 Although Williams’ internalism about reasons has traditionally been interpreted as being motivated (at least in part) by a Humean conception of motivation, by holding that beliefs can have this motivational dimension to them, Williams seems to distance himself from the traditional Humean picture. However, Williams does not abandon Humeanism completely. He says: A man who does believe that considerations of family honour constitute reasons for action is a man with a certain disposition to action, and also dispositions of approval, sentiment, emotional reaction and so forth. (1981, p. 107) Now it is unclear what the connection is between having these dispositions and being motivated by a belief. On Williams’ account, it seems that an agent’s having the relevant dispositions constitutes at least a necessary condition for that agent’s having motivating beliefs. But if this is so, it would be false, or at least grossly misleading to say that an agent’s beliefs can motivate in their own right; rather, they do so with the help of the relevant dispositions. The dispositions Williams explicitly mentions certainly have a non-cognitivist ring to them: ‘approval’, ‘sentiment’, and ‘emotional reaction’ are concepts usually associated with Hume and non-cognitivism. However, dispositions cannot, in any ordinary sense of the word, be called desires. Desires are propositional attitudes, dispositions are not. As far as I can see, cognitivists need not disagree with any of this. Cognitivists can admit that beliefs are not, strictly speaking, capable of motivating in themselves. In order for a belief to motivate, they could argue, we would have to be disposed in such a way that when we come to have the relevant beliefs, we would be motivated to act on them. But, again, this disposition need not be understood as a desire (cf. Parfit, 1997, p. 105; Broome, 1997). On the contrary, according to some philosophers, having the disposition to act in accordance with one’s normative beliefs is simply what being (practically) rational consists in (cf. Scanlon, 1998, pp. 22–30). According to Williams’ proposal, since beliefs about reasons can have motivational force (albeit with the help of certain dispositions), beliefs can

182 How about: Why Reasons Internalists should be Humeans

be members of an agent’s subjective motivational set. More significantly however, as MI tells us, if a belief is a member of an agent’s motivational set, this motivating belief can also provide an agent with a reason to act; that is, these beliefs can actually ground true reasons statements about the agent who entertains them.

8.3 The strange case of Owen Wingrave Let us now apply Williams’ cognitivist proposal to the story of Owen Wingrave. In the original story Owen has no desire whatsoever to join the army. His family, on the other hand, insist that considerations of family tradition and honour provide him a reason to join the army. Owen’s family know fully well that he has no desire to join the army, so when they say he has a reason to join the army, they take an externalist view about reasons. Suppose we change the story so that Owen has now come to believe that he has a reason to join the army because his family tradition bids him to do so; and, as a result, he is now motivated to join the army. If CMI is correct, it is now true of Owen that he has a reason to join the army. This seems relatively straightforward. Consider the following two cases: Case 1: (a) Owen believes that [the fact that family tradition bids Owen to join the army gives Owen a reason to join the army]. (b) Owen’s belief in 1(a) motivates him to join the army. So, according to CMI (c) Owen has a reason to join the army.2 Case 2: (a) Owen believes that [the fact that family tradition bids Owen to join the army gives Owen a reason to join the army]. (b) Owen’s belief in 2(a) does not motivate him to join the army. So, according to CMI (c) Owen does not have a reason to join the army. Are Case 1 and Case 2 correct descriptions of Owen and the reasons he has? According to CMI they are. Remember that according to NRFFB, since motivation resulting from false belief cannot give rise to reasons, it follows that if 1(c) is true, Owen’s belief in 1(a) must be true. But in virtue of what is Owen’s belief true? Because it motivates him seems to be the only answer. What about case 2? If 2(c) is true that must be because Owen’s belief fails to motivate him. But this in turn implies that Owen’s belief in 2(a) is false. And furthermore, Owen’s belief in 2(a) is false because it fails to motivate him. What we end up with if Williams and CMI are correct is the thesis that when an agent’s relevant factual beliefs are correct, his beliefs about what

Kent Hurtig

183

reasons he has are true when and because they motivate him – and false when and because they fail to motivate him.3 Let us sum up this thesis as: TMoNB: A normative belief’s resulting motivation is that belief’s truthmaker. Consider next: Case 3: (a) Owen believes that [the fact that family tradition bids Owen to join the army gives Owen a reason to join the army independently of his motivational state]. (b) Owen’s belief in 3(a) motivates him to join the army. So, according to CMI (c) Owen has a reason to join the army. Does Owen have a reason to join the army in Case 3? If the answer is yes – that is, if 3(c) is true, then, given NRFFB it follows that Owen’s belief in 3(a) is true. But if this is so, it follows trivially that ‘Owen has a reason to join the army independently of his motivational state’ is true. But this of course is nothing short of externalism! Supporters of CMI cannot get around this problem by saying that Owen’s belief in 3(a) is made true by his being motivated by that belief. Owen believes that it is the fact that family tradition bids him to join the army which gives him a reason to join the army; he doesn’t believe that it is his being motivated to join the army which gives him a reason to join the army. Cognitive motivational internalists might claim that is misleading. Owen’s belief in 3(a), they may argue, is really a conjunctive belief consisting of the following two elements: (i) Owen’s family tradition bids Owen to join the army; and (ii) The truth of (i) provides Owen with a reason to join the army. It might then be claimed that what does the motivating is (i) alone, not (ii) – or the conjunction of (i) and (ii). This answer is not convincing however. We could easily imagine that Owen is such that he would never be motivated by (i) unless he also believed (ii). In other words, we could imagine that, in Williams’ own terminology, Owen has a ‘pattern of emotional reaction’ such that he is always motivated only by what he takes to be bona fide external reasons; or that Owen’s ‘disposition of evaluation’ is such that when he believes he has a reason to do something, he believes this, at least in part, because he believes this reason obtains independently of his being motivated by it; If this is a genuine possibility (which it certainly seems to be), then Owen’s ‘externalist commitments’ (expressed in

184 How about: Why Reasons Internalists should be Humeans

part by his belief in (ii)) are central to his motivational set. Denying that beliefs of this kind can motivate does not seem like a plausible move. If beliefs can motivate, then so can beliefs of the kind Owen has in Case 3 (construed conjunctively). If cognitive motivational internalism cannot fault 3(b), it must say that Owen’s belief in 3(a) is false. But if this is so, then CMI has to reject TMoNB. Since Owen is motivated by his normative belief although this belief is false, his being motivated by it obviously does not make the belief true; that is, a normative belief’s motivational output is not that belief’s truth-maker. But if TMoNB is given up, what can remain of CMI? Not much it seems. If this principle is given up, cognitive motivational internalists will be at a loss to explain why, as they would have us believe, Owen has a reason to join the army in Case 1 and not in Case 2. He will have to find some way to allow TMoNB in cases like 1 and 2, but not 3. One cannot help but suspect that any such move will be completely ad hoc. There is another problem for the cognitive motivational internalist. Even if he can find some principled reason why TMoNB should be allowed to hold in Cases 1 and 2, but not Case 3, he has overlooked the fact that Owen’s beliefs in Cases 1 and 2 also display externalist commitments. If we look more closely at the content of Owen’s beliefs in these cases, we can see that what he actually believes is that it is the fact that his family tradition bids him to join the army that gives him a reason to join the army. Again, he does not believe that he has this reason because he is motivated to act on consideration of family tradition.

Conclusion In general, propositions of the form ‘The fact that p gives A a reason to φ’ (which is the kind of proposition Williams considers in the passage quoted above) must be construed as external reasons statements.4 And as such, they cannot ground internal reasons statements – even if they motivate – on pain of contradiction. Even if a proposition of the form just mentioned is true on an internalist reading – for example, if ‘p’ refers to what A is or would be motivated to do – and even if such a proposition motivates an agent, it is hardly going to be made true by its motivating power (if it has any). Rather, such propositions are more likely to be mere restatements of the internalist thesis itself. For the reasons outlined in this essay, I conclude that internalists should reject Williams’ proposal and align themselves with the only theory of motivation which makes sense for them, namely Humeanism.

Notes 1. An interesting question is whether we can actually draw a clear distinction between motivating and non-motivating beliefs. We may even be sceptical about

Kent Hurtig

185

the existence of non-motivating beliefs. According to some dispositional theories of belief, A’s believing that p consists in (among other things) A’s being disposed (motivated) to infer q if A also believes p → q; A’s being motivated to give up his belief that p if presented with good evidence for not-p; A’s being motivated to assert p if sincere, etc. But if this is correct, if beliefs essentially have this motivational aspect, then all of an agent’s beliefs would be in his S. The same point might be generalised to include all mental states that can be given a dispositional analysis. 2. For simplicity, I am going to assume throughout this essay that Owen has no other beliefs (or desires) that motivate him to join the army. 3. It may of course be contentious to call it a ‘fact’ that Owen’s family tradition bids him to join the army. I shall simply ignore this complication. 4. This is not to say that externalists are precluded from grounding reasons statements about an agent in that agent’s S. Externalism, as I understand it, merely denies that reasons statements must be grounded in an agent’s S.

9 Humean Sources of Normativity Herlinde Pauer-Studer

9.1

Introduction

A moral theory should, as many philosophers claim, offer us an account of obligation and motivation. It should address the following questions: How do we specify the reasons that morally require us to do X? How does the recognition that we are required to do X bring us to do X? The task is to provide an answer to the quite plausible claim that moral reasons should be normative as well as motivating. It is nearly a commonplace of contemporary moral philosophy that a Humean theory of morality cannot explain the normative force of moral reasons. Humeanism, so the familiar criticism goes, lacks a conception of practical reason which makes sense of categorical normative requirements. Instrumental rationality in the sense of means-end reasoning is the sole principle of Humean practical reasoning. However, we cannot justify unconditional normative principles or reasons on the basis of the principle of instrumental rationality. Therefore Humeanism misses the core of morality, namely identifying some actions to be obligatory, our desires and inclinations notwithstanding. At best, critics concede, Humeanism provides an account of motivation. Motivation is a matter of having a desire and a belief about the best way to fulfil the desire. Means-end reasoning tells you that you are required to do Y if you desire X, and believe that doing Y is the means to realise X. So Humeans, since they consider means-end reasoning to be a principle of understanding and regard desires to be the driving force of acting, are able to explain how one can be motivated to act. Kantian philosophers take the desire-based character of a Humean conception of agency and its apparently limited understanding of rationality to make their point. Kant, they claim, develops not only a plausible theory of how moral principles are normatively binding; he also offers a more convincing account of motivation than Humeanism. Moral motivation, in the Kantian view, cannot be a matter of just being driven by desires, impulses, and inclinations. Moral motivation presupposes the recognition that certain 186

Herlinde Pauer-Studer

187

considerations and reasons have weight and normative force. Moral motivation includes a cognitive evaluation, namely the insight that one is justified to do X and that, therefore, one has a reason to develop and follow the corresponding desires and inclinations. The most elaborate defence of the Kantian position along these lines has been given by Christine Korsgaard. Kant’s account of a good will, Korsgaard claims, rests on ‘a motivational analysis of the notion of duty or rightness’ (Korsgaard, 1996a, p. 60). This way obligation and motivation do not fall apart since a good-willed person ‘does the right thing because it is the right thing’. Korsgaard offers a specific interpretation of the Kantian position. She defends a ‘constitutive internalism’ according to which the principles of practical reason, the instrumental principle and the categorical imperative, are constitutive of the person as a rational agent. This provides an elegant solution to the questions of obligation and motivation. If moral principles are constitutive of our rational agency, then obligation and motivation are due to internal reflection on the preconditions of our being persons. A practical reason, Korsgaard tells us, ‘must function both as a motive and as a guide, or a requirement’ (Korsgaard, 1997a, p. 219). If normativity is a matter of the truth of external normative facts, the question of motivation, Korsgaard argues, remains mysterious – external truths might leave us cold and fail to have an influence on our ways of acting. A Kantian ‘constitutive internalism’ connects normativity and motivation via the insights of rationally reflecting agents: obligation is due to the recognition that, insofar as we are agents, we are committed to moral principles. Moreover, insofar as reason has an influence on our choices, we are motivated to act on behalf of reasons which have normative force. If a person is rational, then, according to Korsgaard, she ‘must be motivated by her own recognition of the appropriate conceptual connection between the belief and the desire’ (Korsgaard, 1997a, p. 221). Such an insight presupposes autonomous practical reflection, that is, reason in the sense of autonomous self-legislation. Korsgaard’s criticism of the Humean account of morality focuses on two points, namely that Hume’s account of instrumental reason is mistaken and, moreover, that Hume’s conception of moral approbation is insufficient. Her arguments are given here in more detail: First objection: A Humean position cannot provide an account of the normative force of the principle of instrumental rationality. On a Humean desire-based account of agency the principle of means-end reasoning just tells us to follow our desires, whatever they are. So means-end reasoning amounts to a description of our desire-based ways of acting, but not to a normative guideline which could be violated (see Korsgaard, 1997a). Second objection: Though Hume develops a conception of reflective rationality, his account of ‘reflective endorsement’ lacks the power to establish the normative authority of moral principles. Hume’s reflective

188 Humean Sources of Normativity

endorsement method has to be completed by a Kantian conception of autonomous agency that takes the principle of morality to be constitutive of autonomy and practical reason (see Korsgaard, 1996b). If these arguments are successful, then there is indeed not much room left for a Humean approach to morality. My aim in this essay is to assess Korsgaard’s objections against the Humean position. In regard to instrumental rationality, I will reject Korsgaard’s claim that the normative force of the instrumental principle can only be established in case the ends pursued do have unconditional value. A value-neutral reading of the instrumental principle is, as I try to show by referring to John Broome’s work, compatible with understanding instrumental rationality to be a normative requirement. Next, I try to point out some problems with Korsgaard’s argument that a Humean reflective endorsement method is deficient. Hume’s more moderate account of reflective rationality is not as inappropriate an account of practical reason as Korsgaard’s criticism suggests. Humean attempts to offer a conception of normativity in the form of reflective endorsement can avoid some difficulties facing Korsgaard’s strong constitutive internalism.

9.2 Korsgaard’s account of instrumental rationality Humeanism is usually associated with the thesis that the principle of means-end reasoning is the paramount principle of practical reason. Indeed, defenders of a Humean account of rationality and morality consider the principle of instrumental reason to be self-evident: to challenge it makes no sense. For example, James Dreier attributes to the principle of instrumental reason ‘a special status’ due ‘to its being the sine qua non of having reasons at all’ (Dreier, 1997, pp. 98–9). Means-end reasoning is a necessary condition for identifying reasons to act. However, the principle of means-end reasoning is indifferent to the value of the ends; it informs us about the relations between means and ends, given our beliefs about how best to realise particular ends we have adopted. Humeans are careful to draw the line between rationality and morality. They are sceptical towards a conception of practical reason whose core principle is the principle of morality, and the principle of instrumental reason having a diminished status. For Humeans it is clear that the move from rationality to morality demands an additional justification, a justification that cannot be simply provided by the content of practical rationality as such.1 Christine Korsgaard tells a different story about the sine qua non conditions of having reasons. For Korsgaard, the capacity to be an autonomous agent, that is, an agent valuing her or his own humanity as an end in itself, is indispensable for having reasons for action at all (Korsgaard, 1996b, p. 120 ff.). Her ‘truly Kantian strategy’ is to start with an account of rationality as ‘the autonomy of the human mind – and then to define reasons in

Herlinde Pauer-Studer

189

terms of rationality – say, as that which can be autonomously willed, or as those considerations which accord with the principle of autonomous willing’ (Korsgaard, 1997a, p. 243). The principles of autonomous willing are the hypothetical imperatives, namely the principle of instrumental reason and the principle of prudence, and the categorical imperatives.2 Compliance with the principle of instrumental rationality is ‘constitutive of having a will’ and having a will is ‘constitutive of being a person’ (Korsgaard, 1997a, p. 254). The instrumental principle is a normative requirement and a necessary condition of rationality. However, the instrumental principle cannot stand alone, it is not the ‘only requirement of practical reason’ (Korsgaard 1997a, p. 220). Korsgaard links the authority of the requirement of instrumental reason to the normativity of reasons: the instrumental principle only does have normative force if directed to ends we do have good reason or undefeated reason to pursue. As Korsgaard writes: ‘Unless there are normative principles directing us to the adoption of certain ends, there can be no requirement to take the means to our ends (Korsgaard, 1997a, p.220).’3 In the framework of Korsgaard’s account of agency, the adoption of the end has to be a choice based on a person’s free will. Given her conception of the will and rational agency, the ends must have value. Korsgaard’s interpretation of the instrumental principle as a constitutive condition of a free and rational will is due to a modification of Kant’s account which she thinks necessary to overcome the traces of rational dogmatism still affecting Kant’s program.4 Kant, she criticises, reads the principles of practical reason sometimes as external rational restrictions on maxims. Kant’s reconstruction of instrumental rationality presupposes independent normative facts in the form of externally given rational laws. Such an externalism Korsgaard considers deeply problematic. The assumption of independent normative facts in the form of externally given laws of rationality does not, she argues, provide us with a reason why we should conform to these laws. As Korsgaard writes:

[A]ccording to what we might call instrumental realism, facts about the instrumentality of actions to our ends support those reasons. The difficulty with this account in a way exists right on its surface, for the account invites the question why it is necessary to act in accordance with those reasons, and so seems to leave us in need of a reason to be rational. I have an end, and out there in the universe is a law saying what I must do if I have an end (take the means), but the reason why I must obey this law has not yet been given. To put the point less tendentiously, we must still explain why the person finds it necessary to act on those normative facts, or what it is about her that makes them normative for her. We must explain how these reasons get a grip on the agent. (Korsgaard, 1997a, p. 240)

190 Humean Sources of Normativity

The externalism problem confronts us on two levels: on the level of morality and the level of (instrumental) rationality. ‘Why be moral?’ is the sceptical question haunting moral justification. The requirements of morality might leave us cold, because the motivational link between external normative facts and our actions is missing. Korsgaard sees a parallel problem with respect to rationality. ‘Why be rational?’ is the scepticism haunting our accounts of practical reason. If, Korsgaard argues, the principles of rationality in the form of normative requirements amount to external logical truths and normative facts, there might not be any reason to be rational. If external instrumental realism holds, the question ‘why be rational’ remains unanswered. Korsgaard’s solution is, as mentioned already, a constitutive internalism claiming the laws of practical rationality to be the laws of agency. The principle of instrumental rationality is part of the autonomy of the mind, and no longer an external rational truth with no connection to our motivating reasons. So, quite in the same way as Korsgaard internalises the categorical imperative, the principle of instrumental rationality is constitutive of agency and willing: The idea that you could make a maxim and then apply the instrumental principle to it makes no sense. A maxim that does not already at least aspire to conform to the instrumental principle is no maxim at all. So the instrumental principle does not come in as a restriction that is applied to the maxim. Instead, the act of making a maxim – the basic act of will – conforms to the instrumental principle by its very nature. To will an end just is to will to cause or realise the end, hence to will to take the means to the end. (Korsgaard, 1997a, p. 244) Korsgaard considers internalism – apart from its potential to leave behind the legacy of dogmatic rationalism – as the alternative to Humean empricism. Hume, she claims, is not able to establish instrumental reason as a normative requirement. According to her reading of Hume, one is rational in case one has an end and pursues it in an effective way. The ends are set by the desires one has. They are arbitrary, merely dependent upon whatever our passions and inclinations are directed towards. Each end is worthy to be pursued as long as the desire for it is strong enough. Whatever a person desires, the person has reason to take the most effective means to realise that end. Since Hume lacks any conceptual resources to distinguish actual from rational desires, anyone just driven by his or her desires complies with the instrumental principle. If we identify the aims of the person with what the person prefers, one is always guided by what one thinks what one’s reasons are. But if any wish is giving us a reason, independently of its content, we are always practically rational. Korsgaard thus thinks that Hume’s account of

Herlinde Pauer-Studer

191

instrumental reason undermines the concept of practical rationality, since the alternative of irrational acting is missing. The instrumental principle lacks any normative force, because it cannot be violated. So the status of the instrumental principle as a requirement of rationality would be lost – a requirement would compel us to adopt certain desires and give up others (Korsgaard, 1997a, p. 233). Korsgaard’s example of a truly Humean character is Jeremy, who wants to study, then takes a walk, meets a friend, goes to a bar, and in the end comes home without having done anything of what he wanted to do with good reasons (Korsgaard, 1999, p. 19; 1997a, p. 64, n. 64). Jeremy’s problem is that he does not recognise a superior principle which would allow him to distinguish between important aims or reasons and spontaneous impulses and inclinations. To conclude: Korsgaard’s main objection is that Hume does not have a normative conception of instrumental reason, because his radical empiricism makes a normative requirement-interpretation of the principle of instrumental reason impossible. There are according to Korsgaard two readings of the instrumental principle, by which we can reconstruct the empiricist position: The first version reads: (a) ‘If you have a reason to pursue an end, then you have a reason to take the means to that end.’ This is, Korsgaard adds, a normative principle, since it ‘derives a reason from a reason, something normative from something normative’ (Korsgaard, 1997a, p. 223). The second formulation is: (b) ‘If you are going to pursue an end, then you have a reason to take the means to that end’ (ibid.). Korsgaard assumes that only the second interpretation agrees with Hume’s position ‘since it is perfectly clear that he thinks that reason does not play a role in the determination of ends. He would have to believe that the instrumental principle instructs us to derive a reason from what we are going to do’ (ibid.). However, Korsgaard objects, principle (b) contradicts another principle of reasoning that Hume highly cherished, namely that you cannot derive an ‘Is’ from an ‘Ought’. But the validity of an ‘is-ought’ derivation is exactly presupposed by the second formulation of the instrumental principle, that is, formulation (b) above: from the fact that you desire X, you derive that you have a reason to do X. Therefore, Korsgaard concludes, Hume could not have believed in instrumental reason. Hume does not give us a normative account of instrumental reason. Korsgaard considers her criticism of Hume as sufficient to reject Humean accounts of instrumental rationality in general: ‘The instrumental principle, because it tells us only to take the means to our ends, cannot by itself give us a reason to do anything. It can operate only in conjunction with some view about how our ends are determined, about what they are’ (ibid.). Such an assessment of the ends Korsgaard thinks to be possible only in the framework of a Kantian conception of practical reason.

192 Humean Sources of Normativity

9.3 Problems of Korsgaard’s account of instrumental reason I will now address some problems of Korsgaard’s interpretation of instrumental rationality. First, I discuss some shortcomings in her reading of Hume’s views on means-end reasoning. Second, I try to show that Korsgaard’s valuebased conception of instrumental reason is a consequence of her failing to distinguish clearly between the normativity of rationality and the normativity of reasons.5 To read the instrumental principle as a normative requirement does not commit one to the claim that one must have a reason to take the means to an end. Therefore the ends pursued need not necessarily be valuable or worthwhile projects. Hume’s value-neutral reading of instrumental reasoning seems to me plausible. In regard to the first point of my criticism: Korsgaard’s interpretation of Hume’s conception of instrumental reason is based on what Hume says in the famous section ‘Of the influencing motives of the will’ in Book II of the Treatise (T, 2.3.3/413–18, selby Bigge–Nidditch ed.). Hume’s aim in that part of the text is, he tells us, twofold: he tries to show, first, ‘that reason alone can never be a motive to any action of the will’; secondly, he argues ‘that it can never oppose passion in the direction of the will’ (T, 2.3.3.1/413). Often interpreters have read Hume’s discussion of the motives of the will as proving Hume’s complete scepticism in regard to practical reason. Indeed, Hume makes some rather provocative claims in his investigation of the motives of the will, the most famous ones being: ‘’Tis not contrary to reason to prefer the destruction of the whole world to the scratching of my finger. ’Tis not contrary to reason for me to chuse my total ruin, to prevent the least uneasiness of an Indian or person wholly unknown to me. ’Tis as little contrary to reason to prefer my own acknowledg’d lesser good to my greater, and have a more ardent affection for the former than the latter’ (T, 2.3.3.6/416). Considered in isolation, these passages seem evidence for Hume’s hostility to reason. Korsgaard adopts, as far as Hume’s interpretation of instrumental reason is concerned, such a reason-sceptical understanding of Hume. She attributes to Hume a theory of ‘revealed desires’: whatever you do expresses the desires driving and motivating your actions.6 She takes Hume’s allegedly outrageous claims about the bewildering variety of desires rather literally: the only parameter determining the nature of our ends is the subjective intensity of the wish. There is no rational or critical assessment of desires, no distinction between actual and rational desire – since reason, as Hume says, does not have any influence on the motives of the will. It seems to me mistaken to read Hume’s allegedly provocative remarks in that way. Hume, in that passage on the motives of the will, does not discuss reflective rationality, but reason in the sense of understanding, namely abstract or demonstrative reasoning and means-end reasoning. His claim is that these forms of reasoning do not directly influence our actions, but direct our judgements in regard to cause and effect on the one hand, and

Herlinde Pauer-Studer

193

means-end relations on the other hand. In particular cases, reasoning leads us to a judgement that some passions are unreasonable (T, 2.3.3.6/416). However, Hume states that nothing can be judged to be reasonable or unreasonable except by appealing to a standard of truth or reason. We cannot call affections and passions reasonable without a standard of assessment: (N)othing can be contrary to truth or reason, except what has a reference to it, and as the judgments of our understanding only have this reference, it must follow that passions can be contrary to reason only so far as they are accompany’d with some judgment or opinion. (T, 2.3.3.6/416) Hume also informs us that there are ‘two senses, that any affection can be call’d unreasonable’ (T, 2.3.3.6/416). The first one is at stake if our passions rest on false imagination; for example, if we feel ‘hope or fear, grief or joy, despair or security’ in situations where there is no object which could justify those affections. The second instance of displaying unreasonable affections is given when we, blindly following our passions or desires, choose means inappropriate to pursue our ends, and in case we ‘deceive ourselves in our judgment of causes and effects’ (T, 2.3.3.6/416). Understanding has, according to Hume, an influence on our actions – but merely in the way that we check whether our actions follow the rules of reasoning. Hume clearly sanctions the violation of means-end reasoning. Means-end reasoning is a principle of reasoning. It requires us, to take the appropriate means, if we desire an end and believe that a particular meansend connection holds. Hume rejects the famous combat of will and action, of reason and passion, since he considers reason and passion not to be in general opposition. They merely conflict in case a passion can be criticised as unreasonable. However, if affections and passions are deficient in that respect, then reason in the form of a judgement about the proper connections between means and ends tells us to modify them and bring them into conformity with reason’s judgement. Hume emphasises the influence of reflective judgement, when he writes: Since a passion can never, in any sense, be call’d unreasonable, but when founded on a false supposition, or when it chuses means insufficient for the design’d end, ’tis impossible, that reason and passion can ever oppose each other, or dispute for the government of the will and actions. The moment we perceive the falsehood of any supposition, or the insufficiency of any means our passion yield to our reason without any opposition. I may desire any fruit as of an excellent relish; but whenever you convince me of my mistake, my longing ceases. I may will the performance of certain actions as means of obtaining any desir’d good; but as my willing of these actions is only secondary, and founded on

194 Humean Sources of Normativity

the supposition, that they are causes of the propos’d effect; as soon as I discover the falsehood of that supposition, they must become indifferent to me. (T, 2.3.3.7/416–17) In that passage means-end reasoning is clearly described as a principle displaying normative guidance. If I detect that there is no connection between specific means and ends, or that, for example, I assume wrongly that I can realise my end, reflection tells me to modify or give up my end. So Hume’s statement that ‘our passions yield to our reason without any opposition’ does not mean, as Korsgaard supposes, that Hume’s desire-based creatures are following the instrumental principle whatever they do, and that they therefore cannot violate the dictates of instrumental reason. Our passions, Hume tells us, conform to reason, not because reason lacks normative force, but because reason tells us to modify and rethink the passions in case they incline us to take means inappropriate for pursuing an end effectively. Korsgaard proposes a different reading of the passage cited above. She says that the deficiencies mentioned by Hume, namely a false supposition and the choice of inappropriate means, are merely ‘cases of mistake’, not violations of rationality, since ‘the actions that result are not, strictly speaking, irrational (Korsgaard, 1997a, p. 228).’7 However, to admit the possibility of ‘mistake’ in the sense of erroneously choosing inefficient and inappropriate means to realise an end, amounts to admitting that the principle of instrumental reason can be violated. If, for example, you pursue your end to participate successfully in the Vienna City Marathon by just watching marathon competitions on TV, instead of running regularly, then you are instrumentally irrational: you fail to take the means to your end.8 And, given what Korsgaard says about normative strength, making room for the possibility of mistaken means-end reasoning is all Hume needs to establish the normative force of the principle of instrumental reason. It must be possible for a person to will an end, and yet to fail to will the means to that end – otherwise the principle of instrumental reason does not meet the conditions of being a requirement of rationality. Yet, as I tried to show, Hume does not exclude infringements of correct means-end reasoning as such. These infringements can, of course, be merely ‘matters of mistakes’: they can occur in the form of erroneous assumptions about means being effective, and errors about the relations between certain means and ends, and in failures to draw the conclusion about what you are required to do, in case you desire something, and believe that X is the means to realise your desire. Hume in a way tells us that you can fulfil the requirement of means-end reasoning by giving up the end or by taking the means;9 hence a violation of means-end reasoning would be possible in case you intentionally ignore those options. Hume does not hold a ‘revealed desire account’; he leaves room for the reflective agent to reason in accordance with or in violation of the principle

Herlinde Pauer-Studer

195

of means-end reasoning. So, it is not true that ‘Hume has no resources for distinguishing the activity of the person herself from the operation of beliefs, desires, and forces in her’ (Korsgaard, 1997a, p. 233). To make judgements, to assess whether one’s passions are driven by superstition or amount to appropriate responses to given facts or ends, requires that the reflections and evaluative assessments of the person do play a role. In addition to meansend reasoning Hume adopts a conception of ‘reflective rationality’. The principle of means-end reasoning does have normative force. We can respect or violate it in our activities and derivations of understanding. Hume points out the limits of demonstrative reasoning and means-end reasoning. ‘Where a passion is neither founded on false suppositions, nor chuses means insufficient for the end, the understanding can neither justify nor condemn it’ (T, 2.3.3.7/416). Means-end reasoning demands to take the means, in case you have an end and a belief about the appropriate means to realise it. You can violate that principle: you might not choose the means conducive to the end. The function of reason is to help us detect the proper relations between causes and effects, between means and ends. But means-end reasoning does not allow or help one to assess the value of ends and desires as such. Meansend reasoning by itself does not inform us about the truth of our beliefs or the goodness or value of our ends and desires. The assessment of truth and goodness presupposes a judgement in regard to standards of truth and practical reflection. So Hume’s remarks about the bewildering form our desires might take are, besides an affirmation of the wide variety of humans’ wishes, a reminder of the limits of means-end reasoning: by itself, means-end reasoning does not allow an evaluative assessment of the desires; that would be beyond its scope. And Hume is correct to say so. In fact, Korsgaard concedes that point, namely that Hume presupposes the value-neutrality of the means-end principle. Hume, she tells us, ‘thinks we neither ought-to-want nor really-want only those ends which are consistent with our overall good’ (Korsgaard, 1997a, p. 231). She considers that to be a crucial deficiency of Hume’s position, undermining the normativity of the instrumental principle. However, by putting forward that objection Korsgaard makes the normative force of the instrumental principle dependent on something different than the normativity coming with means-end reasoning as a principle of logically correct reasoning. Her critical argument against Hume assumes that instrumental reason only has normative force if the ends pursued do have value. And that is exactly Korsgaard’s position: ‘[F]or the instrumental principle to provide you with a reason, you must think that the fact that you will an end is a reason for the end. … It means that your willing the end gives it a normative status for you, that your willing the end in a sense makes it good’ (Korsgaard, 1997a, pp. 245–6). But this is a problematic objection, since it conflates the normativity of rational requirements with the normativity of reasons. We should distinguish these two forms of normativity. The ‘normativity of rationality’ is the kind

196 Humean Sources of Normativity

of normativity connected with requirements like consistency, coherence, entailment, and means-end reasoning. The ‘normativity of reasons’ is due to the weight reasons have. The source of normativity in regard to logical requirements is a different one than the source of normativity in the case of reasons. Requirements of rationality demand not to violate formal standards of rationality like consistency, coherence, and the rules of valid reasoning. Logical standards are value-neutral. Therefore, normative requirements of rationality do not answer questions of what we should do, all things considered. That is a matter of assessing the weight of reasons we have for doing X or refraining from doing X. The normative power of reasons depends on the justification we can provide for them to count as good or compelling reasons. The reflection on the weight of reasons has to work within the framework provided by formal conditions of rationality – yet the question of fulfilling or violating the conditions of rationality has to be kept separate from the question whether we do have good reasons for acting. The main argument for the separation of the normativity of requirements and the normativity of reasons is the problem of bootstrapping. An account of practical reasoning must, as Michael Bratman has pointed out, avoid ‘bootstrapping’: logically valid practical reasoning cannot create for you a reason to do something. The requirement of means-end reasoning by itself cannot bootstrap a reason into being (Bratman, 1987, pp. 23–7).10 Applied to the principle of instrumental rationality, this means: you have to formulate the principle of instrumental reason in such a way, that it amounts to a normative requirement to take the means, in case you have an end X and believe that doing Y is the means to realise X. The question whether you have a reason to do Y is a separate issue. For example, that you intend to participate in the Vienna City Marathon does not provide you with a reason to start with your daily training sessions, though you believe that running daily is necessary to participate successfully in the marathon. It might well be that you have no reason to do so, since due to a cardiac insufficiency you might not survive running a marathon. The fact that you form an intention to pursue end E does not create a reason for you to take the means to realise E. The point can also be made in regard to Korsgaard’s formulation of the principle of instrumental reason in terms of willing.11 The fact that you will an end X, and believe that if you do Y you will realise X, does not give you a reason to will the end, and hence not a reason to take the means. The claim that your willing an end gives you a reason to will the end, and consequently a reason to will the means, amounts, as John Broome following Bratman points out, to bootstrapping since ‘a reason would be being pulled into existence out of nothing’ (Broome, 2002, p. 93). So if by means-end reasoning a volition gives rise to another volition, the question whether we do have reason to follow the volition derived, has not been answered. Reasoning just works via the logical connections between the contents of propositions, and those connections

Herlinde Pauer-Studer

197

create a normative requirement. The relation of normative requirement is, as Broome expresses it, ‘given by the correctness of the reasoning’ (Broome, 2002, p. 93). Reasoning is a relation between mental states; in the case of means-end reasoning an intention, a belief (about the means to an end), and an intention (as the conclusion of reasoning). It is important to have a neutral notion of instrumental rationality – means-end reasoning as such does not turn the end into something good. Intending an end does not provide one with a good reason to realise it. As Joseph Raz rightly says: ‘(A) would-be murderer cannot create for himself a reason for poisoning his intended victim just by making it his goal to kill him’ (Raz, 2005, p. 3). Broome suggests that the crucial distinction between the requirementrelation and the reasons-relation can be made explicit by focusing on the different scope the ought-operator has, dependent on whether ‘ought’ expresses a reason or whether ‘ought’ indicates a normative requirement. In the case of reasons, ‘ought’ is a narrow-scope operator. The reasons-relation is a slack and detaching relation: if p is a reason for q, it follows that you have a reason to q; you ought to q. In the case of a normative requirement ‘ought’ is a ‘wide-scope ought’, the ought-operator applies to the whole conditional. Normative requirements are for Broome strict non-detaching normative relations. The normativity of instrumental rationality is captured by a ‘wide-scope ought‘, that is, the ‘ought’-operator governs the whole conditional of instrumental deliberation: You ought (to intend to M if you intend to E and you believe your M-ing is a necessary means to your E-ing). (Broome, 2004, p. 29)12 The wide-scope interpretation forbids detachment; and therefore it does not entail that you ought to take the means to end X (in the sense of a narrowscope ‘ought’).13 Since the interdiction of detachment holds for instrumental reasoning in general, it equally holds for the case of a commitment to worthless or morally bad ends. Normative requirements, as Broome states, are based on logical relations between the content of propositions (Broome, 2000, p. 82). The logical relations are the source of normativity. You are required, if you believe the premises, to believe the conclusion; you are required, if you intend the premises, to intend the proposition stated in the conclusion. Therefore, Broome argues, it is wrong to assume that instrumental reasoning provides you with a reason to take the means: ‘But instrumental reasoning does not provide you with a reason to take a means. That is not how it works. Willing (or intending) an end normatively requires you to will whatever you believe is a necessary means to the end’ (Broome, 2000, p. 98). A consequence of keeping the normativity of requirements apart from the normativity of reasons is that instrumental reasoning is completely neutral

198 Humean Sources of Normativity

in regard to the value of the ends. Instrumental rationality can neither confer any sort of value on the ends, nor is its normative status touched in any way by the value of the ends. Instrumental rationality is a sort of rational requirement that is neutral to the good and the bad. As Broome notes: Instrumental reasoning brings you to take appropriate means to your ends, and it is not paralyzed if your ends happen to be ones you should not have. Similarly, your theoretical reasoning works well and in the same way, whether or not it is premised on beliefs you should not have. (Broome, 2000, p. 98) What can we conclude in regard to Korsgaard’s interpretation of the principle of instrumental reason and her criticism of Hume’s account of means-end reasoning? To recall: Korsgaard argues that the principle of instrumental rationality only does have normative authority in the context of ends that do have value, moreover, that do have unconditional value. So, only if there is normative force in regard to the ends, is there a normative requirement to take the means. If you will an end, you ought to take the means, insofar as you are rational. This leaves us with a narrow scope ‘ought’ that expresses that you have a reason to take the means. However, the normativity established by valid reasoning cannot be used to assess the normative status of ends and means. To repeat the point made before: the logical relations between the contents of the propositions at stake in means-end reasoning14 entail normativity in the sense of a requirement to reason correctly; but means-end reasoning does not entail that the ends do have a normative status. Means-end reasoning does not inform us about the worthlessness or goodness of our ends. Whether the ends as such do have normative force is a matter of assessing the value of the ends. Korsgaard’s conception of ‘will’ is a specific one. Due to her adoption of a Kantian notion of the will, willing is connected to a very specific form of autonomous reflection, namely autonomy in the sense of understanding yourself to be bound by the moral law. Take the following passage of her text: ‘The instrumental principle can only be normative if we take ourselves to be capable of giving laws to ourselves – or, in Kant’s own phrase, if we take our own wills to be legislative’ (Korsgaard, 1997a, p. 246). Willing is for Korsgaard good willing; her notion of a will is a Kantian good will based on the moral law.15 Due to her moralised notion of willing – the principle of acting is the moral law – the distinction between the requirement-relation and the reasons-relation is blurred. Since the idea of the autonomy of the will, the freedom to make a decision or a choice, is connected with the moral law, the objects of your willing are by definition good or valuable ends. Therefore, since the end is a good one, you have a reason to pursue the end, and accordingly you have a reason to take the means. Korsgaard’s assumption that the ‘fact that you will an end is a reason for the end’ is

Herlinde Pauer-Studer

199

restricted to a specific form of willing, namely willing those ends which we can justify as being good or valuable. The problem is that Korsgaard extends her moralised conception of the will to the principle of instrumental reason. Korsgaard’s problem is not that the detachment of ‘I ought to will the means’ might not be warranted; in her account of unified agency, which commits a rational agent to unconditionally valuable ends, the detachment would not be a problem. But meansend reasoning as such can also be correctly done in the context of ends we consider to be valueless or simply wrong. Of course, one can claim that good willing, in the sense of acting on a moral principle, might be tied to correct means-end reasoning. But claiming that a good will often presupposes valid means-end reasoning does not mean that means-end reasoning as such amounts to good willing. Kant himself had a different conception of the normative force of the instrumental principle. He claims that the principle of means-end reasoning is analytic and neutral towards values (see Kant, 1994, pp. 27, 25 (Academy edition, 417, 414)). In the case of a principle of skill, it is, for Kant, irrelevant whether the end is reasonable or good; the relevant issue is how we can reach the end. He illustrates the end-neutrality by a rather drastic example: the prescriptions that a doctor needs to cure a man and those that a poisoner needs to kill the man are, as Kant claims, of ‘equal value’ since each prescription ‘serves to bring about its purpose perfectly’ (Kant, 1994, p. 25 (Academy edition, 415)). The hypothetical imperative is for Kant either a principle of skill or a principle of prudence. Kant is careful to distinguish the analytic form of the hypothetical imperative, the principle of means-end reasoning, from the non-analytic version of the hypothetical imperative, namely the principle of prudence. The principle of prudence is addressed to the end of one’s own happiness. Here the value of the means is dependent on their being appropriate to realise a specific end, namely happiness or well-being. In contrast to the principle of means-end reasoning, the imperatives of prudence are not analytic, since the concept of happiness is according to Kant not determinately given – it depends on the circumstances of a life whether long living, being rich, or being successful will bring happiness.16 Kant’s claim about the value-neutrality of the means-end principle matches what Hume says. Hume talks of affections and passions instead of intentions, maxims, and willing. He assumes that we can merely judge or criticise meansend reasoning in regard to the correctness of the belief on which it relies, namely that particular means are in fact appropriate and sufficient to realise an end. His claim is that means-end reasoning is one of the valid principles of understanding – independent from the question whether the affections and passions at stake are good or bad, immediately approved of or not.17 To sum up the discussion so far: Korsgaard’s objection that on Hume’s interpretation the principle of instrumental reason lacks normative force

200 Humean Sources of Normativity

since it cannot be violated, is not valid. Her second critical argument, namely that Hume’s account of means-end reasoning is deficient, since it is neutral in regard to the value of the ends, is not convincing. The principle of instrumental rationality is normative in regard to the process of correct means-end reasoning; the assessment of the value of the ends and the reasons we might have for pursuing ends or for giving up certain ends is beyond its scope. Means-end reasoning cannot provide a normative standard to evaluate the ends; its normativity is not dependent on the ends.

9.4 Hume and ‘Reflective Endorsement’ In her book The Sources of Normativity Korsgaard offers a more sympathetic reading of Hume. She attributes to Hume’s moral philosophy a conception of normativity. Hume, she claims, offers us a method of reflective endorsement that allows us to confirm our moral judgements and assess whether our sentiments of approval and disapproval are rightly held (Korsgaard, 1996b, p. 55). Korsgaard in fact makes Hume’s position very strong: she emphasises that Hume brings a form of objectivity into his conception of moral judgements by the requirements of the general point of view.18 The first of these requirements is that we should not evaluate the character of persons from an idiosyncratic point of view, but from the point of view of our sympathy with the person and the persons of her narrow circle. The second one demands, Korsgaard says, that we must judge the characteristics of a person according to general rules (Korsgaard, 1996b, p. 55). Hume’s method to come to an agreement in regard to the qualities and deficiencies of a character can be extended into a general method of rational agreement on moral principles or judgements. That way, she concludes, Hume offers us an account of normativity. Korsgaard points out that we can even detect a notion of acting from duty in Hume’s conception of morality (Korsgaard, 1996b, pp. 54–8). Though Korsgaard characterises Hume ‘as the major traditional representative of a theory of normativity’, she considers his account of normativity to be deficient. She argues that the only way to make Hume’s reflective endorsement test viable is by extending it into a Kantian account of normativity. As Korsgaard writes: ‘I will end by saying why I think the logical consequence of the theory of normativity shared by Hume … is the moral philosophy of Kant’ (Korsgaard, 1996b, p. 51). Korsgaard’s main objection is that Hume does not apply his method of reflective endorsement universally because he connects the test of reflective endorsement not only with impartiality, but with ‘general rules’, rules that lack specificity. Hume, she concedes, certainly tries to subject our sentiments and dispositions of character to critical moral assessment. Yet since the normative devices for that moral testing are general rules, Hume’s procedure does not, she criticises, cover each and every particular case. In order that

Herlinde Pauer-Studer

201

we can examine not only ‘general tendencies’, but each single disposition, it is necessary, Korsgaard points out, to understand Hume’s appeal to general rules as an appeal to moral principles which we can will to be universal laws for all of us. Seen this way, Kant’s ethics offers the appropriate reformulation of Hume’s test of reflective endorsement and Hume’s conception of morality. The step from Hume to Kant seems inevitable. Korsgaard’s critical argument is a version of those objections familiar from discussions of rule utilitarianism: in certain cases it seems reasonable to violate a rule since the violation produces a better result than the strict adherence to the rule. Korsgaard uses the example of ‘a slightly more attractive version of Hume’s sensible knave’, namely the case of a lawyer who discovers that a recently deceased client has changed his will. The client has eventually decided to leave his money to his careless nephew and not – as originally intended – to medical research. Since it would be far better to give the inheritance to medical research, the lawyer has, Korsgaard concludes, a normative reason to violate the deceased client’s will. However, such a deliberation presupposes a specific weighing of reasons in the particluar case which, Korsgaard says, cannot be provided by general rules which normatively apply merely to dispositions ‘which are themselves tendencies of a general kind’ (Korsgaard, 1996b, p. 88). We need, as Korsgaard puts it, a method of reflective endorsement that extends to all of our dispositions. To have a viable conception of normativity that covers particular cases of inclinations, we must subsume all of our particular motives and inclinations to the test of reflection. However, with endorsing that condition we are on Kantian grounds: to check whether all of our inclinations constitute good reasons or not means to ask whether these reasons can be a law for us. In Korsgaard’s words: [W]hat we need in order to establish the normativity of our more particular motives and inclinations is the reflective endorsement of those. That after all is the whole point of using the reflective endorsement method to justify morality: we are supposing that when we reflect on the things which we find ourselves inclined to do, we can then accept or reject the authority those inclinations claim over our conduct, and act accordingly. But what I have just described is exactly the process of thought that, according to Kant, characterizes the deliberations of the autonomous moral agent. … Since a reason is supposed to be intrinsically normative, we test a motive to see whether it is a reason by determining whether we should allow it to be a law to us. And we do that by asking whether the maxim of acting on it can be willed as a law. (Korsgaard, 1996b, p. 89) So we come to see that autonomy in the sense of self-legislation is the source of normativity and obligation.

202 Humean Sources of Normativity

However, the problem Korsgaard sees at stake in Hume’s appeal to general rules results only when we postulate strict adherence to rules as absolutely prior to particular reasons. Such a form of rule-fetishism would undermine the plausibility of each moral theory. It is a familiar objection, for example, that rule utilitarianism generates counterintuitive results, if we associate it with an absolute priority of rules. In particular cases, so the well-known argument, utilitarian standards tell us to violate the rule.19 The same holds for self-interest versions of contractualism and their vulnerability to ‘sensible knave-strategies’: if a self-interest maximising strategy is the self-evident and incontestable condition of rationality, then it is not merely tempting, but rationally justified to violate an agreement, in case a particular infringement does not endanger the general terms of cooperation, yet increases one’s utility. Even Kantianism is not free from the dangers of rule-fetishism: the idea of following a law could equally deteriorate into a moral rigourism which ignores taking into account the moral relevance of special relations and duties to others. Of course, Kantian testing procedures can do better – they can be used for assessing maxims that include those very specific moral aspects. But the same holds for Hume’s reflective endorsement test: it is a form of reflective assessment that by itself does not commit us to abstract from particular inclinations, sentiments, or dispositions. Our sentiments of approval or disapproval, of ‘blame or praise’, change, Hume tells us, according to nearness or remoteness. Social interaction on any reasonable terms would be impossible, if we judged other persons only from our ‘peculiar point of view’ (T, 3.3.1.15/581). Our moral judgements cannot be based on our immediate feelings; they need to be reflected and corrected if tainted by the stronger effects of pleasure or disapproval, of ‘affection and admiration’, in case other persons and characters are close to us. Our diligent and devoted servant, Hume says, causes stronger immediate sentiments than a historically distant character like Marcus Brutus or a virtuous person in ancient Greece, yet we could reach a more balanced and appropriate judgement by reflection and ‘contemplation … of particular qualities or characters’ (T, 3.3.1.15/581). So we have to adopt ‘some steady and general points of view’ in order to correct the idiosyncrasies of our subjective moral perceptions and feelings. The general point of view is a standpoint of objectivity that indicates what our feelings and sentiments would be, if we were not captives of our immediate impulsive reactions. Our moral judgements are reflective and considerate expressions of our feelings – they are, characteristic for Hume’s conception of morality, still based on sentiments, but not on immediate emotional reactions distorted by contiguity, closeness, and physical and historical distance. It seems worth noting that Hume does not understand general rules as those normative devices we have to follow in our reflections from a general point of view. Hume introduces the notion of general rules to control the

Herlinde Pauer-Studer

203

cases where our passions, easily influenced and driven ‘by degrees of liveliness and strength’, are due to imagined, but not real objects and causes. Sometimes passions are stronger than our beliefs, sometimes they are aroused ‘independent of the real existence of their objects’ (T, 3.3.1.20/585). The general rules are, as Hume tells us, means that remind us of the probability of specific relations between cause and effect; they indicate to us where our imaginations fail to be correct or complete representations of the real connections between our passions and the things causing them. They help us to avoid false inferences from effects to causes and, moreover, to elude mistaken, or at least incomplete, assumptions about causes. ‘General rules’, Hume tells us, ‘create a species of probability, which sometimes influences the judgment, and always the imagination’ (T, 3.3.1.20/585). Sure, Hume does talk of ‘general principles’ which ‘command our judgment and opinion’ (T, 3.3.1.18/584). But these principles, different from the general rules mentioned before, are just those standards of critical reflection that constitute a ‘method of correcting our sentiments, or, at least, of correcting our language, where the sentiments are more stubborn and unalterable’ (T,3.3.1.16/582). These principles are impartial standards which are not biased by our subjective point of view. Impartiality amounts to a considered form of reflection free of the distortions of personal interests; it is not a form of generality that ignores concrete details of moral relevance like specific inclinations and dispositions. To conclude: my aim in this essay has been to show that recent Kantian claims notwithstanding, there is still room for Humean accounts of rationality and normativity. Korsgaard’s criticism that Hume’s conception of reflective endorsement cannot reach the more particular dispositions on which we act, does not do justice to Hume’s views on moral justification. The focus on particular inclinations and dispositions is part and parcel of Hume’s sentiment-based understanding of morality. Thus Korsgaard’s claim that Hume’s method of reflective endorsement finds its completion merely in Kantian moral theory seems unwarranted. There are remarkable resources of moral approbation in Hume’s theory, resources that amount to a conception of normativity that guides not only our thoughts and imaginations but also encourages us to develop into moral beings by a steady correction and refinement of our sentiments. Hume provides us with a conception of normative reasons – they are those judgements which we can rightfully hold. Hume presupposes the autonomy of the reflective mind in forming reliable judgements. But according to his account moral reflection is a form of subtle moral authority that does not commit us to a law-based form of selfconstitution that conceives of moral goodness and moral deficiency merely as a matter of being governed by the right or wrong kind of law. Korsgaard is correct that autonomy is the source of normativity; yet we are not necessarily committed to share her strong notion of autonomy that presupposes the moral law not only as a guideline of willing, but as constitutive of a will.

204 Humean Sources of Normativity

Notes 1. As James Dreier, defending a Humean position, notes: ‘Our skepticism should consist in doubts that the content of practical reason is anything like the content of morality’ (1997, p. 99). 2. The categorical imperatives are: the Universal Law Formula, the Humanity Formula, and the Autonomy Formula, the idea of self-legislation. The step to the Universal Law Formula is provided by Kant’s argument that the form of a law is its universality, hence that the Universal Law Formula fulfils exactly the conditions of a self-legislating will and is therefore the principle of autonomous willing. The Humanity Formula is connected with the Universal Law Formula via the argument that a categorical imperative has to be bound to unconditional values (otherwise it would be a hypothetical imperative, dependent on a conditional value). Humanity is such an unconditional value (in fact the only unconditional value), that is, a value which is not dependent on other conditions. The conditions for an unconditional value are: being valued for its own sake and having the source of value in itself. Cf. Korsgaard, 1996c. 3. The thesis that it depends on the value of the end whether we have a reason to take the means is common ground between Joseph Raz and Korsgaard. In his discussion of instrumental rationality Raz writes: ‘Crucially, the way goals acquire their normative relevance is by being conditions on the applicability or stringency of reasons. Therefore, they can have that effect only if the goals are worth pursuing in the first place. On this point my account is close to Korsgaard’s’ (Raz, 2005, 23). Raz talks here about the normativity of reasons, not about the normative force of requirements. 4. Kant’s allegiances to rational dogmatism are, Korsgaard thinks, observable in Kant’s Groundwork. 5. For the distinction between the normativity of requirements and the normativity of reasons see Kolodny, 2005, pp. 509–12. 6. Korsgaard does not use the term ‘revealed desire’; but in analogy to the concept of ‘revealed preferences’ the formulation seems an appropriate characterisation of Korsgaard’s interpretation of Hume’s views. 7. Hume, Korsgaard claims, allows only for actions being irrational in ‘two derivative ways’ (Korsgaard, 1997a, p. 228). 8. As Jonathan Dancy points out, the principle of instrumental rationality is a meaningful normative principle. It demands of us to take the most effective means to our ends. Yet, Dancy points out, one can be motivated by the recognition that the action X is the most effective means to pursue one’s aims, and one can ignore this fact in a way that one’s irrationality becomes obvious. So instrumental rationality is violated if one does not choose the appropriate means; hence, the principle of instrumental rationality does have normative force independently from the status of the ends. See Dancy, 2000, pp. 46, 47. 9. Compare the following passage (see the quotation above): ‘I may desire any fruit as of an excellent relish; but whenever you convince me of my mistake, my longing ceases. I may will the performance of certain actions as means of obtaining any desir’d good; but as my willing of these actions is only secondary, and founded on the supposition, that they are causes of the propos’d effect; as soon as I discover the falsehood of that supposition, they must become indifferent to me’, Hume, Treatise, p. 417. 10. Broome equally insists on the separation of requirements and reasons. See Broome, 2002, p. 93.

Herlinde Pauer-Studer

205

11. Korsgaard suggests that Kant’s reading of the instrumental principle amounts to the following syllogism: Whoever wills the end, wills the means. I will the end. Therefore: I will the means. That formulation is, as Korsgaard concedes, problematic since someone can will the end, yet fail to will the means. Therefore Korsgaard adds the condition ‘insofar as one is rational’. So the correct syllogism, she argues, is: Whoever wills the end, wills the means insofar as one is rational. I will the end. Therefore I will the means insofar as I am rational, which is equivalent to: Whoever wills the end, ought to will the means. (Korsgaard, 1997a, pp. 238, 239)

12.

13.

14. 15.

This leaves us with a narrow scope ‘ought’. One can make the detachment: you ought to take the means, that is, you have a reason to take the means. That is exactly the point Broome criticises. For Broome’s criticism of Korsgaard’s argument see Broome, 2000, pp. 98–9. Broome’s most recent formulation of the instrumental requirement is: ‘Rationality requires of N that, if N intends that e, and if N believes that e will be so only if m is so, and if N believes that m will be so only if she herself intends that m, then N intends that m.’ See Broome, 2009, forthcoming, p. 2. Broome does not reduce all normative relations to wide-scope normative requirements. His claim is, merely, that the normativity going with instrumental reasoning is best captured by a ‘wide-scope ought’. Therefore his position is not undermined by admitting that there are also requirements expressing a ‘narrowscope ought’, whose normativity is due to the fact that there is strong justificatory evidence for it. In Korsgaard’s case: a volition, a belief, and a derived volition. Korsgaard follows Kant’s functional argument in the Groundwork: the only principle which fulfils the condition of being a law is the categorical imperative. Korsgaard’s argument that brings us to the Kantian laws has two steps: the first part of the argument establishes why we need a law, the second part why this law is the categorical imperative. More specifically: in order to be persons who are able to reflect on what they are doing, we cannot simply follow our impulses, otherwise we would be wantons, not persons. But we can only give reasons for our actions and justify them if there are laws guiding our reflective evaluations. We have to ask whether a certain incentive for action can qualify as a law for us. We have to be a free will, which means to be guided by a self-given law. The principle of a free will is a law, and this condition, to be just a law, is exactly fulfilled by the categorical imperative in the universal law formulation, which brings in the moral law in the form of a law. This completes the first part. The second part of the argument tries to establish the categorical imperative as the only solution to the given problem. Korsgaard brings in a further argument to arrive at a substantive conception of morality: to be a person we need a normative structure. Normative structures are not only supplied by morality but also by our practical identities, that is, those normative codes that result from our social roles and social contexts – whether one is a father, mother, mafioso, or a philosopher. But

206 Humean Sources of Normativity we cannot develop practical identities if we do not attribute value to ourselves, that is, unless we value our humanity. And to value our humanity we have to equally value the humanity of others. That brings us to the second formulation of the categorical imperative, the Formula of Humanity. See Korsgaard, 1996b, pp. 98, 104–5. 16. It is interesting that Korsgaard criticises the ‘self-interest or economic theory of rationality’ for failing to give a justification why it ‘is rational for each person to pursue his overall good’ (Korsgaard, 1997a, p. 230). She argues that the ‘instrumental principle says nothing about our ends, so it is completely unequipped to say either that we ought to desire our overall good or that we ought to prefer it to more immediate or local satisfactions. The self-interest theory of rationality, because it is committed to the principle of prudence, has to go beyond the instrumental theory’ (Korsgaard, 1997a, p. 231). Korsgaard’s criticism of the selfinterest theory is plausible: the apparent validity of the principle of instrumental reason does not provide a justification for assuming the maximisation of one’s advantage to be the goal of a rational agent. The self-evidence of the means-end principle does not support the specific interpretation of instrumental rationality which is tacitly presupposed in the homo economicus paradigm, namely that being rational means to maximise one’s own advantage. There have to be given further arguments why maximising one’s own advantage should be the appropriate end of rational behaviour as such. However, to raise these objections presupposes to distinguish clearly between the normativity of rationality (which ensures merely the logical validity of means-end reasoning) and the normativity of reasons which depends on the assessment of the ends; in this particular case whether we do have good reasons to consider a rational agent by definition to be an agent focusing exclusively on her or his own advantage. 17. Some authors have denied that Hume develops anything like a conception of practical reasoning leading to practical conclusions. The justification for that position is the thesis that Hume’s moral psychology excludes the mental contents which could be the contents of the logical relations at stake in means-end reasoning. Hume, so the criticism goes, assumes that passions are not representational, they are not representations of objects; so passions (which in Hume’s conception would be the mental contents of means-end reasoning) do not have content. Therefore passions cannot be the subject of practical reasoning. Therefore, since the contents are missing, practical reasoning is not possible, given Hume’s assumptions. Such a position is defended by Elijah Millgram (see Millgram, 1995). I consider it to be wrong. Passions, Hume holds, are not copies of objects; in that sense – being copies of objects – they are not representational. However, as mental experiences they do have content (they are directed at ends or disapprove of ends). Moreover, as Hume’s examples show, we can meaningfully talk about the relations between our passions and their ends and the actions we are required to do or not do by our reasoning, given these relations. 18. Korsgaard presents a normative reading of Hume’s appeal to the general point of view. According to the normativist interpretation, we judge from the perspective of the general point of view our sentiments not only in regard to formal aspects like consistency and coherence but also in regard to appropriateness and moral validity. A normativist reading is also developed by Annette C. Baier and Barry Stroud (see Baier, 1991, Chapter 12; Stroud, 1977, Chapter VIII). The normative interpretation is not generally shared by Hume scholars. For example, Rachel Cohon argues that Hume introduces the idea of the general point of view merely

Herlinde Pauer-Studer

207

to make his descriptive moral psychology consistent. Hume explains the origin of moral sentiments with the help of the mechanism of sympathy. Yet sympathy varies with closeness and remoteness. Therefore, Cohon argues, Hume introduces the general point of view as a means to correct these differences and to secure consistency and coherence of our judgements and evaluations. Cohon reduces the general point of view to a psychological mechanism (deflationary account). See Cohon, 1997c. 19. This objection has been raised by Bernard Williams. In his defence of rule utilitarianism John Harsanyi argues that the relevant rules can be formulated in a way that exceptions to the rule are covered by the utilitarian standard: ‘Yet, even though there may often be practical difficulties in deciding what specific list of permissible exceptions to a given moral rule would in fact maximize social utility, I think it is very important from a philosophical standpoint that rule utilitarianism does provide at least a conceptually clear theoretical criterion for the set of morally permissible exceptions’ (Harsanyi, 1985, pp. 129–30).

10 Two Kinds of Normativity: Korsgaard v. Hume Luke Russell

In many disputes about moral theory appeals are made to the normativity of morality. Sometimes theories are dismissed on the grounds that they fail to account for the normativity of morality. What is meant by such claims? What is normativity, and should it be attributed to morality? Through an examination of Christine Korsgaard’s criticism of Humean ethics in The Sources of Normativity, I will argue that there are two distinct notions of normativity in play in contemporary disputes, and hence two different properties which morality may or may not be supposed to possess. The first notion of normativity is connected to Motivation. Morality possesses motivational normativity insofar as those who make moral judgements are moved to act in accordance with them. The second notion of normativity relates to justification and correctness. There is some dispute over what would count as evidence that morality possesses justificatory normativity, but I set out a commonly held view that I shall call the strong requirement for justification. The strong requirement is that, in order to show that morality possesses justificatory normativity, it is not enough to show that morality is practically necessary, nor that moral judgements in fact move us to act. Instead, we must provide positive justifying reasons for all of our moral beliefs and practices. Korsgaard is ambivalent about the relationship between our ability to meet the strong requirement and our capacity for moral deliberation and action. Nonetheless she clearly endorses the strong requirement. Korsgaard rejects Humean ethics precisely because she believes that Humean reflective endorsement fails to provide positive justifying reasons for morality. However, Korsgaard’s own neo-Kantian account of morality is no more able to meet the strong requirement than is the account offered by her Humean opponents. In defending her view Korsgaard implies that the alleged practical necessity of Kantian rational agenthood, plus the fact that we are moved by moral judgements, shows that morality is justified. Yet it is not clear that we must act as Kantian rational agents, and even if we were under such a practical constraint, that would not show that we have 208

Luke Russell

209

a positive reason to be moral. Nor does the admittedly good evidence for the motivational normativity of morality amount to evidence for its justificatory normativity. Despite the fact that both Humean and Kantian ethics fail to meet the strong requirement, a Humean approach is preferable for other reasons.

10.1 Motivational normativity One aim of moral theory is to explain how and why people deliberate, decide and act on the basis of moral considerations. As Christine Korsgaard notes in The Sources of Normativity, this requires an account of the nature of moral concepts, of the things to which those concepts apply and an explanation of how we come to possess those concepts (Korsgaard, 1996b, pp. 10–11). Regardless of whether we favour motivational internalism or externalism about moral judgements, it is evident that people generally act in accordance with their moral judgements (Smith, 1994, p. 7; Brink, 1989, p. 49). Perhaps, along with internalists, we believe that the moral judgement itself necessarily carries some motivational weight. Perhaps, along with externalists, we believe that people usually act in accordance with their moral judgements because people usually desire to do what they believe is right. The demand that we account for the motivational power of moral judgements might be described as the demand that we account for the normativity of morality.1 In this sense, we might explain the fact that S acted in accordance with moral reason R by saying that S was aware of R and that R is normative for S. As we shall see, there is another sense in which moral claims are called normative, so let us call this first notion motivational normativity. How we would respond to a theory that did not attribute motivational normativity to our judgements about what it would be morally right to do? If moral deliberation and judgement lacked motivational normativity, we would expect people to begin deliberating but then never act, or at least not act as a result of and in accordance with the conclusions of their deliberations. Thus the fact that we engage in moral deliberation and then usually act accordingly is very good evidence for the view that morality possesses motivational normativity. Korsgaard agrees that theories that fail to explain how moral considerations move us thereby lack ‘explanatory adequacy’ (1996b, p. 13).

10.2 Justificatory normativity We have seen that one aim of moral theories is to explain how and why almost all people are moved to act in accordance with their moral judgements; that is, why morality is normative for almost all people. There is a quite different notion of normativity which is connected to a second aim of moral theory. That aim is to show that moral behaviour is justified.

210 Two Kinds of Normativity: Korsgaard v. Hume

It is one thing to show that people think they have moral obligations and are steered by moral considerations, but quite another to show that people in fact have moral obligations. Korsgaard notes that this need for what she calls ‘normative or justificatory adequacy’ (1996b, p. 81) is separate from the demand for explanatory adequacy. Even if a theory does explain why people who wield moral concepts and make moral judgements are moved to act in certain ways, it might not justify the making of such judgements or the use of such concepts. As we have seen, internalists claim to account for the motivational normativity of morality by ascribing necessary motivational weight to the moral judgement itself, but Korsgaard maintains that this is only part of the overall project: [I]nternalism in this general sense is correct, [but] internalism in this general sense does not exhaust the question of normativity. Once we see what the motives morality provides us with are – how moral ideas are able to motivate us in a given case – we can still ask whether to endorse those motives. (1996b, p. 81) We are not necessarily stuck with the concepts we inherit, nor with our standard practices of judgement. When it comes to moral and practical deliberation we might do otherwise. Even if we agree that we are moved by moral considerations, we still face what Korsgaard calls the normative question, or what it might be better to call the question of justificatory normativity: Should we be moved by these considerations? Supposing that we possess a motive M, should we be moved by M? There are three importantly different kinds of answer that could be offered to this question. Firstly, we might claim that we have a positive reason to act on M. In such cases we think M is appropriate, and we can say why M is preferable to alternatives. Thus, M appears to us not to be a motive we just happen to have, but one which we ought to have. Secondly, we might claim that we have a positive reason not to act on M; that is, that there is a negative reason that counts against our acting on M. In such cases we think M is not appropriate, and we can say why other alternatives are preferable. For instance, we might believe that we ought not be moved by M because M clashes with our other motives, or because M is immoral. Thirdly, we might claim that we have no positive reason to act on M, but also have no negative reason that counts against our acting on M. In these cases it strikes us that M is a motive that we just happen to have, but since M does not clash with our other motives and is not objectionable for any other reason, there is no reason for us to reject M. In everyday disputes it is common to offer positive reasons and negative reasons for or against our actions and practices, but deeper and more systematic attempts at justification often bring forth the slightly unsettling

Luke Russell

211

claim that at least we have no reason not to endorse our existing motives. Wittgenstein notes: If I have exhausted the justifications I have reached bedrock, and my spade is turned. Then I am inclined to say: ‘This is simply what I do.’ (Wittgenstein, 1953, I. 217, p. 85) In many cases in which we say, ‘This is simply what I do’, we accept that we could do otherwise. What I do is what I just happen to do, and what I could choose not to do. It is, by my own admission, no better than the available alternatives and in that sense arbitrary, so it seems misleading to say that it is what I ought to do. Thus the fact that there is no negative reason counting against M does not show that M is justified. Sometimes, though, the claim that ‘This is simply what I do’ is declared with a kind of firmness, marking a necessary end to practical deliberation over what I ought to choose: This is simply what I do, and I literally could not do otherwise!2 If we do strike a bedrock of mechanical or practical necessity where there are no alternative choices available, have we answered the justificatory normative question satisfactorily? Again, it seems not. Locating a point of practical necessity would nullify the ordinary first-personal benefit of asking for further justification, since in such cases we cannot revise our own practices. Yet recognising that our practices are inevitable is very different to showing that they are right – that it would be a bad thing if they were different, that people whose practices differ are behaving incorrectly, and so on. Practical necessity is no substitute for justification. To suppose that it were such a substitute would be, in another of Wittgenstein’s memorable phrases, to ‘confuse the hardness of a rule with the hardness of a material’ (Wittgenstein, 1956, II. 87, p. 110). What would it take to show that morality possesses justificatory normativity? I have outlined an answer to that question which I shall call the strong requirement for justification. The strong requirement for justification is partly a stipulation of what is not sufficient for justification. Showing that there is no negative reason to endorse morality as a whole is not sufficient to demonstrate that morality is justified, and showing that we are incapable of abandoning morality is not sufficient to demonstrate that morality is justified. People who endorse the strong requirement demand positive justifying reasons for all of morality. Only if we can provide such reasons will we have shown that morality possesses justificatory normativity.

10.3 Korsgaard and the strong requirement for justification What exactly is our predicament if we cannot meet the strong requirement for justification? More specifically, what effect would such a failure have on moral deliberation and action? At various points in The Sources of

212 Two Kinds of Normativity: Korsgaard v. Hume

Normativity, Korsgaard hints at two different answers to that question, one of which is much more plausible than the other. Firstly, let us consider the less plausible answer. This is the bold claim that we must be able to answer the justificatory normative question with a positive reason or we will not act; or, at least, we will not act as a result of and in accordance with deliberation. At one point Korsgaard gives the impression that the upshot of accepting an evolutionary explanation of moral judgements and behaviour would be a moral scepticism that would lead to the rejection of morality and the cessation of moral action: The moral sceptic is someone who thinks that the explanation of moral concepts [and their practical and psychological effects] will be one that does not support the claims that morality makes on us. He thinks that once we see what is really behind morality, we won’t care about it anymore. (1996b, pp. 13–14) Later, Korsgaard echoes this sentiment in the context of the more general claim about the normativity of practical reason: The reflective mind cannot settle for perception and desire, not just as such. It needs a reason. Otherwise, as long as it reflects, it cannot commit itself or go forward. (1996b, p. 93) On this account, if we were unable to answer the justificatory normative question with a positive reason, we would not be able to go forward and act, and hence could not be counted on to perform the actions which we had thought to be morally right. In contrast, if we do go forward we must have provided a satisfactory answer to the justificatory normative question. According to this view, the fact that we reflect and act accordingly indicates that we have convinced ourselves that all of our motives are the right motives, not just the ones we happen to have. This first bold view of the predicament of people who cannot meet the strong requirement for justification depends on what is, at best, a highly controversial model of Motivation. It is obviously true that sometimes we refuse to act unless we have secured a justificatory reason for acting, but is it true that in general we will be moved to act only when there is nothing for which we lack a justificatory reason? If this were the nature of reflective agency, then justificatory normativity would be a precondition for motivational normativity – or, at least, our judgement that our motives are justified would be a precondition for their moving us to act. But why should we believe that this is the case? Surely it is possible that we reflect and act accordingly without having secured justification for all of our motives, and that we act while judging that we have not secured such justification.

Luke Russell

213

When we begin the study of philosophy and are introduced to the regress of justification we realise that, up until that point, we had been acting despite failing to possess positive reasons which would justify many of our beliefs and desires. This newly acquired knowledge may well be disturbing, but it does not prevent us from going about our everyday lives. Thus the mere fact that we reflect and then act accordingly does not count as evidence for thinking that we have answered the justificatory normative question with a positive reason. The right kind of evidence for this achievement would be not merely the fact that we deliberate and act, but our ability to point to the reasons themselves. If we can offer justification for all of our moral actions, sets of motives and standards of assessment, then we have met the strong requirement for justification. If we cannot, we have not. The more plausible view is that people who cannot meet the strong requirement for justification are able to go forward and act morally without having provided justification for the basic moral standards against which they measure actions. Some such people might accept the propriety of the strong requirement and despair at our inability to meet it, but claim to be trapped within an inevitable pattern of moral thought and behaviour. Others might accept that we cannot answer all justificatory normative questions with a positive reason, yet argue that this is compatible with both our going forward and acting and with our being justified. In other words, they might reject the strong requirement for justification as too strong, arguing that there is a weaker, perhaps coherentist, sense in which morality is justified. We have seen that on some occasions Korsgaard suggests that we would be capable of reflecting and acting accordingly only if we had answered all normative questions with positive reasons. Some of Korsgaard’s other comments suggest that she allows for the possibility of action without the agent first having answered all calls for justification. For instance, she maintains it is possible to be compelled to act by naturally selected instincts which we fail rationally to endorse (1996b, p. 15). Thus there is an ambivalence in Korsgaard’s attitude to the relationship between the strong requirement for justification and the possibility of moral deliberation and action. This ambivalence will play a significant role in her assessment of Humean and Kantian ethical theories. Despite this ambivalence about the relationship between justification and action, Korsgaard clearly endorses the strong requirement for justification on several occasions. For instance, Korsgaard rejects the so-called realism of Samuel Clark, Prichard, Ross and Moore because each of them ends the normative regress without offering a justificatory reason to support the end point; that is, they rest without showing why they are resting in the right place. Korsgaard notes: We can keep asking why: ‘Why must I do what is right?’ – ‘Because it is commanded by God’ – ‘But why must I do what is commanded by

214 Two Kinds of Normativity: Korsgaard v. Hume

God?’ – and so on, in a way that apparently can go on forever. This is what Kant called a search for the unconditioned – in this case, for something which will bring the reiteration of ‘but why must I do that?’ to an end. The unconditional answer must be one that makes it impossible, unnecessary, or incoherent to ask why again. The realist move is to bring this regress to an end by fiat: he declares that some things are intrinsically normative. … Having discovered he needs an unconditional answer, the realist straightaway concludes that he has found one. (1996b, p. 33) Nagel also comes under Korsgaard’s fire for supposing that, if we have no reason to think that basic moral standards and values are not reasons, then we should conclude that they are reasons. Nagel’s assertion that it is selfevident that some things are moral reasons is dismissed by Korsgaard as ‘an expression of confidence and nothing more’ (1996b, p. 41). What we need, Korsgaard implies, is something more. We need a positive reason for thinking that our basic moral standards are not merely standards that we happen to find compelling, but are the standards we ought to have. In the remainder of this essay I will argue that Korsgaard dismisses Humean moral theory on the grounds that it does not offer a positive reason in answer to the justificatory normative question, yet that Korsgaard herself fails to offer positive reasons in support of many aspects of her own Kantian view. Instead, Korsgaard attempts to substitute practical necessity for justification, and ultimately treats the mere fact that we reflect and act as evidence of our having answered the justificatory normative question.

10.4 The supposed weakness of Humean ethics To begin, let us consider Korsgaard’s view of Humean ethics. Korsgaard notes that Hume’s own professed intention is to offer an explanatory anatomy of morality rather than an encouragement to be moral (1996b, p. 52). However, she points out that Hume seems concerned with the question of justificatory normativity as well. Firstly, Hume attempts to bring a kind of objectivity to our moral judgements via reflection, claiming that the sentiments, such as benevolence and approbation, that arise in response to specific situations ought not be accepted without first having passed a reflective test. For instance, Hume claims that sentimental moral assessment of actions must take place from a general point of view, so as to guard against our responses favouring our own interests (T, 3.3.1.15/581–2). Moreover, sentimental assessment must take into account the usual consequences of actions, not the consequences they happen to have in any one particular case (T, 3.3.1.21/585). Hume maintains that these mechanisms work to bring our moral judgements into closer alignment with those of our fellows.

Luke Russell

215

Yet, as Korsgaard claims, Hume wants to show not only that we can work towards agreement in our sentimental assessments of particular cases but also that we can endorse our moral nature as a whole. Korsgaard notes: Obviously if we are going to raise the question of whether we can endorse our moral nature, we must appeal to some standard, in terms of which we may judge morality to be good or bad. Morality must be endorsed or rejected from a point of view which itself makes claims on us and so which is itself at least potentially normative. (Korsgaard 1996b, p. 54) Against which standards does Hume measure the moral sentiments? Firstly, against the standard of morality itself. Korsgaard points out that at the end of Book I of the Treatise Hume concludes with despair that the understanding ‘entirely subverts itself’ (T, 1.4.7.7/267), but at the end of Book III claims that the moral sense must certainly acquire new force, when reflecting on itself, it approves of those principles, from whence it is deriv’d, and finds nothing but what is great and good in its rise and origin. (Korsgaard 1996b, 3.3.6.3/619) Thus, Hume thinks the moral sense can pass what Korsgaard calls a test of reflexivity. This test is question-begging, in some sense, but Hume thinks that passing it is no trivial matter, as is demonstrated by his belief that the understanding fails to pass the analogous test. This is not the only standard against which Hume measures the moral sentiments. Korsgaard points out that, especially in the Enquiry, Hume looks for reasons to endorse morality from the point of view of selfinterest (EPM, 9.14/278). Hume thinks that we have positive, self-interested reasons to be moral because we naturally desire to be admired, and virtue is a route to being admired (EPM, 9.19/280). Moreover, Hume thinks sentiments are contagious, that immoral people who initially do not think badly of themselves will ‘catch’ the disapproval of others, and hence will disapprove of themselves in the end (EPM, 9.10/276; Korsgaard, 1996b, pp. 56–60). Measuring morality against self-interest is notoriously controversial. Attempts to show that individual agents always promote their own interests by doing what is morally best arguably turn out to be circular or simply implausible. Even if such attempts were successful it could be argued that we ought to do what is right because it is right, or, at least, not because it secures the outcomes that are prudentially best for ourselves. However, Korsgaard admits that something can be achieved by the method of reflective endorsement employed by Hume. By arguing that morality passes the

216 Two Kinds of Normativity: Korsgaard v. Hume

tests of reflexivity and self-interest Hume aims to settle the nerves of those among us who are worried that the demands made upon us by our various points of view will tear us apart. He does not try to show that our basic standards are themselves the correct standards, but that there is no negative reason which counts against morality. In Korsgaard’s words, the method of reflective endorsement shows that human nature, moral government included, is intrinsically normative, in a negative version of the sense required by the realist Argument: there is no intelligible challenge that can be made to its claims. … It is human nature to be governed by morality, and from every point of view, including its own, morality earns its right to govern us. We therefore have no reason to reject our nature, and can allow it to be a law to us. (1996b, p. 66) Korsgaard notes that this modest programme of reflective endorsement is continued in contemporary ethical theory by Bernard Williams, who thinks the best we can offer as a justification of an ethical system is that it is internally coherent and that it encourages human flourishing (Williams, 1985, p. 154; Korsgaard, 1996b, pp. 67–78). This test is not sufficiently stringent for Korsgaard. After all, the fact that our moral judgements and practices cohere with a bunch of interests and drives with which we happen to be saddled does not show that they are obligatory, nor correct. In accordance with the strong requirement, Korsgaard maintains that it is not enough to have no reason to reject our nature. Instead we must have a positive reason to accept it. She notes: Hume … reverses the realist ordering of things, and argues that vice is bad because we disapprove of it. In a similar way, Williams thinks that ethical value is projected onto the world by our ethical beliefs. Both would deny that it is coherent to ask whether our values are true independently of our own moral or ethical sentiments. The only question left to ask is whether it is good for us to have those sentiments, and that question must be answered from the perspective of the other practical claims that our nature makes on us. (1996b, pp. 75–6) Korsgaard thinks that we need what Hume and Williams fail to provide; namely, a positive reason in answer to the justificatory normative question, an answer that shows us why morality as a whole is right, an answer that does not require us to take anything for granted. Moreover, Korsgaard thinks her own version of neo-Kantianism offers such an answer.

Luke Russell

217

10.5 Korsgaard’s proposed solution Korsgaard’s neo-Kantian solution to the problem of normativity comes in three parts. Firstly, she argues that the categorical imperative provides reasons that are normative. Secondly, Korsgaard claims that these reasons reflect our practical identities; that is, the descriptions of ourselves under which we value ourselves. The realisation that an action passes the test of the categorical imperative, and hence that it accords with one’s practical identity, is supposedly the point at which reflection is brought to a satisfactory end. Korsgaard thinks that this alone is not sufficient to establish that we have reasons to be moral, so she offers the third component of her solution, arguing that each of us necessarily shares a common practical identity – that of a reflective human being – and hence that we have reasons to treat each other with the dignity that is due to rational agents. At each stage of this process Korsgaard fails to give the kind of positive reason that she earlier demanded from her opponents. Indeed, she points to various kinds of practical necessity in order to end the justificatory regress, despite her earlier suggestions that practical necessity is no substitute for positive reasons. 10.5.1

The categorical imperative

Consider the first step of her solution: the appeal to the categorical imperative as the source of reasons which are justified and which require no further justification. As reflective creatures we will ask why we should employ the categorical imperative in the first place. Prima facie, it seems possible that we could choose simply to be moved by desires we happen to have without considering whether our maxims could be willed to be universal laws. If this is possible, why use the categorical imperative to identify our reasons for action? Korsgaard’s answer seems to be that, so long as one is a free will, one has no choice but to do so: [T]he categorical imperative is the law of a free will. It does not impose any external constraint on the free will’s activities, but simply arises from the nature of the will. It describes what a free will must do to be what it is. It must choose a maxim it can regard as a law. (1996b, p. 98) But then, why be a free will? Arguably, in cases such as the toxin puzzle, we might decide that it is worse to be a free will, legislating universally and acting accordingly, and better to allow ourselves to act unreflectively on desires, or, at least, to avoid universal legislation (Kavka, 1983, pp. 33–6). Kantian practical rationality might undermine itself, and hence fail the test of reflexivity. As Peter Railton has argued, there seems to be a coherent, practically significant question regarding whether an agent should continue

218 Two Kinds of Normativity: Korsgaard v. Hume

to be an agent; that is, the kind of being that employs practical reasoning in general, and the categorical imperative in particular (Railton, 1997, pp. 53–79). It is not clear that Korsgaard can provide positive justification for why someone should be a Kantian free will. Instead she denies that we can act on desires rather than reasons: An animal, whose desire is its will, is a wanton. I am arguing here that a person cannot be like that, because of the reflective structure of human consciousness. A person must act on a reason, and so the person who acts like a wanton must be treating the desire of the moment as a reason. (1996b, p. 99 n. 8) Korsgaard’s claim that practical reasoning is inescapable is, perhaps, defensible.3 Her slide from this claim to the conclusion that self-legislation via the categorical imperative is inescapable is certainly more controversial. Even if both of these claims are true, they do not address the question of justificatory normativity. As Korsgaard herself has argued previously, the fact that a person cannot help but think and act in a certain way does not show that a person should think and act in that way. Nonetheless Korsgaard maintains that what passes the purportedly inescapable test of the categorical imperative counts as an unconditioned reason that one is obliged to obey. In contrast to the laws handed down by other sources, she suggests, the laws handed down by oneself need no further justification. Korsgaard claims: The realist objection – that we need to explain why we must obey that [external] legislator – has been answered, for this is a legislator whose authority is beyond question and does not need to be established. It is the authority of your own mind and will. (1996b, p. 104) This claim is puzzling. Read in one way, it seems simply to deny the question that generates the normative regress in the first place: I think I should ø, but is this really what I should do? Yet this is a legitimate question to which Korsgaard thinks we need an answer. So what could Korsgaard mean when she claims that one’s own mind and will has an authority that does not need to be established? Perhaps it is that when one deliberates and finally acts, the best one can do is act on the basis of one’s own mind and will. After all, one cannot be guided by considerations that one has not taken on board, nor moved by reasons of which one is not aware. Hence, there is a kind of necessary limit built into us as reflective agents. We cannot strive to do more than act in accordance with our well-informed and well-considered reflective conclusions. But this practical limit on the capacity of reflective

Luke Russell

219

beings gives us no reason to think that the end product of reflection will be something we really have reason to do, independent of the desires, sentiments and beliefs that we just happen to have. In other words, this practical limit is no substitute for an unconditioned positive answer to the normative question. Korsgaard has not shown that we have reason to reason, nor that we have reason to employ the categorical imperative in our reasoning. 10.5.2

Practical identity

The second step of Korsgaard’s solution is her appeal to practical identity – the description of oneself under which one values oneself – as the legitimate terminus of reflection. In her view, ‘[y]our reasons express your identity, your nature; your obligations spring from what that identity forbids’ (1996b, p. 101). As reflective creatures we will be tempted to ask why we ought to be true to our respective practical identities. Why should we think that our practical identities delineate our permissible or obligatory courses of action? Since Korsgaard thinks it is excessively arbitrary to halt the reflective process, along with Hume, when we reach our reflectively endorsed sets of desires, she must explain why is it not equally arbitrary to halt when we reach the claims of our practical identities. Korsgaard’s response to this demand could be interpreted in one of two ways, neither of which are satisfactory. According to the first interpretation of Korsgaard’s response, she attempts to justify resting on your practical identity by appealing to an instinct of self-preservation. To violate your selfconception, Korsgaard claims, is to ‘lose your integrity and so your identity, and to no longer be who you are’. She continues: That is, it is to no longer be able to think of yourself under the description under which you value yourself and find your life to be worth living and your actions to be worth undertaking. It is to be for all practical purposes dead or worse than dead. When an action cannot be performed without loss of some fundamental part of one’s identity, and an agent could just as well be dead, then the obligation not to do it is unconditional and complete. (1996b, p. 102) In effect, Korsgaard claims that to fail knowingly to be true to oneself is to commit moral suicide. She paints one’s practical identity as a natural and necessary endpoint to the questioning of moral reasons, a point at which one no longer needs to reflect, because the choice is between being true to one’s practical identity and violating it. Such a violation involves a catastrophic loss, as bad or worse than the loss of one’s life. Hence, for Korsgaard, we must act in accordance with our practical identities or perish. It is not at all clear what these claims about self-preservation contribute to the task of answering the justificatory normative question. Perhaps it is

220 Two Kinds of Normativity: Korsgaard v. Hume

true that we flinch at the idea of self-destruction, but that does not mean that we need not ask why we ought not destroy ourselves. Even if it were true that we were practically incapable of deciding to destroy ourselves, that fact would not show that self-destruction is wrong. Yet Korsgaard herself admits that some acts are so horribly immoral that it would be better to die than perform them, and furthermore admits that many people throughout history have willing destroyed themselves rather than perform terrible wrongs (1996b, pp. 161–2). We can destroy ourselves, and Korsgaard thinks that sometimes we should. Thus appealing to an instinctive drive for selfpreservation does not halt the normative regress. Yet, Korsgaard would point out, preserving one’s practical identity is different to mere self-preservation. One’s practical identity is the description under which one values oneself. Hence, it seems, we ought to will our own destruction if doing so is the only way to preserve our values. But if this is the case, it is not clear why Korsgaard has appealed to the notion of selfpreservation in the first place. It is our values, not some allegedly compelling desire for self-preservation, that supposedly provide the ultimate ground in reflection. If it is merely our most deeply held values that are the ultimate ground in reflection, then Korsgaard has offered us no better an answer to the call for justificatory normativity than Hume and contemporary Humeans, who suggest that we can measure our moral practices against our deeply held moral and non-moral values, but cannot show that our most deeply held values are correct. Perhaps, though, Korsgaard does not intend her claims about loss of self to provide justification nor motivational leverage for acting in accordance with one’s practical identity. We could see these claims instead as merely setting out necessary conditions for ownership of action. This second interpretation of Korsgaard’s claims about practical identity suggests that any instances of our behaviour that do not accord with our practical identities cannot count as our actions, since they do not flow from ourselves. On such a reading it is impossible by definition for a person to act in a way that violates her practical identity. Hence, contrary to the first reading of Korsgaard’s claims, loss of self cannot function as a disincentive to violation of practical identity, since we could not violate our practical identities no matter how hard we tried. If this is what Korsgaard means by her claim that practical identity functions as an endpoint to reflection, she has not offered a positive answer to the justificatory normative question. Once again, the fact that it is impossible to act in a way that violates our practical identities does not imply that we ought not violate those identities. Practical necessity is no substitute for justification. In any case, there are additional reasons to reject the claim that behaviour that fails to accord with our practical identities cannot count as our own action. Imagine that a man shoplifts some expensive jewellery even though such behaviour conflicts with his own values. Perhaps he succumbs, in this one case, to overwhelming temptation, and then castigates himself

Luke Russell

221

for having done so. The fact that the theft did not fit with his own values might lead us to soften our reproach or call for a reduced punishment, but it does not lead us to say that the theft was merely behaviour rather than action. His action was out of character, but it was he who acted. Yet the second reading of Korsgaard’s claim denies us this judgement. Either the theft was not his action, since it was not an action at all, or it was not his action, since his practical identity forbids theft. Both of these descriptions are deeply unnatural, and both suggest the counterintuitive conclusion that we should not hold this man responsible for the theft. On neither reading are Korsgaard’s claims about the role of practical identity in grounding normative reflection convincing. On the one hand, an instinctive desire for self-preservation does not end the regress. On the other hand, the claim that it is impossible to act in violation of our practical identities does not address the justificatory normative question, and has counterintuitive consequences regarding responsibility. 10.5.3

The right identity

The final stage of Korsgaard’s neo-Kantian solution to the problem of normativity is her attempt to unify our reasons for action and ensure that they are moral reasons. As Cohen notes, some practical identities – descriptions under which people value themselves – are not appropriate sources of moral behaviour (Cohen, 1996, pp. 183–4). For instance, an earnest Mafioso who values himself as a hit man has a practical identity that, in his eyes, requires him to carry out his purported duties as a hit man. But none of us want to claim that he is morally obliged to be true to that practical identity. How can Korsgaard show that one particular practical identity is the right practical identity; that is, the identity that marks out genuine moral obligations? Once again, instead of offering a positive reason in answer to this question, Korsgaard points to an alleged practical necessity. In this case it is the allegedly practically necessary conception of ourselves as human beings: [A] human being is an animal who needs a practical conception of her own identity, a conception of who she is which is normative for her. But you are a human being and so if you believe my Argument you can now see that that is your identity. You are an animal of the sort that I have just described. And that is not merely a contingent conception of your identity, which you have constructed or chosen for yourself, or could conceivably reject. It is simply the truth. (1996b, p. 123) Yet it is clear that many people do conceive of themselves in ways that, in their own eyes, licence what we judge to be immoral actions. Therefore, if we grant that we must conceive of ourselves as human beings, either such a conception is too thin or too vague to require moral behaviour, or

222 Two Kinds of Normativity: Korsgaard v. Hume

it is a self-conception that might not be held strongly enough to override competing, immoral practical identities. However, Korsgaard believes that our self-conception as reflective human beings is the right practical identity, the identity that always ought to be fully expressed in our behaviour, and she also believes that it is a sufficiently thick and detailed self-conception to produce morally right action. As one’s practical identity becomes more thick and detailed, though, it also becomes more contingent and seems more in need of justification.4 In responding to her critics, Korsgaard admits that what we take to be cogent in moral reflection is deeply contingent, and hence that we cannot pretend to have latched onto independent, eternal moral truths. But in the same breath, Korsgaard maintains that her account is superior to the Humean view offered by Williams. We see the advantage, she writes, when we ‘distinguish our attitudes towards contingency from those towards passivity’. She continues: For contingency itself is something that can be actively embraced or passively endured, and this makes all the difference … Kant urges us to take things to be important because they are important to us. And this means that we must do so in full acceptance of the fact that what specifically is important to us is at bottom contingent and conditional, determined by biological, psychological, and historical conditions that themselves are neither justified nor unjustified, but simply there. In a deep way, all of our particular values are ones we just happen to hold. But the transition from contingency to necessity is sometimes our own work. (1996b, p. 242) At this point it seems Korsgaard has admitted that we cannot meet the strong requirement for justification. If that is the case, it is not clear what kind of advantage over the Humean account Korsgaard is offering. How can Korsgaard’s moral agent, who endorses her own contingent values on no grounds whatsoever, who declares that she simply finds them to be important, be on any firmer justificatory ground than a Humean agent who is moved by the moral inclinations which pass the test of reflective endorsement? The positive justifying reasons Korsgaard demanded earlier are nowhere to be seen. In lieu of those positive reasons, Korsgaard offers a transcendental Argument designed to show that we have constructed an adequate answer to the justificatory normative question: Human consciousness has a reflective structure that sets us normative problems. It is because of this that we require reasons for action, a conception of the right and the good. To act from such a conception is in turn to have a practical conception of your identity, a conception under which you value yourself and find your life to be worth living and your

Luke Russell

223

actions to be worth undertaking. That conception is normative for you and in certain cases it can obligate you, for if you do not allow yourself to be governed by any conception of your identity then you will have no reason to act and to live. … If you had no normative conception of your identity you could have no reasons for action, and because your consciousness is reflective, you could then not act at all. (1996b, pp. 122–3) But only if we assume what I have argued is an implausible view of rational agenthood would the fact that we reflect and then act accordingly count as evidence that we have secured the justificatory normativity of our reasons. At times, Korsgaard adopts this bold view, but elsewhere she allows for the possibility that we might act without having answered every possible call for justification. According to this more plausible alternative view, action in accordance with deliberation is evidence merely that the products of our reflection possess motivational normativity, not justificatory normativity. As Korsgaard herself has argued convincingly, the one is not sufficient for the other.

10.6 Is Korsgaard a Humean in disguise? Korsgaard unfairly criticises Humeans for failing to provide a kind of answer to the justificatory normative question that she herself is equally unable to deliver. Does this mean that Korsgaard is putting forward a Humean moral theory? Is she a Humean in disguise? On occasion in The Sources of Normativity she does express strong sympathy for Hume (1996b, pp. 88–9). Yet, as we have seen, she also makes plenty of claims that run counter to the thrust of the Humean project. There is a modesty about both the powers of reflection and the status of morality in Humean ethics that Korsgaard repeatedly offends. For instance, consider her following assertion: From a third-person point of view, outside of the deliberative standpoint, it may look as if what happens when someone makes a choice is that the strongest of his conflicting desires wins. But that isn’t the way it is for you when you deliberate. When you deliberate, it is as if there were something over and above all of your desires, something which is you, and which chooses which desire to act on. (1996b, p. 100) This is deeply contentious, and is a claim that Humeans should deny. How is the self that is separate from all desires supposed to assess those desires? Against which standard does the measurement take place? On which grounds are the choices made? We cannot escape this problem by appealing to standards of assessment that are not themselves desires – to reasons

224 Two Kinds of Normativity: Korsgaard v. Hume

or beliefs, for example – since the legitimacy of any contentful standard of assessment can be called into question. Could it be that, when you deliberate about which reasons ought to motivate you, there is something over and above all of your awareness of reasons, something which is you, and which chooses which reason to act on? The problem recurs: How is a self that is separate from all standards supposed to assess fundamental standards? While attempting to meet the strong requirement for justification, Korsgaard posits a self that is a contentless, extensionless point; what Charles Taylor has called the ‘punctual self’ (Taylor, 1989, p. 160). The punctual self consists of nothing more than the power to objectify; that is, to step back from beliefs and desires in order to assess them. Such a conception of the self suggests it is possible for reflective agents to take nothing for granted. No charge of dogmatism could ever stick to the punctual self. In some respects, this utter refusal to take things for granted is attractive, and it seems to be what pulls Korsgaard from her sympathy with Hume towards her final Kantian destination. Yet the punctual self seems to be incoherent. Having been stripped of content, it has no means left by which to evaluate its own desires. In contrast, Humeans admit that we must take aspects of ourselves for granted. Humeans think there is no justification for our desires and practices from outside of those desires and practices, and hence that we cannot suppose that people who have a different set of basic evaluative standards are mistaken. Such an admission is hard to square with widespread realist intuitions about moral practice and moral dispute. However, Humeans think that the alternative model of reflection offered by Korsgaard is simply incoherent. This incoherence shines through in Korsgaard’s defence of her neo-Kantian account of practical rationality, in which she demands positive reasons in answer to the justificatory normative question, but, when assessing her own account, is happy to rest on various points of supposed practical necessity. Humeans think that we are better off recognising our inability to meet the strong requirement for justification. Once we recognise that fact, the important task comes to be that of exploring notions of reflection, justification and dispute which can fit within our more limited framework (for example, Hubin, 2001, pp. 445–68).

Notes 1. Korsgaard herself does not explicitly label this notion ‘normativity’, although she does suggest that at least part of the project of explaining the normativity of morality is explaining the motivational power of moral considerations (1996b, p. 81.). Moreover, as I shall argue in Section 10.5, at some points Korsgaard takes evidence of the motivational power of morality to be evidence for its normativity. 2. For instance, in his discussion of the Darwinian origins of ethical sentiments Michael Ruse claims that we have no choice but to be moral (Ruse, 1986, p. 253.).

Luke Russell

225

3. Skidmore (2002) argues that agenthood is escapable through suicide but that the non-existent dead cannot fail to exercise practical reasoning, so there are no cases in which someone exists and has escaped practical reasoning. 4. Cf. Skidmore’s worries about the distinction between agency and ‘fully autonomous agency’ in this context (2002, pp. 138–9).

11 What Kind of Virtue Theorist is Hume? Christine Swanton*

11.1 The nature of virtue for Hume The aim of this essay is to elaborate and defend a claim that the status of a trait as a virtue for Hume is grounded in a plurality of features. Not only are virtues approved as such because of their utility but also because of their immediate impact on our sentiment or taste. This latter general ground of virtue status spawns an interesting plurality of features which make a trait a virtue. To take this latter ground of virtue status seriously, we need to show how traits having this effect can properly be seen as admirable or good. That is the main task of this essay. First a little background is needed. We need to explain in general terms what is a virtue for Hume. Virtues for Hume are stable and enduring qualities of mind which produce a certain kind of pleasure in certain kinds of circumstances. Specifically, a virtue is a quality of mind that has the power to produce love or pride: [T]hese two particulars are to be consider’d as equivalent, with regard to our mental qualities, virtue and the power to produce love or pride, vice and the power of producing humility or hatred. (T, 3.3.1.4/575) As a result of this power, a virtue ‘gives rise’ to sentiments of approbation ‘which [are] nothing but … fainter and more imperceptible love’ (T, 3.3.5.1/614). These sentiments constitute our sense of virtue, but virtues themselves are powers in objects: their ‘view’ causes certain kinds of pleasures or uneasiness (T, 3.1.2.4/471). Sometimes, in emphasising the fact that virtue and vice are akin to (for example) sounds and colours, Hume seems to claim that virtues are nothing but perceptions in the mind: Vice and virtue, therefore, may be compared to sounds, colours, heat and cold, which according to modern philosophy, are not qualities in objects, but perceptions in the mind. (T, 3.1.1.26/469) 226

Christine Swanton

227

But here Hume is claiming only that they are not qualities in objects of the kind discoverable by all rational beings independent of their affective make up. Not all senses of virtue are authoritative. The most obvious way in which our sense can fail to be authoritative is being mistaken about matters of fact, for example consequences of the usual manifestation of traits. Reason can discover truths about our virtues, insofar as such truths can consist in agreement to real existents and matters of fact (T, 3.1.1.9/458). Such truths when grasped, can influence, educate, and refine our moral sense (sense of virtue). We might say then that virtues are powers to produce love and pride in suitably informed, educated, and refined persons in whom distortions of ignorance, crudity, and partiality are absent. What reason cannot do is discover ‘eternal’, ‘immutable’, ‘natural’ fitnesses and unfitnesses of things which provide the foundations of morality (and self-interest) in every rational being regardless of the kinds of sentiments (if any) that form their constitution. For morality in humans cannot exist without an ‘original passion’ of benevolence, any more than prudence can exist without an ‘original passion’ of self-love. This is the point of Hume’s claims that ’[t]is not contrary to reason to prefer the destruction of the whole world to the scratching of my finger. (T, 2.3.3.6/416) and ’[t]is not contrary to reason for me to chuse my total ruin, to prevent the least uneasiness of an Indian or person totally unknown to me. (T, 2.3.3.6/416) The condition of the possibility of morality for Hume then is the existence of certain sentiments and emotional capacities which are part of our constitution as normal human beings – benevolence and sympathy. Likewise, the condition of the possibility of prudence is the existence of the passion of self-love. ‘Reason’, as understood here, is that faculty which is common to all rational beings (regardless of their affective constitutions): a faculty which cannot on its own discover facts about prudence, virtue, and beauty. Hume makes exactly the same point in regard to aesthetics. Just as vice is not ‘in the object’ as a property accessible to the understanding (understood as a rational faculty independent of emotional make up), or even to the sense of sight, so beauty is not a property of the circle accessible to the understandings of Euclid or the five senses: Euclid has fully explained all the qualities of the circle; but has not in any proposition said a word of its beauty … the beauty is not a quality

228 What Kind of Virtue Theorist is Hume?

of the circle. It lies not in any part of line whose parts are equally distant from a common centre … it is only the effect which that figure produces upon the mind whose peculiar fabric or structure renders it susceptible of such sentiments. (EPM, App. 1.14/291–2) In other words for beauty to be a property at all it needs to be accessible to a mind capable of appreciating a circle as beautiful. ‘Mathematical reasonings’ and the ‘senses’ are not sufficient for such appreciation; beauty is not a property of the circle as such. Thus Hume’s denial that there are ‘eternal’ and ‘immutable’ natural fitnesses proper to ethics, aesthetics, and prudence is perfectly consistent with his affirmation (in ‘The Standard of Taste’, for example) that properties may be ‘naturally fitted’ to produce approbation in suitable observers: A clear and distinct sentiment attends him through the whole survey of the objects; and he discerns that very degree and kind of approbation or displeasure, which each part is naturally fitted to produce. (Essays, 1.23/237) For the quoted claim is simply an example of the following thesis, which is compatible with Hume’s attack on the doctrine of ‘eternal fitnesses’, namely that such properties as virtues and beauty are qualities in objects which are powers naturally fitted to produce certain sentiments in creatures constituted as we are constituted. That is, these qualities only exist as (let us say) virtues relative to beings with constitutions like ours. Virtues, qua virtues, are not ‘in the objects themselves’ independent of our moral sense, but the properties which cause approbation in suitable observers are naturally fitted to cause this.1 Hence ‘we attain knowledge’ of virtues only by a ‘finer internal sense’; unlike our knowledge of those sorts of facts which are discoverable by methods common to ‘every rational intelligent being’ (EPM, 1.3/170; see also T, 3.1.1.3/456). The ‘internal sense’ is the ‘finer’ to the extent that it exhibits the qualities which make it authoritative (see below). Note too that virtues are qualities of mind, but that other qualities can also excite love and pride, namely ‘advantages’ of the body (such as athletic prowess) and fortune (such as riches and fine equipages) (T, 3.3.5.1/614). Certainly Hume claims that natural abilities can be virtues, but only those that are qualities of mind. Among these he includes good sense, knowledge, and judgement, which give ‘lustre’ to the other virtues. We turn now to a summary of the circumstances in which senses of virtues are authoritative. These circumstances are multilayered. First, the approver must satisfy the condition of the possibility of a moral sense.

Christine Swanton

229

Were that condition not satisfied in the human species as a whole, there would be no morality (for that species): Extinguish all the warm feelings and prepossessions in favour of virtue, and all disgust or aversion to vice: render men totally indifferent towards these distinctions; and morality is no longer a practical study, nor has any tendency to regulate our lives and actions. (EPM, 1.8/172) However, this is not all there is to the moral sense. For what if we had ‘warm feelings’ towards cruelty and callousness, and ‘aversion’ to benevolence? For the existence of genuine moral properties – virtue and vice – another condition is required. We also need as a species to possess benevolence as an ‘original passion’, one that is part of our very ‘frame and constitution’, and to have a capacity for a benevolence-driven sympathy. In ‘The Sceptic’ Hume describes this condition in terms of conditions which disqualify an agent from possession of a moral sense: [W]here one is born of so perverse a frame of mind, of so callous and insensible a disposition, as to have no relish for virtue and humanity, no sympathy with his fellow-creatures, no desire of esteem or applause; such a one must be allowed entirely incurable, nor is there any remedy in philosophy. (Essays, 1.18/169) Third, the approver must possess the moral point of view: that is, be possessed of the capacity for ‘extensive sympathy’ (T, 3.3.1.22/586). It is not enough that she have passions of benevolence, or even ‘limited generosity’: the desire for the good of another must be transferable by the mechanism of sympathy sufficiently and extensively.2 Fourth, the moral point of view, to be appropriate, must satisfy a condition of sufficient stability and impartiality – described in the Treatise as the steady and general point of view (T, 3.3.1.15/581). Fifth, to be authoritative, a moral sense must be educated: the approver must both have an understanding of the general tendencies of traits from the point of view of their ‘usefulness’ (see, for example, EPM, App. 1/285–6), and be sufficiently discriminating: But in order to pave the way for such a sentiment, and give a proper discernment of its object, it is often necessary, we find, that much reasoning should precede, that nice distinctions be made. (EPM, 1.9/173) As we shall explain further below, the approver must be able to distinguish, for example, genuine tenderness from phony or affected tenderness, what is genuinely dazzling from what is dazzling to a person whose mind is

230 What Kind of Virtue Theorist is Hume?

disordered by excessive enthusiasms, and so forth. Note however that there can be different authoritative judges: they may have different ‘humours’ depending on (for example) whether they are young or old, and there may be different but legitimate weightings of the sources of virtue: [D]ifferent customs have also some influences as well as different utilities; and by giving an early bias to the mind, may produce a superior propensity, either to the useful or the agreeable qualities; to those which regard self, or to those which extend to society. (EPM, ‘A Dialogue’, 37/336) Whether such ‘early bias’ constitutes disqualifying prejudice will be a matter of debate. The nature of the subject matter itself may also allow for latitude. Note too, that an authoritative moral sense cannot be replaced by a reliance on rules and general reasoning: It is allowed on all hands, that beauty as well as virtue, always lies in a medium; but where this medium is placed is the great question, and can never be replaced by general reasonings. (Essays, 1.20/194) Experience, practice, comparison, and correct attitudes, such as the elimination of prejudice (due to such factors as rivalry, envy, hatred), are all necessary for a correct judgement when we descend to the particular. The various conditions of appropriateness of moral sentiment give us a standard of objectivity which is not in Hume tantamount to intersubjective agreement in moral judgements. For such judgements may be highly disputatious as Hume recognizes, and most people can be wrong, if they do not know relevant facts. However without a general basic agreement in sentiment (or, indeed, some basic agreement in standards of argument), morality cannot get off the ground (EPM, 9.5/ 272). The notion of a virtue as a power to produce certain kinds of sentiments in certain circumstances – sentiments of certain kinds of pleasure and pain – does not give us a definition of virtue which captures a virtue’s point and function. When Hume argues that a virtue is a power of a certain kind, he is intending to argue that a virtue would not exist qua virtue unless it had this power. Certainly Hume claims in the Treatise that the ‘very essence of virtue … is to produce pleasure, and that of vice to give pain’ (T, 2.1.7.4/296), but he is referring to their (ontological) nature as powers. The point and function of a virtue in the living of our lives, by contrast, is not to excite certain sentiments of pleasure: these are factors which enable us to detect good qualities in human beings. So we need to discuss what, for Hume, is the point of the virtues. Certainly for Hume the point of many virtues is to

Christine Swanton

231

promote human happiness. But we must be careful to note that for Hume human happiness does not just consist in pleasure and the absence of pain. In ‘Of Refinement in the Arts’ Hume claims the following: Human happiness, according to the most received notions, seems to consist in three ingredients; action, pleasure and indolence: And though these ingredients ought to be mixed in different proportions, according to the particular disposition of the person; yet no ingredient can be entirely wanting, without destroying in some measure, the relish of the whole composition. (Essays, 2.2/269–70; cf. EHU, 1.6/8) Indolence is necessary for rest, both from pleasure and activity. A life of activity with no regard for pleasure is not good, but nor is a life of hedonism where we are seen as mere receptacles for pleasure. The virtues speak to all three of these ingredients: for example, justice, benevolence, and courage, to activity; temperance, charm, appreciation of beauty and the arts, to pleasure. Industriousness and lack of self-indulgence (by, for example, avoiding excessive devotion to luxury) ensure that the necessary indulgence to the ‘weakness of human nature’ (our need for rest) does not degenerate into self-indulgence and laziness as vices. The virtues themselves are stable dispositions of the passions and motives which make for excellence or goodness in these spheres: so the latter are in proper balance and directed at the ‘medium’.3 Although we can assume that many virtues speak to the elements of human ‘happiness’: activity, pleasure, and indolence, we shall show in the next section that this does not entail either that all virtues are directed at human happiness, or that those which are directed at human happiness are necessarily designed to promote it. In the remainder of this essay I shall argue that there are two general characterizations of what makes a trait a virtue, and that if this duality is taken seriously we shall find in Hume a richly pluralistic account of virtue. This latter claim is based on an account of what sorts of properties Hume would regard as naturally fitted for our approval. However, there are those who think that there is only one fundamental criterion of virtue – the standard of utility – which Hume describes as follows: (B) ‘qualities acquire our approbation, because of their tendency to the good of mankind’. (T, 3.3.1.10/578) On the other hand this apparently single standard appears to be at odds with the well-known pluralistic account: (A) Virtues are ‘useful’ or ‘immediately agreeable’ to oneself or to others. (T, 3.3.1.30/590–1)

232 What Kind of Virtue Theorist is Hume?

Unfortunately, it appears, the individual features specified in (A) are not separately sufficient for virtue, for, surely one may think, some immediately agreeable traits are not virtues. This point is made by Rosalind Hursthouse (1999). She objects to an account of virtue that uses (A) as a disjunctive criterion of virtue: (A) would count a disturbing number of vices as virtues. The problem might be thought resolvable if we took (B) to be the genuine criterion of virtue. I shall argue on the contrary that Hume’s criteria of virtue are fundamentally plural, and in Section 11.3 argue further that a second broad criterion, understood at an abstract level as features of persons that appeal to ‘immediate taste or sentiment’, is itself resolvable into several substantive criteria of virtue, not reducible to (B). Where does this leave the above problem with (A)? (A) should be read as a taxonomy of virtue only: according to (A), if V is a virtue it must belong to one of the four categories specified in (A). (A) is not a pluralistic disjunctive criterion. Nonetheless given that I believe that Hume operates with a pluralistic criterion of virtue, (B) on my view, offers at best a sufficient condition of virtue only. Our task then is to propose a pluralistic criterion of virtue which does not share the weakness of (A) read as a disjunctive criterion. Before a pluralistic interpretation is defended however, we need to note in the next section that (B) is susceptible to more than one interpretation. Even in regard to (B) it is not obvious how to read Hume as a normative ethical theorist of virtue. Three quite different important interpretations are possible. Though I shall not argue that any is superior to the other two, I shall claim that (B), however interpreted, does not offer a necessary condition of virtue for Hume.

11.2 Interpretations of ‘good of mankind’ in Hume The three interpretations of (B) are: (a) A utilitarian interpretation. (b) A eudaimonist interpretation. (c) A Bauhausian interpretation.4 The utilitarian view is described by Sayre-McCord thus: On the utilitarian interpretation what the four categories [of the virtues] have in common, over and above their all giving rise to approbation thanks to sympathy, is their subsumability under a utilitarian principle. (1996, p. 286) This interpretation depends on the idea that ‘trait having a tendency to the good of mankind’ should be read as offering a consequentialist account of the virtues, which in the actualist (as opposed to the expectabilist) version,

Christine Swanton

233

is described thus by Julia Driver: a virtue is a character trait that ‘systematically produces more actual good than not’ (2001, p. 68). A utilitarian interpretation of actual good would yield the following interpretation of (B): (B1) What would be approved in the right circumstances as virtues are those traits which systematically produce more actual utility than not. It is easy to see, given standard axiological claims, how (B) can be interpreted in a utilitarian way. These claims have two central features, which are: (1) (2)

The replacement of the ‘good of’ (mankind) by ‘the good’ or ‘value’ (notably utility of one or other sorts). The espousal of what I have called the thesis of the aretaic independence of value. (see Swanton, 2003)

How do (1) and (2) make for a utilitarian interpretation of (B)? Consider first, (1). In normal usage, the idea of tending to the good of mankind should be understood as tending to what is good for mankind or to its benefit. Furthermore, on normal understandings, what is good for mankind is what is good for (benefits) individual human beings. However, according to (1), the idea of benefit, or the good for, is replaced by the idea of value. If the notion of the good for is replaced by the idea of value, then value can be more readily seen as a kind of stuff that has to be promoted or produced, usually maximally. Thus are born aggregative consequentialist theories about what is good for mankind. Consider now claim (2). According to (2), the concept of value is allied to the thesis of the aretaic independence of value. That is, the idea of value does not have aretaic value or excellence built into it. What has value, for example, is pleasure rather than pleasure handled excellently, friends or friendship rather than friends or friendship handled excellently. Hence, given (2), a utilitarian interpretation of (B) would seem more natural, since value could then be understood as pleasure or welfare understood in a non-aretaic way. Virtue would then be understood as a disposition to promote something that is of independent value (such as pleasure). By contrast, if a virtue is understood in an Aristotelian way, as being well disposed to items in its field (domain), and if, for example, pleasure is in the field of the virtue of temperance, then the virtue of temperance is understood as the disposition to handle pleasure excellently. What is of value, or is good without qualification as Aristotle would put it, is not pleasure as such, but pleasure handled excellently. Thus if (1) and (2) are rejected, (B) will not be understood in a utilitarian way. Rather, the good of mankind will be understood as the good for mankind, and the good for mankind – what is of benefit to individual humans in general – will comprise not having pleasure as such, but the excellent

234 What Kind of Virtue Theorist is Hume?

handling of such things as pleasure. In fact (B) is compatible with eudaimonism, which rejects both (1) and (2) above, and which we discuss next. According to a eudaimonist interpretation of Hume, (B) should be understood as follows: (B2) What would be approved in the right circumstances as virtues are those traits that are of benefit to individual human beings, where the idea of benefit is given a suitable aretaic reading. For Hume to be understood as a eudaimonist as normally understood, he would also have to subscribe to the following theses. (a) (b)

It is a necessary condition of her flourishing that a human being possess and exercise at least the core virtues. It is a necessary condition of a trait being a virtue that it characteristically (partially) constitute (or contribute to) the flourishing of the possessor of the virtue.

It is questionable that (B) should be interpreted as the claim that nothing is a virtue unless it characteristically benefits its possessor. (B) is quite clearly a claim about the connection between virtue and the benefit of mankind, understood either in terms of particular persons or of mankind in general, as Hume makes clear at T, 3.3.1.27/589–90, quoted below. It is possible that such virtues also characteristically benefit their possessor, but whether or not Hume believed this is a question I will not address. Finally, Geoffrey Sayre-McCord proposes an interesting and plausible interpretation of Hume, which he calls the Bauhaus theory of ethics. According to this view, as applied to Hume’s criterion (B) of virtue status, we have: (B3) A trait of character ‘commands approbation, when it does, in virtue of its being well suited for the achieving of certain ends or the solving of certain problems’. Particular evaluations, of (for example) traits as virtues ‘are driven by a concern for specific problems and are determined by what is well suited to solve the problems in question’. (Sayre-McCord, 1996, p. 282) ‘Dispositions apt for the solving of mankind’s problems’ is a plausible reading of ‘dispositions tending to the good of mankind’. Indeed (B3) is an improvement on (B1), the utilitarian reading, since it does not assume that mankind has, fundamentally, just one big problem – how to maximise utility. Hence SayreMcCord’s claim that ‘What the theory avoids … is any commitment to their being a single overarching standard for evaluating all solutions’ (1996, p. 287). However the unifying feature of the Bauhausian account lies in its general claim that virtues are so because they are well suited to the solving

Christine Swanton

235

of ‘problems posed by the circumstances and necessities of mankind’. Paradigm examples are justice (the problem being the need for cooperation when there are tendencies to selfishness) and benevolence (the problem being the existence of neediness and vulnerability). Even if (B1), (B2), or (B3) are good candidates for the interpretation of (B), I am not convinced that (B), interpreted in any of these ways, provides a necessary condition for a trait being a virtue. Here is some textual evidence for my view that (B) is not a necessary condition of virtue: Moral good and evil are certainly distinguish’d by our sentiments, not by reason: But these sentiments may arise either from the mere species or appearances of characters and passions, or from reflexions on their tendency to the happiness of mankind, and of particular persons. My opinion is, that both these causes are intermix’d in our judgments of morals; after the same manner as they are in our decisions concerning most kinds of external beauty: Tho’ I am also of opinion, that reflexions on the tendencies of actions have by far the greatest influence, and determine all the great lines of our duty. There are, however, instances, in cases of less moment, wherein this immediate taste or sentiment produces our approbation. (T, 3.3.1.27/589–90) This passage reveals not just one but two higher level characterizations of what makes a trait a virtue. The second such characterization, to be set along side (B), can be described as follows: (C) Qualities (species and appearance) arouse our approbation, because of their immediate impact on our (suitably educated) taste or sentiment. Consider what is being said in the above quotation. There are two general causes of the moral sentiment: reflections on the tendencies of traits for the happiness of mankind or particular others, and the immediate impact of species or appearance. The reason the second cause of the moral sentiment produces immediate pleasure is that the approbation does not arise from considerations of future beneficial consequences. Yet the felt quality of the moral sentiment is the same though arising from different sources. However, that the felt quality of the moral sense is the same does not entail that what makes a trait a virtue is, at a fundamental level, the same. But nor does the fact that there are two basic causes of the moral sense entail that what makes a trait a virtue is, at a fundamental level, plural. To show that the grounds of virtue are indeed plural, more work needs to be done. Let us then examine (C), the second, more elusive, ground of status as a virtue. What kind of traits have the power to produce immediate (pleasurable) taste or sentiment which in turn ‘gives rise to’ approbation, despite

236 What Kind of Virtue Theorist is Hume?

their not having tendencies to the good of mankind? And what can make such traits virtues rather than vices? They are traits whose expression or manifestation in various passions (such as joy, love, esteem), afford a certain kind of delight or pleasure. This delight is not mere pleasure: it is pleasure in a quality which both causes the pleasure and makes it reasonable or fitting. For what gives rise to that kind of pleasure is the fact that the pleasurable qualities are perceived by appropriate, discriminating judges as meriting honorific descriptions, such as ‘tender’, ‘noble’, ‘worthy’, ‘honorable’, ‘wellbred’, ‘splendid’, ‘dazzling’, ‘decent’, ‘cultivated’, ‘engaging’, ‘refined’, ‘delicate’, ‘proud’, and so forth. For example, the right sort of charm is engaging, honesty is honourable, proper deference as opposed to (for instance) servile deference is well-bred, attentiveness can be delicate. Such honorific terms have normativity built into them: what is really tender (as opposed to cloying, or phony, or affected, and so on), what is genuinely dazzling (as opposed to glitzy, cruel, or vulgar), is what gives immediate pleasure to suitable persons under suitable conditions. Take for example the joy that is produced by a quality said to be dazzling. Is the joy manifested an expression of a virtue of joyfulness or is the joy some kind of ‘deformity’? That depends on whether the quality described as dazzling is genuinely so, and thereby merits the joyous response, and that depends in turn on whether the person whose passion of joy is excited is a true judge, or is, to the contrary, someone whose mind is ‘disordered by the frenzies of enthusiasm’ (Essays, 2.2/268). Again, take the pleasure elicited by a display of apparent love described as ‘tender’. It may be that the witness of such a display lacks a refined sensibility, and cannot discriminate between phony and genuine displays of affection. In that case her immediate sentiments of pleasure would not be authoritative for judgements of virtuous love. It is clear that Hume suggests that there may be virtues conforming to (C) – for example, those that arouse pleasure by their ‘dazzling’ qualities, or by their ability to ‘seize the heart’ by their ‘noble elevation’ or ‘engaging tenderness’ (EPM, 7.22/258). Such virtues are ‘excessive bravery’ and ‘resolute inflexibility’, for we take pleasure in ‘splendour and greatness of appearance’ (EPM, 7.24/258; T, 3.3.2.15/601). What we need to do now is explain why such virtues should be understood as such. For the fact that immediate agreeability (as well as reflections on utility) both cause approbation is compatible with the claim that at bottom, it is only the latter that grounds virtue status.

11.3 Hume’s pluralism Though we may admire and take delight in some, indeed most, virtues because they are effective in promoting some end – the good of mankind – some may be delighted in and admired for other reasons. Their rationale

Christine Swanton

237

as virtues does not lie in their promoting anything. Why then should such traits be admirable? Shouldn’t their admirability be grounded in something worthwhile? The task now is to account for Hume’s acceptance of virtues which are immediately agreeable as opposed to possessing tendencies for the good of mankind or of particular persons. I shall argue that attention to Hume’s discussion of the passions makes it quite plausible to assume that the admirability or goodness of certain traits is grounded in the excellent and stable expression of a variety of passions. The complexity and plurality of the account of virtues arises naturally from the complexity and plurality of the passions discussed by Hume, provided that we do not assume that the approbation elicited from our survey of virtues based on these passions is targeted and grounded in one presumed end of all those virtues – promotion of the good of mankind. That is why the second higher level characterization of virtue status – (C), can lead to a rich and fascinating pluralism in Hume’s account of the virtues. I shall develop a pluralistic account of virtue status in Hume along the following lines. The best judge of what counts as virtuous expression of the passions is the judgement of a ‘true’ judge, who has a delicacy of appreciation, is educated, and knowledgeable in the relevant respects, whose mind is cleared of prejudice and inappropriate partiality, and is ‘perfected by comparison’.5 So, for example, a true judge may take delight in a manifestation of love, and that delight elicits the description ‘tender’. The status of that manifestation as excellent or good is vouchsafed by a true judge’s description as ‘tender’, which marks the fact that her pleasure is authoritative. Of course, we cannot assume from the one case that a virtue is present, since the single manifestation may not be of a trait that is stable or enduring. As we have seen, we also need to know what it is about the trait that is ‘naturally fitted’ to produce such responses, so that we can make sense of the true judges’ determinations. My claim will be that features of our constitution recognized by Hume, such as the fact that we are creatures who bond, creatures who ascribe status to objects in virtue of certain qualities, and creatures who appreciate certain values (regardless of their tendencies to promote our good), explain the ‘natural fittingness’ of the relations detected by a true judge. Let us now consider the various passions which are the bases of what may be called virtue clusters – (the love-based virtues, joy-based virtues, pridebased virtues, and so on) and show how their status as virtues conforms to (C) as opposed to (B). Consider, first, love. The object of love is always someone related to us: Virtue and vice when consider’d in the abstract; beauty and deformity, when placed on inanimate objects … excite no degree of love or hatred, esteem or contempt towards those, who have no relation to them. (T, 2.2.1.7/331)

238 What Kind of Virtue Theorist is Hume?

Love is distinct from benevolence. Benevolence as a virtue is a disposition to desire the good of another (and act on that desire). Benevolence as a (direct) passion is simply the desire for the good of another. Love, by contrast, is an indirect passion. Benevolence as Jane L. McIntyre explains in ‘Hume’s Passions: Direct and Indirect’ ‘is not inherent to the emotion of love itself’ (2000, p. 80). She elaborates thus: Benevolent desires, which are themselves direct passions, arise in the complex causal context created by the indirect passion of love. We take pleasure, for example, in another person’s witty or intelligent conversation, and, through the double association of impressions and ideas, that pleasure gives rise to a pleasurable feeling of love for that person. It is this indirect passion of love, according to Hume, that causes the desire of that person’s good. (McIntyre, 2000, p. 80) Such love is able to be extended to others not immediately connected with us by the principle of sympathy. For example, esteem/love of the rich and powerful is originally occasioned by our expectation of succour and protection from them. The esteem/love is transferred to others from whom we can expect no advantage, by the principle of sympathy, and the ‘influence of general rules’ (T, 2.2.5.12/362). Even though the desire for another’s good is not essential to love (T, 2.2.6.5/367), Hume does claim that we can never love any person without desiring his happiness. However other emotions such as the indirect passion of envy can be mixed with love, contaminating it. In resentmentfilled individuals, any desire for the welfare of individuals loved may be contaminated by and indeed overwhelmed by, malice. Hume indeed explains this phenomenon: The predominant passion swallows up the inferior and converts it into itself. (T, 2.3.4.2/420) Hume’s own example concerns the conversion of hatreds to love: When a person is once heartily in love, the little faults and caprice of his mistress … however related to anger and hatred; are yet found to give additional force to the prevailing passion. (T, 2.3.4.3/420) In a sports-loving country, prone to resentment of the rich, conversion of love to hate may be displayed towards the most successful professional sportsmen, to the point where, for example, the attacking genius of a great

Christine Swanton

239

player is downgraded (‘no real sidestep – relies on his strength’) and defensive frailties so exaggerated (‘no real technique in the tackle – relies on his strength’) that resentment-filled individuals do not want him in the national side.6 His strength is still related to love, but yet ‘gives additional force’ to the ‘prevailing passion’ of hate. In the taxonomy of the virtues, many love-based virtues, such as attentiveness and tenderness, come under the category of the immediately agreeable. When Hume states that our hearts are seized by ‘engaging tenderness’ he is focussing on the bonds of love rather than the mere promotion of good (or value) by acts of tenderness. What affords pleasure is not in this instance the thought that the consequences of such tenderness are likely to be generally beneficial, or that such tenderness is liable to play a role in the solving of problems. Rather, the pleasure is excited by the immediate, emotionally charged perception of a bond, which can be recognized as itself a basis of a moral response (see further Swanton, 2003, Chapter 2). Attentiveness has the same kind of rationale for Hume: ’Tis remarkable, that nothing touches a man of humanity more than any instance of extraordinary delicacy in love or friendship, where a person is attentive to the smallest concerns of his friend, and is willing to sacrifice to them the most considerable interest of his own. (T, 3.3.3.5/604–5) He goes on to make it clear that the merit in such virtue is not its tendency to the good of mankind: Such delicacies have little influence on society: because they make us regard the greatest trifles: But they are the more engaging, the more minute the concern is, and a proof of the highest merit in any one, who is capable of them. (T, 3.3.3.5/605) A similar point can be extracted from Hume’s discussion of grief in T, 3.3.3.6/605. Here Hume is claiming that ‘excessive’ grief for a close friend, though disutile, can be seen as virtuous for it is expressive of a strong bond. Of course not all excessive grief is virtuous grief: it may be pathological or not expressive of a genuine bond. The outpourings of grief on the death of Princess Diana may be classed in this category, but correct judgements in this area are not clear cut, since theories of grief themselves are subject to controversy. My claim then is this. As a substantive basis of virtue we can add to (B): (C1) Traits acquire our approbation because they are expressive of bonds of (non-pathological) love.

240 What Kind of Virtue Theorist is Hume?

Hume lists some of these virtues: Love may shew itself in the shape of tenderness, friendship, intimacy, esteem, goodwill. (T, 2.3.9.32/448.) We turn now to pride. What are we to make of our finding immediately agreeable traits such as excessive bravery and dazzling spendour? Pleasure in such traits is founded on such virtues as loyalty, patriotism, and appreciation of heritage. To understand these as virtues we need to turn to another passion, pride. According to Hume, the object of pride is self, but we can take pride in the praiseworthy qualities of others provided there is a connection between those others and self. Indeed pride can be taken in all manner of objects, but these must somehow reflect on ourselves as Hume makes clear in this passage: We found a vanity upon houses, gardens, equipages, as well as upon personal merit and accomplishments; and tho’ these external advantages be in themselves widely distant from thought or [of?] a person, yet they considerably influence even a passion, which is directed to that as its ultimate object. This happens when external objects acquire any particular relation to ourselves, and are associated or connected with us. A beautiful fish in the ocean, an animal in the desert, and indeed any thing that neither belongs, nor is related to us, has no manner of influence on our vanity, whatever extraordinary qualities it may be endowed with. (T, 2.1.9.1/303.) So the beautiful fish spotted by me on a snorkelling trip to Vanuatu cannot be an object of pride. However, when I, a New Zealander, am affected by the dazzling splendour of the royal processions in Pall Mall, London, I am affected not simply by their ‘impersonal’ value but by their connection with those of us who value (take pride in) our British (or perhaps English) heritage. This point is made by Hume in T, 2.2.4.8/355: The mind finds a satisfaction and ease in the view of objects to which it is accustom’d and naturally prefers them to others, which tho’, perhaps, in themselves more valuable, are less known to it. The partialism rationalised by pride of course has to be subjected to the general and steady point of view for such ‘satisfaction and ease’ may be a mark of complacency, prejudice, laziness, fear, and other faults. But Hume is drawing our attention here to a feature of human nature ignored at our peril. Like Heidegger, Hume emphasizes that we are creatures who need to

Christine Swanton

241

belong, who need a heritage, as opposed to merely needing to appreciate the value of things (from a disinterested perspective) independent of their relation to our own particular dwelling (as Heidegger would put it). This fact accounts too, I think, for excessive courage being immediately agreeable: we take pride in it insofar as its manifestation in others benefits the defence of our country, our homes, our children. One’s nation or country can be a source of pride for Hume, and wartime reinforces the sense of unity and associated partialistic perspective (T, 2.2.3.2/348). Hence excessive courage or resoluteness in the prosecution of war is admired from a sense of national pride. Indeed, far from it being the case that utility is the sole determinant of such admiration for Hume, he says something surprising to modern sensibilities: The excessive bravery and resolute inflexibility of Charles the XIIth ruined his own country, and infested all his neighbours; but have such splendour and greatness in their appearance, as strikes us with admiration; and they might, in some degree, be even approved of. (EPM, 7.24./258) In other words ‘species and appearance’ can be a source of approval as a virtue, but note that such approval must be educated and be capable of discrimination. For in the above passage Hume qualifies his claim: if the traits are expressive of (‘symptoms’ of) ‘madness and disorder’ they would not be approved. This kind of reason for withdrawal of approval is common in Hume: if joy is expressive of a ‘disordered’ mind it is not approved, and nor is excessive cheerfulness manifested in ‘dissolute mirth’ approved when it is seen to be a ‘sure symptom and characteristic’ of folly (EPM, 7.24/258, n. 1). A third ground of virtue status can now be provided: (C2) Traits acquire our approbation because of the (proper) pride we take in them. There is not space to develop what might count as proper pride – what are the limits of pride-based virtues. But discussion of such limits would yield a rich account of virtue: for example Annette Baier draws our attention to the fact that modesty is a virtue for Hume, on account of the fact that ‘modesty is simply a recognition of the limits of one’s grounds for pride’ (1991, 207). Modesty can then be seen as a pride-based virtue, but one that recognizes its proper limits. I have claimed then that for Hume there are two groups of virtues – love-based and pride-based – whose status as such is grounded on a feature of the human constitution: our need for bonding (including bonding to self). If we cannot take proper pride in our achievements and so forth, then that is expressive of a lack of self-love. The notion of proper bonding explains why approbation

242 What Kind of Virtue Theorist is Hume?

of love- and pride-based virtues is sourced in the immediately agreeable rather than in reflections on utility, and why they are naturally fitted to produce such a response in true judges. A puzzle remains: how are we to account for virtues related to appreciation of the beautiful fish in the ocean when the fish is not connected with me directly or indirectly? If such a thing cannot be a source of pride or love, how can we account for the environmental virtues, or the virtues of connoisseurship, or the love of wisdom? What of the ability to take delight in ‘virtue in the abstract’, or ‘beauty in inanimate objects’ entirely unrelated to me? The ability of a person to ‘overflow’ from an ‘elevate’ or ‘humane’ disposition (T, 2.2.2.8/335) when taking pleasure in wonderful or pleasing objects is sourced in our capacity for joy, a direct passion. Joy is a pleasure taken in a good when it is certain or probable, while uncertain good gives rise to hope only (T, 2.3.9.5–6/439). We have then another ground of virtue: (C3) Traits acquire our approbation insofar as they (appropriately) express joy. Such virtues are joyfulness, curiosity (disposition to take joy in the process of discovery), wonder (tendency to take joy in mystery or the mysterious), and cheerfulness. Corresponding vices are apathy, tendency to boredom, and curmudgeonliness. The response of joy and related passions are features of our constitution which explain how we can take delight in valuable things independent of their consequences for our good. Again, we have an explanation about what makes certain traits naturally fitted to produce immediate pleasure. Joy-based virtues can also be linked to those virtues which speak to our need for bonding. Wit and charm, much maligned as virtues relevant to morality, but endorsed as virtues by Hume, can therefore be understood as moral virtues, for they are expressive of bonds between friends and acquaintances, and that is the source of the pleasure we take in them. If one were indifferent or misanthropic, one wouldn’t bother with either wit or charm, cultivating instead Aristotle’s conversational vices of sourness, or buffoonery. But as Hume recognizes, to class them as virtues is not necessarily to class them as expressing the great lines of our duty. There remains for us to consider the moral response (or responses) particularly associated with the ‘artificial’ virtues such as justice and honesty. I shall explain these by appeal to Hume’s sensitivity to various kinds of status properties which are not reducible to those of utility. These properties give rise to respect for those persons or institutions which possess them. The immediate impact the expression of such virtues has on our sentiments is marked by such terms as ‘decent’ or ‘noble’ or ‘honourable’ or ‘fitting’; with corresponding pejorative descriptions of the manifestations

Christine Swanton

243

of the correlative vices. But what is it about acts of justice and honesty that gives rise to such responses? It is well known that for Hume, there is no natural sentiment constituting the motive of justice, such as love or benevolence, and that this caused a problem for Hume’s doctrine that ‘no human action can be virtuous, or morally good, unless there be in human nature some motive to produce it, distinct from the sense of its morality’ (T, 3.2.1.7/479). Hume rejects a number of possible candidates for such a motive of justice. There is no motive of love of mankind as such, ‘private benevolence’ may fail, and even when present will not necessarily deliver the right, just, results. For justice may be owed to someone not loved or hated. For ‘what if [a person] be my enemy, and has given me just cause to hate him? What if he be a vicious man, and deserves the hatred of all mankind? What if he be a miser and can make no use of what I wou’d deprive him of? What if he be a profligate debauchee, and wou’d rather receive harm than benefit from large possessions? What if I be in necessity and have urgent motives to acquire something for my family?’(T, 3.2.1.13/482). However, it seems to me that the immediate pleasure aroused by the manifestation of justice (for example) can be caused by the proper display of relevant passions recognized and discussed by Hume. Is there not a desire, associated with respect, which is different from benevolence (desire for the good of another)? Rather than respect being the desire for the good of another or institutions, we should understand respect as fundamentally the desire to leave things and persons alone; to abstain from interfering with, or violating them, either capriciously, or for their own or other’s ends (including their good). This kind of respect, which I shall call justice-based in order to distinguish it from another sense of ‘respect’ (discussed below) in which respect is opposed to contempt, is grounded in the passion of compassion or pity.7 Could Hume account for the approbation of justice towards persons to whom we are indifferent or whom we hate, by means of an appeal to a desire not to violate? At first sight it would appear not, given that hatred is causally annexed to the desire for the misery of another (T, 2.2.6.6./368). How then can hate be coupled with respect in the requisite sense? It can be done in the following way. Note first that the desires for the misery of those hated ‘arise only upon the ideas of the … misery of our enemy being presented to the imagination’ (T, 2.2.6.5/367). But if these ideas are not presented to the imagination (for example, by incitement, the media, and so forth), then the above condition of the desire for the misery of others will not be satisfied. Given this, the following becomes possible. The imagination allows us, through the operations of pity combined with sympathy, to not desire the misery of those hated. In this way Hume can assert that passions of love and hate ‘may subsist for a considerable time without our reflecting on the happiness or misery of their objects’ (T, 2.2.6.5/368). I believe then that Hume could account for a justice-linked desire within his theory of the passions, by appeal to the passion of compassion or pity.

244 What Kind of Virtue Theorist is Hume?

The suggestion is this. Hume claims that all human creatures are related to us by resemblance, and that we can therefore pity even those who are indifferent to us. Can the operations of the imagination allow us to pity those in ‘grief and affliction’ even where they are hated? If desires for the misery of another are in abeyance, the operations of pity may take over. As love is causally associated with benevolence (desire for a person’s good), pity is causally associated with a desire not to cause ‘grief and affliction’ – a desire based on sympathy with humanity. This desire is not the same as that of benevolence: it is causally based on sympathy with humanity based on resemblance, rather than on praiseworthy qualities in which we take pleasure. As a result, pity can exist without benevolence: the sympathies may be unconnected with the love that causes benevolence. The misery of another can cause me to take a sympathetic interest in him; but ‘as I am not so much interested as to concern myself in his good fortune, as well as his bad, I never feel the extensive sympathy, nor the passions related to it’ (T, 2.2.9.14/ 386). Thus justice, founded on the desire not to cause grief and affliction – a desire based on sympathy with a person by virtue of his humanity, as opposed to esteemed qualities – is to be distinguished from benevolence. Nor is it simply reducible to the regard for justice itself. We have then another basis for the immediate agreeability of a trait: it is immediately agreeable because it displays justice-based respect for persons. It can also be seen that respect for another person’s property can be based on this pity or compassion-based sympathy, since one can imaginatively be aware of the personal affliction caused by loss or damage to a person’s entitlements including his property. Accordingly we offer another potential ground of virtue status in Hume based on (C): (C4) Traits acquire our approbation, because they display justice-based respect for persons and their property, and that respect is caused by our capacity for compassion or pity. Not all respectful demeanours are founded on our capacity for pity. The respect of deference, to those in authority, to the aged because of characteristic wisdom and experience, and to others to whom we defer, must be founded on another passion. Indeed Hume discusses that passion which he calls esteem or respect, and which is opposed to contempt. He claims: A sense of superiority in another breeds in all men an inclination to keep themselves at a distance from him, and determines them to redouble the marks of respect and reverence, when they are oblig’d to approach him; and where they do not observe that conduct, ’tis a proof they are not sensible of his superiority. (T, 2.2.10.10/393)

Christine Swanton

245

Such failure is a mark of ‘ill-breeding’. Consider the following example. When a Prime Minister of Australia violated rules of distance by putting his arm round the Queen’s waist, that behaviour provoked a reaction of extreme immediate disagreeability on the part of many (especially in the UK) which had nothing to do with beliefs about the good of mankind. For respect has to do with treatment according to status, as opposed to good (see Swanton, 2003). Korsgaard calls (deference-based) respect a modified, blended passion: one that ‘results when the pleasing qualities in a person that cause … love also at the same time cause humility and fear’ (1999, p. 19). Both kinds of respect (justice-based and deference-based) involve keeping a distance, but for entirely different kinds of reasons. I suggest too, that Hume can account for the respect for authoritative rules in the same sort of way. It may be that breaking a rule will not cause harm, and would promote good. But we defer to these rules because they emanate from authoritative persons who are superior to us in that regard. Hence it is arrogant of us to usurp the status of the rules, taking it upon ourselves to break them in the common interest. Virtues of deference and obedience – virtues of proper demeanour according to status-rankings – are based on respect (as opposed to contempt). Such virtues, including appropriate obedience and deference, are opposed to a range of vices, including servility, blind obedience, sycophancy, authority complex (sourced in anger against those in authority), and envy-based resentment (sourced in anger against those who are superior to oneself).8 Not all status-related virtues involve deference to superiority in a certain respect: rather the respect owed may simply express itself in virtues of decency, genteelness (EPM, 9.18/280), or decorum: Decency, or a proper regard to age, sex, character, and station in the world may be ranked among the qualities which are immediately agreeable to others. (EPM, 8.12/266) However, whether the respect (esteem) is hierarchical or whether it is based on a more general notion of proportionality (EPM, 8.12./266), a general ground of virtue status can be seen to depend on status properties: (C5) Traits acquire our approbation because they mark proper appreciation of and respect for status properties (whether or not that respect is deferential), and that respect is caused by our capacity for esteem.

11.4 Conclusion The problem addressed in this essay was not only to defend a claim that for Hume (C), as well as (B), provides a substantive ground for the status

246 What Kind of Virtue Theorist is Hume?

of a trait as a virtue but also to show how this could be so. The task was to discover what makes traits ‘naturally fitted’ to arouse, in suitable observers, approbation – (faint and more imperceptible love) as a result of immediate presentation to taste or sentiment. I claimed that, for Hume, the ‘immediately agreeable’ virtues, like other virtues, speak to ‘ingredients’ of our happiness, namely activity, pleasure, and indolence; but that it does not follow from this that all virtues are such because of their promotion of the good of mankind. The general idea of virtues speaking to elements of our ‘happiness’ needs to be understood as excellences relative to these ingredients in the following sense: they can be understood as excellences when we understand the nature of our constitution, and thereby, what it is that makes such virtues immediately agreeable to the refined educated sensibility. I argued that properties naturally fitted to arouse the requisite responses relate directly to aspects of our constitution which address our needs for bonding (to other people, our nation, etc.), our respect for status (for example, as authoritative, inviolable, superior/inferior in relevant respects), and our appreciation of the kind of value (such as beauty in nature and art for example) expressed in the wonder and joy in nature experienced by those of ‘elevate’ disposition, and in the delight in artistic quality experienced by the sophisticated art appreciator. Notice however that properties such as proper bonding, appropriate status, and value, like virtue itself, are ‘in the object’ only insofar as the human constitution sees qualities in objects in a certain light: a light intrinsically related to our responses and constitution. No talk of ‘eternal’ ‘immutable’ fitnesses is permitted. On the view proposed then, appreciation of art or nature, for example, relates to the ingredient of our happiness which is pleasure, but it does not follow that virtues of connoisseurship are so because they tend to the pleasure of mankind. Those virtues speak directly to the perception of quality in art, or value in nature. Similarly, one may say, in a Humean spirit, that creativity in art relates to the activity element of human happiness, but that does not entail that creative virtue is such only because its operation promotes human pleasure or happiness. A true judge may rather take immediate pleasure in the quality of the created piece or the nature of the creative process. And these grounds for approval need not be based on consequences for human good. Hume is interesting for a virtue theorist because his careful discussion of the passions whose refinement according to education, maturity, experience, and grasp of ‘general rules’ produces habits of virtue, shows that virtues do not need to be justified as virtues in terms of their promoting the ‘good of mankind’. For the passions display multiple grounds of immediate agreeability. Hence Hume’s substantive ethics can be seen as pluralist in a sense defined originally by Urmson (see his 1975). There is a plurality of ultimate principles for determining whether a trait is a virtue, and there is no lexicographical principle or decision procedure for determining which has priority

Christine Swanton

247

when they conflict. Certainly, Hume seems to suggest that for certain virtues, (B) may have priority across time as opposed to at a time. Although, for example, respect for the status of a rule as authoritative provides a constraint on breaking that rule for the sake of beneficial consequences at a given time, were a system of rules to undermine the good of mankind, that system in Hume’s view should over time be replaced. But this view does not imply that what makes justice a personal virtue is simply its tendency to promote the general good. And nor does it imply that the environmental virtues, for example, need be grounded in human good, even across time. Human activity and human pleasure are aspects of human happiness, but excellence in environmental activity and taking pleasure in the environment need not be determined by what is beneficial or pleasurable for human beings. Admittedly there is a real question for example whether (putative) virtues sourced in pride such as excessive courage or resolute inflexibility should be replaced by prudent courage and sensible levels of resolution, on the grounds that these latter tend more clearly to the good of mankind. However there are two issues of weighting here. One is the weighting of differing grounds for virtue status at a time. The other is not so much a matter of individual ethics but of larger scale policymaking and societal change. For example should our need for heritage – a need for bonding – be gradually replaced by a more cosmopolitan attitude, where regional pride is replaced by something else? As far as Hume is concerned one thing is clear. If (B) is to ground long-term change it cannot ride roughshod over features of human nature. Though, as Michael Gill claims, Hume has a progressive rather than an originalist theory of human nature, which allows that ‘original motives can evolve into other motives of different kinds’ (2000, p. 100), it is not totally plastic.

Notes * I wish to thank Maria Merritt for her helpful comments on an earlier draft, members of the audience at The International Hume Conference, Dunedin, New Zealand January 2003, the Ethics Circle UNC Chapel Hill, the colloquia at the University of Texas at Austin, and University of Reading; and, especially, Charles Pigden for his extremely useful suggestions. 1. Thus I disagree with Wiggins, 1991, p. 194, who claims that passages such as the one quoted above have ‘no clear place in his official theory’. 2. See further Abramson, 1999, who distinguishes between the moral point of view and the general point of view. 3. This notion should not be understood in terms of moderation, nor should we assume that ‘the medium’ is discoverable by formula. Avoidance of excess is obviously one feature but by no means the only one. 4. A term used by Geoffrey Sayre-McCord, 1996. 5. See ‘The Standard of Taste’. It is beyond the scope of this essay to discuss in detail the notion of a true judge. 6. Such a fate befell a New Zealand rugby player, Jonah Lomu – adored in Europe but belittled shamefully in his home country.

248 What Kind of Virtue Theorist is Hume? 7. Compare Darwall, 1977, where he distinguishes between recognition respect and appraisal respect, claiming that: ‘The distinction between appraisal respect and recognition respect for persons enables us to see that there is no puzzle at all in thinking both that all persons are deserving of respect just by virtue of being persons and that persons are deserving of more or less respect by virtue of their personal characteristics’ (p. 192). I believe the same kind of view can be attributed to Hume. 8. I have suggested that Hume’s understanding of respect-type passions can allow us to understand justice and deference-based virtues as respect-based. If so, then Hume could resolve the inconsistency discussed by Cohon, 1997b: that between his claim that there is in human nature some natural non-moral motive that elicits our approval and renders traits suitably expressing such motives or passions virtues, and that for the ‘artificial’ virtues there are no such motives.

12 Kinds of Virtue Theorist: A Response to Christine Swanton Annette Baier

Swanton refers to Hume as ‘a normative ethical theorist of virtue’. What breed of theorising beast is that,1 and is Hume, who is certainly some sort of theorist of virtue, a ‘normative ethical’ one? What other species of virtue theorist are there? Locke wrote that ‘These Names, Vertue and Vice, in the particular instances of their application, through the several Nations and Societies of Men in the World, are constantly attributed only to such actions, as in each Country and Society, are in reputation or discredit’ (Essay, 2, 28, 10/353). He notes that what passes for vice in one country can count in another as virtue, or at least not as vice. In a footnote to the fourth edition, replying to ‘the learned Mr Lowde’, who had objected to the apparent relativism of this account of virtue, he says that he is only reporting matters of fact. Is he disavowing being, in this footnoted passage at least, a ‘normative ethical theorist of virtue?’ Is he an anthropological theorist of virtue? For Locke, most of our moral vocabulary consists of names of ‘mixed modes’, such as sacrilege, adultery, incest, murder, parricide, obligation, justice, gratitude, hypocrisy, ideas which are made by the human mind for its own purposes and convenience. That convenience varies from one society and social condition to another, so terms from the moral vocabulary of one language often lack one-word equivalents in another. (Locke cites the Romans’ ‘versura’ and the Hebrew ‘corban’.) Our virtue-names, names such as ‘generosity’, ‘honesty’, ‘fairness’, name traits approved in our society, relating them to what Locke calls ‘the law of reputation’, as the names of what we call ‘sins’, such as lust and greed, relate actions to the law of God, and the names of what we call ‘crimes’, such as treason and manslaughter, relate to the civil law. Is Locke a virtue theorist, in giving us this account of what virtues essentially are? If so, he seems a descriptive not a normative theorist, and as much a law theorist as a virtue theorist. By contrast, that other mixed law-and-virtue theorist, Hobbes, who lists 19 virtues corresponding to his 19 ‘articles of peace’ or ‘laws of nature’, is clearly reserving the title ‘virtue’ for conformity to laws that he himself thinks have rational and normative force. The learned 249

250

Kinds of Virtue Theorist

Mr Lowde seemed to suspect Locke of being a sceptical virtue theorist. Is Hume’s account of the virtues a sceptical one, a descriptive one, a normative one? Or might it be all of these? Hume originally put forward his version of the virtues we recognise as confirmation of what he had claimed in the first two books of the Treatise was the way our minds work, and in particular, the way our various passions interact and determine our actions and reactions. He analysed human inference and belief, and our belief-informed passions and motives, before recording what particular forms of these mental abilities, these passions, and these motives have been admired. Sometimes, but not always, he joins in the recorded praise. In EPM, he records some once-admired things that he himself seems to find abhorrent. He is curious about the way our minds, passions, and our mutual admirings and abhorings show themselves in different cultural conditions, pastoral and agrarian, highland and lowland, rural and urban, barbaric and civilised, monotheistic and polytheistic, religious and secular, monarchical and republican. His essays explore some of these differences, and culminate, as far as this question goes, in ‘Of the Standard of Taste’. For morality, to him, is a matter of taste in character traits. Is Hume a mixed law-and-virtue theorist? There is no doubt that he agrees with Locke that virtues are whatever mental traits are praised, and that concern for reputation is the main encouragement to virtue. In EPM he tells us that, to compile a list of virtues, all we need do is ask what traits we would like others to find in us. This is pretty clear acceptance of Locke’s ‘law of reputation’ as the determinant of virtue. And, like Locke, he notes that different societies have recognised different virtues, that ‘the martial temper of the Romans, enflamed by continual wars, raised their esteem of courage so high, that, in their language, it was called virtue, by way of excellence and of distinction from all other moral qualities’ (EPM, 7.13/254–5). (Of course it was seen as the most excellent manly quality, and a feminist theorist of virtue might well ponder the etymology and suspect history of the word ‘virtue.’) Is Hume, as a theorist of virtue, a temperament-relativist and a cultural relativist? More, I think, a sceptic, putting the case both for and against the idea of there being one true list of genuine virtues. He not merely notes the cross-cultural and cross-temperament variations in what gets counted as a virtue, but holds out some hope that we can all agree on what to praise, if we can get ourselves to adopt a ‘steady and general’ point of view, the ‘viewpoint of humanity’. To do this, he thinks we must free ourselves, if we can, from ‘the delusions of superstition and false religion’. Locke said (at least to Mr Lowde and others like him) that it is, of course, to God’s unchanging Law that we must turn, to find a fixed version of right and wrong. That law, he says, is concerned with ‘the General Good of Mankind’. Hume’s point of view of humanity is a secular descendent of Locke’s God’s-eye-view of what is for the human good.2 Both Hobbes and Locke see virtues as obedience to some set of laws. With his refusal to join the religious in restricting virtues to what is shown in our

Annette Baier

251

voluntary action, for which we might be rewarded or punished, with his determinism, and his defiant inclusion of the ‘natural abilities’ in his list of virtues, Hume cannot assimilate all virtue to law-obedience. He is an antivoluntarist virtue theorist, in contrast to Locke. But one important subset of Hume’s virtues, those he calls ‘artificial’, does consist of voluntary conformity to rules, to the varying rules of property, contract, government, and marriage that we find in different societies. Virtue theorists, these post-Kantian days, often are anti-deontological. Hume lived in a society in which law versions of morality were standard, and were standardly supplemented by lists of supporting virtues. He reverses this. He follows Aristotle and Cicero in taking the concept of a virtue as the main ethical concept, and then supplements his account of the virtues with a recognition of our need for agreed rules in some vital matters,3 so that conformity to these rules comes to count as having the ‘artificial virtues’. What Hume termed the ‘natural’ virtues need supplementation by what he is willing to call the ‘laws of nature’, since it is natural for all communities to invent some version of them. Like the ancient moralists, he just takes the centrality of the concept of a virtue for granted, and goes on from there. Unlike them, he lived in a society where, at least since Aquinas, divine law versions of morality had been dominant. We find anti-religious arguments in Hume, and by implication these are antidivine law arguments, but his attitude to the secular Natural Law tradition, to Grotius for example, is not unfriendly.4 Hume however believes that, by nature, the human mind does not restrain itself ‘by any general and universal rules, but acts on most occasions as it is determined by present motives and inclinations. As each action is a particular individual event, it must proceed from particular principles, and from our immediate situation within ourselves, and with respect to the rest of the universe’ (T, 3.2.6.9/341). He also thinks that ‘we naturally desire what is forbid, and take a pleasure in performing actions, merely because they are unlawful. The notion of duty, when opposite to the passions, is seldom able to overcome them, and when it fails in that effect, is apt rather to encrease them’ (T, 2.3.4.5/421). The ‘inviolable’ duty to refrain from taking the possessions of others arises out of an ‘alteration of direction’ of the very passion it restrains, avidity, not from its frustration. None of the duties arising out of what Hume calls ‘artifices’ oppose self-interest; they merely redirect it, and he thinks we are reasonable enough to accept inflexible rules when ‘a sense of common interest’ shows a need for them. Inflexible rules, then, always call for justification, and, if they are not to prove counterproductive, should not be needlessly multiplied. Hume is not anti-deontological, but he does limit the place for inviolable duties and inflexible rules. So far I have suggested that as a virtue theorist Hume is descriptive as much as normative, that he may be a sceptical virtue theorist, that he is anti-voluntarist, and that he is not completely anti-deontological. I have also noted that a concern for human happiness is what he thinks should

252

Kinds of Virtue Theorist

guide our selection of what mental traits are virtues, if we are to expect other human beings to agree with us. Does this last fact make him a utilitarian virtue theorist? Hume in EPM lists roughly 67 virtues, and claims they all are either ‘useful’ or ‘agreeable’ to their possessors or their fellows (where neither ‘or’ is exclusive). Given the role of pleasure in Hume’s moral psychology, the agreeable, that which pleases, can scarcely get swallowed up in the useful. As Hume put the matter in the first appendix to EPM, ‘Something must be desirable on its own account, and because of its immediate accord or agreement with human sentiment and affection.’ Such intrinsically valued things are what Hume terms ‘directly agreeable,’ as contrasted with the indirectly agreeable, the ‘useful’. Hume can ask why utility pleases, a question that would not make sense if ‘utility’ just were pleasure, or even if, as for Bentham, it included what gives pleasure directly. As Henry Sidgwick noted in his History of Ethics (p. 209, note 2), Hume does not use ‘utility’ in the inclusive sense given to the term by Paley and Bentham, but restricts it to what has ‘tendency to ulterior good’. (Sidgwick also notes that this is its colloquial sense.) Swanton writes ‘immediate agreeability and utility can be understood as separate and justifiable grounds for virtue status within Hume’s system’. I agree that the ‘agreeable’ does not imply Hume-style ‘utility,’ but it is not so clear to me that Hume’s ‘utility’ is fully separable from what he terms the ‘agreeable’. The useful is the indirectly agreeable, the expected means to the directly agreeable. This seems to make the agreeable the more basic category, ‘agreeable’ the name for the more ultimate good. Hume is an Epicurean theorist of virtue. It is easy to see how he inspired Bentham, but Hume offers us nothing like a hedonic calculus. (The dubious honour of having been the inventor of a primitive version of that should go to Hutcheson.) The agreeable is separable from the useful, but Hume does stress that we like to persuade ourselves that many things that we do because we enjoy doing them are also useful, and have agreeable end products. (Hunting and ‘the love of truth’ are his examples.) And he thinks that we tend to find whatever is well designed for some useful purpose to be also beautiful and pleasing. This so-called Bauhaus effect is the reason given by Hume for why utility, to others as much as to ourselves, pleases us. ‘We enter, I shall suppose, a convenient, warm, well-contriv’d apartment: We necessarily receive a pleasure from its very survey, because it presents us with the pleasing ideas of ease, satisfaction and enjoyment’ (EPM, 5.2.19/220). There is a natural mental slide from discerning valued utility, to finding what it has to be pleasant and directly agreeable. In the Treatise, Hume would have called this an example of the ‘double relation of impressions and ideas’ (the passion/impressions being related by hedonic resemblance, the ideas by the relation of cause and effect), but by the time, in EPM, he came to address the question of why utility pleases, he had decided to leave out the associationist mental mechanics, and with them his original reservation of the

Annette Baier

253

term ‘sympathy’ for what communicates passions from person to person. In EPM, ‘sympathy,’ which still plays a big role, is as much the communicated passion, whatever it be, as the principle that does the communicating, as it had been in the Treatise. According to Herodotus, cited by Hume, the Scythians, ‘after scalping their enemies, dressed the skin like leather, and used it as a towel: and whoever had the most of those towels was most esteemed by them’ (EPM, 7.14/255. De Selincourt, in the Penguin translation of Herodotus, has a ‘kind of handkerchief’.). Herodotus tells us that the Scythians attached these human leather cloths, which turned out white and supple, to their horses’ bridles, and sometimes made patchwork cloaks out of them. It was not so much that the tanned skin was valued primarily for its usefulness, as that it signified a killed enemy, so was pleasing, and in addition could be put to use. This grisly example is given while Hume is discussing ‘qualities immediately agreeable to ourselves’. Should we ourselves be of the Scythians’ martial temper, we could be proud of our large human scalp collections, and proud on a plurality of grounds. Swanton, in defence of her pluralist reading of Hume, does not discuss this example, which, like ‘A Dialogue’, could well cause one to wonder just how normative Hume’s virtue theory really is (cf. Mazza, 2005). The agreeable need not be useful, but often, like the Scythians, and like the hunter who insists that his wife cook and serve the birds he has enjoyed slaughtering, we find a use for it. The useful need not be agreeable, and Hume stresses how disagreeable, especially in the hands of a Cato, are some aspects of that very useful virtue, justice. He also emphasises the disagreeability, to women of normal sexual appetite, of the virtue of chastity, deemed in some societies not only useful but essential in women. In other societies, such as France in Hume’s time, adultery is condoned, at some social cost. ‘We must sacrifice somewhat of the useful, if we be very anxious to obtain all the agreeable qualities’ (EPM, ‘A Dialogue’, 47/339). But the utility of something that avoids sacrifice of the agreeable, such as a well-made piece of furniture, or a benevolent act, can make a difference to how pleasing it looks to us. Utility can please, and often does. And did it not hold out the promise, once ‘cashed’, of pleasure or avoidance of distress, it would not even count as utility, as that which serves our interests. ‘But useful? For what? For someone’s interest, surely’ (EPM, 5.1.15/218). Our well-served interests rarely get fully ‘cashed out’ in pleasure, or even in lessened pain. (Most of us die with ‘money in the bank’, and insurance policies paid up.) Our interests are not reducible to our pleasures and pains, but nor are they independent of them. ‘Ulterior goods’ presuppose immediate ones, or at least the prospect of them. Hume in EPM calls the special pleasure that is taken in virtues, both the agreeable ones and the useful ones, ‘the sentiment of humanity’. The phrase might seem to do little except to emphasise that it is usefulness and agreeability to human beings, rather than gods or supermen, that is the morally

254

Kinds of Virtue Theorist

decisive thing. But, as Swanton notes, ‘humanity’ in the Treatise was the name for a virtue, or group of virtues, involving passions not far removed from pity. How far removed? Pity or compassion is, for Hume, the result of fellow feeling, or ‘sympathy’, with the suffering or misfortune of others; it is literally suffering with them. It is pain or distress taken in others’ pain or distress. He thinks women are especially prone to it, and prone to feeling it for those on their way to the scaffold. These women, he says, not merely ‘contract a kindness’ for condemned criminals, but ‘readily imagine them to be uncommonly handsome and well-shap’d’ (T, 2.2.9.18/388). Pity, Hume had earlier said, when it leads on to benevolent action towards the pitied, ‘counterfeits’ love, and both these tender-hearted women’s ‘kindness’ towards the pitied criminals, and their finding them uncommonly handsome, are expressions of counterfeit love. I do not think Hume is admiring the women he is describing in this passage. Nor do I find proneness to feel pity on his list of virtues. Benevolence heads his EPM list; kindness is on it, and pity of course can be what leads to these. Humanity the virtue is, I think, broader than the pity that can lead to kind and benevolent actions. ‘Here is a man that does many benevolent actions; relieves the distress’d, comforts the afflicted, and extends his bounty even to the greatest strangers. … We regard these actions as proofs of the greatest humanity’ (T, 3.2.1.6/478). Pity may be what motivates the relieving of the distressed and the comforting of the afflicted, but sharing one’s bounty with strangers, say by inviting passers-by to join one’s celebratory feast, is a more general benevolence, a wish to share one’s good fortune not just with loved ones and the less fortunate, but with anyone. It is what Hume terms the ‘overflowings of an elevate and humane disposition’ (T, 2.2.2.8/335). Benevolence as such, not pity-motivated benevolence, is essential to humanity the virtue. Capacity for the ‘sentiment of humanity’, that special sentiment that discerns any genuine virtue, may require that one possess some degree of humanity, the virtue. Certainly it requires both sympathy and the ability to correct its natural bias. It is an informed and impartial concern for all those represented by ‘the party of humankind’. So the sentiment of humanity will discern, among other virtues, the virtue of humanity. Hume is proud of this virtuous circularity. The moral sense ‘must certainly acquire new force, when, reflecting on itself, it approves those principles, from whence it is deriv’d’ (T, 3.3.6.3/619). Swanton, following Cohon (1997b), thinks that Hume’s account of our natural motivation to be in his sense ‘just’, and to approve of justice as an ‘artificial virtue’, needs a bit of help, and that what he says about ‘status,’ and about our capacity for pity, might supply it. I disagree. For Hume, justice is giving each person his conventionally recognised due, respecting his conventionally recognised rights and entitlements. The status of the rich and the politically powerful (and of the poor and the powerless) depends on the ‘artifices’ of property and authority, so it would be arguing in a bad

Annette Baier

255

circle to appeal to respect for status5 as the approved motive for respecting property rights or rights to command, or any entitlements. Pity is not a form of respect. It is what we might have for beggars and the very poor, for condemned criminals who have forfeited their rights, and for slaves who are denied them, but it is not in place for those claiming their just due. They might well say, ‘I do not want your pity. I demand my rights’. Justice is respect for rights, and I think Hume follows Hobbes in seeing this as, in the first instance, a matter of reflective self-interest, aided by some willingness to cooperate with like-minded others. It is ‘the interested passion’ correcting itself, the better to satisfy itself. When, eventually, moral approval is given to this ‘obliquely’ self-interested conformity to the accepted scheme of conventional rights, on the story Hume gives us, it is a matter of our ‘sympathy with public interest’, an interest in which we have a share, bolstered by our sympathy with victims of theft, fraud, breach of contract, law-breaking. We want our own rights to be respected. ‘By the laws of society, this horse, this coat is mine, and ought to remain perpetually in my possession: I reckon on the secure enjoyment of it: By depriving me of it, you disappoint my expectations, and doubly displease me, and offend every bystander’ (EPM, App. 3.11/310). The bystanders will be indignant on Hume’s behalf, when his coat is stolen, but may have little reason to pity him, if they know he still has several fine coats. There is one natural virtue that Hume does associate with justice, and that might have been brought in, along with self-interest, to supplement his version of what motive is approved when justice is recognised as a moral virtue. That is equity. Equity forbids us to exempt ourselves from rules we want others to keep. We are also able to see that some oncerecognised rights, such as those of slave-owner over slave, offend against humanity and natural equity, so we deny them the title, ‘moral rights’. To say that they offend against humanity is not just to say that we have reason to pity slaves, let alone that we love them or feel counterfeit love for them. It is to judge that we see them as fellow-human-beings, as those who, in equity, deserve, as much as we ourselves do, to be free. As Swanton notes, Hume in the Treatise explicitly denies that there is any ‘love of mankind, merely as such’, which could be what motivates us to keep the rules of our society that we expect others to observe, so ‘public benevolence’ is out, as the approved motive to justice. He also argues that ‘private benevolence’ cannot be what makes us repay a loan to a vicious man. Self-love, ‘when it acts at its liberty’, would often motivate dishonesty, not honesty. Swanton thinks that these arguments lead up to Hume’s acceptance of the conclusion that justice has to be its own motive, so that doubt is being cast on Cicero’s ‘undoubted maxim’. This is contentious interpretive territory, and here I content myself with saying that I have argued elsewhere for another interpretation of these difficult early sections of Treatise 3.2 (see Baier, 1991, Chapter 10).6 I think Hume is arguing dialectically there, and his conclusion is not seen till we attend to what he says in (T, 3.2.2)

256

Kinds of Virtue Theorist

about the intelligent self-correction of the ‘interested passion’ that he thinks leads to acceptance of a sequence of ‘conventions’. Once these are invented and accepted, there is the consequent creation of a ‘public interest’ in conformity to them. (A sense of public interest is of course not the same as ‘public benevolence’.) Sympathy with this newly created interest, in which each has a share, makes us, from a moral point of view, approve of conformity to such conventions as really can be seen to be in the interest of ‘the whole and every part’.7 I disagree with Swanton in her separation of prudence, one admirable form for self-interest to take, from (other) moral virtues. Justice, like prudence and frugality, involves an admirable and intelligent shaping of natural self-interest. ‘The interested passion’, just as much as natural benevolence and natural sympathy, takes both more and less admirable forms, and caring about the public interest is one of its admirable forms. Swanton sees ‘clusters’ of Humean virtues around love, pride, and joy.8 If we must revise Hume’s taxonomy, I see no reason not to include a fear and hope cluster, a self-interest cluster, an anger cluster (what he at one place in the History terms ‘splenetic virtues’), and several natural ability clusters, including ‘whimsical talents’. (History, vol. 6, Chapter 68, p. 379. The whimsical talents and the splenetic virtues are attributed to the Earl of Radnor, successor to the first earl of Shaftesbury as head of the Privy Council to Charles II.) Hume himself, in his philosophical works, seemed content with the four overlapping ‘clusters’ of ‘agreeable or useful to self or to others’. (In the History there is often a distinction between private virtues, shown in domestic life, and the different virtues that are needed in public office.) Given the number of virtues Hume named, there are bound to be a plurality of ways of clustering and re-clustering them. Hume is not just a pluralist theorist of virtue, he inspires plural regroupings in his ‘catalogue’ of virtues. Maybe that is the kind of virtue theorist he is, the cataloguing kind. He did, after all, go on to be a librarian. Books can be catalogued in different ways, and so can virtues.9 As a librarian, Hume got into trouble for ordering ‘indecent’ books, unfit to be in the library, and some, such as Balfour, found some of the entries on his virtue catalogue equally unsuitable ‘for families’. Balfour found Hume’s catalogue to contain ‘an unnatural assemblage of a strange variety of features, the most inconsistent and ill proportioned imaginable. Some of them are beautiful, and taken from the true original; others are borrowed from vice, and sometimes from what in vice is most shameful and deformed’ (1989, p. 161). Balfour devotes a whole chapter, one of the six that make up his book, to chastity, and Hume’s classification of it as a merely useful virtue is its main target. Hume’s list, and his fourfold classification, enabled him to raise some puritan hackles, and that I think was part of his aim. He was, in his native culture, an offensive theorist of virtue. It is also worth noting that his list of virtues is unisex, unlike Aristotle’s list for freeborn men, and unlike

Annette Baier

257

the lists to be found in The Whole Duty of Man and its supplement, The Ladies’ Calling. Hume has what may look like sexist glosses on some entries in his catalogue, but they can generously be taken as mere sociological reports of what is expected in some societies. In his History, he reports how offensive to her accusers, those who burned her alive, was Joan of Arc’s presuming to wear men’s dress, and to command men in battle. But to Hume she is an ‘admirable heroine’, to whom the ancient Greeks and Romans, with their ‘more generous superstitions,’ would have erected altars. The admired and the admirable need not coincide, and it is not always easy to tell when Hume is reporting on the former, when he is expressing his views on the latter. It is easy enough to tell what he thought of those who condemned Joan as unwomanly and a heretic. And he minces no words about the monkish ‘virtues’. But for much of the rest, he is content to tease us, perhaps to provoke us into forming our own tentative catalogue, complete with its own revised cataloguing system, and its own sceptical epilogue. So I think he would find Swanton’s response to him very agreeable, as long as someone supplies the sceptical epilogue. I have tried to do that.

Notes 1. For her answer to that, see Swanton, 2003. 2. Neither Locke’s nor Hume’s view counts as what Swanton counts, in her book, calls ‘eudaimonist’ since neither view rules out the possibility that the just person may fall victim to the unjust, and so fail to flourish. I will not go into the vexed question of how sincere Locke is when he invokes God, God’s law, and God’s purposes (see Waldron, 2002, Chapter 8, Tolerating Atheists?). 3. See Taylor, 1998, Baier, 1988, and Baier, 1991, Chapters 8 and 10, for slightly varying accounts of why this supplement is so vital. 4. See the interpretation of Hume by Braybrooke, 2002. 5. What Swanton means by ‘status’ is made clear in Virtue Ethics, 2003, especially pp. 43–7. It includes age, gender, and expertise as well as social status. Respect for it then becomes more what Hume termed ‘decency’ than any ‘artificial’ virtue. ‘Decency, or a proper regard to age, sex, character, and station in the world, may be ranked among the qualities which are immediately agreeable to others, and which, by that means, acquire praise and approbation’ (EPM, 8,12/266). 6. Cicero’s maxim is not voiced in EPM, and there is no discussion there of the approved motive or motives to justice. See my ‘Why Hume asked us not to read the Treatise’, in 2008, forthcoming, pp. 257–264. 7. I did not, in Progress, 1991, sufficiently allow for Hume’s final remarks in T, 3.2.2 about the need, in large societies where the coincidence of private and public interest may have become hard to discern, of new educational and political artifices, which create a sense of honour that shores up the interested motive to observe the rules of justice. This in a way confirms his earlier prediction that it is education, as well as convention, that must create the sense of justice and injustice, if nature is not to be seen to have ‘establish’d a sophistry.’ 8. Her wish to recognise a joy cluster is linked with her own views on the plural sources of value. Joyful response to the sight of the unowned fish in the

258

Kinds of Virtue Theorist

subtropical waters becomes, for her as for Hume, an important human response to non-human beauty. 9. I myself did a bit of dubious re-cataloguing, in Progress, 1991, Chapter 9, ‘A Catalogue of Virtues’, when I subdivided Humean virtues into the great and the good, and into those of the head and the heart. Many of them, such as considerate forgetting, straddle that latter supposed divide, which Hume himself casts doubt on in EPM, App. 4/284.

13 Reply to Baier Christine Swanton

Annette Baier quite properly asks two questions: (1) What kind of theoretical beast is a ‘normative ethical theorist of virtue’? (2) How can one inflict such a creature on Hume, especially in relation to his conception of justice? In ‘What Kind of Virtue Theorist is Hume?’ I was concerned to show that Hume’s moral theory should be understood as a pluralistic virtue theory. The emphasis was on the pluralistic character of his criteria of virtue rather than on arguments for reading him as a virtue ethicist per se. (I argue for a virtue ethical reading of Hume in ‘Can Hume be Read as a Virtue Ethicist?’ in Hume Studies, April 2007, Vol. 33, No.1, 91–113.) Nonetheless, implicit in the paper is a virtue ethical reading, so it is useful to say something about such a reading. It is important first to distinguish virtue ethics as a genus of moral theory from virtue ethics as a species. Virtue ethics has often been defined in terms of one of its species, Aristotelian virtue ethics. Such a definition makes a virtue ethical reading of Hume problematic, unless of course one interprets Hume as some kind of Aristotelian. It is my view that Hume should not be so interpreted, so a virtue ethical reading of him has to be non-Aristotelian. So the question arises, what is virtue ethics as a genus? Here are the central tenets. (1) In a variety of ways, character traits, as relatively durable dispositions, are central to ethics. (2) Character traits go very deep. They are not just tendencies to act, but they are also affective and motivational dispositions. (3) Hence, emotion is central to ethics. (4) However, though emotion is central to ethics, virtue ethics is a class of objectivist and not subjectivist moral theory. (5) The emotional component of virtue as an excellence of character, needs to be reliable, and informed by ‘reason’ or reasonableness, through 259

260

Reply to Baier

refinement, training, experience and so on. Such reason we may say is not ‘pure reason’ but incorporates in large part, what is now called emotional intelligence. Hume’s moral thought, I believe, satisfies all the above features of virtue ethics as a genus, but we are still left with Baier’s question 2. Since virtue ethics is a modern category, it may be asked, why call Hume a virtue ethicist, or attempt to give a virtue ethical reading of him? The fact that virtue ethics is a modern category does not make it inappropriate to apply that category to past philosophers: we do not feel uncomfortable, for example, in calling Aristotle a virtue ethicist. Indeed, the category ‘virtue ethics’ is historically grounded to the point that it is a reductio ad absurdum of those definitions of virtue ethics which remove Aristotle from the genre. Furthermore, theorizing under the rubric ‘virtue ethics’ has become instructive in reminding us of the centrality of virtue as a moral category in past thinkers, and reminding us of a lack in the current moral theoretic landscape, which is dominated by such categories as values and rights. However, nestled in Baier’s second question is another issue. Central to Hume’s ethical thought is the topic of justice, and it is often thought that virtue ethics cannot handle that important notion. So, to call Hume a virtue ethicist is to saddle him with this difficulty. What is the difficulty? It is sometimes said that Aristotle does not have, and indeed cannot have, a ‘virtue ethics of justice’ because ‘law is the immediate standard of just acts’ (see Santas, 1997, p. 277). Similarly, it is claimed, Hume cannot have a ‘virtue ethics of justice’ because Hume’s notion of justice, too, is law-centred. Since a central virtue, justice, cannot be understood in a virtue ethical way, it is claimed, virtue ethics as such is in trouble as a type of moral theory. To assess this claim, we need to distinguish between a virtue as a disposition of character, and a virtue’s field (its domain of concern). Similarly, we need to distinguish between a virtue as a disposition of character and its target (which is to do well in relation to its field, in line with the point and function of the virtue). True, the field of the virtue of justice for Hume, as for Aristotle, is law-centred. A person with the virtue of justice respects laws, property rights and so forth. Hence the field of justice as a virtue goes beyond ethics to institutions of law and politics. The target of the virtue is precisely to respect these things, by obeying laws, not stealing property and so on. A full theory of justice then, needs a normative theory of law and politics. Only then will we understand the nature of the field of the virtue of justice. But why is this thought to be a problem for virtue ethics? The point that understanding the field of a virtue may require theorizing about institutions, applies to a host of virtues. Consider obedience as a virtue. To understand fully the nature of obedience as a virtue we need to understand its field, and to understand that in turn we need to know what rules and so forth are to be

Christine Swanton

261

obeyed. That in turn requires that we know by what authority the rules, etc. are promulgated, and what counts as legitimate authority. This demands knowledge of proper procedure and, at a deeper level, the point or function of relevant institutions, be they medical, educational, political, legal. Only then will we be able fully to understand whether an act of whistle blowing, for example, is expressive of virtue or expressive of a vice, of (for example) an authority complex. Indeed, this general point applies to all the role virtues. A claim that X has a role virtue of benevolence as a business executive requires a knowledge of the point and function of business institutions. A CEO who goes over the heads of relevant authority in promoting a cause, which he holds dear, by using business funds to donate to that cause in a business capacity, is not exercising an appropriate role virtue. I agree then with Baier that for Hume, justice is ‘giving each person his conventionally required due’ (assuming the conventions are themselves beneficial to mankind in the long term), and ‘respect for rights’, specifically property rights. But all this tells us is the nature of the field of justice as a virtue: it does not tell us much about the nature of the disposition itself, and it is pre-eminently the task of virtue ethics to do this. For remember, it is a central claim of virtue ethics that character traits go a ‘long way down’. And here, Hume has a problem – a problem not foisted on him by a virtue ethical reading of him, but one which he recognizes himself. Here is the problem. To understand the nature of justice as a personal virtue we need to understand its emotional components. In particular, we need to know in what the ‘sense’ of justice consists. In the case of the ‘natural virtues’ such as parental love, genuine virtue requires the motive of love for one’s children as opposed to caring for them from a sense of the morality of caring, the duty of care. Hume claims explicitly that all genuine virtue or virtuous action (action expressing virtuous motive) should comprise robust emotional motivation of this ‘natural’ nature. But in the case of justice he finds none. The sense of justice is neither more nor less than the sense of the morality of justice. My suggestion was that Hume gave up too easily. A person with the virtue of justice has more than mere dutiful motivations. That person is motivated in a more primordial way, we might say, by humanity, generally called by Hume, pity. Humanity or pity is a passion constituting a natural motive distinct from benevolence. How so? ‘Pity’ is defined as ‘a concern for … the misery of others, without any friendship or enmity to occasion the concern. … We pity even strangers, and such as are perfectly indifferent to us’ (T, 2.2.7.1/369). Two important features of the notion are evident. First, pity in this sense has an impartial quality: it is not dependent on bonds or enmity. In this sense, pity may be understood as humanity. Second, a concern for misery may manifest in two ways: compassion (a ‘suffering with’ someone already suffering), and a desire not to inflict suffering by

262

Reply to Baier

violating a person. It is this second notion that might provide the passional foundation of justice as a virtue. This second way of being concerned at the suffering of others is better understood as humanity, rather than pity, and the notion of humanity, too, better reflects the impartial nature of Hume’s notion of pity. Pity (humanity) in this sense provides the emotional base of respect for persons and their entitlements by providing a motive not to violate them. In Kant’s language, such a motive is not the ‘coming close’ – helpfulness and care – of beneficence. It is the ‘keeping distance’ – non-violation – of respect. This natural motive is perfectly in keeping with the artificial nature of the field of the virtue of justice. For in order for the ‘natural’ motive to be part of virtue, the agent must have her natural humanity refined by education. She must be trained to attach this humanity to a treatment of persons in accord with the laws and rules of society. These determine a person’s entitlements, both to property and to his bodily integrity. Only when this education has occurred, does the agent understand the (law-centred) field of justice. However, in Hume’s system, a person, who possesses only the knowledge of the morality of justice (grounded in his knowledge of the public interest), but no natural humanity, is all too likely to become a ‘sensible knave’, unless the moral education to which he is subjected is sufficiently stringent and pervasive. Where this is lacking, he will have no natural humanity to fall back on – a humanity which provides a natural disgust at violating other persons for his own ends or pleasure. Baier is correct to be concerned at ‘pity’ being the motive of justice, if pity is understood as the kind of compassion one has for the suffering of ‘beggars and the very poor’ and for ‘condemned criminals’. But ‘pity’ defined as ‘a concern for … the misery of others, without any friendship or enmity to occasion the concern’ (T, 2.2.7.1/369) can have a broader meaning as we saw above. Certainly, Hume claims that ‘when I relieve a person in distress, my natural humanity is my motive’ (T, 3.3.1.12/579), and argues that just acts cannot be motivated by such an unadulterated natural motive. However, firstly, humanity as concern for misery also motivates a refraining from violating, and secondly, if that natural motive is to be part of a virtue of justice, it must be educated to form a developing conception of the field of justice. As part of the virtue of justice, humanity is not unadulterated raw concern. But nor does benevolence as a virtue consist simply in concern for others: concern as desire for the well-being of others needs to be educated, and such education involves education about institutions. Similarly, a natural motive of ‘doing as one is told’ or a natural desire to follow leaders has to be educated if it is to be part of a virtue of obedience. Otherwise it may become servility or blind obedience. Thirdly, in claiming that there must ‘be in human nature’ a natural motive for a virtue, Hume does not imply, I think, that such a motive must motivate or even be part of the motive for each virtuous act. Affection for

Christine Swanton

263

example is not the immediate motive of a parent administering punishment to her child, though affection may constrain and shape the nature of the punishment. I am not arguing that Hume himself thought there could be a natural motive of justice. Rather his discussion of the passions furnishes him with the tools to argue thus.

14 Hume on Justice Rosalind Hursthouse

14.1 Introduction As everyone knows, Hume begins his discussion of justice in the Treatise by claiming ‘that no action can be virtuous or morally good, unless there be in human nature some motive to produce it, distinct from the sense of its morality’ (T, 3.2.1.7/479). He recognises that this makes for a ‘great difficulty’ because, in the case of just actions, it seems that my motive for them is my regard for justice, which is not distinct from my sense of their morality, their justice. Why does he think that this seems to be so? Well, he might simply take it as obvious; a number of other people have done so. After all, when we try to think of a reason for doing something that is just other than the (putative) fact that it is, we seem to get stuck. Bernard Williams, for instance, discussing what Aristotle can mean by saying that the virtuous agent chooses virtuous actions ‘for their own sake’, considers whether ‘A V person chooses V acts qua V acts’ and rejects it (taken in a de dicto sense) as largely false. ‘Courageous people’ he says, ‘rarely choose acts as courageous, and modest people never choose modest behaviour as modest’ (Williams, 1995, p. 16). But, he adds casually in passing, ‘[j]ustice is about the only case in which it clearly holds’ – that is, just people choose just acts because they are just. What other reason or motive could they, after all, have? As Hume points out, the obvious candidate – with regard to the public interest – won’t do, and in some contexts, especially perhaps the contexts in which we are trying to think in Humean terms, nothing else seems to come to mind. But shouldn’t we be able to think of another candidate? After all, hundreds of philosophers are currently writing, confidently, about respecting others’ autonomy as rational agents and what we owe to each other. Does nothing they say suggest a reason or motive for doing what is just beyond the putative fact that it is? We might say it suggests the following: what motivates the just agent to do what is just is regard for another’s rights. Do I hear you cry, scornfully, 264

Rosalind Hursthouse

265

‘But talk about the regard for others’ rights is simply talk about justice in different terms!’ True, but I might reply ‘So what?’ What motivates the benevolent or charitable agent is regard for another’s good or well-being, but talk about regard for others’ good or well-being is simply talk about benevolence or charity in different terms. Yet Hume clearly holds that the regard for another’s good is a motive to produce benevolent acts that is distinct from a sense of their benevolence. So what is the difference? ‘Well’, one might say, ‘intuitively, rights are very different from wellbeing.’ Yes indeed. And that, I shall contend, is Hume’s point. (I return to this intuition below.) My general contention is that Hume’s discussion of justice is in fact an attack on that concept of rights we think of as having motivated the American and French revolutions, a concept that is still (deplorably in my view) prevalent. I think the very fact that it is so prevalent and so sanctified by its position in the American Constitution has blinded most readers to the power and plausibility of Hume’s attack, which I shall try to defend. On the face of it, linking Hume on justice with the American and French revolutionary talk about rights must seem odd, because, as we all know, the latter is concerned with natural, or as we now say, human rights, whereas Hume concentrates exclusively on property. Surprisingly, few commentators say anything about this but surely it cries out for comment. After all, it is beyond belief that Hume could have been ignorant of the ‘natural rights’ talk that was going on before and around him – Locke is full of it, discussing not only the right to property but also the rights to life and liberty; Grotius and Pufendorf discuss these three and more, they are in the Encyclopédie. We have to suppose that it is common talk at least among the educated. So how can Hume argue for the ‘artificiality’ of justice solely on the basis of property rights and leave himself wide open to any critic to attack his general claim simply by appealing to the other ‘natural’ rights, as Thomas Reid was quick to do? Reid claims that there are six branches of justice, four that pertain to certain natural or innate rights, viz ‘an innocent man has a right to the safety of his person and family, a right to his liberty and his reputation’ (Reid, 1977, 425) and two others, one regarding goods or property, and the other, contracts. And he says that, when we read Hume ‘it appears evident, that he had in his eye only two particular branches of justice. No part of his reasoning applies to the other four. He seems, I know not why, to have taken up a confined notion of justice, and to have restricted it to property and fidelity in contracts’ (Reid, 1977, 427). So there is the question – why would Hume, knowingly, take up such a ‘confined’ notion of justice? A theoretically possible answer is that he was too stupid to see the objection coming. But that too is beyond belief. So we are left with only one possibility. He thought that his reasoning concerning property applied equally well to the other putative natural rights and,

266

Hume on Justice

moreover, that this was so obviously the case that he need not bother to mention it. This may seem beyond belief too, so I will try to show that it is not.

14.2 Hume on private property Let us start with the example in Section I of ‘Of Justice and Injustice’: I suppose a person to have lent me a sum of money, on condition that it be restor’d in a few days; and also suppose, that after the expiration of the term agreed on, he demands the sum: I ask What reason or motive have I to restore the money? (T, 3.2.1.9/479) We might observe that, to anyone with the virtue of justice, the answers come readily enough: the money must be restored because it is his property, he has a right to it, it is his, I am under an obligation to restore it to him, I owe it to him. And Hume would be happy with the idea that such answers are intelligible to human beings in their ‘civilised state, and when train’d up according to a certain discipline and education’ (T, 3.2.1.9/479). But, he wants to say, if we imagine ourselves in our ‘rude and more natural condition’, namely before we have united into society and entered into the convention to abstain from the possessions of others, the answers would be ‘perfectly unintelligible’. I might just as well say, unintelligibly, ‘the money (of course there wouldn’t be money in the rude and natural condition either, but let that pass) must be restored because it is his cloperty, he has a smight to it, I am under a gobligation to restore it to him, I powe it to him’. What turns cloperty into property? The existence of laws of society. ‘Our property is nothing but those goods, whose constant possession is establish’d by the laws of society; that is’ (as Hume perhaps too quickly remarks) ‘by the laws of justice’ (T, 3.2.1.11/491). These laws turn cloperty into property and simultaneously turn smights-to-money into rights, gobligations-to-restore money into obligations and powing into owing. Now let us return to Hume’s claim, let us call it – (C) That ‘no action can be virtuous … unless there be in human nature some motive to produce it, distinct from the sense of its morality’. How does he arrive at it? His basis for it is an earlier conclusion: [T]hat the first virtuous motive, which bestows a merit on any action, can never be a regard to the virtue of that action, but must be some other natural motive or principle. (T, 3.2.1.4/478)

Rosalind Hursthouse

267

And he gets to that from the premise that ’[t]is evident, that when we praise any actions, we regard only the motives that produced them, and consider the actions as signs or indications of certain principles in the mind and temper. The external performance has no merit … all virtuous actions derive their merit only from virtuous motives, and are consider’d merely as signs of those motives. (T, 3.2.1.2, 4/477–8) So let us begin with that premise. We should, I think, be struck by the fact that, at first sight, the premise sounds much more like Kant than like Bentham or Mill. It is the utilitarians who say that an action is right – has merit or is praiseworthy in that sense – if and only if it has the right consequences, but Kant who says that, although an action that aids others may be right, it has no genuinely moral worth – lacks merit in that sense – unless it is done from a sense of duty. And this is very odd, given that Hume is going to move from this premise to the claim that seems so obviously un-Kantian, namely claim (C). How does the move come about? I take it that the premise has to be read carefully, as an aspect of his virtue ethics. It is a premise about actions as virtuous – that is, benevolent, just, courageous, kind, honest, etc. – rather than as right. It is not invariably true that an action cannot be given a virtue description (benevolent, courageous, kind, grateful, friendly, generous, etc.) if it was not prompted by the motive proper to the virtue in question, but it is certainly true enough to make Hume’s premise quite plausible – certainly much less tendentious than it would be if we construed it as being about actions that are right and had to read him as prefiguring Kant. If I do what appears to be benevolent or generous for ulterior motives, then, I think, I do not do what is benevolent or generous at all; and if I face danger through fear of punishment I do not do what is courageous. If my apparently honest action of telling you the truth is intended to deceive you long term, it does not count as honest and so on. So the correct reading of the premise does not make it about the necessary conditions for morally worthy action in the Kantian way, but about the grounds for praising actions by describing them in virtue terms. If my action regarding you is to count as genuinely benevolent, my motive must be a concern for your well-being. What about the intermediate conclusion he derives from this – ‘that the first virtuous motive, which bestows a merit on any action, can never be a regard to the virtue of that action, but must be some other natural motive or principle’ (my italics)? What would be examples of virtuous motives? He gives us a parent’s concern for her child; a concern for the well-being of others – both of which, we may note, are passions that occur in us naturally. But what about a regard to the virtue of the action? Well, suppose I want to do a benevolent action

268

Hume on Justice

because it is benevolent, that is, virtuous. As Hume notes, (in a very Kantian passage) this may happen (T, 3.2.1.8/479). Feeling my heart devoid of concern for the well-being of others, I may hate myself on that account and perform the action ‘from a certain sense of duty’. But although this may happen on occasion, it could not always happen among human beings in general The very reason I hate myself when I find my heart thus devoid is that the benevolent motive, though lacking in me is ‘common in human nature’, and grounds the praise given to benevolent actions. For benevolent actions are so-called, and merit bestowed on them thereby, because they are taken as signs of a benevolent motive, the naturally occurring concern for the well-being of others. If there were no such concern in human beings, common in human nature, but only a ‘concern to do (as it were) benevolent actions’ there would not be any benevolent actions, and hence the ‘concern’ to do them would lack an object. (Cf. Wittgenstein, 1956, §337: ‘If the technique of the game of chess did not exist, I could not intend to play a game of chess.’) So, before – logically ‘before’ – there can be a concern to do benevolent actions as such, there must first be a natural concern for the well-being of others. That concern is logically prior to benevolent actions – the ‘first’ virtuous motive as far as benevolent actions are concerned. Hence (C). Quite generally, no action can be virtuous unless, somewhere in human nature – not necessarily operative in the moment of action, but somewhere – there is some motive (or motives or ‘distinct principles’) capable of producing it. But there is not, and could not be, a natural concern, common in human nature, to abstain from others’ property, or respect others’ property rights, or honour one’s obligations regarding others’ property any more than there could be a natural concern to play chess. Before there can be that concern, there has to be chess, with its rules, which is something that human beings invent or contrive – an artifice. And before there can be property, property rights and obligations, there has to be a society with laws governing the stability and transference of external goods, which is something else human beings invent or contrive. The artifice or convention of the laws of justice, which establishes which acts in relation to possessions are just, comes ‘first’. So just actions with regard to property are logically prior to the motive(s) proper to them. They come first. So there is something wrong with saying that what motivates the just agent to do what is just (that is, restore the money), is regard for another’s right to his property and that this is parallel to saying that what motivates the benevolent agent to do what is benevolent is regard for another’s good or well-being. The intuition above was at least partially correct; well-being is indeed different from property and property rights. I turn now to the question of whether property and property rights are significantly different, in the context of Hume’s discussion, from other,

Rosalind Hursthouse

269

‘natural’ rights, such as the right to life or the right one has over one’s own body, and argue that they are not.

14.3 Grotius, Pufendorf, Locke and Hume on property The first point I want to make is that ‘property rights’ would have been for Locke, and I think is still for Hume, a pleonasm. In Locke’s day, saying that something was one’s property was a way of saying that it was something you had a right to or in. Hence his claim, in the Essay: ‘Where there is no property there is no injustice’ is a proposition as certain as any demonstration in Euclid: for the idea of property being a right to anything, and the idea to which the name injustice is given being the invasion or violation of that right; it is evident that these ideas being thus established. … I can as certainly know this proposition to be true as that a triangle has three angles equal to two right ones. (4.3.18/549) So saying you have a ‘property right’ would have to be saying something like ‘you have a right to or in something you have a right to or in’. We can see this in Two Treatises of Government, where Locke discusses rights concerning life, limbs, body and actions all under the heading of ‘property’. When he says there that ‘the chief end of civil society is the preservation of property’ (Locke, 1970, 2.4/85), this is not the claim it became in the mouths of anti-socialists who thought that nationalisation was going to bring about the downfall of civilisation. The preservation of a man’s ‘property’, in Locke, is the preservation of not only his private property but also his life and liberty: ‘to preserve his property – that is, his life, liberty and estate, against the injuries and attempts of other men’ (1970, 2.4/87, my italics). We find this very odd nowadays, because for us ‘property’ by and large covers only what Hume refers to as ‘external goods’ – my house, my car, my money, those external goods that are mine. But that use of ‘my’ and ‘mine’ gives us the clue that can take us back to understanding the seventeenth and eighteenth century use of ‘property’. The very word ‘property’ is derived from propria – that which belongs to someone, is ‘of him’ or ‘his’ (suum) – and in Grotius and Pufendorf, in Locke and Filmer and Hooker, the debates and arguments about ‘property’ are debates about rights – not particularly property rights, but rights to what is one’s own (suum) such as one’s life, limbs, body, liberty, actions, virtue and reputation. So let us just use ‘my’ and ‘mine’ (yours, his) for the moment and look at Hume on private property just in those terms. Here we are, let us suppose, in our rude and more natural condition before we have united into society. Wandering around one day, I come upon a rather nice horse and take it home with me. The horse and I get on very well for several

270

Hume on Justice

months – I feed it, it pulls my plough and gets me from A to B, everything’s fine, and then one day someone passing by says ‘Hey! That horse is mine’. What can he possibly mean? That he was ploughing with it and riding it at some earlier time and wants to again? But so what? So shall I say that it is not his but mine? But what can I mean by that? That I have been ploughing with it and so on recently and want to go on doing so? We can both shout and fight about what we have been doing with the horse and what we want, but clearly, in the absence of any laws, neither of us can claim that the horse is our property and that we have a right to it. If a third person, considerably stronger than either of us, comes along and says, ‘I fancy that horse, I’ll take it’ and does so, he has not violated the property right of either of us because, as Hume says, in this ‘state of nature’, ‘there was no such thing as property’. (I note here, because I shall want to come back to it, that the fuller quotation is ‘tho I assert, that in the state of nature, … there be neither justice nor injustice, yet I assert not, that it was allowable, in such a state, to violate the property of others. I only maintain that there was no such thing as property’ (T, 3.2.2.28/501, my italics).) This seems obvious enough in the case of private property where, as Hume nicely puts it in the Enquiry ‘uninstructed Nature’ surely never made any such distinction as that between mine and yours. But, we might say, nature surely has drawn a distinction between my life, limbs, body and actions, and yours. Even in the rude state before we unite into society, surely my life is mine, my property, I have a right to it, and, thereby, if you take my life or injure my body, you have failed to abstain from what is another’s and thereby violated my rights and hence done me an injustice. But, enlightened by Hume, we can deny that in the rude state you have done me an injustice by violating a right of mine by, for example, killing me. For we can plausibly maintain that the obvious or natural sense in which my life or body is mine is not and could not be the meum and suum, ‘anything I have a right to or in’, sense, that is, the Lockean property sense. This is simply shown by the fact that a life (limbs, body, actions) may be mine or another’s in one sense but not in the other. My life, limbs, body and actions are, in the natural sense, necessarily mine, inalienable. There is nothing I, or anyone else can do, to make them not mine. But in the meum-suum sense they are not. We should recall that the slave trade was still thriving in Hume’s day and that, in the context of slavery, it made clear sense (I can’t bring myself to say ‘perfect’ sense) to claim that someone else’s life, limbs, body and actions were mine – naturally his, indeed, but my property, what I had a right to and in because he was my slave. (Recall that when Locke claims that something becomes mine when I have mixed my labour with it, he is actually talking about the labour that someone else has done at my behest, for which I have paid some (notional) amount and thereby made it mine.) Perhaps some of the rationalist moral philosophers against whom Hume waxes so sarcastic attempted to argue

Rosalind Hursthouse

271

that what was wrong with slavery could be discerned by pure reason, namely that it embodied a contradiction – how could his life be mine! But it would have been a bad argument and moreover contrary to the common acceptance of the contemporary view that my life isn’t actually mine in the meum-suum sense but God’s. Those of us who do not believe that my life is God’s in the meum-suum sense tend to recognise an obvious way in which I can make what is naturally my life not mine in the meum-suum sense. I do it by explicitly waiving my right to life. What is naturally my life then ceases to be mine in the sense of something I have a right regarding, and so (given some other restrictions) you can kill me without violating any right of mine or doing me an injustice, as in cases of voluntary (but not involuntary) euthanasia. Similarly I can give you one of my kidneys (I don’t know that it is yet possible for me to give you one of my limbs, but no matter). In the natural sense, it will always be mine, but once I have given it to you, it is yours in the meum-suum sense; you are the one who has rights regarding it. Or consider an example that was also of interest to Hume’s predecessors. In the natural sense of the possessives, a man’s semen is his, a woman’s ovum is hers, and any child a woman bears as a result of intercourse with a man, hers, his and theirs. But in the meum-suum sense, none of these are necessarily so. Many societies have made sense of the idea that it is the begetter of a child who has rights regarding it, not its mother; the child is his, not hers nor theirs. And the ancient phenomenon of adoption has always allowed the people who brought up a child to say that it is theirs in the meum-suum sense without claiming that it is naturally so. And so, contrary to first impressions, we see that uninstructed nature can no more draw a distinction between what is mine – life, body, limbs, actions, child, say – and yours in the meum-suum sense than it can between my piece of private property and yours. The very idea of mine and yours in this sense can arise only after we have united into society and established laws that fix their content. Does this mean, you may cry in scandalised tones, that, according to Hume there would be nothing wrong with your killing or raping me in the state of nature? No indeed, for remember I quoted him as saying ‘tho I assert, that in the state of nature, … there be neither justice nor injustice, yet I assert not, that it was allowable, in such a state, to violate the property of others. I only maintain that there was no such thing as property.’ Of course murder and rape are not ‘allowable’ in the state of nature because they are ruled out by benevolence. The natural motive of benefiting, or not harming others is all that is needed. We are so accustomed to thinking of murder and rape as violations of rights that I constantly find myself having to reiterate Foot’s point that killing someone when their life is a good to them – as it normally is – is as contrary to charity as it is to justice. And hence Hume’s oft-repeated claim that if our benevolence were not limited,

272

Hume on Justice

if ‘cordial affection, compassion, sympathy, were the only movements with which the human mind was … acquainted, (T, 3.2.2.15/494), ‘if every man had a tender regard for another’, (T, 3.2.2.16/494), there would be no need for justice. (In fact, I think he is wrong about that. He is right that strong universal benevolence or charity would stop us from murdering and raping and enslaving each other, and from meanly depriving another of something he would feel the lack of, all with no need for justice. But living together in perfect benevolence, we might still find the need for some justice to protect each of us from the well-intentioned efforts of others to benefit us where we saw no benefit. That, after all, is Foot’s point in ‘Euthanasia’ (Foot, 1977). If I am absolutely convinced that you would be better off dead, charity or benevolence bids me kill you, regardless of your firmly expressed wish to be left alive. It is only justice – the fact that your life is yours in the meum-suum sense and thereby that you have the right to decide whether you live or die, not me – that stays my hand.) So my right to life, my right to determine what happens to my body, my right to my child, all these rights are as ‘artificial’, in Hume’s sense, as my right to my car. But Hume’s sense is not, I want to argue, a sense that excludes the common intuition that my right to life, at least, is a ‘natural’ or human right whereas my right to my car is not. To do this I have to backtrack.

14.4 Hume on natural Law and natural rights Earlier, considering the question ‘What turns cloperty into property?’ – or, as we now might say, the natural ‘mine’ into the meum ‘mine’ – I gave Hume’s answer as ‘The laws of society’; and noted that ‘perhaps too quickly’ he adds ‘that is, the laws of justice’ (T, 3.2.2.12/491). Given that, in illustration of the fact that the natural ‘mine’ and the meum ‘mine’ have clearly distinguishable uses, I have alluded to societies whose laws were far from being laws of justice, such as those allowing slavery, I can hardly leave this casual move of Hume’s unexamined. Why – how – can he possibly say ‘the laws of society, that is, the laws of justice’? On the strength of this remark, and many others, and his claim that justice is not a natural but an artificial virtue, some people take Hume to be a conventionalist or ‘positivist’ about law (and thereby rights) as Hobbes was (‘No law can be unjust’, Hobbes says). But this is surely a mistake. Hume is perfectly aware of the fact that we have no difficulty in annexing the idea of injustice to that of an actual civil law, and moreover has firm views about what can justify our doing so. He has no objection to the contrast between civil (or actual or conventional) law and so-called natural law understood, that is, as a set of just laws. Hume’s assumption that we are talking, not about any particular legal code – some civil law – but about civil law as it should be, that is as

Rosalind Hursthouse

273

enshrining laws of justice – is perfectly comprehensible when we keep to the forefront of our minds the fact that he is talking about the virtue of justice, that thing we approve of. And everyone knows what it is to approve of some actual law in one’s society and to disapprove of another because, say, it serves the interest of some members of the society, but not all. So his search for the origins of justice is a search for the origins of something to which we give moral approbation, and he already has in place his views about the sort of mechanisms that can generate this rather special thing. I think myself, that in the Treatise, his detailed attempts to describe its aetiology lead him into ascribing a power to indoctrination which, though rightly ascribed, threatens at times to put him back in the conventionalist camp, but however far he strays in that direction, he never fails to return to his fundamental view that justice – real or ‘natural’ justice, not conventional or civil or legal justice – is approved of because it is useful – ‘absolutely requisite to the well being of mankind and existence of society’ as he says in the second Enquiry (EPM, 3.2.159/199). It was, and is, a commonplace in natural jurisprudence that the natural laws, being just, are the criterion against which civil laws are to be judged. The way you work out whether a particular civil law is just is by working out whether or not it coincides with a law in the ideal code of natural law. And, if Hume is right, the way in which you work that out is by working out whether or not the particular civil law is useful – whether it is ‘requisite to the public interest, and to that of every individual’ (T, 3.2.2.20/496). Now what does Hume’s view contrast with? Well, one might say, with at least the following line or lines of thought. Suppose someone were to say ‘Justice is a constant and perpetual will of giving every one his due’ (T, 3.2.6.2/526) or, as we might say nowadays, ‘of giving everyone what one owes to him’ or ‘of giving everyone his rights’. Those suggest that, in order to find out what is just, we must first determine what is due to everyone, what we do owe to each other, what are our rights. So we come up with some supposedly incontestable premises about our rights, say that we have rights to life and liberty, and then infer ‘and so a just law protects/enshrines these rights’. This way of arguing encapsulates the concept of rights that I referred to at the outset as the one Hume is attacking and, in my view, rightly. For, as he remarks of the first (‘Justice is giving every one his due’) and might equally well remark concerning the others, ‘In this definition ’tis suppos’d, that there are such things as right and property [meum and suum] independent of justice and antecedent to it; and that they wou’d have subsisted tho’ men had never dreamt of practising such a virtue’ (T, 3.2.6.2/526–7). For Hume, trying to get to just laws from such premises is a logical howler and has things back to front. If you want to claim that we have a right to life or liberty, you have to argue for that from a claim about just laws, not from it to what the just laws would be or are. You argue for it by arguing that laws

274

Hume on Justice

protecting or enshrining those rights are or would be just laws, and then you have to argue that the laws are or would be just because they are or would be useful, not because they enshrine rights. No wonder Bentham and Mill claimed him as a utilitarian, and no wonder Bentham, taking himself to be following Hume, said that talk about rights was nonsense and talk about natural rights nonsense on stilts. In the absence of talk about useful laws, actual or hypothetical, established in society, both of them are nonsense, just as talk about owing someone $5 is nonsense in the absence of talk about currency. So talk about Humean rights and natural rights does make sense in the context of talk about useful laws, but I think it is worth noting (given the prevalence of the nonsensical concept) that the talk about rights has to change in certain respects – all of which, I might say, I regard as distinct improvements. Here are some examples. The question ‘What are my/your/Maori’s, (real) rights?’ is a way of asking ‘What laws establishing certain rights ought NZ to have?’ and that is a way of asking ‘What laws establishing certain rights would be useful in NZ society?’ The question ‘What are my, or the, natural rights?’ is a way of asking ‘What laws establishing certain rights ought to be part of any society’s legal system?’ and that is a way of asking ‘What laws establishing certain rights are or would be useful to any society?’ The question ‘Who (or what) are the bearers of rights?’ is a way of asking ‘With respect to whom (or what) would useful laws establish rights?’ ‘Do men and women have equal rights?’ is a way of asking ‘Are there any useful laws (for our society or many or any) which would establish different rights for men and women?’ And faced with the question ‘Do you believe we have a right to autonomy?’ anyone who has accepted Humean rights will sensibly say, ‘I haven’t a clue. Give me some examples of laws that you think would establish such a right and I’ll tell you when I have worked out whether or not I think they would be useful.’ And none of these ways of talking is quite what we are accustomed to in discussions of rights. How are rights established in law? Hume illustrates with some glee the ways in which ‘those rules, by which property, right and obligation are determin’d, have in them no marks of a natural origin, but many of artifice and contrivance’ (T, 3.2.6.6/528), describing ‘the variations, which all the rules of property receive from the finer turns and connexions of the imagination, and from the subtilties and abstractions of law-topics and reasonings’ (EPM, 3.2.161/203). In determining ‘what is a man’s property’ (what he has a right to or in) we have ‘recourse to statutes, customs, precedents, analogies and a hundred other circumstances; some of which are constant and inflexible, some variable and arbitrary’ (EPM, 3.2.158/197). This brings out two significant features of the Humean concept of rights. One is that what ‘having a (real) right to or in x’ consists in, may be relative to a particular society at a particular time. Customs and precedents vary

Rosalind Hursthouse

275

between societies with different histories. Some circumstances (for instance the local weather, the concentration of population) vary; given different cultures, what people find ‘analogous’ varies. So different societies may formulate different laws determining different rights and there may be no question as to which society has secured the ‘real’ rights. For if the different laws that different societies come up with are equally useful to their members, then the laws are equally just, and the real rights of the members of each society will be determined by those laws, despite being different. And that does not seem to me quite how we tend to talk nowadays about real (as opposed to ‘merely conventional’) rights. The second feature is that what ‘having a right to or in x’ consists in may be made up of a mass of details – a whole set of useful laws. A Humean concept of natural rights makes sense on the supposition that there are some useful laws – say a law prohibiting murder – which (a) establish a right and (b) are ‘constant and inflexible’; that is, such that they would be useful to any society no matter what its history, culture, circumstances or whatever. But no society (as far as I know) ever has a single short sentence law establishing the right to life. It has, say, laws forbidding different sorts of murder, laws or statutes about killing in self-defence and/or through recklessness or through letting die, and/or about capital punishment, infanticide, suicide, euthanasia, abortion and so on – a whole set of (perhaps) useful laws. When we talk about rights nowadays we do not, I think, tend to talk about them as things you can have to some degree, especially where the natural rights are concerned. I have the right to life or I don’t; a foetus has a right to life or it doesn’t; there’s not much room for saying, well I have more of the right to life than a foetus does. But when we think of ‘having a right to x’ as supervening on a fair number of useful laws, (and remember that, in order to be useful, laws have to be consistent with other laws so that they are all applicable, and that they have to be applicable in the ordinary life of human beings) the idea does not seem so strange. As Hume says, ‘Half rights and obligations … seem so natural in common life’ (T, 3.2.6.8/531). (Notwithstanding that quote, I must confess that I am going against Hume on this point. He maintains that ‘abstract reasoning, and the general maxims of philosophy and law’ do establish that ‘property, and right, and obligation admit not of degrees’ and that it is in our ‘common and negligent way of thinking’ that we think otherwise; this is yet another of his arguments which purports to show that justice is artificial not natural. But the argument looks to me like a confused attempt to argue against Grotius’ distinction between perfect and imperfect rights.) Finally – and briefly – in what sense, according to Hume, is justice useful? Useful to what end? It is clear in Hume that the end is human well-being. What is nothing like as clear is where Hume stands on human well-being – with, at one extreme, the hedonistic Bentham or, at the other extreme with Aristotle, the Epicureans and Mill in On Liberty – and consequently it

276

Hume on Justice

is unclear what he means by ‘useful’. I have argued elsewhere (Hursthouse, 1999) that he is not consistent on this, but that he tends strongly towards the latter, eudaimonistic account; here, I will close with just one point in its favour. Hume claims in the Enquiry, ‘Examine the writers on the laws of nature [he means on ‘natural law’]; and you will always find, that, whatever principles they set out with, they are sure to terminate here at last, and to assign, as the ultimate reason for every rule which they establish, the convenience and necessities of mankind’ (EPM, 3.2.156/195). From the little I know of the natural law theorists, that seems to me true. But the human well-being that they all thought natural law served or preserved – as Hume must have known – was, of course, the sort of human well-being that the God of Christianity intended when He laid down natural law; not the sort that consists in mere animal pleasure or the satisfaction of desires, but the eudaimonistic sort that is found only in virtue. So we must surely take it that this moralised conception of ‘the convenience and necessities of mankind’ is what Hume is talking about here. And, insofar as that is true, we have the final Humean nail in the coffin of the modern concept of rights, insofar as that is predicated on the Rawlsian assumption that ethics is not prior to politics and that we can limn the laws of a just society without committing ourselves to any substantive ethical views on the good for human beings. This is far from being the Humean picture of rights.

Bibliography K. Abramson (1999) ‘Correcting Our Sentiments About Hume’s Moral Point of View’, The Southern Journal of Philosophy, Vol. 37, No. 3, 333–61. G. E. M. Anscombe (1957) Intention (Oxford, Blackwell). G. E. M. Anscombe (1981) Ethics, Religion and Politics (Oxford, Blackwell). P. S. Árdal (1966) Passion and Value in Hume’s Treatise (Edinburgh, University of Edinburgh Press). Aristotle (1962) Nichomachean Ethics, Martin Ostwald (trans.) (Indianapolis, BobbsMerrill Co.). J. Austen (2004) [1818] Persuasion (Oxford, Oxford University Press). J. L. Austin (1962) How to Do Things with Words (Oxford, Oxford University Press). J. L. Austin (1971) ‘Performative-Constative’, in J. Searle (ed.) The Philosophy of Language (Oxford, Oxford University Press), pp. 13–22. J. L. Austin (1990) [1970] ‘Performative Utterances’, in A. P. Martinich (ed.) The Philosophy of Language (New York, Oxford University Press), pp. 105–14. A. J. Ayer (1971) [1936] Language, Truth and Logic (New York, Penguin Books). A. J. Ayer (1980) Hume (Oxford, Oxford University Press). A. J. Ayer (1992) Hume, in J. Dunn, J. O. Urmson and A. J. Ayer The British Empiricists: Locke, Berkeley, Hume (Oxford, Oxford University Press), pp. 179–279. A. Baier (1988) ‘Hume’s Account of Social Artifice: Its Origins and Originality’, Ethics, Vol. 98, 757–88. A. Baier (1991) A Progress of Sentiments (Cambridge, Harvard University Press). A. Baier (2008) ‘Acting in Character’, in C. Sandis (ed.) New Essays on the Explanation of Action (Basingstoke, Palgrave Macmillan). A. Baier (2008) ‘Why Hume asked us not to read the Treatise’, in Death and Character (Cambridge, Harvard University Press). A. Baier (2010) ‘Hume’s own deduction of an Ought from an Is’, in C. Pigden (ed.) Hume, Is and Ought: New Essays (Basingstoke, Palgrave Macmillan). J. Balfour (1989) [1753] A Delineation of the Nature and Obligation of Morality (Edinburgh, Hamilton, Balfour and Neill; Bristol, Thoemmes Antiquarian Books). J. Balguy (1733) The Foundations of Moral Goodness, reprinted in Raphael (ed.) (1991) The British Moralists, Vol. 1 (Indianapolis, Hackett). S. Beardman (2007) ‘The Special Status of Instrumental Reasons’, Philosophical Studies, Vol. 134, 255–87. J. Bennett (2001) Learning From Six Philosophers: Descartes, Spinoza, Leibniz, Locke, Berkeley, Hume, Vol. 2 (Oxford, Oxford University Press). J. Bentham (1988) A Fragment on Government (Cambridge, Cambridge University Press). J. Bigelow, S. Dodds and R. Pargetter (1988) ‘Against the Will’, Pacific Philosophical Quarterly, Vol. 69, 307–24. J. Bigelow, S. Dodds and R. Pargetter (1990) ‘Temptation and the Will’, American Philosophical Quarterly, Vol. 27, 39–49. S. Blackburn (1984) Spreading the Word (Oxford, Clarendon Press). S. Blackburn (1993) Essays in Quasi-Realism (Oxford, Oxford University Press). S. Blackburn (1996) A Dictionary of Philosophy, 2nd edn (Oxford, Oxford University Press). 277

278

Bibliography

S. Blackburn (1998a) ‘Projectivism’, in E. Craig (ed.) (1998) Routledge Encyclopedia of Philosophy (London, Routledge), pp. 737–40. S. Blackburn (1998b) Ruling Passions (Oxford, Clarendon Press). S. Botros (2006) Hume, Reason, and Morality: A Legacy of Contradiction (London, Routledge). M. Bratman (1987) Intention, Plans and Practical Reason (Cambridge, Harvard University Press). D. Braybrooke (2002) Natural Law Modernized (Toronto, University of Toronto Press). F. Brentano (1969) The Origin of our Knowledge of Right and Wrong, in R. Chisholm (ed.), R. Chisholm and E. Schneewind (trans.) (London, Routledge). J. Bricke (1996) Mind and Morality: An Examination of Hume’s Moral Psychology (Oxford, Clarendon Press). D. O. Brink (1989) Moral Realism and the Foundations of Ethics (Cambridge, Cambridge University Press). D. O. Brink (1997) ‘Moral Motivation’, Ethics, Vol. 108, 4–32. C. D. Broad (1930) Five Types of Ethical Theory (New York, Harcourt Brace). J. Broome (1997) ‘Reason and Motivation’, Proceedings of the Aristotelian Society, Supplementary Vol. 71: 131–46. J. Broome (2000) ‘Normative Requirements’, in J. Dancy (ed.) Normativity (Oxford, Blackwell), pp. 78–99. J. Broome (2001) ‘Normative Practical Reasoning’, Proceedings of the Aristotelian Society, Supplementary Vol. 75, 175–93. J. Broome (2002) ‘Practical Reasoning’, in J. Bermùdez and A. Millar (eds) Reason and Nature: Essays in the Theory of Rationality (Oxford, Oxford University Press), pp. 85–111. J. Broome (2004)‘Reasons’, in J. Wallace, M. Smith, S. Scheffler and P. Pettit (eds) Reason and Value: Themes from the Moral Philosophy of Joseph Raz (Oxford: Oxford University Press), pp. 28–55. J. Broome (2009, forthcoming) ‘The Unity of Reasoning?’ in S. Robertson, J. Skorupski, and J. Timmerman (eds) Spheres of Reason (Oxford, Oxford University Press). C. Brown (1988) ‘Is Hume an Internalist?’, Journal of the History of Philosophy, Vol. 2, 69–87. R. Carnap (1935) Philosophy and Logical Syntax (London, Kegan Paul, Trench, Trubner & Co. Ltd.). R. Chisholm (1986) Brentano and Intrinsic Value (Cambridge, Cambridge University Press). G. A. Cohen (1996) ‘Reason, Humanity and the Moral Law’, in C. Korsgaard (ed.) The Sources of Normativity (Cambridge, Cambridge University Press), pp. 167–88. R. Cohon (1997a) ‘Is Hume a Noncognitivist in the Motivation Argument?’, Philosophical Studies, Vol. 85, 251–66. R. Cohon (1997b) ‘Hume’s Difficulty with the Virtue of Honesty’, Hume Studies, Vol. 23, 91–112. R. Cohon (1997c) ‘The Common Point of View in Hume’s Ethics’, Philosophy and Phenomenological Research, Vol. 57, 827–50. D. Coleman (1992) ‘Hume’s Internalism’, Hume Studies, Vol. 18, 331–47. A. W. Collins (1997) ‘The Psychological Reality of Reasons’, Ratio, Vol. 10, Iss. 2, 108–23. D. Copp (2001) ‘Realist-Expressivism’, Social Philosophy and Policy, Vol. 18, 1–43. G. Cullity and B. Gaut (eds) (1997) Ethics and Practical Reason (Oxford, Oxford University Press).

Bibliography

279

J. Dancy (1993) Moral Reasons (Oxford, Blackwell). J. Dancy (1995) ‘Why there is Really no such thing as the Theory of Motivation’, Proceedings of the Aristotelian Society, Vol. 95, 1–18. J. Dancy (2000) Practical Reality (Oxford, Oxford University Press). J. Dancy (2004) Ethics Without Principles, (Oxford, Oxford University Press). S. L. Darwall (1977) ‘Two Kinds of Respect’, Ethics, Vol. 88, 36–49. S. L. Darwall (1983) Impartial Reason (Ithaca, NY, Cornell University Press). S. L. Darwall (1995) The British Moralists and the Internal ‘Ought’ (Cambridge, Cambridge University Press). D. Demetriou and G. Oddie (2007) ‘Review of Moral Fictionalism, by Mark Eli Kalderon’, Mind, Vol. 116, Iss. 462, 439–46. S. Dimock (1992) ‘Calling all Knaves: Hume on Moral Motivation’, Eidos: The Canadian Graduate Journal of Philosophy, Vol. 10, 179–97. J. Dreier (1997) ‘Humean Doubts about the Practical Justification of Morality’, in G. Cullity and B. Gaut (eds) Ethics and Practical Reason (Oxford, Oxford University Press), pp. 81–99. J. Dreier (2005) ‘Meta-Ethics and the Problem of Creeping Minimalism’, Philosophical Perspectives, Vol. 18, 23–44. J. Driver (2001) Uneasy Virtue (Cambridge, Cambridge University Press). J. Dunn, J. O. Urmson and A. J. Ayer (1992) The British Empiricists: Locke, Berkeley, Hume (Oxford, Oxford University Press). J. Fieser (ed.) (2005) Early Responses to Hume, 2nd edn, Vol. 1, Early Responses to Hume’s Moral, Literary and Political Writings (Bristol, Thoemmmes Continuum). S. Finlay (2004) ‘The Conversational Practicality of Value Judgement’, The Journal of Ethics, Vol. 8, 205–23. S. Finlay (2006) ‘The Reasons that Matter’, Australasian Journal of Philosophy, Vol. 84, 1–20. S. Finlay (2009) ‘Oughts and Ends’, Philosophical Studies,Vol. 143, 35–40. R. Firth (1952) ‘Ethical Absolutism and the Ideal Observer’, Philosophy and Phenomenological Research, Vol. 12, 336–41. A. Flew (1963) ‘On the Interpretation of Hume’, Philosophy, Vol. 38, 178–81. A. Flew (1986) David Hume; Philosopher of Moral Science (Oxford, Blackwell). P. Foot (1972a) ‘Morality as a System of Hypothetical Imperatives’, Philosophical Review, Vol. 81, 305–16. P. Foot (1972b) ‘Reasons for Action and Desires’, Proceedings of the Aristotelian Society, Supplementary Vol. 46, 203–10. P. Foot (1977) ‘Euthanasia’, Philosophy and Public Affairs 6, 85–112, reprinted in Foot (2002) Virtues and Vices and Other Essays in Moral Philosophy (Oxford, Oxford University Press), pp. 33–61. P. Foot (1978a) ‘Hume on Moral Judgment’, in Virtues and Vices and Other Essays in Moral Philosophy (Berkeley, University of California Press), pp. 74–80. P. Foot (1978b) Virtues and Vices and Other Essays in Moral Philosophy (Berkeley, University of California Press). P. Foot (2002) Moral Dilemmas and Other Essays in Moral Philosophy (Oxford, Clarendon). H. Frankfurt (1971) ‘Freedom of the Will and the Concept of a Person’, reprinted in Gary Watson (ed.) (1982) Free Will (Oxford, Oxford University Press). G. Frege [1892] (1997) ‘On Sense and Reference’, in M. Beaney (ed.) The Frege Reader (Oxford, Basil Blackwell), pp. 151–71. G. Frege [1897] (1997) ‘Logic’, in M. Beaney (ed.) The Frege Reader (Oxford, Basil Blackwell), pp. 227–50.

280

Bibliography

R. Frey and C. Morris (eds.) (1993) Value Welfare and Morality (Cambridge, Cambridge University Press). D. Garber and S. Nadler (eds.) (2005) Oxford Studies in Early Modern Philosophy, Vol. 2 (Oxford, Oxford University Press). D. Gauthier (1992) ‘Artificial Virtues and the Sensible Knave’, Hume Studies, Vol. 18, 401–27. M. Gill (2000) ‘Hume’s Progressive View of Human Nature’, Hume Studies, Vol. 26, 87–108. J. D. Greene and J. Haidt (2002) ‘How (and Where) does Moral Judgment Work?’, Trends in Cognitive Sciences, Vol. 6, 517–23. J. D. Greene, L. E. Nystrom, A. D. Engell, J. M. Darley and J. D. Cohen (2004) ‘The Neural Bases of Cognitive Conflict and Control in Moral Judgment’, Neuron, Vol. 44, 389–400. J. D. Greene, R. B. Sommerville, L. E. Nystrom, J. M. Darley, and J. D. Cohen (2001) ‘An fMRI Investigation of Emotional Engagement in Moral Judgment’, Science, Vol. 293, 2105–8. P. Grice (1989) Studies in the Way of Words (Cambridge, Harvard University Press). J. Haidt (2001) ‘The Emotional Dog and its Rational Tail: A Social Intuitionist Approach to Moral Judgment’, Psychological Review, Vol. 108, 814–34. J. Hampton (1995) ‘Does Hume Have an Instrumental Conception of Practical Reason?’, Hume Studies, Vol. 21, No. 1, 57–74. J. Hampton (1998) The Authority of Reason (Cambridge, Cambridge University Press). R. M. Hare (1952) The Language of Morals (Oxford, Oxford University Press). R. M. Hare (1999) Objective Prescriptions and Other Essays (Oxford, Oxford University Press). G. Harman and J. J. Thomson (1996) Moral Relativism and Moral Objectivity (Oxford, Blackwell). J. Harrison (1976) Hume’s Moral Epistemology (Oxford, Clarendon Press). R. Harrison (ed.) (1980) Rational Action: Studies in Philosophy and Social Science (Cambridge, Cambridge University Press). J. Harsanyi (1985) ‘Rule Utilitarianism, Equality, and Justice’, in E. F. Paul (ed.) Ethics and Economics (Oxford, Blackwell), pp. 115–27. R. Heinaman (ed.) (1995) Aristotle and Moral Realism (London, UCL Press). Herodotus (2003) The Histories, translated by A. De Sélincourt, revised by J. Marincola (New York, Penguin Books). T. E. Hill Jr. (1992) ‘The Hypothetical Imperative’, in Dignity and Practical Reason in Kant’s Moral Theory (Ithaca, Cornell University Press). T. Hobbes (1994) Leviathan, E. Curley (ed.) (Indianapolis, Hackett). D. Hubin (1996) ‘Hypothetical Motivation’, Noûs, Vol. 30, 31–54. D. Hubin (2001) ‘The Groundless Normativity of Instrumental Rationality’, The Journal of Philosophy, Vol. 98, No. 9, 445–68. W. D. Hudson (ed.) (1969) The Is-Ought Question: A Collection of Papers on the Central Problem in Moral Philosophy (London, Macmillan). L. Humberstone (1992) ‘Direction of Fit’, Mind, Vol. 101, 59–68. D. Hume [1757] (1898) ‘A Dissertation on the Passions’, in T. H. Green and T. H. Grose (eds) The Philosophical Works of David Hume, Vol. II (London, Longmans, Green, and Co.), pp. 139–66. D. Hume (1978) Enquiries Concerning Human understanding and Concerning the Principles of Morals, edited by L. A. Selby-Bigge and P. H. Nidditch , 3rd edn (Oxford, Oxford University Press).

Bibliography

281

D. Hume [1739–40] (1978) A Treatise of Human Nature, edited by L. A. Selby-Bigge and P. H. Nidditch, 2nd edn (Oxford, Clarendon Press). D. Hume (1983) The History of England, W. Todd (ed.), 6 volumes (Indianapolis, Liberty Fund). D. Hume [1762] (1986) ‘A New Letter to Hugh Blair from July 1762’, Mind, Vol. 95, 411–16. D. Hume (1987) Essays, Moral, Political, and Literary, D. Miller (ed.) (Indianapolis, Liberty Fund). D. Hume [1742] (1996) ‘The Sceptic’, in S. Copley and A. Edgar (eds) David Hume: Selected Essays (Oxford, Oxford University Press), pp. 95–113. D. Hume [1757] (1996) ‘Of the Standard of Taste’, in S. Copley and A. Edgar (eds) David Hume: Selected Essays (Oxford, Oxford University Press), pp. 133–154. D. Hume [1741] (1998) ‘Of the Delicacy of Taste and Passion’, in S. Copley and A. Edgar (eds) David Hume: Selected Essays (Oxford, Oxford University Press), pp. 10–13. D. Hume [1751] (1998) An Enquiry Concerning the Principles of Morals, T. L. Beauchamp (ed.) (Oxford, Oxford University Press). D. Hume [1760] (1998) ‘Of the Standard of Taste’, in S. Copley and A. Edgar (eds) David Hume Selected Essays (Oxford, Oxford University Press), pp. 133–54. D. Hume (2000) An Enquiry Concerning Human Understanding, T. L. Beauchamp (ed.) (Oxford, Oxford University Press). D. Hume [1739–40] (2000) A Treatise of Human Nature, D. F. Norton and M. J. Norton (eds) (Oxford, Oxford University Press). G. Hunter (1962) ‘Hume on Is and Ought’, reprinted in Hudson (1969) The Is-Ought Question: A Collection of Papers on the Central Problem in Moral Philosophy (London, Macmillan), pp. 59–63. R. Hursthouse (1999) ‘Virtue Ethics and Human Nature’, Hume Studies, Vol. 25, 67–82. F. Hutcheson (2002) An Essay on the Nature and Conduct of the Passions and Affections with Illustrations on the Moral Sense, A. Garrett (ed.) (Indianapolis, Liberty Fund). F. Hutcheson (2004) An Inquiry into the Originals of Our Ideas of Beauty and Virtue, W. Leidhold (ed.) (Indianapolis, Liberty Fund). T. Jefferson (1999) Political Writings, J. Appleby and T. Ball (eds) (Cambridge, Cambridge University Press). M. Johnston, (2001) ‘The Authority of Affect’, Philosophy and Phenomenological Research, Vol. 63, No. 1, 181–214. R. Joyce (2001) The Myth of Morality (Cambridge, Cambridge University Press). R. Joyce (2002) ‘Expressivism and Motivation Internalism’, Analysis, Vol. 62, 336–44. R. Joyce (2006) The Evolution of Morality (Cambridge, MA, MIT Press). R. Joyce (2007a) ‘Moral Anti-Realism’, The Stanford Encyclopedia of Philosophy, http:// plato.stanford.edu/entries/moral-anti-realism/. R. Joyce (2007b) ‘Morality, Schmorality’, in P. Bloomfield (ed.) Morality and Self-Interest (Oxford, Oxford University Press), pp. 51–75. R. Joyce (2008) ‘What Neuroscience can (and cannot) Contribute to Metaethics’, in W. Sinnott-Armstrong (ed.) Moral Psychology, Volume 3: The Neuroscience of Morality: Emotion, Brain Disorders, and Development (Cambridge, MA, MIT Press), pp. 371–94. R. Joyce (2009) ‘Is Moral Projectivism Empirically Tractable?’ Ethical Theory and Moral Practice, Vol.12, 53–75. M. Kalderon (2005) Moral Fictionalism (Oxford, Oxford University Press). I. Kant (1911) Grundlegung zur Metaphysik der Sitten, P. Menzer (ed.) (Berlin, Koenigliche Preussische Akademie der Wissenschaften).

282

Bibliography

I. Kant (1994) Groundwork of the Metaphysics of Morals, translated and edited by J. Ellington, in Kant, Ethical Philosophy: The Complete Texts of Grounding for the Metaphysics of Morals, and Metaphysical Principles of Virtue, 2nd edn (Indianapolis, Hackett Publishing Company). I. Kant (1997) Critique of Practical Reason, M. Gregory (trans. and ed.) (Cambridge, Cambridge University Press). G. Kavka (1983) ‘The Toxin Puzzle’, Analysis, Vol. 43, 33–6. N. Kemp Smith (1941) The Philosophy of David Hume (London, Macmillan). N. Kolodny (2005) ‘Why Be Rational?’, Mind, Vol. 114, 509–63. C. Korsgaard (1986) ‘Skepticism about Practical Reason’, Journal of Philosophy, Vol. 83, No, 1, 5–25. C. Korsgaard (1996a) ‘Kant’s Analysis of Obligation: The Argument of Groundwork I’, in C. Korsgaard (ed.) Creating the Kingdom of Ends (Cambridge, Cambridge University Press), pp. 43–76. C. Korsgaard (1996b) The Sources of Normativity (Cambridge, Cambridge University Press). C. Korsgaard (1996c) ‘Two Distinctions in Goodness’, in C. Korsgaard (ed.) Creating the Kingdom of Ends, pp. 249–74. C. Korsgaard (1996d) Creating the Kingdom of Ends (Cambridge, Cambridge University Press). C. Korsgaard (1997a) ‘The Normativity of Instrumental Reason’, in G. Cullity and B. Gaut (eds) (1997) Ethics and Practical Reason (Oxford, Oxford University Press), pp. 215–54. C. Korsgaard (1997b) ‘Skepticism about Practical Reason’, in S. Darwall, A. Gibbard and P. Railton (eds) (1997) Moral Discourse & Practice (New York, Oxford University Press). C. Korsgaard (1999) ‘Self-Constitution in the Ethics of Plato and Kant’, The Journal of Ethics, Vol. 3, No. 1, 1–29. K. Korsgaard (2009) Self-Constitution (Oxford, Oxford University Press). J. Lenman (1996) ‘Belief, Desire and Motivation: An Essay on Quasi-Hydraulics’, American Philosophical Quarterly, Vol. 33, No. 3, 291–301. D. K. Lewis (1989) ‘Dispositional Theories of Value’, Proceedings of the Aristotelian Society, Supplementary Volume 63, 113–37. X. Liu (2003) Mencius, Hume and the Foundations of Ethics (Ashgate, Brookfield). Y. S. Lo (2001) ‘A Humean Argument for the Land Ethic?’, Environmental Values, Vol. 10, 523–39. J. Locke (1970) Two Treatises of Government, Peter Laslett (ed.), 2nd edn (Cambridge, Cambridge University Press). J. MacFarlane (2005) ‘Making Sense of Relative Truth’, Proceedings of the Aristotelian Society, Vol. 105, 321–39. J. L. Mackie (1977) Ethics: Inventing Right and Wrong (New York, Penguin Books). J. L. Mackie (1980) Hume’s Moral Theory (New York, Routledge). N. Malcolm (1958) Ludwig Wittgenstein: A Memoir (Oxford, Oxford University Press). E. Mazza (2005) ‘Cannibals in “A Dialogue,” (In search for a standard of morals)’ in E. Mazza and E. Ronchetti (eds) Instruction and Amusement: Le Ragioni dell’illuminismo Britannico (Padova, Il Poligrafo), pp. 45–66. E. Mazza and E. Ronchetti (eds) (2007) New Essays on David Hume (Milano, Franco Angeli). J. McDowell (1978) ‘Are Moral Requirements Hypothetical Imperatives?’, Proceedings of the Aristotelian Society, Supplementary Volume 52, 13–29. J. McDowell (1979) ‘Virtue and Reason’, Monist, Vol. 62, 331–50.

Bibliography

283

J. L. McIntyre (2000) Hume Studies, Vol. 26, No. 1, 77–86. D. McNaughton (1988) Moral Vision (Oxford, Blackwell). A. Meinong (1972) On Emotional Presentation, M. Schubert (trans.) (Evanston, IL, Northwestern University Press). A. R. Mele (1992) The Springs of Action (Oxford, Oxford University Press). E. Millgram (1995) ‘Was Hume a Humean?’, Hume Studies, Vol. 21, 75–93. J. Moll, R. de Oliveira-Souza, P. J. Eslinger, I. E. Bramati, J. Mourão-Miranda, P. A. Andreiuolo and L. Pessoa (2002) ‘The Neural Correlates of Moral Sensitivity: A Functional Magnetic Resonance Imaging Investigation of Basic Moral Emotions’, Journal of Neuroscience, Vol. 22, 2730–6. R. Monk (1996) Bertrand Russell: the Spirit of Solitude (London, Jonathan Cape). G. E. Moore (1942) ‘A Reply to my Critics’, in P. A. Schilpp (ed.) The Philosophy of G. E. Moore (Evanston, Ill., Northwestern University), pp. 535–677. K. Mulligan (1998) ‘From Appropriate Emotions to Values’, The Monist, Vol. 81, No. 1, 161–88. T. Nagel (1970) The Possibility of Altruism (Princeton, Princeton University Press). T. Nagel (1997) The Last Word (Oxford, Oxford University Press). D. Nolan, G. Restall and C. West (2005) ‘Moral Fictionalism Versus the Rest’, Australasian Journal of Philosophy, Vol. 83, 307–30. R. Nozick (1993) The Nature of Rationality (Princeton, Princeton University Press). G. Oddie (2001) ‘Scrumptious Functions’, Grazer Philosophische Studien, Vol. 62, 137–56. G. Oddie (2005) Value, Reality and Desire (Oxford, Oxford University Press). G. Oddie and D. Demetriou (2007) ‘The Fictionalist’s Attitude Problem’, Ethical Theory and Moral Practice, Vol. 10, 485–98. D. Owen (1999) Hume’s Reason (Oxford, Oxford University Press). D. Owen (2003) ‘Locke and Hume on Belief, Judgment and Assent’, Topoi, Vol. 22, No. 1, 15–28. D. Parfit (1997) ‘Reasons and Motivation’, Proceedings of the Aristotelian Society, Supplementary Volume 71, 99–130. C. Pigden (1989) ‘Logic and the Autonomy of Ethics’, Australasian Journal of Philosophy, Vol. 67, No. 2, 127–51. C. Pigden (2007) ‘Hume, Motivation and “the Moral Problem”’, in E. Mazza and E. Ronchetti (eds) New Essays on David Hume (Milano, Franco Angeli), pp. 199–221. C. Pigden (2009a) ‘A Niggle at Nagel’, in C. Sandis (ed.) (2009) New Essays on the Explanation of Action (Basingstoke, Palgrave Macmillan). C. Pigden (ed.) (2010) Hume, ‘Is’ and ‘Ought’: New Essays (Basingstoke, Palgrave Macmillan). Plato (1952) Phaedo, R. Hackforth (trans.) (Cambridge, Cambridge University Press). Plato (1977) Phaedo, D. Gallop (trans.) (Oxford, Oxford University Press). H. Price (1988) Facts and the Function of Truth (Oxford, Blackwell). Prinz, Jesse (2007), The Emotional Construction of Morals, (Oxford University Press). W. Quinn (1993) ‘Putting Rationality in its Place’, in R. Frey and C. Morris (eds) Value Welfare and Morality (Cambridge, Cambridge University Press). E. Radcliffe (1996) ‘How Does the Humean Sense of Duty Motivate?’, Journal of the History of Philosophy, Vol. 34, 383–407. E. S. Radcliffe (ed.) (2008) A Companion to Hume (Oxford, Blackwell). P. Railton (1997) ‘On the Hypothetical and Non-Hypothetical in Reasoning about Belief and Action’, in G. Cullity and B. Gaut (eds) Ethics and Practical Reason (Oxford, Oxford University Press), pp. 53–79.

284

Bibliography

F. P. Ramsey (1931) The Foundations of Mathematics and Other Logical Essays, R. B. Braithwaite (ed.) (Patterson, Littlefield Adams & Co.). D. D. Raphael (ed.) (1991) The British Moralists, 2 volumes (Indianapolis, Hackett). J. Raz (1986) The Morality of Freedom (Oxford, Clarendon Press). J. Raz (2005) ‘The Myth of Instrumental Rationality’, Journal of Ethics and Social Philosophy, Vol. 1, No. 1, http://www.jesp.org/PDF/6863_Raz-vol-1-no-1-rev.pdf. T. Reid (1843) Essays on the Active Powers of the Human Mind (London, Thomas Tegg/ Whitefish, Kessinger). T. Reid (1977) ‘Whether Justice be a Natural or an Artificial Virtues’, Chapter 5, Essay 5, Essays on the Active Powers of the Human Mind (New York, Garland Publishing). M. Ridge (2006) ‘Sincerity and Expressivism’, Philosophical Studies, Vol. 131, 487–510. W. D. Ross (1930) The Right and the Good (Oxford, Oxford University Press). M. Ruse (1986) Taking Darwin Seriously (Oxford, Blackwell). B. Russell (1999) Russell on Ethics, C. Pigden (ed.) (London, Routledge). B. Russell (2000) Autobiography (London, Routledge). C. Sandis (2008a) ‘Gods and Mental States: The Causation of Action in Ancient Tragedy and Modern Philosophy of Mind’, in C. Sandis (ed.) New Essays on the Explanation of Action (Basingstoke, Palgrave Macmillan). C. Sandis (ed.) (2008b) New Essays on the Explanation of Action (Basingstoke, Palgrave Macmillan). C. Sandis (2010a) ‘Hume on the meaning of “necessity”’, in Keith Allen and Tom Stoneham (eds) Causation: 1500–1800 (London: Routledge). Sandis C, (2010b), ‘Action, Reason, and the Passions’, in Alan Bailey and Dan O’Brien (eds) The Continuum Companion to Hume (London, Continuum) G. X. Santas (1997) ‘Does Aristotle Have a Virtue Ethics’, in D. Statman (ed.) Virtue Ethics: A Critical Reader (Edinburgh, Edinburgh University Press). G. Sayre-McCord (1986) ‘The Many Moral Realisms’, Southern Journal of Philosophy Supplementary, Spindel Conference on Moral Realism, 1–22, reprinted in G. SayreMcCord (ed.) (1988), pp. 1–23. G. Sayre-McCord (ed.) (1988) Essays on Moral Realism (Ithaca, Cornell University Press). G. Sayre-McCord (1994) ‘On Why Hume’s “General Point of View” Isn’t Ideal – and Shouldn’t Be’, Social Philosophy and Policy Foundation, Vol. 11, 202–28. G. Sayre-McCord (1996) ‘Hume and the Bauhaus Theory of Ethics’, in P. A. French, T. E. Uehling Jr. and H. K. Wettstein (eds) Midwest Studies in Philosophy 20, Moral Concepts (Notre Dame, University of Notre Dame Press), pp. 280–98. G. Sayre-McCord (1997) ‘Hume’s Representation Argument Against Rationalism’, Manuscrito, Vol. 20, 77–94. T. Scanlon (1998) What We Owe to Each Other (Cambridge, Harvard University Press). T. Scanlon (2003) ‘Metaphysics and Morals’, Proceedings and Addresses of the American Philosophical Association, Vol. 77, 7–22. M. Scheler (1973) Formalism in Ethics and Non-formal Ethics of Value, M. Frings and R. Funk (trans.) (Evanston, IL, Northwestern University Press). M. Schroeder (2004) ‘The Scope of Instrumental Reason’, Philosophical Perspectives, Vol. 18, 337–64. M. Schroeder (2007) Slaves of the Passions (Oxford, Oxford University Press). J. R. Searle (1969) Speech Acts (Cambridge, Cambridge University Press). K. Setiya (2007) ‘Cognitivism about Instrumental Reason’, Ethics, Vol. 117, 649–73.

Bibliography

285

R. Shafer-Landau (2005) Moral Realism (Oxford, Oxford University Press). H. Sidgwick (1902) Outlines of the History of Ethics, for English Readers, 5th edn (London, Macmillan and Co.). H. Sidgwick (1907) The Methods of Ethics (Chicago, Chicago University Press). J. Skidmore (2002) ‘Skepticism about Practical Reason’, Philosophical Studies, Vol. 109, 121–41. A. Smith (1976) The Theory of Moral Sentiments, D. D. Raphael and A. L. MacFie (eds) (Oxford, Oxford University Press). M. Smith (1987) ‘The Humean Theory of Motivation’, Mind, Vol. 96, No. 381, 36–61. M. Smith (1988) ‘On Humeans, Anti-Humeans and Motivation: A Reply to Pettit’, Mind, Vol. 97, 589–95. M. Smith (1991) ‘Realism’, in P. Singer (ed.) A Companion to Ethics (Oxford, Blackwell), pp. 399–410. M. Smith (1994) The Moral Problem (Oxford, Blackwell). F. Snare (1975) ‘The Argument from Motivation’, Mind, Vol. 84, 1–9. F. Snare (1991) Morals, Motivation, and Convention (Cambridge, Cambridge University Press). C. L. Stevenson (1937) ‘The Emotive Meaning of Ethical Terms’, Mind, Vol. 46, 14–31. C. L. Stevenson (1963) Facts and Values (New Haven, CT, Yale University Press). B. Stroud (1977) Hume (London, Routledge). N. Sturgeon (2008) ‘Hume’s Metaethics: Is Hume a Noncognitivist’ in E. Radcliffe (ed.) A Companion to Hume (Oxford, Blackwell) pp. 513–528. S. Svavarsdóttir (2006) ‘How do Moral Judgments Motivate?’ in J. Dreier (ed.) Contemporary Debates in Moral Theory (Oxford, Blackwell), pp. 163–81. C. Swanton (2003) Virtue Ethics: A Pluralistic View (Oxford, Oxford University Press). C. Swanton (2007) ‘Can Hume Be Read as a Virtue Ethicist?’, Hume Studies, Vol. 33, No. 1, 91–113. C. Taylor (1989) Sources of the Self (Cambridge, Cambridge University Press). J. Taylor (1998) ‘Justice and the Foundations of Social Morality in Hume’s Treatise’, Hume Studies, Vol. 24, No. 1, 5–30. M. Timmons (1999) Morality Without Foundations (Oxford, Oxford University Press). J. R. R. Tolkien (1995) The Lord of the Rings (London, Harper/Collins). J. O. Urmson (1975) ‘A Defence of Intuitionism’, Proceedings of the Aristotelian Society, Vol. 65, 111–9. G. H. von Wright (1978) ‘On So-Called Practical Inference’, in J. Raz (ed.) Practical Reasoning (Oxford, Oxford University Press). J. Waldron (ed.) (1987) Nonsense on Stilts: Bentham, Burke, Marx and the Rights of Man (London, Methuen). J. Waldron (2002) God, Locke and Equality (Cambridge, Cambridge University Press). R. J. Wallace (1998) ‘Moral Motivation’, in E. Craig (ed.) The Routledge Encyclopedia of Philosophy (London, Routledge), pp. 22–8. R. J. Wallace (2003) ‘Practical Reason’, Stanford Encyclopedia of Philosophy, http://plato. stanford.edu/entries/practical-reason/. G. Watson (ed.) (1982) Free Will (Oxford, Oxford University Press). D. Wiggins (1987) ‘A Sensible Subjectivism’, in D. Wiggins (ed.) Needs, Values, Truth (Oxford, Basil Blackwell), pp. 185–214. D. Wiggins (1991) ‘A Sensible Subjectivism?’, in D. Wiggins (ed.) Needs, Values, Truth, 2nd edn (Oxford, Basil Blackwell), pp.185–214. D. Wiggins (1998) Needs, Values, Truth: Essays in the Philosophy of Value, 3rd edn (Oxford, Oxford University Press).

286

Bibliography

B. A. O. Williams (1980) ‘Internal and External Reasons’, in R. Harrison (ed.) Rational Action (Cambridge, Cambridge University Press). B. A. O. Williams (1981) Moral Luck (Cambridge, Cambridge University Press). B. A. O. Williams (1985) Ethics and the Limits of Philosophy (London, Fontana). B. A. O. Williams (1995) ‘Acting as the Virtuous Person Acts’, in R. Heinaman (ed.) Aristotle and Moral Realism (London, UCL Press). L. Wittgenstein (1953) Philosophical Investigations (Oxford, Blackwell). L. Wittgenstein (1956) Remarks on the Foundations of Mathematics (Oxford, Blackwell). L. Wittgenstein (1958) Philosophical Investigations, G. E. M. Anscombe (trans.), 2nd edn (Oxford, Blackwell).

Index Abramson, K. 247 actions 8, 15, 17, 21, 22, 31, 33, 34, 42, 43, 45, 47, 62, 66–8, 70, 72, 73, 70, 72–73, 82–85, 87, 91, 93, 95, 100, 101, 103, 104, 106–108, 113, 114, 117, 121, 144–154, 155–160, 162–166, 168–177, 181, 186, 189–190, 192–194, 204–206, 208, 210–214, 217, 219–223, 229, 231, 235, 243, 249–251, 254, 264, 267–271 and affections 4, 22, 23, 31, 57, 82, 83, 100, 101, 106 benevolent 265, 267, 268 causing approbation 6, 26, 89 from duty 267, 268 excited by morals 58, 59, 73, 74–6, 84, 85, 87, 105, 116, 124, 147 impelled to or deterred from by moral opinion 21, 22, 23, 57, 82, 84, 85, 101, 105, 106 just 264, 268 mental 60 moral 264 not excited by reason 58, 59, 73, 74–76, 85, 86, 87, 93, 103 of the will 7, 189 producing or preventing 7, 8, 10, 11, 24, 26, 27, 31, 57, 82, 103, 105, 148–153 reasons for 16, 17, 19, 20, 143, 151, 153, 179, 181 theory 4 virtuous or vicious 12, 25, 60, 62, 65, 67, 69, 80, 96, 99, 100, 110, 112, 113, 264, 266–268 adultery 249, 253 agents 24, 55, 81, 82, 104, 105, 109, 116, 179, 182, 184–185, 155, 157–160, 162, 164, 166–170, 172–174, 176, 177, 187–189, 194, 206, 208, 213, 215, 217–219, 222–225, 229, 264

autonomous 188, 201 benevolent 268 charitable 265 as existentially threatened by irrationality 168 human 13, 26 just 264, 268 rational 12, 13, 14, 15, 16, 17, 18, 21, 116, 187, 199, 206, 208, 217, 223, 264 reflective 218, 224 time-slice conception of 169 virtuous 89, 264 agreeable 77, 230–232, 237, 239–242, 244–246, 252–253, 256–257 relation to the useful 230, 239–241, 246 to the person himself or to others 6, 230–232, 237, 239–242, 244, 245 anthropologist 33 alien 98 Martian 25 anthropology, and its relevance to virtue theory 249 anti-realism moral 121, 148 of Hume 78, 148 anti-socialists (their fears about nationalization) 269 American Constitution 265 American Revolution 265 Amnesty International 89–90 amoralism 43–45, 55 amorousness 92, 103 Anscombe, G.E.M. 27, 28, 127 appetite to good, and aversion to evil 8, 16, 17 appetite sexual 253 approbation 12, 13, 25, 28, 60, 61–65, 67–69, 80, 96, 99, 100, 102, 110, 113–116, 187, 203, 214, 226, 228, 231, 232, 234–237, 239–245, 257, 273 pleasing sentiment of 6, 26, 60, 78 287

288

Index

approval 26, 72, 73, 77, 101, 112, 115, 116, 131, 133, 181, 200, 202, 226, 228, 229, 231, 233, 234, 241, 246, 249, 254–257, 273 Aragorn’s principle 19 Árdal, P.S. 53, 79 Argument from Queerness (AFQ) 18, 53, 126, 127, 129 Aristotle 27, 29, 131, 141, 147, 233, 242, 251, 256, 259, 260, 264, 275 confused with St Paul 41–42 army 182–185 atheists – should they be tolerated? 257 attentiveness 236, 239 Austen, J. 13 Austin, J.L. 32, 40, 41, 47 Ayer, A.J. 22, 30, 34, 39, 53, 54, 78, 96 backgammon 51 bad 90, 122, 123, 139, 140, 141, 197–198 Baier, A. 28, 104, 152, 154, 206, 241, 255, 257, 259–262 Balfour, J. his objections to Hume’s account of morality 256 Balguy, J. 12, 18 ‘bastard’ 44 as used by Bill Woodfull, Australian cricket captain 45, 55 Bauhaus (interpretation of Hume) 232, 234, 252 Beauchamp, T. 96 beauty 1, 51, 61, 132, 227, 228, 230, 231, 235, 237, 240, 242, 246, 252, 256, 258 Beardman, S. 178 beer (in the fridge) 92 being of God 5 being dead 219 beings 13, 15, 67, 94, 96, 218, 228 believing/desiring 16 human 10, 13, 14, 21, 25, 27, 28, 62–66, 68, 69, 98, 112–115, 138, 217, 221, 222, 227, 230, 233, 234, 247, 252, 253, 255, 266, 268, 275, 276 moral 203

rational 14, 19, 31, 57, 70, 77, 227, 228 reflective 219 superhuman 64 talking, thinking 19 beliefs (see also opinions) 7–9, 15–17, 20, 23, 24, 26, 31, 33, 36, 37, 39, 42, 44, 47, 49, 51, 53, 55, 81, 82, 87, 91–92, 94, 96–98, 101, 103–104, 105, 107, 109, 111, 112, 116, 118 119, 120, 123–130, 133, 135, 137, 143–154, 164, 169, 171, 172, 174, 179–185, 186–188, 195, 197–199, 203, 205, 213, 219, 224, 245, 250 analytic 10, 86, 93, 97, 105, 117, 119 as based on impressions or sentiments 23, 97, 99, 100–102, 110, 111, 145, 146, 149 as causes of actions or desires 9, 23, 105, 117–119, 151, 152 as composed of or derived from ideas 9, 145, 146 as constituents of subjective motivational sets 20, 180–182 as constituents of reasons 19, 144, 154 genuine 82–87, 91, 95–98, 117, 118 as incapable of motivating by themselves 7, 10, 11, 13, 14, 18–19, 23, 82, 84–87, 94, 95, 97, 99, 100, 102, 103, 104, 144 as impressions 26, 101, 102, 110, 111 moral 12, 23, 24, 25, 26, 60, 82–85, 87, 89, 91, 95–103, 109–113, 115, 118, 119, 148, 149, 150, 208, 216 normative 180–181, 183–184 perceptual 110–112, 118–119 as products of reason 7–11, 13, 23, 82–84, 86, 93, 96, 97–100, 102, 103, 110–112 synthetic 10, 86, 87, 91, 95, 97, 103 about value 132–135, 137–139, 140 benevolence 12, 27, 214, 227, 229, 231, 235, 238, 243, 244, 253, 254, 256, 261, 262, 265, 267, 268, 271, 272 contrasted with love 238 Bennett, J. on Hume on belief 104

Index Bentham, J. 27, 29, 267, 274, 275 Bigelow, J. 116 Blackburn, S. 31, 78, 109, 147, 154 BO, its passion-killing effects 101 bodies 269–272 bodyline 55 bootstrapping 196 Botros, S. 144 Bratman, M. 196 bravery excessive 236, 240 and Charles XII 241 Braybrooke, D. on Hume and natural law 252 Brentano, F. 141 Bricke, J. 53, 77 Brink, D.O. 143, 209 Broad, C.D. 131, 141 Broome, J. 20, 143, 159, 176, 181, 188, 196–198, 204, 205 Brown, C. 77 Buddhism 168 Burke, E. 29 Bush, G. 128 his lack of curiosity 9 cancelability 41, 42, 43, 45, 46, 55 carbon footprint 8 Carnap, R. 39, 53 Categorical Imperative 155, 176, 186, 187, 189, 190, 204–6, 217–219 Formula of Universal Law 204, 205 Humanity Formula 188, 204, 206 cause 8, 10–11, 32, 33, 34, 63, 68, 77–8, 82, 83, 85, 86, 87, 91–93, 95, 97, 100, 101, 103, 110, 117–119, 137, 144–146, 150, 161, 168, 169, 171, 176, 177, 181, 190, 190, 192–195, 202–204, 226, 228, 235, 236, 238, 243, 244, 245, 252 charity 88, 124, 125, 265, 272 as precluding killing 271 Charles II (of England) 256 Charles XII (of Sweden) 241 charm 231, 236, 242 chastity, Hume’s account of 27–28, 253, 256 cheerfulness 241, 242 Cheney, R. 125 chest of drawers 27

289

Chisholm, R. 131, 141 chocolate – a liking for 9, 92, 103 Christianity 276 Cicero 251 his maxim about motives to virtue 255, 257 cigarette (just one more) 9, 165, 166 cigarette packets 87 Clarke, S. 7, 213 cognition 72–74 cognitive motivational internalism (CMI) 179–184 cognitive processes 60, 67, 71, 73–76 cognitivism 55 moral 19, 30, 31, 34, 39, 42, 47, 48, 49, 52, 53, 56, 108, 109, 112, 113, 115, 116, 118–120, 121, 124–127, 130–131, 137, 141, 149 about motivation 179–184 Cohen, G.A. 221 Cohon, R. 22, 54, 77, 206, 207, 254 on Hume’s artificial virtues 248 cold 5, 52, 80, 96, 100, 104 Collins, A.W. 143 ‘colouring’ (as used by Frege) 46, 56 colours 5, 50, 52, 62, 63, 64, 80, 100, 104, 134, 226 concepts, moral 6, 209, 210, 212, 251 thin 27 nonsensical 274 of rights 274, 275, 276 value 122, 233 of virtue 251 conceptual analysis 26 (Hume’s of morals) 25, 60, 64, 65, 66, 68, 69, 70, 74, 79 conscientiousness 88–90, 103 common experience 23, 57, 82, 85, 87, 105, 107 common sense 51, 88 Common Law 17 Common View (on motivation), CV 143, 153 compassion 12, 243, 244, 254, 261, 262, 272 complacency 240 conative state 31, 34, 36, 38, 39, 42, 45, 46, 47, 48, 49, 55, 160

290

Index

convention 29, 261, 266, 268, 272, 273, 275 of joking 44 linguistic 44, 45, 46, 47, 48 Copp, D. 41, 42, 43, 46, 47, 48, 49, 55, 56, 174 copulations of propositions 5 other 5 courage 250, 264, 267 criminals 254, 255, 262 ‘cur’ versus ‘dog’ (in Frege) 41, 46 curiosity 9, 103, 150, 242 cycling to work 8 Daddies 90 Dancy, J. 143, 149, 153, 204 The Dark Knight 8–9 Darwall, S.L. 12, 104, 107, 248 daughters 137–139 insisting that they eat they their vegetables 167 decency 257 decision 61, 161, 172, 198, 235 (as different from willing) 160 moral 61, 64, 67, 68 procedure 246 Declaration of Independence (American) 29 deduction 5 deference 236, 244, 245, 248 lack of and Paul Keating 245 deformity 51, 236, 237 Demetriou, D. 122, 141 desire 78, 261, 262 as basis for agency (eg in Hume) 186, 187, 194 as the basis for normativity (Desiredependence thesis) 155, 156, 168, 170, 171,173, 174 for our good on the whole 8 desires 8, 9, 11, 12, 13, 14, 15, 16, 18, 19, 20, 21, 24, 25, 26, 31, 78, 81–83, 86, 88, 91, 94, 101, 104, 122–134, 136–141, 143–147, 150–151, 153, 157–160, 167–177, 180–182, 185, 186, 187, 190–195, 204, 212, 213, 215, 217–219, 221, 223, 224, 243, 244, 276 as passions 9, 10, 100

calm 9, 94 de dicto 88, 89 de re 88 as dispositions 91 for esteem or applause 229 for our good on the whole 94 for the good of another 229, 238, 243, 244 for the misery of another 243, 244 moral 23, 87–90, 103, 104, 209 new 8, 10, 11, 82–86, 93, 94, 101, 104 perverse 140, 176 pre-existing 8, 11, 13, 14, 18, 24, 81, 83–85, 87, 91, 92, 94, 95, 97, 98, 100–103 as propositional attitudes 91 rational 190 revealed 192, 194, 204 second-order 160 Vulcan 94 desirability 26, 97, 111–112 dining 50 dinner party distracted thanks for 36 and the nice plates 43–44 Dimock, S. 77 disapprobation 6, 13, 25, 48, 60, 62, 63, 65, 67, 68, 69, 78, 99, 102, 113–116 disapproval 26, 34, 62, 67, 112, 115, 159, 200, 202, 206, 215, 216, 273 dispositional analysis of value 64, 67, 78 dispositions 13, 15, 17, 26, 61, 64, 113, 115, 118, 152, 164, 169, 200–203, 229, 229, 231, 233, 234, 238, 242, 254, 259–261 amorous 9, 92 constitutive of rationality 15–16, 119, 181 of evaluation 183 dispositions to acquire desires (DTADS) 9–10, 13–16, 19, 20, 91–95, 103, 118–119, 181 Dmitri 98 and the free drinks 92 in his new tuxedo 101, 103 Dreier, J. 109, 175, 188, 204 Driver, J. 233

Index driving to work, a case of decision without volition 161 drowning girls 137, 139 Dunya her feelings for Dmitri 98, 101 why she does not want free drinks 92 duty sense of 267, 268 dwarves 19 Ebbs, G. 174 economists 12, 206 efficiency 158 ego, his big (less important than the Party) 17 Elliot, Mr – as sensible knave 13 elves 19 emotional capacities 227 emotions 9, 27, 3, 33, 34, 36, 67, 70, 71, 82, 99, 122, 123, 133, 140, 147, 170, 181, 183, 202, 238, 239, 259–261 emotivism 82 metaethical (as opposed to psychological) 31, 34, 39, 53, 54 psychological (as opposed to metaethical) 33, 34, 39, 48, 52, 53, 54 Encyclopédie 265 ends 8, 9, 15, 16, 17, 18, 19, 20, 21, 24, 81–83, 85, 94, 104, 117, 120, 151, 155–177, 186–200, 204–206, 234, 262, 269, 275 unconditional 188, 198, 199, 204 value of 188, 192, 195, 198, 199, 200, 204 end-setting entities 8, 20 Engell, A.D. enlightenment 64, 270 Enquiry Concerning Human Understanding (EHU) 13, 96, 146, 215, 231 Enquiry Concerning the Principles of Morals (EPM) 5, 14, 26, 48, 50, 51, 57, 60, 61, 64, 68, 78, 79, 100, 145, 147, 228–230, 236, 241, 245, 250, 252–255, 257, 258, 270, 273, 274, 276 Epicureans 275

291

error theory (see also nihilism) 18, 19, 24, 49–53, 56, 95, 110, 121 error theory (about reasons for action) 151, 170 error-theoretic expressivism 49 esteem 229, 236–238, 240, 244, 245, 250, 253 Euclid 93, 104, 117–119, 227 euthanasia 271, 272, 275 evil 8, 16, 17, 23, 42, 43, 57, 58, 59, 60, 63, 65, 67, 68, 69, 70, 72, 73, 74, 76, 77, 122, 123, 133, 140, 150, 173, 235 ‘evil be thou my good’ 45, 55 existentialism 168, 177 experience 123, 129, 133–142, 168, 230, 244, 246 expressing 41, 45, 46, 48, 62, 82, 124, 125, 135, 136, 171 commands 155, 171, 173 a mental state 32, 33, 40, 41, 42, 44, 45, 47, 49, 55, 64 the expressing relation 32, 33, 36, 40, 43 as opposed to being caused by 32, 33, 37 expressivism 24, 25, 30, 31, 32, 34, 35, 36, 38, 40, 48, 49, 52, 53, 55, 82, 121, 130 as non-cogntivism for wimps 14 mentalistic construal of 36, 39, 53, 54 externalism 143, 179, 180, 182, 183, 185, 189, 190 motivational 209 Smith’s arguments against 88–90 external commitments 184 external goods 268, 269 external objects 51, 86, 240 external reasons 179, 180, 183, 184 Egyptian prisoner 90 facts 95–96, 103–104, 112, 113, 118, 126, 133, 136, 139, 143, 172, 191, 195, 196, 198, 204, 227, 228, 230 matters of 86, 98, 103, 156, 227, 249 of the matter 53 moral 12, 13, 15, 18, 19, 106, 110 normative 187, 189, 190 objectively prescriptive 12, 14, 18, 19

292

Index

false religion, its delusive glosses 25, 100, 250 family 243, 265 family tradition (Owen Wingrave’s) 181–185 Faustus, Dr. 3 fear 240, 245, 256 Fessler, D. 33 fetishism, moral 89–90 rule 202 Filmer, Sir R. 269 finger, scratching of 11, 15, 128, 147, 192, 227 Finlay, S. 16, 19–20, 154 Firth, R. 78 fish 240, 257 in Vanuatu 240 startling of 163, 164 fittingness 130, 131, 140, 236, 242 eternal and immutable natural 237 Flew, A. 22, 30 Foot, P. 27–28, 77, 272 football player (his irrational foul) 171 French Revolution 265 free will (the benefits of being one) 217–218 Frege, G. 46 Frege-expression 46, 47, 49 friendship 233, 239, 240, 261, 262 Gauthier, D. 56 general point of view (Hume) 61–64, 112–115, 200, 202, 206, 207, 214, 229, 240, 247, 250 general rules (Hume) 14, 200–203, 238, 246, 251 Gibbard, A. 42 Gill, M. 247 God 5, 12, 20, 56, 90, 93, 213, 214, 249, 250, 257 good 39, 56, 57–60, 63, 65, 67–70, 72–74, 76, 77, 89, 90, 94, 118, 122–25, 129–137, 139, 140, 193, 195, 197, 198, 199, 203, 204, 206, 215, 216, 222, 226, 228, 233, 234, 237, 239, 242–247, 250, 252, 253, 264, 271, 276

of another 229, 238, 243, 244, 265, 268 assassin 44 lesser to greater – not against reason to prefer 12, 175, 177, 192 on the whole 118 person/people 44, 88, 89 qualities 230 reason(s) 151, 153, 189, 191, 196, 197, 199, 201, 206 sense 48, 228 time tonight 8–9 will 187, 198, 199 good of mankind 231–234, 236, 237, 239, 245–247, 250 as criterion of virtue 231–232 three interpretations 232–236 goodness 88–89, 131, 133, 134, 136–137, 240 goods 265, 266, 268, 269 goodwill 240 gratitude 249 Greene, J.D. 33 Grice, P. 42, 46, 56 Grotius, H. 251, 265, 269, 275 Haidt, J. 33, 34 Hampton, J. 150, 155, 156, 169, 170, 175, 177, 178 Hare, R.M. 35, 55 Harman, G. 35 Harsanyi, J. 77, 207 heat 5, 52, 80, 100, 104 Helen of Troy 3 Herodotus 253 high voltage equipment 87 Hill, T.E. 176 history 29, 62 a necromantic art 2 Hitler, A. (evil of) 42, 43, 45 Hobbes, T. 93, 95, 100, 250, 255, 272 falling in love with geometry 93, 117–119 his mixed law and virtue theory 249 and the Slavery of Reason Thesis 104 honest gentlemen of England (in Hume) 51 honesty 14, 236, 242, 243, 249, 255 Hooker, R. 269 Hubin, D. 116, 160, 224

Index Hudson, W.D. 22, 79 human affairs 5 human good 246, 247, 250 human happiness 63, 65, 231, 246, 247, 251 three ingredients of 231, 246, 347 humanity 64, 70, 188, 206, 229, 244, 250, 254, 255, 261, 262 sentiment of 253, 254 human mind 50, 63, 66, 69, 79, 188, 249, 251, 272 causal configuration of 8, 9, 10, 11, 16 human nature 63, 64, 66, 68, 78, 112, 152, 216, 231, 240, 243, 247, 248, 262, 264, 266, 268 Humberstone, L. and direction of fit 127 Hume, D. 1–29, 30, 31, 32, 34, 35, 39, 40, 47, 53, 54, 57–79, 80, 82–89, 91, 93, 95–104, 105–120, 121–124, 128, 132–134, 140, 142–154, 155, 156, 160, 164, 165, 170, 171, 174, 175, 181, 187, 188, 190–195, 198–204, 206, 207, 226–232, 234–248, 249–258, 260, 262, 264–276 critical attitude towards 1–2 as degenerate son of science 29 dominant figure in recent debates 4 as expressivist 29, 30, 40 as grandfather of non-cognitivism (or not) 6, 14, 19, 22, 24, 25, 26, 81–82, 105, 108, 109, 111, 112, 115, 120 as judgement-internalist 107–108, 116 as metaethically indeterminate 30, 34, 47–53 as psychological emotivist 34 and reason 84 scepticism of as traitor to his fellow men 29 as virtue theorist 27–29, 226–247, 249–254, 256, 259–261 Hume’s canonical method (in morals) 66, 68, 70, 71 Humeanism (about motivation) 15, 16, 17, 20, 106, 143–145, 152–153,

293

155, 159, 171, 175, 179–181, 184, 186, 187, 191, 192, 204 Humeanism about morality 186–188, 201, 209, 213, 214, 223 Humeanism about normativity 188, 203 Humean themes (not necessarily Hume’s) 3–4, 121, 123, 126, 131, 137, 144, 190, 222 Humean virtues 256, 258 Humeans 20, 64, 71, 94–95, 100, 208, 220, 224 is Korsgaard one in disguise? 223 Hume’s fork 146, 149, 150 Hume’s Law (see also No-OughtFrom-Is) 79, 106, 122, 151 Hursthouse, R. 29, 154, 232 Hurtig, K. 20 Hutcheson, F. 6, 12, 18, 22, 80, 95, 96, 100, 109, 110, 112, 252 hypothesis, the one which we embrace 5, 60 hypothetical imperative 150, 159, 167, 170, 172, 175–177, 189, 199, 204 principle of prudence 189, 199, 206 principle of skill 199 ideas 7, 9, 10, 57, 64, 72, 77, 78, 97–100, 102, 110, 111, 122, 145, 147, 149, 151, 238, 243, 249, 252, 269, 271, 272 adequate 131, 132 relations of 10, 68, 72, 77, 86, 97, 103, 147, 150, 156 ideal conditions 62–65, 67, 75, 78 immediate agreeability 231, 232, 236, 237, 239, 240–242, 244–246, 252, 253, 257 impartiality of observer (Hume) 12, 25, 28, 69, 200, 203, 239, 254 imperatives 150 hypothetical 150, 159, 167, 170, 172, 175–177, 189, 199, 204 implicature conventional 46, 56 conversational 46, 135–136 impressions 26, 78, 61, 64, 68, 71, 97–102, 122, 144, 145, 149, 151, 238

294

Index

inclinations 12, 13, 14, 18, 89, 90, 160, 186, 187, 190, 191, 201–203, 222, 251 Independence principle 126 indecent books 256 infanticide 275 inferences 5, 7, 68, 70, 77, 79, 93, 97–99, 203 inference to the best explanation 24, 85, 95, 103, 104, 169 inflexibility 236, 247, 251 resolute and Charles XII 241 Influence Argument 144, 147, 148 injustice 21, 257, 266, 269–272 inquisitors 89 insincerity 32, 33, 36, 37, 39, 40, 41, 54 internalism 20, 38, 58–60, 69–71, 73–77, 77, 125–127, 130, 137, 143, 148, 187 constitutive 187–189 about normative reasons about practical reason 19, 20, 25 belief-internalism/judgement internalism 88, 106, 107–108, 116, 119, 125, 130–131 cognitive 149 fact-internalism 12 motivational internalism 24, 25, 31, 33, 34, 35, 38, 39, 179–184, 209, 210 normative 35 simple 32, 33 simple promising 32 sincerity 35 instrumentalism about practical reason 19, 20, 25, 116 Hume 15–16, 19, 116 instrumental principle 187–92, 194 195, 198, 199, 205, 206 as analytic 16, 162, 163 instrumental norm 15, 16, 156–166, 168–176 as constitutive of agency 16, 168 instrumental (ir)rationality 15, 154, 157, 159–166, 169–171, 173, 175–177, 186–192, 194, 196–198, 200, 204, 206 Hume’s account of 15

Kant’s account of 16 Korsgaard’s account of 19–21 instrumental reason 187–192, 194, 196–199, 205, 206 intention 32, 33, 37, 38, 161–164, 169,170, 176, 196, 197, 199 as automatic targetting 164 as identical to willing 160–161 occurrent vs. dispositional 169 opacity of 163–164 irrationality (practical) 13, 15, 17, 35, 107, 108, 116, 119, 156–163, 166–167, 170–171, 173, 175–177, 191, 194, 204 ordinary judgments of 170, 171 as diachronic inconsistency 166, 169, 176 rejected by Hume 156 as confused with theoretical irrationality 166, 169 as caused by passion 165–167, 169, 171 as existentially threatening to agents 168 instrumental 158–160, 162, 165–166, 169, 171, 175–177 is-ought 5–7, 79, 106, 122, 151, 191 Jefferson, T. his attittude to Hume 29 and human rights 29 Joan of Arc, moral heroine 257 Johnston, M. 121 joy 132, 193, 236, 237, 242, 246, 256, 257 disordered 236, 241 Joyce, R. 24, 36, 49, 50, 53, 56 Joyce, T. 153 judgements 24, 122, 124, 125, 130, 145–152, 170–174, 193,195, 203, 208–210, 212, 214, 216, 221, 230, 236 concerning causes and effects 86, 87, 93, 95, 103, 144, 192, 193 moral 3, 10, 14, 18, 22, 24, 31, 32, 34, 35, 37–39, 40, 41, 43–46, 47, 49, 50, 52, 53, 54, 55, 81, 82, 88, 95, 104, 110, 114, 115, 148, 149, 151, 200, 202, 207, 230, 235, 237

Index as caused by emotions 33, 34 hybrid 122 of the understanding 87, 193 value/evaluative 26, 121–127, 130–131, 137, 141 judges 236, 237, 242, 246, 247 authoritative 230 justice 27, 231, 235, 242–245, 247, 248, 249, 253–257, 259–262, 264–266, 268, 270–273, 275 as problem for Hume 243 as difficulty for virtue ethicists 260 as source of rights 29, 264, 265, 268, 270 Kail, P. 154 Kalderon, M. 50, 141 Kant, I. 16, 18, 155–156, 159–163, 186–189, 198–202, 204–205, 208, 209, 213, 214, 216–218, 222, 224, 251, 262, 267, 268 and search for unconditioned 214 Kantians 15, 20, 21, 150, 203 Kavka, G. 217 Kemp Smith, N. 96 kindergartens, (who should stay away from them) 176 kindness 132, 254 to children 9 killing 270–272, 275 against charity or justice? 271–2 knowledge 112–113, 115, 213, 261, 262 of future passions 8 moral 58–61, 66–77, 228 of value 121, 123, 138–139 Kolodny, N. 177–178, 204 Korsgaard, C. 15, 19, 20, 21, 108, 146, 150, 155, 160–161, 165–168, 170, 172, 175–176, 178, 187–192, 194–196, 198–206, 208–224 and her neo-Kantian solution to the problem of normativity 208–224 and practical identity 217, 219–222 The Ladies’ Calling 257 law 271–276 civil 272, 273 moral 272 natural 272, 273, 276 useful 274, 275

295

laziness 231, 240 Lenman, J. 143 Lewis, D.K. 55, 56, 67 Liu, X. 77 Lo, Norva Y.S. 25–26, 79, 106, 108, 109, 110, 112, 113, 115 Locke, J. 6, 7, 28, 250, 251, 257, 265, 269, 270 on virtues and vices, and their relation to laws 249, 251 logic 2, 5, 6, 71, 93 language of 5 logical priority 268 Lomu, Jonah 247 Lowde, the learned Mr, and Locke’s reply to him 249–250 love 131, 139, 226–228, 236–246, 254–256, 261 McDowell, J. 15, 143 MacFarlane, J. 55 Macias, R. McIntyre, J.L. 238 Mackie, J.L. 18, 49, 56, 77, 78, 144, 151, 177 Mafioso – his purported duties as a hitman 221 McNaughton, D. 143 make-beliefs 67 Malcolm, N. 54 mantis from Mars 19 Maori 274 Mazza, E. 253 meaning 39, 41, 42, 43, 44, 47, 52, 53, 54, 175 (of moral claims) 34, 39, 62, 131 means-end reasoning 8, 15, 94, 104, 151, 186–188, 192–200, 206 means to ends 8, 9, 15, 16, 20, 24, 117, 120, 146, 151, 155–166, 168–169, 171–177, 186–200, 204, 206, 207, 253 attempting vs. performing 158, 161, 162 constitutive vs. causal 8–9 disjunctive 158–161 necessary 158–166, 169, 171, 172, 175, 177

296

Index

means to ends–(Continued) sufficient 158, 162, 176 temporally dislocated from ends 160–162 modality (alethic vs. deontic) 172 Meinong, A. 131, 140, 141 Mele, A.R. 143 Mephistopheles 3 men 19, 29, 64, 249, 269, 271, 273, 274 as governed by their duties 21, 23, 57, 82, 84, 85, 87 mental contents 99, 102, 103 mental abilities 250 entities or states 8, 20, 32, 33, 34, 36 37, 38, 39, 40, 126, 127, 134, 152, 161, 167, 169, 184, 206 expressing of 40–42, 46, 47 qualities 226 traits 250, 252 merchant his interest in his accounts 86 Merteuil, Mme de 101 meta-ethics 4, 21, 24, 32, 33, 34, 37, 38, 39, 42, 43, 47, 48, 49, 52, 53, 54, 58, 60, 64–67, 80, 81, 89, 95, 119, 114 metaphysics of value 121, 156, 163, 170 meum/suum 270–273 Mill, J.S. 267, 274, 275 Millgram, E. 15, 19, 108, 150, 156, 175, 206 Milton, J. 45 mind 1, 8, 9, 11, 16, 50, 51, 80, 96, 145, 152, 162, 166, 168, 169, 170, 176, 177, 188, 190, 203, 218, 226, 228–230, 236, 237, 240, 241, 249, 250, 251, 272 its propensity to spread itself on things 51 reflective 212 mind-indepedence 110 Moll, J. 55 Monk, R. 83, 117 Moore, G.E. 40, 54, 213 his paradox 40–41, 42–43, 54, 135–136 his shadow 135–136

Moore-express 41, 42, 43, 46, 47 moral cognitivism 48, 81, 108, 109, 112, 115–120, 137, 141 moral conceptual framework 53 moral considerations 102–103, 209, 210 moral decisions 61, 67, 68, 99 moral deliberation 208–211, 213 moral disagreement 31 moral discourse 49 moral distinctions 51, 80, 124 deniers of 50 are they derived from reason? 4, 23, 76, 77, 81, 98, 99, 102, 104 moral eliminativism 50 moral language 34, 39, 47, 49, 50 moral nihilism 50, 121 moral philosophy 50, 186, 200 Hume’s 49, 200 moral point of view 229, 247, 256 moral projectivism 52, 121 Moral Problem, the 24, 26, 80, 81 moral properties 12, 52, 54, 58–60, 62, 64–67, 69, 70, 72–74, 77, 79, 80, 96, 99, 110, 229 as secondary qualities 12, 22, 80, 96, 99–100 moral psychology 142–143, 252 moral qualities 52, 64, 250 moral realism 30, 95, 110 moral reasoning 33 moral reasons 186, 209, 219, 221 moral response 236, 239, 242, moral scepticism 30, 49, 50, 212 moral sense 12, 22, 25, 213, 227–230, 235 moral sentiments 62, 63, 65–70, 75, 78, 79, 207, 215, 216, 230, 235 moral standard or norm 213–214 moral terms 45 moral theory 259, 260 moral vocabulary 249 moral worth 267 morality 43, 44, 45, 50, 51, 53, 55, 56, 57, 59, 60, 64, 76, 77, 96, 98, 101–103, 110, 113, 121, 122, 186–188, 190, 200–205, 213–216, 223–224, 227, 229, 230, 235, 242, 243, 250, 251, 261, 262, 264, 266

Index evolutionary explanations of 212 a Humean anatomy of 214 a Humean justification of 208, 214, 220, 224 its influence on human passions and actions 31, 32, 57, 87, 101, 107 its rules no the conclusion of our reason 31, 57, 99, 105, 147 justification of 208–214, 216–224 more properly felt than judged of 34, 99 not derived from reason 58, 71, 76, 99, 147, 148 its normativity 208–211, 217 moralists, ancient 251 morals 60, 66, 67, 75, 87, 96, 98, 110, 147, 235 as exciting passions and actions 39, 57, 58, 77, 84, 101, 102, 105, 106, 147 Hume’s conceptual analysis of 60–66, 69–71, 74, 76, 79 Hume’s epistemology of 66–68 as products of a sentimental faculty 48 seeming harmony of 48 motivation 15, 19, 20, 25, 29, 38, 39, 54, 55, 58, 59, 66, 69, 70, 72–76, 78, 81–104, 106,107, 109, 115–119, 123, 124, 126, 129, 131, 137, 142–150, 152–154, 155, 160, 171, 175, 179–184, 186, 187, 190, 192, 204, 208–213, 220, 223–224, 254, 255, 259, 261, 262 Motivation Argument (see also Influence Argument) 4, 5, 9, 12, 14, 22–27, 34, 39, 40, 54, 57–60, 69–77, 80, 81, 84, 90, 97, 98, 105–120 as an argument for non-cogntivism or expressivism 14, 19, 23, 24, 25, 26, 48, 80–82, 84, 85, 90, 95, 96, 103–104, 105–120 as an argument for sentimentalism 19, 23, 24, 25, 26, 80, 81, 96, 98, 100–104 as dead dog 95 as dragon in search of a slayer 95 in need of repair 24 its obvious invalidity 23

297

motivational internalism 32, 34–35, 38, 39, 58–59, 73, 74, 143, 148, 209 cognitive 148, 179–184 normative 35 simple 32–33 sincerity 35 Motivation, theory of (TOM) 142, 143, 152 motive 231, 243, 247, 248, 250, 251, 255, 257, 261–263, 264–268, 271 Mulligan, K. 131, 141 Mummies 90 murder 271, 272, 275 mussels – as a breakfast food 97, 101, 111 Nagel, T. 15, 143, 155, 181, 214 Nazis 89 ‘nigger’ – abusive? 45 nihilism (see also error theory) 121 about practical reason 151 non-cogntivism 6, 14, 18, 19, 20, 22–26, 54, 55, 56, 91, 95, 96, 99, 100, 103, 104, 105–116, 120, 121, 125, 130, 134, 141, 145, 147, 149 about motivation 180–181 ‘nonsense on stilts’ 29, 274 No-Ought-From-Is (see also Hume’s Law) 5–7, 79, 81, 106, 122, 150, 191 not the prime topic of this volume 4–5 ‘no reasons from false beliefs’ (NRFFB) 180, 182, 183 normal conditions 62, 64, 132, 136 norm 44, 45, 49, 108, 116 normativity 21, 155, 156, 161, 162, 164–175, 177, 186–192, 195, 196, 198–206, 208, 214–218, 220–224, 232, 236, 259, 260 as brute/mysterious 170, 173 of Hume’s virtue theory 249–251, 253 justificatory 208–214, 216–224 motivational 208–212, 223 possible violation: criterion of 16, 20, 156, 162 reasonable expectation: criterion of 156, 157, 159, 160, 162, 165, 168, 169

298

Index

normative requirements 155, 156, 161, 186, 188–191, 194, 196–199, 204, 205 Norton, D.F. 76 Nowell-Smith 55 Nozick, R. 17 obedience 245, 250, 251, 260, 262 obligation 6, 17, 21, 23, 27, 57, 58, 77, 78, 82, 84, 85, 96, 101, 105, 186, 187, 201, 210, 219, 221, 249, 266, 268, 274, 275 observer (ideal/suitable) 12, 13, 22, 25, 30, 62, 64, 70, 96, 139, 238, 243 Oddie, G. 26, 122, 129, 134, 141 Ogden, C.K. 39 one thought too many 88–90 opacity/intensionality of intention 163–164 opinions 23, 97, 100, 144, 146, 193, 203 of injustice 21, 57, 58, 96, 101, 105–107 moral 23, 64, 106 of obligation 23 Otago, University of 4, 154 ought 4, 5, 6, 27, 44, 55, 89, 149, 150, 155–160, 162, 168, 171, 173, 174, 175, 177, 178, 197–199, 205–206, 210, 211, 214, 215, 219, 220, 222, 224, 274 as end-relative 171–172 as instrumental 160, 172 moral 171 narrow scope 197, 198, 205 prudential 171 rational 163, 170, 171, 173 wide scope 197, 205 ovum 271 Owen, D. 77, 79, 148 Oxfam 88–89 pain 17, 28, 77, 78, 87, 94, 95, 101, 122, 129, 132, 137, 146, 150, 253, 254 Parfit, D. 143, 181 parthenogenesis (of actions) 83 Party, the (as against ‘his big ego’) 17 party, the (as in ‘the party tonight’) 32

passions 4, 7, 8, 9, 10, 11, 13, 22, 23, 24, 26, 27, 31, 32, 39, 57–59, 69, 72–77, 81, 82, 84, 86, 87, 96, 97, 99, 100, 102, 105, 107, 111, 114, 116, 121–123, 128, 134, 140, 145, 147, 150–152, 156, 170, 171, 190–195, 199, 203, 206, 227, 229, 231, 235–240, 242–248, 250–256, 267 calm 9, 95, 171, 177 as cause of irrationality 165, 166, 167, 169 as desires 8, 9, 10, 11, 100, 101 as DTADS 9, 10 as stable Pauer-Studer, H. 20 pejorative language (‘kike’, ‘slut’ etc) 41, 44, 46, 47 Peruvian philosopher 89 philosophers 264, 270 philosophizing 50 philosophy 50, 52, 213, 229, 275 modern 226 of psychology 143 Pigden, C. 5, 15, 25, 26, 106, 108–115, 117–120, 154, 174, 247 pity (see also compassion) 243, 244, 261, 262 Plato 53, 146 pleasure 17, 28, 94, 122, 132, 146, 202, 226, 230, 231, 232, 234–240, 242–244, 246, 247, 251–253, 262, 276 pluralism 28, 257 in theories of virtue 28, 231, 253, 256, 259 in aspects of Hume’s nonconsequentialist criterion of virtue 23, 232, 236, 237, 246 practical necessity 21, 208, 211, 214, 217, 220, 221, 224 practical identity 21, 217, 219, 221 and fear of one’s own destruction 219–220 and ownership of action 220 as a human being 217, 221, 222 prescriptivism 31, 82 Price, H. 30 Prichard, H. 213

Index pride 77, 78, 123, 226–228, 237, 240–242, 247, 253, 256 proper 28, 241 Princess Diana (outpourings of grief at the death of) 239 projectivism 51, 52, 56, 121, 216 promising 28, 32, 33, 41, 43, 47 promising internalism (simple) 32 properties 22, 52, 53, 227, 228, 242, 245 moral 12, 52, 54, 58–60, 62, 64–67, 69, 70, 72–74, 77, 79, 80, 96, 99–100, 110, 227, 228, 242, 246 natural 31 response-dependent 22, 246 secondary 12, 25, status 235, 242, 245 value 121 visual 135 property 28, 244, 251, 254, 255, 260, 262, 265, 266, 268–274 and cloperty 272 rights 29, 255, 260, 261, 265, 266, 268–270, 274, 275 rights as pleonasm 269 propositions 5, 9, 18, 39, 58, 81, 82, 92, 124–126, 128, 129, 130, 134–136, 143, 148, 184, 169, 172, 173, 197, 198, 227, 269 propositional attitudes 9, 91, 92, 163 propositional content 124, 126, 197 prudence 16, 171, 189, 199, 206, 227, 228, 256 psychologism (wrt motivation) 143, 152–4 psychology 143 moral 143, 206, 207, 252 Pufendorf 265, 269 punctual self 224 and its inability to evaluate 224 puritan hackles 256 Queen Elizabeth II 245 ‘queer’ – insulting or otherwise? 44, 45 queerness (or moral properties or facts) 53, 126–127, 129 Quinn, W. 143

299

Radcliffe, E. 77 Radnor, earl of, his splenetic virtues and whimsical talents 256 Railton, P. 217, 218 Ramsey, F.P. 143 rape (in state of nature) 271–272 rational criticism 127–128, 171, 173, 174, 177 rationalism 59, 60, 75, 76, 81, 165, 169, 170, 171, 172, 174 dogmatic 190 as opposed by Hume 60–70–72, 74, 124 in ethics 7, 12, 14, 18, 25 moral 76 vs sentimentalism 12–13 support from common sense Rationalist Argument (against sentimentalism) 12–13, 18 rationalists 60, 77, 81, 89 eighteenth century 12, 18, 270 modern 14, 18 rationality 16, 17, 19, 186–192, 194–196, 202, 205, 206 instrumental 161, 186–192, 196–198, 200, 204–206 desires or DTADS as constitutive of 14, 15, 16, 94 Humean accounts of 188, 191, 203 norms of 16, 17, 18 practical 17, 107–108, 115–116, 119, 188, 191, 217, 224 self-interest theory of 202, 206 Rawls, J. 8, 20, 276 Raz, J. 143, 177, 197, 204 realism moral 95 of Hume 74 realist-expressivism 49 reason 4, 5, 6, 7, 8, 9, 10, 11, 22, 31, 34, 48, 50, 51, 57–60, 67, 71–77, 79, 81, 82, 84, 87, 96, 98, 99, 102–104, 121–124, 128, 141, 144–54, 155, 155–159, 162, 164, 165, 167, 169, 170, 175,177, 179–185, 187, 193–5, 227, 235, 259, 260 as belief-forming faculty 7, 82, 84, 86, 96 as discursus 98

300

Index

reason–(Continued) impotence of 31, 82, 84, 85, 87, 105, 116–117, 119–120, 144, 147 as inferential organ 7, 98, 99 as inferential organ operating on ideas 7, 98, 99, 110–111 instrumental 150, 187–189, 191–192, 194, 196–199, 205, 206 moral beliefs not derived from 147 practical 17, 107–108, 115–116, 119, 155, 156, 167, 171, 175, 186–192, 204, 206, 212 products of 82, 84, 85, 87, 110–112 pure 271 as slave of the passions (see slavery of reason thesis, SRT) 2, 4, 7, 10, 11, 12–21 ultimate, for every rule 276 and wantons 218 reasoning 5, 58–60, 68–77, 86, 100, 192, 195–198, 206, 219, 229, 230 abstract or demonstrative 10, 86, 93, 95, 144, 192, 195, 228, 275 moral 33, 98 practical 218, 225 reasons 19, 21, 186–192, 196–198, 200–202, 204–206, 211, 213, 223, 224, 245, 268 for action 16, 17, 19, 166, 212, 217–219, 221–223, 264–266 justifying 208, 211–214, 222, 223 motivating 19, 142–154 moral 186, 209, 214, 219 negative 210, 211, 216 normative/normativity of 143, 151, 152, 186, 192, 195, 196, 197, 201, 203, 204, 206, 217 positive 209–214, 216, 221, 222, 224 to be moral 208, 209, 211, 215, 217, 221 to be rational 190 unconditioned 218 red 5, 52, 62, 96, 97, 100, 110, 132–133, 135 red button (which will destroy the world) 15 red dress – as worn by Zena 97, 111–112 red rose 132–133, 136

reflection 212, 214, 217, 218, 219, 220, 221, 222, 223, 224, 235, 236, 242 its power to influence conduct 219, 223 reflective agents/beings/creatures 212, 217, 218, 219, 222–224 reflective endorsement 187, 188, 200, 201, 203, 208, 214, 215, 216, 219, 222, 254 reflective equilibrium Reid, T. 94, 265 and the desire for our good on the whole 94, 118 and the Vulcan valediction 94 relativism 250 in theories of virtue 228, 246, 249 requirements 56, 200, 261 rational 189, 191, 194–196, 198, 205 moral 190 normative 186–192, 196, 197, 204 Richards, I.A. 39 Ridge, M. 36, 37, 54 right 21, 57, 77, 81, 82, 88–90, 95–96, 100, 103, 104, 110, 149, 187, 209, 211–213, 215, 216, 222, 250, 267 righteousness, hunger and thirst after 96, 103 rights 29, 254, 255, 261, 264–266, 269–276 human 29, 90, 265 innate 265 natural 265, 269, 272, 274, 275 property 29, 261, 265, 266, 268–270, 275 real 274, 275 women’s 90, 274 Roeser, S. 154 roller coaster, Korsgaard’s irrational rider of 165, 166 Ross, W.D. 27–28, 213 Ruse, M. 224 Russell, B (third earl) 93, 95, 103, 104, 119 falling in love with geometry 93, 117–118 and the Slavery of Reason Thesis 104, 118 Russell, F (second earl) 93 Russell, L. 21

Index St Paul, as confused with Aristotle 41, 42 sapphire ring, stolen, as an anniversary present 159 Sandis, C. 19, 146 Satan (Milton’s) 45 Save the Children Fund 9 Sayre-McCord, G. 77, 78, 110, 114, 232, 234, 247 Scanlon, T. 143, 177 Sceptic, the (Hume’s essay) 50, 52, 229 scepticism 178 in Hume’ 49–50, 52, 156, 170, 175, 250, 251, 257 Scheler, M. 131, 141 scholastic headpieces 29 Schroeder, M 176 Scythians unusual towels of 253 Searle, J.R. 32, 33, 36 Searle-sincerity 36, 37 self-conception (Korsgaard) 219, 221–224 self 171, 176, 220, 230, 240, 241, 256 reflective 168 self-destruction 220 ‘and flinching 220 self-interest 215, 216, 227, 231, 255, 256 self-love 227, 241, 255 selfish/selfishness 17, 63, 78, 235 self-preservation 219–221 semen 271 sensible knave, the 13, 14, 23, 201, 202, 262 sentimentalism 25, 48, 148, 149 Sentimentalist Argument (against rationalism) 13 sentimentalists 12, 13, 14, 18 sentiment of humanity 255, 254 sentiments 58–70, 73–75, 77–79, 145, 147, 148, 151, 200, 202, 203, 206, 207, 214, 216, 226–228, 230, 235, 236, 242, 252, 253 moral 62, 63, 65–70, 75, 78, 79, 113, 114, 116, 214–216, 224, 226 Setiya, K. 175 Shafer-Landau, R. 35 Shaftesbury, A.A. (1st earl of) 256

301

shoelaces tied, not grounds for irrationality 158 Sidgwick , H. 27, 167, 170, 176, 252 simple-MI (see also internalism, motivational) 32, 35, 39 simple promising internalism 32 Simpsons, on TV 92 sincerity 36, 37, 38, 39, 124, 129, 130, 135 Ridge on 36, 37 of speech acts 36, 38 Searle on 36, 37 sincerity-MI 35, 36, 38 Skidmore, J. 225 slavery/slaves 255, 270–272 Slavery of Reason Argument 81, 86–87, 90–92, 95–98, 100, 101, 103 Slavery of Reason Thesis (SRT, reason the slave of the passions) 7, 9, 10, 14, 15, 16, 18, 19, 22, 23, 81, 82, 86, 87, 94–96, 103, 121, 150 as premise to the Argument from Queerness 18 as premise for Motivation Argument 14, 22–23 as premise for Sentimentalist Argument 13 as threat to rationalism 14–18 ‘slut’- pejorative? 44, 46 Smith, A. 27 Smith, M. 3, 24, 25, 31, 35, 80, 82, 83, 93, 100, 104, 126–128, 143, 153, 154, 209 his arguments against externalism 88–90 on desire 16, 91 and the Moral Problem 24, 80, 81 on Norva Lo 26, 106, 108, 109, 110, 112, 113, 115 on Charles Pigden 26, 106, 108–115, 117–120 as Kantian 15 smoker 165, 166 reformed 9 smoking – its effects 87 Snare, F. 148 Sommerville, R.B. sounds 5, 51, 52, 63, 80, 100, 104, 226

302

Index

spectator, suitably qualified, judicious etc (see also observer, ideal) 6, 12, 22, 26, 28, 60–64, 78–79, 80, 96, 100, 110 Spinoza, B. 7 Spooner, the reverend William (confusing Aristotle with St Paul) 41 SRT. See the Slavery of Reason Thesis states of affairs 143, 157 as objects of desire 91, 92 status (in Hume) 242, 245, 246, 254, 255, 257 stealing – its moral wrongness 31, 35, 36, 44, 55, 260 Stevenson , C.L. 31, 39, 53, 55 strangers 132, 137, 139, 261 their daughters 137, 139, 140 their pain 137, 140 Strawson, G. 153 Strong requirement, the 208, 209, 211–13, 216, 223, 224 Stroud, B. 206 Sturgeon, N. 53, 54, 175 subjective motivational set 20 subjectivism 30, 34, 53, 121, 259 suicide 17, 225, 275 moral 219 swimming the Hellespont 16 Svavarsdóttir, S. 35 Swanton, C. 28–29, 233, 239, 245, 249, 252–257 sympathy 227, 229, 232, 238, 243, 244, 253–256, 272 talk 19, 265, 274, 275 Taylor, C. 224 Taylor, J. 257 temperance 231, 233 tenderness 229, 236, 237, 239, 240 test of reflexivity 214, 215, 216, 217, 222 test of self-interest 216 theft 255 uncharacteristic but culpable 221 Thomism, creeping 28 Thomson, J.J. Thrasymachus 13, 43, 44 Timmons, M. 35 Tolkien, J.R.R. 19 TMoNB 183–184

total ruin, my 11, 17, 128, 192, 227 Transcendental argument 222 Treatise of Human Nature (Hume) 5, 6, 7, 13, 27, 29, 39, 50, 51, 52, 57, 60, 80, 81, 84, 97, 98, 105, 121, 204, 215, 229, 230, 250, 252–255, 257, 264, 273 treason 249 truthmaker(s) 17, 183–184 turning on the light 163–164 as a means to spotting the kitten not startling the fish 163–164 Two Treatises on Government (Locke) 269 understanding, the 10, 12, 68, 79, 86, 87, 97, 186, 192, 193, 195, 199, 215 uneasiness of an Indian 11, 128, 192 Urmson, J.O. 246 useful 6, 27, 229, 230, 231, 247, 252, 253, 256, 273–276 utilitarianism 27, 30, 114, 201, 202, 207, 232, 233, 234, 252 utilitarians 27, 28, 88, 113–114 Hume as 274 utility 26, 27, 202, 207, 226, 230, 231, 233, 234, 236, 241, 242, 252, 253 utterances 32, 36, 37, 40, 41, 42, 43, 44, 46, 52, 54, 56, 125, 135, 136, 170, 171, 172, 177 moral 34, 36, 39, 42 Valmont, vicomte de 9 value(s) 121–127, 130–134, 136–141, 160, 188, 189, 192, 195, 198, 199, 200, 204, 206, 214, 216, 220, 221, 222, 233, 237, 23–241, 246, 257, 260 agent-neutral 140 experience of 132–4, 136–140 realism 12, 121, 122, 124, 127, 131, 140 vegetables, forced on children 167 Viagra 91 vice 5, 6, 28, 48, 51, 52, 54, 60, 61, 63, 65, 67, 68, 77, 78, 80, 98, 99, 100, 102, 104, 110, 113–115, 152, 216, 226, 227, 229–232, 236, 237, 242, 243, 245, 261

Index Vienna City Marathon 194 virtue 12, 14, 22, 25, 28, 48, 51, 52, 54, 57, 60, 63, 65, 67, 68, 73, 74, 77, 78, 79, 80, 89, 90, 98, 99, 100, 102, 104, 10, 113–115, 152, 215, 226–227, 229–239, 241, 242, 244–248, 249, 250, 252, 254, 255, 256, 259–262, 266, 267, 269 as a causal concept 60–61 its original sense as a manly excellence 250 sense of 226–227, 256, 257 virtue clusters 237, 256 joy 237, 242 love 237, 239, 241 pride 241, 242 respect 248 virtues 27, 28, 29, 226–234, 236, 237, 239–242, 245–248, 250–254, 256, 258, 260–262, 267 artificial 29, 54, 242, 248, 251, 254, 257, 272 as powers in objects 226–228, 230, 235 authoritative sense of 227 criteria of 231, 232, 234 natural 261 splenetic 256 taxonomy of 232, 239 virtue theory/ethics 27–29, 249–253, 256, 259–261, 267 volition 72, 144, 149, 156, 160–163, 176, 196, 205

303

(as distinct from willing) 161 Vulcan valedition 94 Waldron, J. 257 Wallace, R.J. 155–156, 175 wanton (and the purported impossibility of being one) 218 wants 8, 82, 93, 121, 128, 139 well-being 265, 267, 268, 273, 275, 276 whole world, the destruction of 11, 128, 147, 192, 227 The Whole Duty of Man 257 Wiggins, D. 78, 247 will 7, 8, 10, 11, 75, 75, 77, 149, 153, 156, 167, 187, 189, 190, 192–199, 201, 203–205, 217, 218 constant and perpetual 273 Williams, B.A.O. 20, 143, 179–184, 207, 216, 222, 264 Wingrave, Owen 182–185 wit 238, 242 Wittgenstein, L. 54, 268 and the turning of his spade 211 and the hardness of rules and materials 211 women 253, 254, 274, 271, 274 von Wright, G.H. wrongness 31, 57, 77, 87, 88, 90, 94, 95, 103, 148, 199, 220, 271 as metaphysically problematic 31 Zena in a red dress 26, 97, 111, 112