Oxford Studies in Normative Ethics [5] 9780191062100, 0191062103

Table of contents :
Cover
Oxford Studies in Normative Ethics Volume 5
Copyright
Contents
Acknowledgments
List of Contributors
Introduction
References
Chapter 1: The Asymmetry of Good and Evil
1. On Actions and Their Effects
Options
Acts or Deeds
Actions or Doings
2. Doing Good
Disposition-Independent and Disposition-Dependent Benefits
Respect
Love and Friendship
Honesty, Justice, and other Virtues
3. Doing Evil
Do Evildoers Exercise Malevolent, Self-sacrificing Control?
Answer: Not in General, Only in Exceptional Cases
4. Explaining the Asymmetry Between Good and Doing Evil
The Role of Standards
Historical Antecedents
5. The Knobe Effect
Acknowledgments
References
Chapter 2: A Particularist but Codifiable Virtue Ethics
1. Introduction
2. The Virtue Rules
3. A Target-Centered Virtue Ethical Notion of Right Action
4. The Thick Concepts, V-Reasons, and V-Acts
5. V-Rules Codify Default Reasons
6. The V-Rules, Particularism, and Codifiability
7. The Burden of Proof
8. Conclusion
Acknowledgments
References
Chapter 3: My Welfare and Yours
I
II
III
IV
V
Acknowledgments
Chapter 4: Moral Worth and Normative Ethics
General Implications
Altruism
Misguided Altruistic Actions and Types of Reasons
References
Chapter 5: Ideas of the Good in Moral and Political Philosophy
Introduction
Sidgwick´s Dualism of Practical Reason
Marx, Rawls, and Williams on Acceptance
Answering the Question of Acceptance
Acknowledgments
Chapter 6: Scrupulous Judgments
The Issue
What Obsessive-Compulsive Disorder Is
What Scrupulosity Is
Abnormality
Charity
Anxious Morality
Concluding Thoughts
Chapter 7: Obligations of Gratitude and Correlative Rights
1. The Obligation Platitude
2. The Directionality Platitude
3. The No-Right Platitude
4. Correlativity Reconsidered
Acknowledgments
Chapter 8: Moral Anxiety and Moral Agency
1. Varieties of Anxiety
2. Social Anxiety as Template
3. Moral Anxiety: What it is and What it Does
3.1 The Basic Model
3.2 Moral Anxiety as Distinct
4. Moral Anxiety, Deliberative Virtue, and Agency
4.1 Premise 2: Moral Decision Making and Moral Metacognition
4.2 Premise 3: Moral Anxiety and Moral Metacognition
4.3 Premise 5: Moral Anxiety Contributes Productively
4.4 The Pay-Off: Moral Anxiety as an Emotion to Cultivate
5. Three Implications
5.1 Cultivation
5.2 The Significance of Psychological Conflict
5.3 A Challenge to Existing Accounts
6. Conclusion
Acknowledgments
References
Chapter 9: The Narrative Calculus
1. From Temporal Distribution to Narrative Shape
2. A Narrative Calculus
What Is a Narrative?
When Is a Story Prudentially Good?
Progressivism vs. Achievementism
3. Testing the Theory
4. Objections
5. Conclusion
References
Chapter 10: Morality Within the Realm of the Morally Permissible
0. Introduction
1. Is There a Non-Trivial Realm of Moral Permissibility?
2. What is the Relationship Between Moral Permissibility and All-Things-Considered Permissibility?
3. Morally Permissible Moral Mistakes
4. Rejecting Morality Within the Realm of the Morally Permissible
5. Other Moral Categories
6. Conclusion
Acknowledgments
Chapter 11: Fairness, Participation, and the Real Problem of Collective Harm
1. Introduction
2. Two Examples
3. Weak Participation
4. Strong Participation
5. The Fairness Approach
6. Fairness and Difference-Making
7. A Common Problem
8. Can´t We Just Apply a Non-Consequentialist Theory and Call it a Day?
9. The Problem of Collective Harm (and How To Solve It)
Acknowledgments
Chapter 12: The Virtue of Authenticity
1. Introduction
2. Three Popular Conceptions of Authenticity
3. Refining the Three Popular Conceptions of Authenticity
4. Authenticity as A Personal Virtue
5. Summary
Acknowledgments
References
Chapter 13: What Normative Terms Mean and Why It Matters for Ethical Theory
1 Introduction
2 Weak and Strong Necessity
3 Endorsing and Non-Endorsing Uses
4 Dilemmas
5 Supererogation
6 Judgment Internalism and ``The Normative´´
7 Conclusion
Acknowledgments
References
Index

Citation preview

Oxford Studies in Normative Ethics

Oxford Studies in Normative Ethics Volume 5 EDITED BY

MARK TIMMONS

1

3

Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © the several contributors 2015 The moral rights of the authors have been asserted First Edition published in 2015 Impression: 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2015932779 ISBN 978–0–19–874466–5 (hbk.) ISBN 978–0–19–874467–2 (pbk.) Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

Contents

Acknowledgments List of Contributors Introduction

vii viii 1

MARK TIMMONS

1.

The Asymmetry of Good and Evil

15

PHILIP PETTIT

2.

A Particularist but Codifiable Virtue Ethics

38

CHRISTINE SWANTON

3.

My Welfare and Yours

64

MARK LEBAR

4.

Moral Worth and Normative Ethics

86

NOMY ARPALY

5.

Ideas of the Good in Moral and Political Philosophy

106

T. M. SCANLON

6.

Scrupulous Judgments

129

JESSE S. SUMMERS AND WALTER SINNOTT-ARMSTRONG

7.

Obligations of Gratitude and Correlative Rights

151

TONY MANELA

8.

Moral Anxiety and Moral Agency

171

CHARLIE KURTH

9.

The Narrative Calculus

196

ANTTI KAUPPINEN

10.

Morality Within the Realm of the Morally Permissible

221

ELIZABETH HARMAN

11.

Fairness, Participation, and the Real Problem of Collective Harm JULIA NEFSKY

245

Contents

vi 12.

The Virtue of Authenticity

272

ERNESTO V. GARCIA

13.

What Normative Terms Mean and Why It Matters for Ethical Theory

296

ALEX SILK

Index

327

OUP CORRECTED PROOF – FINAL, 17/7/2015, SPi

Acknowledgments

Versions of the articles in this collection were presented at the fifth annual Arizona Workshop in Normative Ethics that took place in Tucson, Arizona on January 9–11, 2014. I would like to thank the Center for the Philosophy of Freedom and the Department of Philosophy at the University of Arizona for their continuing generous financial support of the workshop. Thanks to Sameer Bajaj who assisted me in running the workshop, to Chris Howard for preparing the volume’s index, and to Betsy Timmons for her gracious help with workshop details. I would also like to express my sincere thanks to the following philosophers for serving as de facto program referees: Dale Dorsey, Brad Hooker, Tom Hurka, Eleanor Mason, Doug Portmore, and Sarah Stroud. Two anonymous referees for Oxford University Press offered very helpful, constructive advice to our authors. Thanks finally (and again) to Peter Momtchiloff, my OUP editor, for his support. Mark Timmons Tucson, AZ

List of Contributors

Nomy Arpaly is Professor of Philosophy at Brown University Ernesto V. Garcia is Assistant Professor of Philosophy at UMass Amherst Elizabeth Harman is Associate Professor of Philosophy and Human Values at Princeton University Antti Kauppinen is Research Fellow at the Academy of Finland Charlie Kurth is Assistant Professor of Philosophy at Washington University in St. Louis Mark LeBar is Professor of Philosophy at Florida State University Tony Manela is Adjunct Lecturer at Georgetown University Julia Nefsky is Assistant Professor of Philosophy at University of Toronto Philip Pettit is Laurence Rockefeller University Professor of Politics and Human Values at Princeton University and University Distinguished Professor of Philosophy at the Australian National University T. M. Scanlon is Alford Professor of Natural Religion, Moral Philosophy, and Civil Polity in the Department of Philosophy at Harvard University Alex Silk is Lecturer in the Department of Philosophy at University of Birmingham Walter Sinnott-Armstrong is Chauncey Stillman Professor of Professional Ethics in the Department of Philosophy at Duke University Jesse S. Summers is Post-Doctoral Fellow, Kenan Institute for Ethics Christine Swanton is Honorary Research Fellow in the Department of Philosophy at the University of Auckland

Introduction MARK TIMMONS

This fifth volume of Oxford Studies in Normative Ethics features some of the best contemporary work in the field of normative ethical theory. This volume includes thirteen essays, covering a broad range of topics: the asymmetry of good and evil, particularism and virtue ethics, personal welfare, moral worth and normative theory, ideas of the good in moral and political philosophy, moral scrupulosity, gratitude and rights, moral anxiety and moral agency, prudential value in an individual’s life, moral theory and the category of the morally permissible, fairness and the problem of collective harm, the virtue of authenticity, and the significance of the meanings of moral terms for normative theory. In “The Asymmetry of Good and Evil” Philip Pettit explains the asymmetry in terms of the so-called robustness of doing good that is lacking in doing evil. In explaining the idea of robustness, Pettit uses the term ‘action’ to refer to a sufficiently rich sense of one’s doings that includes one’s intentions or motives and distinguishes two kinds of effects of actions: disposition-dependent effects, which they have essentially in virtue of one’s intentions or motives (one’s dispositions), and dispositionindependent effects including, for example, what one’s actions bring about causally. Actions are contrasted with acts or deeds—the latter being one’s behavior considered independently of one’s dispositions. Thus, my act of benefitting you will count as a genuinely kind action depending on my motive. When I am so disposed that in benefitting you I not only confer a benefit in the actual circumstances, but I am disposed to do so across a range of non-actual circumstances, including those which involve some level of self-sacrifice, then I am controlling benevolently for positive effects on your well-being and thereby realize a kind of good (e.g., kindness). Such goods are thus properly described as robustly demanding, and include respect, attachments to be found in love and friendship, as well in actions that express such virtues as honesty, justice, fidelity, and kindness. If doing

2

Mark Timmons

good in this robust manner involves controlling for benefitting others, what about doing evil; does doing evil require malevolently controlling for acts that affect others negatively in a way parallel to doing good? Pettit’s answer is no; performing an evil act does not require acting from a robustly demanding self-sacrificial disposition; typical cases of evil result from pursuing self-interest, heedless of the harm done to others. The explanation Pettit offers for this asymmetry between doing good and doing evil is that the standards of respect, attachment, and virtue are demanding standards, whose rationale concerns the fact that humans are of limited altruism living in a world of scarcity. So by adhering to and acting from such standards everyone is generally better off than otherwise. In contrast, there are no parallel standards for doing evil, and no basis for having them. One implication Pettit draws from this difference is that it makes sense of the “Knobe-effect” (named after Joshua Knobe) which refers to the fact that people are generally more willing to ascribe intentionality when people do evil than they are in similar cases in which people do good. One interpretation of this effect (offered by Knobe) is that the folk concept of intentionality is value-laden and not purely explanatory. As Pettit argues, his explanation of the asymmetry between good and evil and what is required to meet standards of good enables one to resist this interpretation and preserve the explanatory character of the notion of intentionality. A commonly voiced objection to virtue ethical accounts of right action (virtue ethics) is the “action guidingness objection.” The objection is simply that such accounts of right action fail to provide agents with sufficient action guidance and so are of little use to ordinary morally conscientious individuals. The apparent root cause of this alleged problem for virtue ethics is that such views deny that the deontic realm is “codifiable.” In her “A Particularist but Codifiable Virtue Ethics,” Christine Swanton addresses this objection by defending a “target centered” virtue ethical account of rightness that explains the deontic status of actions in terms of what she calls “v-reasons”—reasons pertaining to action that involve “thick” ethical concepts, such as generosity, honesty, and beneficence. Very roughly speaking, on this view, an action’s rightness depends on and is explained by v-reasons in favor of the action. Such reasons are featured in virtue rules (v-rules, such as “be generous”) that serve to codify default reasons, which Swanton understands as working holistically in the manner featured in the version of moral particularism defended by Jonathan Dancy and others.

Introduction

3

The upshot of this combination of a virtue ethical account of right action in terms of v-reasons that operate holistically (particularism)—Virtue Ethical Particularism—is a view according to which the deontic realm is weakly codifiable by virtue rules, and so an account of right action that answers the action guidingness objection. Of particular importance in Swanton’s virtue ethics is a question about default reasons: why certain reasons have the status of defaults—as typically “switched on.” As Swanton explains, although an account of the nature of v-reasons and their default status is going to be complicated, in general terms the account will refer to the point and function of various virtues in relation to the goodness of a life. Mark LeBar’s “My Welfare and Yours” explores the nature of wellbeing, or what is good for persons as social creatures, against the background of an Aristotelian conception of practical rationality called “Virtue Eudaimonism” (VE) that LeBar (2013) has recently developed and defended at length. According to LeBar’s VE, one’s practical rationality is understood in terms of one’s end-seeking nature; some ends we seek both for themselves (final ends) and for the sake of further ends, but like Aristotle, the end of eudaimonia or well-being is an end we seek only for itself and something that is impossible to achieve without virtue. The picture of good life, then, is one of adopting a network of ends, forming a hierarchy, with those higher up in the network constraining one’s pursuit of other ends, and in a life well lived, shaping as well as contributing to one’s Ultimate End of well-being. LeBar’s particular focus in his contribution is how one’s social nature influences one’s hierarchy of ends and, in particular, how other individuals can figure among the final ends one adopts. The vehicle for LeBar’s exploration is the life of Dr. Josef Breuer, a character in Irvin D. Yalom’s 1992 novel, When Nietzsche Wept. The novel is a fictionalized account of a relationship the real Dr. Breuer—a wealthy and successful 19th century Viennese physician and mentor to his younger colleague Sigmund Freud—might have had with Friedrich Nietzsche, who, in the novel, challenges Josef to examine his own life, a life in which he feels trapped by his marriage to Mathilde, his family, and his career. In the novel, Yalom has Josef (under hypnotic suggestion) imagine what his life would be like freed from his present commitments to family, and Mathilde in particular. To bring into focus the significance of the well-being of others in one’s own life, LeBar explores some of the contours of the life of Josef in which his ends (as imagined) and

4

Mark Timmons

those of Mathilde fit well together, and contrasts this with the case in which their respective ends threaten to pull apart—the situation that Yalom has Breuer confront. Of particular significance, according to VE, is the “authority” that Mathilde, as Josef ’s spouse, has to influence Josef ’s living well. In exploring this issue, LeBar illustrates the manner in which the welfare of those in a committed relationship is entangled— an important aspect of living well that VE helps to illuminate. According to the view of moral worth that Nomy Arpaly defends, a morally right action has moral worth just in case and because the agent performs the action for the right reasons—the reasons that are in fact right-making and that make such actions praiseworthy. In her chapter, “Moral Worth and Normative Ethics,” Arpaly explores some implications of this understanding of moral worth, particularly as they bear on questions about the correctness of normative moral theories. A guiding idea here is that assuming people sometimes do perform morally praiseworthy actions for a variety of what are intuitively right-making reasons, one should expect a correct moral theory to pick out such reasons as right-making features (or trivially derivable from the theory’s rightmaking features). One implication Arpaly draws from this claim is that one should expect a correct moral theory not to be too “lofty” in the sense that the right-making features it picks out are features that philosophically unsophisticated agents (who are not steeped in moral theory) can understand and act on. Another implication for moral theory, as Apaly explains, concerns altruistic actions—actions motivated by a noninstrumental concern for another’s well-being. Intuitively, such actions have moral worth. If so, then advocates of moral theories that deny moral worth to such actions must either deny the intuition in question, or somehow find a way to accommodate the moral praiseworthiness that such altruistic actions merit. Given that altruistic actions have moral worth because the agent is responding to moral reasons provided by facts about another’s well-being, what about other reasons other than altruistic concern? In response, Arpaly notes that, on occasion, acting on a purely altruistic motive can lead a person to do what is in fact a morally wrong action. A paternalistic lie, for instance, can be motivated by altruism, yet be the wrong thing to do. In such cases, considerations having to do with well-being that provide one with reasons to help others are sometimes overridden by other moral considerations. If this is right, then arguably a correct moral theory should feature a plurality of basic

Introduction

5

right-making features of action. Reflection, then, on moral worth, leads to one or another form of a pluralist moral theory that recognizes more than one fundamental right-making feature of actions. If answers to questions about moral and political principles involve appeal to what individuals have reason to want, how are those reasons related to reasons an individual rightly has in making personal decisions about his or her own life? This is the central question addressed in T. M. Scanlon’s “Ideas of the Good in Moral and Political Philosophy.” Scanlon distinguishes two conceptions of individual good—what individuals have reasons to want—that he claims figure in answers to questions about the content and the acceptance of principles that moral and political philosophy are concerned with. The conceptions in question involve (i) claims about what one has reason to want in general and (ii) claims about what one has reason to want only because of how it affects one’s own life—what is good for an individual, including but not restricted to considerations of one’s welfare. Scanlon embraces what is called an “individualist” answer to the content question: answers to questions about the content of moral and political principles are properly based on considerations of what is good for individuals. With regard to the question of acceptance (reasons one has to accept moral principles as binding and to care about political principles) Scanlon’s view is that a complete answer to this question will require going beyond considerations of what is good for individuals to include considerations regarding what individuals have reason to want in general. Returning to the opening question, perhaps what individuals have reason to want in general and for themselves can serve to address questions about content and acceptance as well as form the basis of an individual’s personal decisions about his or her own life. However, Scanlon argues that this is not that case; that the conception of what is good for an individual that figures in arguments about the content of moral requirements and how it so figures differs from the conception that individuals rightly appeal to in guiding their own lives. But if so, this then raises the issue of “how individuals can be justified in accepting the requirements of morality and justice if the content of these requirements take account of how their lives are affected in ways that are alien to them.” In exploring this question, Scanlon considers how the acceptance problem arises in the thought of Sidgwick, Marx, Rawls, and Williams before sketching his own solution to this problem, which involves

6

Mark Timmons

(i) accepting a pluralistic account of the good for individuals, (ii) drawing conclusions about the relation between such goods and what an individual has reason to want all things considered, and (iii) explaining how these conclusions bear on questions about the acceptance of moral and political principles. Roughly speaking, morally praiseworthy agents are praiseworthy because of their appropriate sensitivity and responses to morally relevant considerations. But is it possible for an individual to pay too much attention to moral considerations, so much so and in a manner that is not morally praiseworthy? This is the question Jesse S. Summers and Walter Sinnott-Armstrong address in “Scrupulous Judgments.” Their particular focus is on cases of what is called “scrupulosity,” a recognized form of Obsessive Compulsive Disorder (OCD) that involves having “unwanted, anxiety-induced intrusive thoughts that may be secular or religious.” After explaining what OCD is and why scrupulosity is one of its forms, Summers and Sinnott-Armstrong focus on three characteristic features of moral scrupulosity. Moral perfectionism (not to be confused with perfectionist theories of value and politics) involves an individual having extremely rigid, high moral standards and a heightened sensitivity to falling short of them. Thought-action fusion is where an individual either believes her thoughts alone can increase the likelihood of certain events, or fails to distinguish doing something immoral and thinking about doing something immoral, treating the thinking and doing as almost morally equivalent. And, third, there is chronic moral doubt about the moral status of one’s actions and one’s own character. Given this conception of the morally scrupulous, besides being particularly morally sensitive, what distinguishes them from individuals whose moral sensitivity and responses are judged to be normal? According to the authors, the main difference between these groups “lies in the amount of anxiety or chronic doubt that plague people who suffer from Scrupulosity.” As they go on to explain, because of the manner in which the judgments and concerns of the scrupulous either produce or are driven largely by excessive anxiety, such individuals are arguably less praiseworthy from both epistemic and moral points of view. One conclusion that they draw from their depiction of the scrupulous individual is that moral motivation by anxiety, in the manner characteristic of such individuals, appears to be pathological and consequently undermines their reasoning and behavior as being genuinely praiseworthy.

Introduction

7

It is generally agreed that genuine acts of beneficence call for gratitude on the part of the beneficiary. However, the very idea of a duty of gratitude (sometimes referred to in the literature as a “debt” of gratitude) is puzzling. Tony Manela in “Obligations of Gratitude and Correlative Rights” articulates the puzzle by formulating three apparent platitudes regarding gratitude which, taken together, are in prima facie tension. According to the obligation platitude, gratitude entails a set of strict obligations—something a beneficiary must do. The directionality platitude says that obligations are owed to one’s benefactor, while according to the no-right platitude benefactors do not have a right to the fulfillment of these obligations. But if one has a strict obligation of gratitude to one’s benefactor, how can it be that the benefactor has no right against the beneficiary? After all, there is a tradition in law and moral philosophy according to which strict directed obligations always correlate to rights held by the person to whom the obligation is owed. Call this the correlativity thesis. Perhaps, then, duties of gratitude are an exception; the correlativity thesis does not hold for them. But, then, why don’t they? Manela’s aim in this chapter is to examine each of the platitudes and propose a way of dispelling the tension just described. In pursuing this aim, Manela first argues that the obligation and directionality platitudes are in fact true: at least many cases where gratitude is an appropriate response to one’s benefactor, one has a strict and directed obligation of gratitude toward one’s beneficiary. If the obligation and directionality platitudes hold with respect to gratitude, and if the tension in question is to be dissolved, then one might re-consider the no-right platitude. However, as Manela argues, this platitude is also true. After all, as ordinarily understood, having a moral right against someone that they do or refrain from action involves having the standing to demand that the one who is obligated comply, which arguably does not apply to the case of gratitude. But then, if we understand moral rights in this way, then together with the correlativity thesis we are left with the original tension. Nevertheless, as Manela points out, even if a genuine benefactor does not have the moral standing to demand an act of gratitude, it remains true that at least in some cases, she has standing to express justified resentment toward the beneficiary—a form of expression that falls short of the forcefulness of issuing a demand. (He illustrates the point with a case of grateful nonmaleficence.) To capture this form of justified response by the benefactor, Manela introduces the idea of an

8

Mark Timmons

imperfect right as a kind of standing characterized “by license to resent and remonstrate in the face of noncompliance,” arguing that it is an important but overlooked component of self-respect; a feature it shares with perfect rights that ground demands. In light of the notion of an imperfect right, Manela argues that we should revisit the correlativity thesis revising it to say that strict directed obligations entail rights that may be either perfect or imperfect. This then allows us to hold on to the no-right platitude, interpreted as claiming that a benefactor does not have a perfect right to the fulfillment of an obligation of gratitude, though she does have a correlative imperfect right against her beneficiary that it be fulfilled. Manela concludes with some brief remarks about the political and moral implications of recognizing the category of imperfect rights. One experiences moral anxiety when one is both uncertain and troubled about some moral issue. But what exactly is moral anxiety? And of what value (if any) is it to those who experience it? Moreover, is it something persons ought to cultivate? In “Moral Anxiety and Moral Agency,” Charlie Kurth addresses these questions. As he explains, generally speaking the so-called formal object of experiences of anxiety—what they are about—is problematic uncertainty; to feel anxious about some situation is to respond to the situation as involving one or another type of problematic uncertainty. Different species of anxiety are distinguished by the nature of the uncertainty in question. Kurth proposes that social anxiety, which has been extensively studied by psychologists, can serve as a model for understanding moral anxiety. Social anxiety involves experiences of problematic uncertainty about how others will view one’s social worthiness and worry that one may be the target of social criticism. To have such experiences is to respond to one’s situation in an evaluatively loaded way, normally prompting one to engage in certain forms of social caution, and thus serving to help one properly negotiate one’s social world. In this way, social anxiety is a valuable emotion. Moral anxiety is in some ways similar. One experiences moral anxiety as an adverse emotional response when, for instance, one is uncertain and troubled over whether breaking a particular promise is permissible. The formal object of moral anxiety, then, is problematic uncertainty about the correctness of a moral judgment (and associated decision), and thereby to see one’s moral judgment and corresponding decision as a possible target of moral criticism. Seeing one’s circumstances in this particular evaluatively loaded way will normally prompt one to engage

Introduction

9

in certain epistemic behaviors including further reflection and deliberation about the uncertain matter in order to resolve one’s uncertainty. In this way, like social anxiety, moral anxiety is a valuable emotion. Given this description of moral anxiety—what it is and the valuable role it can play in one’s life—we can now ask our normative question: is such anxiety something one ought to cultivate? In addressing it, Kurth sets forth an argument for the claim that moral anxiety is something one ought to cultivate by locating it as playing a central role in good moral decision making and agency. He then proceeds to defend the premises by way of fleshing out the argument, appealing in particular to recent work in social and clinical psychology. Kurth concludes with remarks about the cultivation of moral anxiety, the significance of psychological conflict that accompanies such anxiety, and about a distinctive picture of moral psychology that emerges from his account of moral anxiety. How and in what way do the prudentially relevant factors in an individual’s life contribute to the overall prudential value of that individual’s life, either over a particular span of time or over the course of her or his entire life? According to localism, the prudential values in a person’s life at some moment in time supervene on how things are at a moment for an individual in isolation from events at other times, and so an individual’s welfare is entirely a function of local goods and evils. In opposition to this view is globalism, the view that the prudential value of an individual’s life supervenes on select global features of that life, such as the arrangement of the parts of that life or the “shape” a life takes. Antti Kauppinen in “The Narrative Calculus” defends narrativism, a version of globalism, according to which, roughly, narrative relations among the events and segments of a person’s life contribute to that life’s prudential value. As Kauppinen remarks, “the core idea is that narrative value is the measure of how we fare as temporally extended planning agents vulnerable to fate. We are not just subjects of better or worse experiences at various moments, but active shapers of our stories.” But what are narrative relations? And when are they prudentially good? Kauppinen addresses these questions arguing that an individual’s life is narratable because individuals are planning agents, and events in one’s life have narrative significance in virtue of how they figure in goaldirected pursuits. Further, three main factors contribute (though not exclusively) to the narrative value of events: (1) an Instrumentality Factor—the positive or negative contribution of those events to one’s

10

Mark Timmons

individual goals and pursuits, (2) a Value Factor—the value of those goals, and (3) a Desert Factor—the degree to which success in achieving a goal is deserved in virtue of the exercise of the individual’s agential capacities. Kauppinen proposes an admittedly rough heuristic—a narrative calculus—for determining the prudential value of an event in one’s life that he expresses as a multiple of these three factors. Of course, as Kauppinen notes, in addition to our interest in the narrative value of a single life event, we are generally more interested in the value of a sequence of life’s events. He thus proposes a straightforward additive characterization of the narrative value of the sequences of events during a particular time period of an individual’s life, which generalizes to a narrative value for the whole of a person’s life. Since his narrative calculus assigns a value to events on the basis of their contribution to giving one’s life a progressive shape, his view represents a progressivist version of narrativism. In defending his narrativist view, Kauppinen argues that it is superior to temporal distributivism—a competing globalist view according to which mere temporal arrangement of local benefits contributes to life’s prudential value. He concludes by defending his version of narrativism against various objections. A common gloss on moral theory is that it purports to provide a unified account of which actions (and practices) are morally required— an account that not only correctly distinguishes the morally required from actions that are morally permissible and those that are not, but explains why action and practices belong in one or another of these deontic categories. Moreover, it is normally taken for granted that a moral theory should recognize a full range of morally important distinctions among actions. So, for instance, assuming that supererogatory actions are possible, one should revise the above gloss to countenance actions that are not morally required but are nevertheless morally good to perform. In “Morality Within the Realm of the Morally Permissible,” Elizabeth Harman argues that this more accommodating conception of moral theory is still too narrow; it does not recognize other morally significant distinctions within the category of the morally permissible— distinctions that do not merely follow from a theory’s account of the morally required. How then should moral theory be represented? In tackling this question, Harman argues for a conception of moral theory that recognizes the category of what she calls “morally permissible moral mistakes.” A morally permissible moral mistake is characterized as an

Introduction

11

action which, although it is morally permissible (and thus not morally required), it is nevertheless an action one should perform for purely moral reasons. Such actions belong in the category of the supererogatory and so are not required, but because of the moral reasons that favor the action, one would be making a moral mistake in not performing them. After making an intuitive case for this category and then responding to possible objections, Harman concludes by formulating a conception of moral theory that explicitly allows for instances of the category in question. Cases of collective harm, as Julia Nefsky describes them in “Fairness, Participation, and the Real Problem of Collective Harm,” are ones in which people collectively engaging in or refraining from a certain activity bring about a morally significant outcome, but no individual act appears to make a difference to that outcome. In light of one’s apparent superfluous contribution to the desired outcome, one can reason as follows: Since my action will not make any difference to the outcome, I have no reason to act. The challenge, for those who disagree with the conclusion, is to find the argument’s flaw. There would seem to be only two options: deny the premise and argue that (perhaps surprisingly) a single individual’s action does make a difference to the outcome, or deny the implication and claim that an individual does have reason to participate even if her action (taken alone) will have no effect on the outcome. Nefsky’s chapter explores this second option, by considering three proposals, two of which appeal to considerations of participation and the third to considerations of fairness. She argues that all three proposals suffer from a common problem, namely, that although all of them identify considerations other than the difference to the desired outcome in collective harm cases, it is not clear that the considerations in question can plausibly serve as reasons for participation, given that an individual’s lone action makes no difference to the desired outcome. As Nefsky observes, as long as one’s act would be instrumentally superfluous, the reasons these proposals are offering don’t seem to apply; they all face what she calls the “Superfluity Problem.” One might suppose that this problem only afflicts consequentialist moral theories, and so theories such as virtue ethics or Kantian moral theory that are not exclusively concerned with outcomes of actions will have the resources to plausibly deny the problematic implication and account for one’s obligation to participate. But Nefsky argues that the Superfluity Problem generalizes and besets even non-consequentialist

12

Mark Timmons

moral theories. In light of apparent failures to plausibly deny the problematic implication, one might think that if the explanatory challenge can be met, it would have to be by denying the premise and arguing that an individual’s act is not superfluous after all. Elsewhere, Nefsky (2011) has argued against going this route. If she is right about the two routes, it may seem we have come to a dead end. But this is not Nefsky’s position. She concludes by noting that reflection on the failure of the two options suggests a move that one could make, namely, argue for the claim that even if an individual action cannot make any difference to the outcome in question, such actions can still play a significant instrumental role in bringing about an outcome; a task she leaves for another paper. Ernesto V. Garcia in “The Virtue of Authenticity” defends a conception of authenticity as a personal value according to which, whatever the correct conception of one’s true or authentic self turns out to be, being authentic involves relating to one’s self in the right kind of way. Garcia develops his view by first critically examining three popular conceptions of authenticity and then developing his own conception, which he labels, “existential authenticity.” In developing this conception, Garcia leaves open how precisely to understand the notion of one’s true self, taking it to be constituted by “those features which, objectively speaking, have a more significant or wide-ranging impact on how I think, feel, and act, than those which do not.” With regard to relating to oneself in the right kind of way, Garcia sets forth and defends the following three conditions that he claims must be met in order to achieve (to some degree) his reflective form of existential authenticity. One must: (1) have an accurate understanding of one’s true self, (2) live in conformity with one’s accurate self-understanding, and (3) care about what type of person one is and live in conformity with one’s accurate selfunderstanding, at least to some extent, out of such concern. As Garcia explains, one can understand existential authenticity as fitting with an Aristotelian conception of virtue in two ways. First it represents a mean between the deficiency of not caring enough or in the right way about oneself and the excess of narcissism. And second, because it does not impose moral constraints on what is to count as one’s true self (which allows for cases in which one’s true self is morally evil), it follows that in order to constitute a truly virtuous character trait, authenticity, so understood, requires the ‘unity’ of the virtues in order to count as unconditionally good.

Introduction

13

In the final chapter, “What Normative Terms Mean and Why it Matters to Ethical Theory” Alex Silk explains how clarifying the language used in moral evaluation can contribute importantly to ethical theorizing in ways that often go unappreciated by moral philosophers. For instance, it is quite common in ethics to treat necessity modals used for normative evaluation as equivalent. However in descriptive linguistics a distinction in strength is drawn between such modals as ‘ought’ and ‘should’ (weak) and ‘must’ and ‘have to’ (strong). Another distinction among modals concerns “objective” and “subjective” uses of them. One uses a necessity modal subjectively when one endorses a norm that grounds the modal claim. Objective uses of modals do not commit the speaker to such endorsement. As Silk notes, these distinctions in strength and subjectivity cut across one another. For instance, while the strong modals, ‘be required to’ and ‘have to’ tend to be used objectively, other strong modals such as ‘have got to’ and ‘must’ tend to be used subjectively. To illustrate the importance of these varied uses of modal language in normative evaluation, Silk turns to disputes over moral dilemmas, supererogation, and judgment internalism. Regarding moral dilemmas, Silk points out that on the one hand there is a tradition (associated with rationalism) that denies the very possibility of genuine moral dilemmas. On the other hand, there is a tradition in logic that supports the thought that such dilemmas cannot be ruled out on purely logical or semantic grounds and so such dilemmas are at least coherent possibilities. Silk’s proposal, based on linguistic evidence, is that by carefully attending to differences among modals, it is possible to capture both of these seemingly opposing views. The sentence, “I must help Alice and I must help Burt, but I can’t help them both” is naturally taken to be inconsistent, given the strong modal used subjectively. By contrast, the sentence, “I have to help Alice and I have to help Bert, but I can’t help them both” is naturally understood as consistent, given the strong modal used objectively. So, the difference in one’s sense of speaker consistency illustrated by these two sentences is nicely captured in terms of objective versus subjective uses of modals together with the differences in their strength. An important upshot here is that genuine dilemmas cannot be ruled out simply on linguistic grounds by framing the issue in terms of strong modals used subjectively. Rather, as Silk explains, what this linguistic evidence helps to highlight is that questions about genuine moral dilemmas are properly addressed by substantive normative theory, not by logic or semantics.

14

Mark Timmons

Regarding the controversy over the coherence of supererogation, Silk points out that the so-called paradox of supererogation, appealed to by anti-supererogationists, tends to get its intuitive grip by being framed in terms of strong modals, while those intuitions are not as strong if one expresses the alleged paradox in terms of a weak modal like ‘ought.’ Again, the point is not that the pro-supererogationists are correct; rather, either side of this debate will have to make their case on grounds that are independent of normative inferences funded by inattention to normative language. Silk makes similar observations about the dispute over judgment internalism: conflicting intuitions over the truth of this thesis are often reflected in the kinds of differences in strength and subjectivity of necessity modals that parties to this dispute employ in disputing whether there is a necessary connection between sincere moral judgments and motivation. The subjective use of the strong modal ‘must,’ as in “I must reply to the invitation” makes internalism compelling, while this is not the case with judgments employing a weak necessity modal such as ‘should.’ Thus, as Silk recommends, careful attention to normative language should be used when conducting this debate. Having demonstrated the importance of attending to linguistic and related semantic considerations for ethical theorizing, Silk concludes by discussing a worry about the terms that are used to characterize the class of normative judgments which, for example, would seem to make a thesis like judgment internalism either clearly empirically false or trivial. REFERENCES

LeBar, M. 2013. The Value of Living Well (Oxford and New York: Oxford University Press). Nefsky, J. 2011. “Consequentialism and the Problem of Collective Harm: A Reply to Kagan,” Philosophy & Public Affairs 39: 364–95. Yalom, I. D. 1992. When Nietzsche Wept (New York: Harper Perennial).

1 The Asymmetry of Good and Evil PHILIP PETTIT

The thesis of this paper is that there is a marked difference between doing good and doing evil—in particular, doing good and doing evil to others—as we conceive of these actions in everyday life. Doing good to others often means doing them good robustly—controlling for good effects—whereas doing evil rarely has such a robust or modal character. This asymmetry thesis has a number of implications and I will concentrate in particular on one.1 This is that it enables us to make sense of the Knobe effect, so-called: the effect, as it appears, whereby we are generally more willing to ascribe intentionality in cases where people do evil than in otherwise similar cases where they do good. The paper is in five sections. In the first I look at some important issues in the ontology of behavior, distinguishing between the acts or deeds whereby we produce effects and the actions or doings whereby we control for effects. In the second I show how we expect those who do good to others to control robustly for beneficial effects, being prepared at the limit to pay a cost in self-interest for securing that result. In the third I argue that we do not have the same expectation with those who do evil to others; they may aim at bringing about bad effects but not generally at any cost in selfinterest. In the fourth I offer an explanation of this asymmetry. And then in the fifth section I show that the asymmetry between doing good and doing evil helps to explain the Knobe effect in a particularly satisfying way. 1. ON ACTIONS AND THEIR EFFECTS

In order to do good, in particular to do good to another, you have to perform actions that have good effects on that other, contributing to 1 For other implications see (Pettit 2015). The argument in this paper makes use of material in that book, in particular material from chapters 5 and 6.

16

Philip Pettit

their welfare in some sense. And in order to do evil, in particular to do evil to another, you have to perform actions that have bad effects on that other, reducing their welfare in some sense. But if we are to understand what either involves we need to be a little clearer about our conception of actions and about the variety of effects that actions may have. Options Whenever you act, you choose from among a set of options—mutually exclusive and jointly exhaustive possibilities—that are accessible to you. The options are possibilities such that, depending on your preference, you are able in the circumstances on hand to realize or not realize any one of them. Suppose that you have a choice in acting between turning to the left, turning to the right, or standing still; or between sending a message by post or email or fax; or between speaking your piece or opting for silence. Each option in such a choice is something that you have the wherewithal and the knowledge to implement or enact. And in the normal case this will be manifest to you: you will not be deceived about the alternatives between which you have to choose. On this standard conception of an option, all that determines whether or not the possibility in question is to become actual is your will or your wish; you need only try to realize it and, given how the world is, success is guaranteed. Suppose then that it is a matter of chance whether your flipping the switch will turn on the light or not; suppose that there is an element of indeterminacy about this. In that case, the option before you is that of flipping the switch, not that of turning on the light; whether or not the light comes on is an outcome associated under a certain probability with that option. If things are contingently such that flipping the switch will turn on the light, however, then turning on the light is indeed an option for you; it is something that you can just do, although you will do it only by flipping the switch. Acts or Deeds With this conception of an option in place, we can identify an act or a deed as the enactment or realizer of an option: an event that has the effect of actualizing the possibility represented by the option. If we think of the act as a particular event that is capable of bearing many properties, as I shall do here, then we can think of it as having a behavioral character, given by the option it realizes; an instrumental character, given by the

The Asymmetry of Good and Evil

17

way in which it realizes the option; and a variety of circumstantial effects in virtue of that behavioral and instrumental character. Under this way of thinking, then, the act with the behavioral character of turning on the light is the same act as that of flipping the switch—this gives its instrumental character—and the same act as that of surprising the burglar.2 Considered as a particular event, the act may have a variety of effects and in ascribing an effect we naturally take the act under a description that does not presuppose the effect and that highlights a property relevant to its occurrence. We do this when we describe the act as one of flipping the switch in ascribing to it the effect of turning on the light. The property of flipping the switch is causally relevant to closing the electrical circuit and offers the appropriate aspect under which to take the act in the ascription of that effect. We follow the same logic in describing the act as one of turning on the light when we ascribe the effect of surprising the burglar. It is under its aspect as an act of turning on the light, not under its aspect of an act of flipping the switch, that it surprises the burglar; after all, the burglar would not have been surprised if you had flipped the switch but the light had not come on. The effects of an act or deed, taken as the realizer of an option, come in two broad varieties. First there are causal effects of the kind illustrated so far: the effect of the act in closing the electrical circuit, thereby putting on the light, and the effect of the act in surprising the burglar. But an act may also have effects of a non-causal or constitutive kind. This sort of effect consists in the instantiation of a distinct property by the act, where the instantiation does not come about as a result of any causal process. Thus in turning on the light you may be fulfilling a promise—say, a promise that you made to your partner to go downstairs and check on whether there is a burglar in the house. The effect whereby your act consists in keeping that promise does not materialize causally; it is logically bound to materialize in the event of your actually going downstairs to look around, given what you had said to your partner prior to descending. Actions or Doings All of the effects illustrated so far, causal and constitutive, are circumstantial effects of what you produce in enacting an option; they are 2 In taking this line, I follow (Anscombe 1957; Davidson 1980). For an alternative way of individuating actions, see (Goldman 1970).

18

Philip Pettit

effects such that in the circumstances the act involved is sufficient, in virtue of its behavioral and instrumental character, to generate them. But there are other effects that you may have in acting that are not due to the character of the act as such but rather to a more complex antecedent: the combination of that act with the particular disposition out of which it was performed. Suppose that despite having a power of interfering with me at will you exercise restraint in our interactions, allowing me to choose as I will in the normal range of personal discretion: in what we generally think of as the basic liberties that any autonomous agent ought to enjoy. Does this mean that you give me respect, as we say? Not necessarily, since any of a variety of ulterior motives may lead you to exercise restraint. It may be that you want to impress a third party, or that you are just interested in seeing how I use my discretion, when given my head. And in any such case it would not be true that you show me respect or that I enjoy respect at your hands. What I would enjoy rather is good luck or indulgence. In order for your restraint to have the effect of showing me respect— in order for it to constitute a display of respect—you must exercise restraint out of a particular disposition. You must restrain yourself because of having renounced any act of interference in my basic liberties, putting the option of interference more or less beyond your reach. You must perform the act out of a respectful, self-denying disposition.3 Disposition-dependent effects of the kind illustrated by respect are bound to be constitutive rather than causal in character. Going downstairs to look around constitutes the keeping of a promise in the context of your earlier utterance to your partner. And exercising restraint constitutes a display of respect in the context of manifesting a disposition to renounce any form of interference with my basic liberties. The property of displaying respect attaches to the act in virtue of its origin in that disposition, not in virtue of being causally generated by the act on its own. Thus you cannot be said, strictly speaking, to give me respect by avoiding interference as you can be said to turn on the light by flipping the switch; you give me respect by avoiding interference only on the 3

That may not be enough for this effect to materialize; it may be necessary, in addition, that you are externally constrained to act in accord with the disposition, say by the law (Pettit 2015). But it is certainly necessary for giving me respect that you act out of such a disposition. The effect is disposition-dependent.

The Asymmetry of Good and Evil

19

extra assumption—perhaps salient from context—that you act out of a suitable disposition. Given that disposition-dependent effects are not strictly effects of your acts, why should we think of them as effects that you bring about? The answer is that they are effects of what you do, not in the sense of being effects of the acts you produce, but in the sense of being effects of the control you exercise in acting. In giving me respect you actually produce an act of restraint, avoiding interference in my exercise of a basic liberty. But what is also true is that you do this because of controlling for restraint: because of bringing it about that no matter how circumstances vary in certain ways—strictly, vary according to your presumptively reliable beliefs—you will still avoid interference in my exercise of that type of choice. You might not restrain yourself if the cost were a loss of life or fortune for someone, whether that be you, me or a third party. But you would restrain yourself across a large range of scenarios where it was more or less inconvenient to avoid interference. You display restraint robustly—actually and across relevant scenarios— and in that sense control for displaying it. Given that we describe what you produce in performing as an agent as your acts and deeds, we may speak of the controlling interventions that lead to such acts or deeds as your actions or doings. A dispositiondependent effect of giving me respect may not be an effect of your act of restraint but it certainly is an effect of your action. It is something that comes about as a result of what you control for and as a result of what you do in that richer sense. For every action in this sense there will almost always be a corresponding act and it will share disposition-independent effects with that act.4 But while the action will have disposition-dependent effects essentially, those effects will be irrelevant to the act: they will not be effects of the act in the sense in which disposition-independent effects are. In performing the action of showing me respect, you produce the act 4 Although true in most cases, even in cases of omissions, this does not hold universally, because you may control in a virtual or standby manner for a certain effect. Think of the cowboy who rides herd on his cattle. He does nothing that helps lead his cattle in the direction of the railhead but he does control for their taking that direction. He controls for their doing so in the sense that while he does nothing in the actual case, where they go in the desired direction, he is ready to intervene should any of them wander off track. For the notion of virtual control see (Pettit 1995).

20

Philip Pettit

of refraining from interference. And while the action and the act will have the same disposition-independent effects, say in allowing me to choose as I wish, they will be distinguished by the fact that the respectful disposition is essential to the action but quite irrelevant to the act. In adopting this way of thinking about human behavior, I join forces with Christine Korsgaard (2009, 12). She takes herself to follow Aristotle and Kant—and to oppose ‘many contemporary moral philosophers’—in making a distinction very like that which we have made here. She uses the term ‘act’ to pick out the option realized by the agent, as we have done, and she takes the action involved in realizing it to vary with any variation in intention or motive or policy and so with any difference in disposition-dependent effects. ‘So for instance,’ she says, ‘if you choose to dance for the sheer joy of dancing, then dancing is the act, and dancing for the sheer joy of dancing is the action. We might contrast it to the different action of someone who dances in order to make money, or to dodge the bullets being shot at his feet.’5 What is the point of adopting this language of action or doing in addition to the language of act and deed? It reminds us that to perform an action is not just to produce an act or deed, it does not matter by what causal route. To perform an action is to produce the act or deed in manifestation of this or that intention or motive or policy. It is to produce the act in response to the consideration or set of considerations that become relevant in the presence of such a disposition. To perform the action, as we put it, is to control for the production of the act in any circumstances where the relevant considerations are registered. Thus, to show me respect is to control for the production of restraint in any circumstances, the actual ones included, where you register that my exercise of a basic liberty is at stake. Acting always involves this sort of control and the control exercised reflects the dispositions by which you are moved.6

5 I am grateful to Alec Walen for pointing out that the distinction was already made, and made in the same terms, by David Ross (2002 New Edition) in 1930. 6 Thinking of actions as controllers in this way may guard against a tendency in the moral philosophy of action to ignore the disposition-dependent properties of what we do and to treat questions of disposition as bearing only on the evaluation of agents, not on the evaluation of their behavior (Bennett 1995; Scanlon 2008). Barbara Herman (2011, 89–90) opposes that tendency, for example, in arguing that in ethics we should ‘make motive relevant to determinations of wrongness’.

The Asymmetry of Good and Evil

21

2. DOING GOOD

Disposition-Independent and Disposition-Dependent Benefits There are many ways in which you may do me good. You may choose acts or deeds with a welcome behavioral or even instrumental character. And, even more saliently, you may choose acts or deeds with a variety of congenial effects, causal and constitutive: in the nature of the case, congenial, disposition-independent effects. Thus your acts or deeds may give me pleasure or satisfaction, thereby scoring high in utilitarian terms. They may provide me with money or opportunity or other such resources. They may testify to my standing, increasing the esteem in which I am held in our society. And they may also do me good on the basis of constitutive linkages rather than causal connections. Your communication may constitute truth-telling, for example, thereby increasing my access to knowledge. And your action may constitute promise-keeping, thereby giving me the satisfaction of an associated claim. But apart from these straightforward ways in which you may do me good, there are many ways of doing me good that involve dispositiondependent effects. You do me good in these ways, not just because of the acts or deeds you perform, but because of your actions or doings. You do me good because of the dispositions out of which you act: because of what you control for in producing your acts or deeds. Respect The example of respect serves to illustrate the possibility. You do me important good by almost any account of value in showing me respect. And, as we have seen, you show me respect only insofar as you exercise restraint in your dealings with me out of a suitable disposition. You do not act out of a motive of opportunism or whim or whatever, but out of a respectful disposition. You act out of a disposition to exercise restraint across a range of possible circumstances—you control for displaying restraint in those scenarios—including the circumstances that actually materialized. Suppose I have a choice, within the sphere of the basic liberties, between doing either of two things: speaking out against the government, speaking in favor of the government, or remaining silent. And suppose that I choose to speak in favor of the government and that you

22

Philip Pettit

exercise self-restraint, allowing me to say my piece. Do you show me respect? Well, you do not show me respect if your restraint comes only of the fact that you are a government supporter yourself. You show me respect in your exercise of restraint only if it is the case that you would also have exercised restraint if I had chosen one of the other options: only if you controlled for the exercise of such restraint across those possibilities as well as in the possibility actually realized. Putting this from my perspective rather than yours, I enjoy your actual respect only if you act in a certain way as things actually are—you exercise restraint—and only if you act a certain way—again you exercise restraint—under certain unactualized possibilities. The good of respect that I can enjoy at your hands requires you to exercise restraint, not just actually, but with a certain robustness: that is, across a certain range of possible variations on actual circumstances. But how can you provide me with this robustly demanding good: this good that requires things to be thus and so in non-actual as well as in actual circumstances? The obvious answer is: only by performing the required act—the exercise of restraint—in manifestation of a suitable disposition; only by adopting a respectful course of action. Controlling for the exercise of restraint in the sense associated with respect means controlling for its realization even in scenarios that are quite unlikely to materialize. It means controlling for restraint in the scenario where I wish to speak against the government, for example, even if it is highly improbable that I should want to do so. Thus controlling for restraint is very different from maximizing expected self-restraint. If you wanted to maximize expected self-restraint in dealing with me, you could ignore scenarios that were almost certain not to materialize, such as the scenario where I wish to speak against the government. You would need to ready yourself to provide restraint across various scenarios only in the degree to which they were likely to materialize; to devote resources to unlikely scenarios would be utterly wasteful. Making restraint suitably robust is more demanding that making it suitably probable, then; controlling for restraint is more demanding that maximizing its expectation. I now want to suggest that as the delivery of respect requires you to control for restraint in this sense, acting out of a suitable disposition, so the delivery of other important goods requires you to control for acts associated with them in the way that the act of self-restraint is associated

The Asymmetry of Good and Evil

23

with respect. There are any number of goods you can give me that, like respect, are robustly demanding in character: they require that you perform a certain sort of act, not just actually, but in some non-actual circumstances. Like respect, these are goods that you can put my way only by virtue of the dispositions out of which you act and so only by virtue of what you control for. The feature of these goods that is of particular importance, as we shall see, is that they require you to perform certain beneficial acts, not just when it is personally suitable, but also when this requires a certain frustration or even sacrifice on your part. There are two categories where such goods figure prominently. One involves attachments like those of love and friendship, where you are bound to another person in their individuality, albeit under the aspect of a lover or friend or whatever. The second involves virtues of a substantive kind—for example, honesty or justice, tolerance or fidelity—that are targeted at another’s need for information or acceptance or whatever. They are different from respect in the sense that the need involved is not conditional on the agent’s making one or another choice: the need, should I choose X, not to be interfered with in that choice; the need, should I choose Y, not to be interfered with in that choice; and so on. Love and Friendship Take the good of love or friendship that is celebrated under almost any conception of human welfare. When you show me love or friendship, as we generally put it, you must perform a certain act or service: you must give me a form of care in the case of love, let us say, a form of favor in the case of friendship. But providing that actual service—performing a caring or favoring deed—is not enough for you to give me the good of love or friendship. After all, you might have performed it out of an opportunistic sense of your own self-interest or even out of a disinterested sense of what you are impersonally required to do. In order to provide me with love or friendship you must provide me with that service out of a disposition to do so across a range of possibilities, including possibilities where doing so impacts on your self-interest or frustrates certain cherished projects. You must provide me with the care or favor more or less robustly, not just contingently on its fitting well with your prudential or moralistic projects. It is by controlling for giving me care and favor in those possibilities, as well as in the actual scenario, that you give me love or friendship.

OUP CORRECTED PROOF – FINAL, 17/7/2015, SPi

24

Philip Pettit

Thus consider the situation charted in Oscar Wilde’s comedy, ‘The Importance of Being Earnest’. Jack Worthing uses the pseudonym ‘Ernest’ on his visits to London, as he wishes to retain a certain anonymity in the big city. Under that pseudonym he attracts the attentions of Gwendolen, the cousin of his friend, Algernon, and she declares her love for him in response to his own confession of attachment. But does she really offer him love? In explaining her attachment, Gwendolen suggests that it is as contingent as his pseudonym: ‘my ideal has always been to love some one of the name of Ernest’, she says. ‘There is something in that name that inspires absolute confidence’. And as if that were not sufficiently bewildering, she adds: ‘The moment Algernon first mentioned to me that he had a friend called Ernest, I knew I was destined to love you’. Jack remonstrates with her, of course, explaining that he would much rather be called ‘Jack’. But Gwendolen will have none of it, expounding with enthusiasm on the charms of ‘Ernest’. ‘It suits you perfectly. It is a divine name. It has a music of its own. It produces vibrations.’ The comedy of this situation derives from the fact that if Gwendolen is to count as giving Jack her love, then the care she purports to give him must not be contingent on something as incidental as his having a name that appeals to her. To give someone love is to give them care—to perform deeds of care—with a degree of robustness that rules out such sensitivity to personal taste. ‘Love is not love’, as Shakespeare puts it, ‘Which alters when it alteration finds’. To love someone is to provide them with care across a range of possibilities that includes the possibility where their name changes, as well as various possibilities where there are other uncongenial changes in the beloved, or the care is more inconvenient for the lover in other regards. It is to provide them with care out of a disposition that ensures a high degree of robustness; in particular, a degree of robustness under which the care is not provided contingently on its fitting the precise tastes or advancing the personal interests of the lover. We need not rule on the exact range of possibilities over which you must give me care in order to count as loving me, as we need not rule on the exact range of possibilities over which you must give me favor in order to count as my friend. Our common conceptions of love and friendship are vague on those margins. You will love me to the extent that you give me suitably robust care and you will be treat me as a friend

The Asymmetry of Good and Evil

25

to the extent that you give me suitably robust favor. You will love me or treat me as a friend to the extent that you control in a relatively benevolent measure for giving me care or favor across those various possible situations where the considerations that trigger love or friendship obtain. But however vague at the margins, the possible situations across which your care or favor must persist include many where there is a certain, self-sacrificing cost to you in providing it. Love and friendship require a disposition to care for the lover and favor the friend, even when doing so is inconvenient or goes in other ways against the grain. By all accounts you do me good insofar as you show me love or friendship, as by all accounts you do me good insofar as you show me respect. And that means that by common consensus you do me good insofar as you act out of the dispositions associated with love and friendship, controlling in a benevolent measure for giving me the associated care or favor. The love or friendship you provide comes about as a disposition-dependent effect of your providing care and favor. It is an effect of your action but not, in our sense, of the act that you produce in the course of taking that action. Honesty, Justice, and other Virtues This claim in the case of attachments like love and friendship is parallel to a similar claim that can be made in the case of various virtues, in particular virtues that govern how people treat one another. As you do me good in giving me love or friendship so, by our received ideas, you do me good in giving me the benefit of your honesty or justice, your tolerance or acceptance, or your fidelity as a promise-keeper. And as love or friendship require you to control in a suitable measure for certain good effects like care and favor—to produce them, in particular, with a suitable degree of self-sacrificing robustness—so these virtues require you to control in a corresponding degree for parallel effects. The point is readily illustrated. In order to give me honesty you must tell me the truth, in order to give me justice you must satisfy my claims against you, in order to give me tolerance or acceptance you must treat me decently, and in order to give me fidelity you must keep your promises. But it will not be sufficient in any of these cases that you contingently provide such truth-telling or claim-satisfaction, decent treatment or promise-keeping. You must perform the required acts or deeds, not just as things actually happen to be, but also across a range of

26

Philip Pettit

unactualized possibilities, including possibilities where it is relatively costly for you to do so. You must control in a relatively benevolent degree for realizing the good options in question. As in the case of attachments like love and friendship, our common conceptions of the virtues will dictate an indeterminate but demanding threshold that you must meet in controlling for the appropriate effects if you are to count as honest or just, tolerant or accepting or faithful. And as in the other case, we need not worry particularly about where this threshold lies. The important point to note is that the threshold is sufficiently demanding to imply that you do not count as honest unless you are prepared to tell the truth when it is inconvenient, you do not count as faithful unless you are prepared to keep a promise that you would prefer not to have made, and so on. This means that in the area of virtue, as in the areas of respect and attachment, we routinely think that doing good requires that you control for taking good options, and do not just happen to produce them or produce them for contingent reasons of convenience or interest. You do me good as a result of the dispositiondependent effects of the actions you perform. 3. DOING EVIL

Do Evildoers Exercise Malevolent, Self-sacrificing Control? We have seen that while doing good may sometimes mean just realizing good options—doing good deeds or acts, as we may say—it routinely means controlling for their realization. It means ensuring for a range of possibilities, not just for the actual situation on hand, that you take those options, where the possibilities include some in which the performance runs against your self-interest. This observation raises a straightforward question about doing evil as distinct from doing good—specifically, for doing evil as distinct from doing good to another human being. Does doing evil require controlling for evil deeds in the way in which doing good often means controlling for good deeds? In particular, does it require controlling for evil deeds even across possibilities where those deeds impose a self-sacrificial cost on the agent? There can be little doubt but that the effects in virtue of which we might say that you do evil, imposing some costs on me, can include disposition-dependent effects. If you hurt me out of thoughtlessness that is not so bad as hurting me out of indifference, which in turn is not so

The Asymmetry of Good and Evil

27

bad as hurting me out of enmity. But the question in the case of doing evil is whether there are cases parallel to those illustrated by attachments and virtues where the evils involved require controlling in a similar, selfsacrificing measure—controlling malevolently, not benevolently—for the performance of harmful deeds. Do various forms of evil-doing require that you should perform certain evil deeds with a similar degree of robustness? Do they require in the same way, not only that you actually perform those evil acts, but that you would perform them across a range of possibilities, including possibilities where the performance imposes a cost in self-interest? Doing good deeds with the robustness required by attachment, virtue or respect has a somewhat heroic character, since the control involved demands some self-sacrifice on your part; it requires that you set aside your self-interest in a certain measure, being prepared to do those deeds even when that impacts on your own interests. The question here is whether there are robustly demanding evils, parallel to the robustly demanding goods of attachment, virtue and respect, that require a similar sort of self-sacrifice and a correspondingly heroic character. Answer: Not in General, Only in Exceptional Cases The answer, intuitively, is that there are not many evils of this robustly demanding kind. When you do evil you certainly perform an evil deed, realizing an option that has negative effects on me, whether of a disposition-independent or disposition-dependent character. But the sort of evil pursued does not often require that you should control in a heroic degree for harming me in that way. It does not often require that you should be willing to impose the harm, even in scenarios where it is costly for you to do so. There are certainly a limited number of cases where evil-doing involves controlling malevolently for imposing a harm on another, whatever form that harm may take. One example might be the case of acting out of envy or spite where you are required by the nature of the evil you pursue to hurt another, even if it should cause you great loss as well. Another might be the case of revenge where that goal requires you without heed for your own well-being—say, without heed for the retaliation you may trigger—to do harm to another. And a third might be the sort of racial hatred promoted in Himmler’s 1943 Potsdam speech, when he encouraged his troops to maintain their genocide against the Jewish

28

Philip Pettit

people at whatever cost to their own feelings or sensitivities. But the striking thing about these cases is that they are so extreme. In each example, you are required as an evil-doer to be ready, in an old phrase, to cut off your nose to spite your face. You are required, as that phrase suggests, to behave in what is taken to be a self-defeating, even irrational manner. To be an heroic evil-doer of the kind envisaged here would be to espouse the sort of attitude ascribed to Aaron in the final scene of Shakespeare’s Titus Andronichus: ‘ If one good deed in all my life I did, I do repent it from my very soul’. This kind of maleficence is demonic in character, fitting the slogan of Milton’s Satan: ‘Evil, be thou my good’.7 But the heroic character of the exercise suggests that it is likely to be very rare. There are lots of bad things that we human beings do to one another, but there are few plausible evils that require self-sacrificial action (Baumeister 1997; Russell 2006). Doing evil is generally a much more banal phenomenon—even among Nazis, as Hannah Arendt (1963) famously suggested. Hobbes (1994 [1668], Ch 15) recognizes the non-heroic aspect of evildoing, in particular injustice, when he says in Leviathan that whereas a just man is ‘he that taketh all the care he can that his actions may all be just, an unjust man is he that neglecteth it’.8 The unjust man’s will is framed, as he goes on to explain, not by desire for injustice as such, but ‘by the apparent benefit of what he is to do’. On this account, the evil of injustice does not consist in imposing damages on others out of a pure, selfless desire to do so. On the contrary, it consists in the failure to honor people’s claims of justice in how you treat them—the failure that consists in an intentional or at least negligent action—because of a self-seeking search for reward or benefit.9 4. EXPLAINING THE ASYMMETRY BETWEEN GOOD AND DOING EVIL

The Role of Standards What Hobbes directs our attention to is the fact that whereas doing good often means controlling for the performance of good deeds, even at some 7 The line in this paragraph gives us reason to be reluctant to endorse Thomas Nagel’s (1980, 132) claim: ‘To aim at evil, even as a means, is to have one’s action guided by evil’. 8 Charles Pigden (1988, 29) puts the point nicely in endorsing a view he finds in Aquinas: ‘the just man is not one who makes a policy of injustice — merely one who does not make a policy of justice’. 9 For similar lines of thought in the medieval period, see (Anselm 2000). I am grateful to Calvin Normore for pointing this out.

The Asymmetry of Good and Evil

29

cost to yourself—taking all the care you can that your deeds be good— doing evil has a different character. It is more likely to consist, not in controlling with a corresponding robustness for the performance of evil deeds, but in performing evil deeds as a means to satisfying your selfinterest or as a side-effect of satisfying your self-interest. The person who does good often controls for performing good deeds, even at some cost to their self-interest. The person who does evil typically controls for the satisfaction of their self-interest, where the cost of doing so involves imposing some harm on another: doing an evil deed. There should be no surprise about this asymmetry between doing good and doing evil. The standards that we hold up for one another, including standards of respect, attachment and virtue, direct us to robustly demanding goods, in particular goods that often demand a degree of self-sacrifice. And there are no standards of the same kind that would direct us to evils of an equally demanding character: evils such that doing them would require you to control malevolently, even self-sacrificially, for imposing certain costs on others. Why do we have standards of goodness that we effectively promulgate, being ready to approve of conformity to the standards and to disapprove of deviance—and being ready to do this on public grounds, not just on grounds of private interest? We are creatures of limited altruism who have to live in a world of scarce resources and these circumstances often put us in a situation where we each do better in self-interested terms to the extent that we all abide by standards that limit our direct pursuit of self-interest. We do better by sticking to the standards prescribed for lovers and friends, since we can thereby build mutually advantageous relationships rather than finding ourselves isolated from one another. We do better by sticking to standards of honesty and justice and fidelity in our dealings with one another, even with strangers, because this makes for a peaceful world in which we can plan our lives with a degree of assurance. There are many stories about how we human beings come to identify such standards and implement them more or less effectively as social norms but we need not look at the relative merits of those accounts here. It suffices for our purposes to assume that, however their emergence and maintenance is explained, the standards are commonly recognized, with cultural variations, across most societies. In each of our communities, and even across different communities, we acknowledge collectively

30

Philip Pettit

beneficial standards of good behavior to which we hold one another, approving of conformity and disapproving of non-conformity. And we hold one another to these standards in full recognition that they are often individually burdensome, requiring people to put the good of acting as the standards require above their immediate self-interest. It is no surprise that given the presence of suitable standards, we expect people who count as doing good to be disposed to comply with those standards and to act out of that disposition. We expect them to control for acting as lovers or friends, or as honest, just and faithful brokers, when such action is required by the needs of others, according to the shared standards. And of course we expect them to do this even across possible scenarios where their immediate self-interest is frustrated; after all, the standards are designed precisely to overcome the problem of limited altruism. The existence of standards that are designed to cope with our limited altruism makes sense, then, of why we associate various forms of doing good with controlling benevolently for good acts like those of care and favor, truth-telling and claim-satisfaction, promise-keeping and selfrestraint. But the striking thing on the side of evil is that there are generally no parallel standards there. At least within the organization of a given society, there is no basis for wanting people to overcome their immediate self-interest in imposing certain costs on others. It is not as if such a pattern would be collectively beneficial. And so there are no standards of evil-doing that serve the role associated on the side of good with standards of attachment, virtue and respect. If there are any exceptions, they involve standards governing people’s behavior towards outsiders or those cast as outsiders. Himmler’s promulgation of racial hatred offers a chastening example. If most societies promote standards of good but few if any societies promote standards of evil, it is understandable that while we have relatively heroic standards for doing good—those that require controlling benevolently, even self-sacrificially, for performing certain good deeds—we do not have heroic standards for doing evil: standards that would require controlling malevolently, even self-sacrificially, for performing certain evil deeds. What evil-doing is likely to be taken to involve, in the presence only of standards for good, is the imposition of costs in a way that breaches those standards: an imposition of costs, presumably, that is prompted by self-interest. And that is precisely what

The Asymmetry of Good and Evil

31

Hobbes suggests when he says that to do justice is to take all possible care that you act justly whereas to do injustice is to neglect the requirements of justice—to act unjustly—because of the benefit that that promises for yourself. To do justice, and to do good of any of a variety of kinds, is to control benevolently, even self-sacrificially for doing just or good acts. To do injustice, and to do evil of any of a range of forms, is to breach the requirements of good because of controlling for self-interest. Historical Antecedents The line emerging from these considerations does not find support only in the thought of Hobbes. It fits with the long tradition of thinking, associated with Augustine and Aquinas, that rejects the Manichaean idea that good and evil are of equal metaphysical importance and that the moral life involves a struggle between the two principles. Without delving into the metaphysics associated with the position, it should be clear that the observations we have rehearsed may give some plausibility to this line of thought. The primary struggle in moral life is to live up to the demands of our standards of good, such as the standards associated with various attachments and virtues. What it means for good to lose out, and evil to triumph, is merely that you fail to do good: you act in a way that does not live up to the standards of good, being tempted away from the high path charted in those standards by the temptations of immediate selfinterest. Thus good is the positive goal and evil means acting in a way that is contrary to that goal: failing in that sense to live up to the goal. In the received Latin phrase, malum est privatio boni; evil is merely the absence of good, not something positive in itself. Not only do our observations offer some support for this antiManichaean way of thinking. They may also help to make sense of another medieval principle of moral thought, which received particular prominence in the work of Aquinas. This principle holds that anything desired is desired under the aspect of good: quidquid appetitur sub specie boni appetitur. It is often presented as a doctrine governing the very possibility of intelligible desire and intentional action and it is often criticised in that role (Velleman 1999). But it may be better taken as an affirmation that whereas we often desire the good of others for its own sake, controlling benevolently for conferring associated benefits, we generally seek the bad of others in order to promote our own

32

Philip Pettit

self-interest or at least on the proviso that our own self-interest is preserved. We do not desire the bad of others for its own sake but only as a means or a complement to our own good. 5. THE KNOBE EFFECT

Joshua Knobe (2003; 2006; 2010) introduced the effect now named after him by testing a simple vignette on a variety of subjects, identifying a difference in their willingness to describe a would-be good action and a would-be bad action as intentional (see too Pettit and Knobe 2009). Knobe’s vignette involves two stories about the chairman of a company board. First story: asked to consider a program that he is told will increase profits but harm the environment, the chairman answers that the company’s only concern is to increase profits—the environment does not matter—and that the program should be put in place. Second story: asked to consider a program that he is told will increase profits and help the environment, he answers again that the company’s only concern is to increase profits—the environment does not matter—and that the program should be implemented. Does the chairman intentionally harm the environment when he acts as in the first story? Does he intentionally help the environment when he acts as in the second? Most people, so it turns out, think that he intentionally harms the environment in the first story but that he does not intentionally help the environment in the second. The Knobe effect consists in the fact that this asymmetry in the ascription of intentionality characterizes the responses of most human beings in this sort of vignette. The effect is challenging, because it suggests that the folk psychology on which we rely in distinguishing between cases where people do things intentionally and cases where they don’t is warped by our evaluative viewpoints. The idea would be that despite the fact that there is nothing of psychological significance to distinguish the position of the chairman in the two scenarios, still we find a psychological difference there. We draw a psychological distinction, it appears, not in response to considerations that are germane to a presumptively explanatory enterprise, but rather in response to irrelevant, ethical considerations. We practice a psychology that is essentially value-dependent. Knobe himself appears to be willing to live with this valuedependence, for he explains the effect by hypothesizing that ascriptions

The Asymmetry of Good and Evil

33

of intentionality are sensitive to moral considerations, and cannot be taken to be exclusively or even primarily explanatory. But it turns out that our observations about the asymmetry of good and evil make sense of people’s intuitions about ascriptions of intentionality, without resort to any such hypothesis.10 Why do we readily think in the first story that the chairman does intentionally harm the environment but in the second story that he does not intentionally help the environment? The asymmetry that we have noticed between doing good and doing evil supports a natural suggestion: that helping involves controlling for compliance with environmental standards, whereas harming does not; that this makes it harder to count as helping rather than harming; and that consequently there are stricter requirements that the chairman must meet if he is to be cast as intentionally helping the environment than if he is to be cast as intentionally harming it. Most of us think that there are standards that formulate the preconditions for a healthy environment. Indeed the very questions raised in the vignette presuppose an assumption of this kind; without that assumption the idea of helping or harming the environment would otherwise make little sense. What is to count as helping the environment, against the background of such standards, and what is to count as harming it? It will certainly be enough to count as harming the environment that you breach environmental standards, say for reasons of self-interest; you do not have to be an environmental vandal who controls self-sacrificially for flouting those standards. But it will hardly be enough to count as helping the environment that you happen to conform to the standards, again for reasons of self-interest. Here the natural suggestion is that to help the environment—to do good in that respect—it is essential that you control for conformity to environmental standards, even at some loss to yourself (see too O’Brien forthcoming). This means that if the chairman is to count as intentionally helping the environment he has to cross a high bar, intentionally controlling for 10 The line taken here parallels in some measure the approach championed by Richard Holton (2010) although his approach is defended in quite different terms. For an alternative approach, but with some affinities, see (Hindriks 2013). Quite independently, Lillian O’Brien (forthcoming) has found empirical support for an idea close to that defended here: viz., that respondents are triggered in Knobe-like experiments by the different conceptions that they hold of the requirements of helping, on the one hand, and harming on the other.

34

Philip Pettit

conformity to environmental standards. And obviously the chairman in the first vignette does not reach that bar, so that people are quite right to say that he does not help the environment intentionally. If the chairman is to count as intentionally harming the environment, however, he merely has to cross the bar of breaching those standards intentionally. And that is easily met: breaching those standards does not require controlling for breaching them, in the spirit of a vandal; and, in ordinary usage, it is enough for him to breach those standards intentionally that he is aware of what he is doing (Scanlon 2008; Pettit 2015).11 Thus people are quite right in this case to assert that the chairman does harm the environment intentionally. There is no mystery about the difference of attitude they assume in the two cases. Knobe and his colleagues tried many variations of the original vignette on people, with broadly similar results. Instead of asking whether the chairman intentionally helped or harmed the environment, for example, people were asked whether he decided to help or decided to harm, whether he favored helping or favored harming, whether he advocated helping or advocated harming.12 Unsurprisingly from our point of view, the results retained much the same asymmetry as in the original case. The chairman certainly doesn’t decide to help, or favor or advocate helping, given that helping requires controlling for conformity to environmental standards. But it is quite plausible to think that he decides to harm, or even favors or

11 For the record, I think that the requirements for doing X intentionally are weaker than the requirements for acting with the intention to X and that they in turn are weaker than the requirements for forming the advance intention to X. In order to form the advance intention to X it is necessary to do more than just act out of a desire to X, taking whatever steps are believed to be suitable; you have to let X-ing at the appropriate time assume the place of a relative fixture in your decision-making (Bratman 1987). In order to act with the intention of X-ing you do not have to act on the basis of a pre-formed intention; you need only act out of a desire to X, adopting the steps you believe to be appropriate. And in order to do X intentionally you do not have to act out of a desire to X; you need only act out of a desire to Y in a case where you foresee that Y-ing will involve X-ing and you are prepared to live with that result: as it may be, with that cost. 12 They also asked whether the chairman can be said to desire harming or to intend to harm—that is, to act with the intention of harming. To say either would be to suggest that he is attracted to the option under all its aspects—see the previous footnote—whereas to say he harmed intentionally is consistent with his desiring or intending the action under another aspect only—say, as a way of making profits; in this case he merely foresees and accepts that the harm will occur (Scanlon 2008; Pettit 2015). It is not so surprising, then, that in these cases the results are less decidedly asymmetrical: that people tend to say that he did not desire or intend to harm, as he did not desire or intend to help (Pettit and Knobe 2009).

The Asymmetry of Good and Evil

35

advocates harming, given that harming requires only breaching environmental standards. The challenge raised by the Knobe effect is to vindicate the explanatory role of ascribing intentionality by showing that the ascription is not value-dependent. Our story meets that challenge, because it traces the effect to the different sorts of actions represented by helping on the one side and harming on the other. We do not take different lines in the two cases because one action has positive value, the other negative. We take different lines because of a difference between the requirements that must be satisfied if an action is to count as helping or as harming. But is the difference we trace still value-dependent insofar as helping is a positive action, harming a negative one? No, it is not. The fact that there is a high bar for helping and a low bar for harming is what explains the difference in our ascriptions of intentionality and it is incidental that the explanation for the difference in the bar required goes back to matters of good and evil. Thus we would expect to find an analogue of the Knobe effect wherever there is a high-low difference in the bar to be crossed, and not just when that difference has moral significance. Take the difference in the bar associated with following the rules of grammar and breaking the rules of grammar. And now imagine a vignette similar to Knobe’s. The chairman is told in one story that the advertising branch has come up with a great company slogan but one that is ungrammatical and in the other that it has come up with a great slogan and one, moreover, that is grammatical. In each case he says: I don’t mind about the grammar; if it’s a great slogan, let’s use it. Does he intentionally breach the rules of grammar in the first story? Surely he does. Does he intentionally follow the rules of grammar in the second story? Surely he doesn’t. What then is the difference? Clearly, I would say, the fact that there is a high bar for following the rules of grammar— you must control for conforming to them—and a low bar for breaching them: you must certainly breach the rules but you need not control for breaching them. We saw in the first section that the things we do have both causal and constitutive effects and that among their constitutive effects are ones that depend on the dispositions out of which we act. We saw in the second section that these disposition-dependent effects are very important in doing good to others, as when we act out of attachment, virtue, or respect; in particular, when we do so in neglect of our own self-interest.

36

Philip Pettit

We saw in the third section that in doing ill to others, such dispositiondependent effects are not of the same importance and, in particular, that most forms of evil-doing do not require us to neglect our own selfinterest in the same way. But there is no mystery in this asymmetry, as we saw in the fourth section. The explanation is that doing good generally involves living up to certain standards that guard against limited altruism whereas doing evil generally involves acting in breach of such standards, not living up to any counterparts on the dark side. With these observations in place, we can now see that the Knobe effect, which is so prominent in recent discussions of moral psychology, is not mysterious either. If helping the environment means acting up to environmental standards, and harming the environment means failing to act up to those standards, then it is much harder to count as helping than it is to count as harming. And so it is unsurprising that the company chairman in Knobe’s vignette can be thought to harm the environment intentionally but not to help the environment intentionally. This asymmetry simply reflects the more general asymmetry between helping and harming, doing good and doing evil. Acknowledgments I am grateful for many helpful discussions of this material. I used it in an Inaugural Lecture at the Australian National University, the Dan and Gwen Taylor lecture at the University of Otago, the Taylor-Lewis lecture at the University of Melbourne, and the Kendrick lecture at the 2014 Arizona Workshop in Normative Ethics. I am also grateful for very useful comments received from an anonymous referee, and from the editor, Mark Timmons. REFERENCES

Anscombe, G. E. M. (1957). Intention. Oxford, Blackwell. Anselm of Canterbury. (2000). ‘The Fall of the Devil’, in The Complete Philosophical and Theological Treatises of Anselm of Canterbury. J. Hopkins and H. Richardson (eds). Minneapolis, Arthur J Banning Press. Arendt, H. (1963). Eichmann in Jerusalem: A Report on the Banality of Evil. New York, Viking Press. Baumeister, R. (1997). Evil: Inside Human Violence and Cruelty. New York, Henry Holt. Bennett, J. (1995). The Act Itself. Oxford, Oxford University Press. Bratman, M. (1987). Intention, Plans, and Practical Reason. Cambridge, MA, Harvard University Press.

The Asymmetry of Good and Evil

37

Davidson, D. (1980). Essays on Actions and Events. Oxford, Oxford University Press. Goldman, A. (1970). A Theory of Human Action. Englewood Cliffs, NJ, Prentice-Hall. Herman, B. (2011). ‘A Mismatch of Methods’, in Derek Parfit: On What Matters, vol. 2. S. Scheffler (ed.). Oxford, Oxford University Press: 83–115. Hindriks, F. (2013). ‘How to Explain the Knobe Effect and its Relatives’. Mind and Language 28. Hobbes, T. (1994 [1668]). Leviathan. E. Curley (ed.). Indianapolis, Hackett. Holton, R. (2010). ‘Norms and the Knobe Effect’. Analysis 70: 1–8. Knobe, J. (2003). ‘Intention, Intentional Action and Moral Considerations’. Analysis 64: 181–7. Knobe, J. (2006). ‘The Concept of Intentional Action’. Philosophical Studies 130: 203–31. Knobe, J. (2010). ‘Person as scientist, person as moralist’. Brain and Behavioral Sciences 33: 315–65. Korsgaard, C. (2009). Self-Constitution: Agency, Identity, and Integrity. Oxford, Oxford University Press. Nagel, T. (1980). ‘The Limits of Objectivity’, in The Tanner Lectures on Human Values, vol. 1. S. McMurrin (ed.). Salt Lake City, University of Utah Press. O’Brien, L. (forthcoming). ‘Side-Effects and Asymmetry in Act-Type Attribution’. Philosophical Psychology. Pettit, D. and J. Knobe (2009). ‘The Pervasive Impact of Moral Judgment’. Mind and Language 24: 586–604. Pettit, P. (1995). ‘The Virtual Reality of Homo Economicus’. Monist 78: 308–29. Expanded version in U. Maki (ed.), The World of Economics, Cambridge University Press, 2000; reprinted in P. Pettit (2002), Rules, Reasons, and Norms, Oxford, Oxford University Press. Pettit, P. (2015). The Robust Demands of the Good: Ethics with Attachment, Virtue and Respect. Oxford, Oxford University Press. Pigden, C. (1988). ‘Anscombe on “Ought” ’. Philosophical Quarterly 38: 20–41. Ross, D. (2002 New Edition). The Right and the Good. Oxford, Oxford University Press. Russell, L. (2006). ‘Dispositional Accounts of Evil Personhood’. Philosophical Studies 149: 231–50. Scanlon, T. M. (2008). Moral Dimensions: Permissibility, Meaning, Blame. Cambridge, MA, Harvard University Press. Velleman, D. (1999). ‘The Guise of the Good’. Nous 26: 3–26.

2 A Particularist but Codifiable Virtue Ethics CHRISTINE SWANTON

1. INTRODUCTION

Moral Particularism of the kind developed by Jonathan Dancy1 is treated as a topic in meta-ethics. Until it is applied to a suitable type of normative theory, criticisms which have assailed it are difficult to rebut. This chapter aims to apply Dancy’s particularism to a type of virtue ethics, showing how many of these criticisms are off the mark. At the core of these criticisms is an alleged problem on which I focus here: a problem shared by virtue ethics. For both virtue ethics and particularism, apparently, ethics is uncodifiable. In the case of virtue ethics that criticism takes the following form. A normative ethical theory is supposed to provide action guidance, but virtue ethics is unable to do so. The objection (call it the action guidingness objection) has been around for years, has been addressed, but never quite goes away. The objection is directed at the traditional ‘qualified agent’ (Swanton 2003: 227) virtue ethical account of right action: an action is right if and only if it would be chosen or performed by a virtuous agent choosing or acting in character (Hursthouse 1999). This form of virtue ethics is vulnerable to the action guidingness objection for the following reason. Moral reasons are discerned through the sensitivities of a virtuous agent. They are thus viewed as inaccessible to the non-virtuous, opaque, and even ineffable. Indeed one might say that discernment of this kind is tantamount to no guidance at all. In the case of Dancy-style particularism the “uncodifiablity” objection takes a more sinister form. It is expressed thus by Lance and Little 2006: 20–1

1

Primarily in his 2004.

A Virtue Ethics

39

( . . . ) killing and the infliction of pain have no more intimate connection to wrongness than do truth-telling, healing and the giving of pleasure. After all, each, in the right context, can have a positive, negative or neutral moral import. But the morally wise person, one might have thought, is someone who understands that there is a deep difference in moral status between infliction of pain and shoelace color, even if both can, against the right narrative, be bad-making.

Particularists thus ‘flatten the moral landscape’ (McKeever and Ridge 2006: 48, 2007 passim). A major argument against Dancy’s particularism then is that moral education and moral predictability are impossible since the particularist, unlike the generalist, cannot account for ‘general moral minuses and moral plusses’ (Hooker 2000:19).2 I shall argue that both virtue ethics and particularism are codifiable in a suitable sense, and that both these types of criticism can be rebutted. Most importantly the objection that particularists ‘flatten the moral landscape’ is overcome by applying the particularist thesis to virtue ethics. Virtue ethics, I shall argue, is codifiable through the “virtue rules” (‘v-rules’) (Hursthouse 1999) in a way that displays the ‘deep difference in moral status’ between types of reasons to which Little alludes. The v-rules codify reasons specified in terms of virtue concepts as applied to action (v-reasons). To argue in addition that virtue ethics is codifiable in a way compatible with Dancy style particularism, we need to understand the notion of v-rules, and their relation to v-reasons.

2. THE VIRTUE RULES

It is odd that virtue ethics is often thought to be uncodifiable. Despite the frequently expressed claim that virtue ethicists have no truck with rules, in her pioneering work On Virtue Ethics Rosalind Hursthouse allows for what she calls the ‘v-rules’. These are prescriptions, recommendations, and prohibitions ‘generated’ by each virtue and vice concept respectively, for example, ‘Do what is courageous, do what is honest, do what is loyal’ (80); ‘Do not do what is dishonest, uncharitable, mean’ (36). Indeed the notion of a v-rule is a notion introduced in the context of combating the 2 In similar vein McNaughton and Rawling (2000: 273) claim that on Dancy’s view (which they call ‘thin intuitionism’) ‘thick moral properties have no more intrinsic moral significance than non-moral properties.’

40

Christine Swanton

action guidingness objection, by means of an important distinction between strong and weak codifiability. Strong codifiability is offered by a set of rules providing a ‘decision procedure’ (39–40). Weak codifiability describes a set of rules which apply ‘for the most part’.3 By now the demand for strong codifiablity in the form of an ‘algorithm’ (54) has diminished, but weak codifiability seems to be a desideratum. As I shall argue, it enables us to conceptualize and identify types of reason which resolves ‘the flattening of the moral landscape’ problem. In this way the v-rules may offer guidance possibly unavailable in the aesthetics case where an analogous “rule” such as ‘Be bold’ or ‘Be bold with strokes of red’ may offer no guidance: ‘the bold stroke of red that helps balance one painting would be the ruin of another ( . . . )’ (Little 2000: 280). But now there is a puzzle. What is the relation between conforming to the v-rules and the qualified agent criterion of right action? Is it the case that to do what is honest, for example, is simply to act as the virtuously honest agent acts? Let us call doing what is honest, what is generous and so forth ‘v-acts’ (Hursthouse 1999: 80). It may seem obvious that to perform a v-act is to act virtuously, and to act virtuously is to act as a virtuous agent acts. Unfortunately, as I shall argue, these relationships do not hold, and the appearance that they do depends on ambiguities in the notion of “virtuously”. For a start, as Hursthouse makes clear, doing what is just, for example, need not involve acting in the way a virtuous agent acts in the strong sense of expressing virtue. Might it not be the case however that performing v-acts entails doing what a virtuous agent would do in the fields respectively of justice, honesty, courage, in a sense weaker than expressing or manifesting virtue? This seems to be the view of Hudson who claims that an ‘act gets rightly called courageous when and only when it is such that the courageous person would perform’ in the type of circumstances where courage is characteristically called for.4 But as I argue below a virtuous agent may rightly run away, lie, or do something unjust, and we cannot call such actions courageous, honest, or just, simply because they are performed by such an agent in the relevant fields. The argument for a conceptual link between conforming to v-rules and the qualified agent criterion of right action trades on an ambiguity in the notion of acting 3 4

Hursthouse (1999: 58), citing McDowell (1979: 337). Hudson (1986: 42–3) cited in Hursthouse (1999: 80).

A Virtue Ethics

41

virtuously between performing an act properly described in terms of a thick virtue concept (doing what is kind, just etc.), and acting as (in whatever sense) a virtuous agent acts. What then is the relation between conforming to v-rules and a virtue ethical criterion of right action? I shall argue for the following view. V-rules specify v-reasons (that an action is just, generous and so forth)—reasons which at least defeasibly favour actions. What is held to make actions right (or more accurately help make actions right) are v-reasons which count in favour of acts,5 and these v-reasons are (at least characteristically) codifiable through the v-rules. The real task, undertaken below, is thus giving an account of v-reasons and their relation to right acts. I shall conclude that at least most v-reasons are default reasons which, though not having invariant positive valence nonetheless apply “for the most part” in a sense characterized below. The v-rules provide weak codifiability compatible with Dancy-style particularism. To show this, it is first necessary to provide an account of right action according to which what is held to (help) make actions right are the v-reasons counting in their favour.

3. A TARGET-CENTERED VIRTUE ETHICAL NOTION OF RIGHT ACTION

An objection to the orthodox virtue ethical conception of right action offered by Hursthouse is that what makes actions right is not that they would be chosen by virtuous agents, but that the actions themselves are describable in terms of v-reasons: they themselves are just, kind, caring, beneficent, and so forth (see further van Zyl 2014). At best the Hursthouse account offers an extensional equivalence between right action and actions chosen by virtuous agents acting in character, but I have called this equivalence into question (Swanton 2003). Virtuous agents are wise but not necessarily right. Indeed in her (2006) Hursthouse rightly claims that for Aristotle ‘the phronimos is not infallible’ (292) even though

It is not being claimed here that “natural” or “descriptive” factors such as its being green can’t in a context make an action such as wearing this shirt right, but what ultimately makes the action right is a fact specified by a thick concept, such as its being green is not in the context offensive (unlike some other salient colour), or in the context shows moral courage. 5

42

Christine Swanton

practical wisdom is a property of someone who has ‘‘full virtue’ or ‘virtue in the primary sense’ (1144b13) [Nicomachean Ethics]’ (285–6). Indeed she denies that the phronimos needs to have ‘encyclopaedic’ knowledge, for to demand that of practical wisdom ‘seems not only absurd but elitist, guaranteeing that practical wisdom, and thereby full virtue, is open only to the academically intelligent, those who are capable of absorbing such a vast amount of varied information’ (308). In other words practical wisdom cannot be defined in terms of ‘perceiving the details of situations correctly’ as some appear to accept (such as Pettigrove 2011: 202). Given this feature of practical wisdom a virtuous agent can get things wrong.6 A natural solution is to idealize. (Ideally) virtuous agents do have encyclopaedic knowledge; are expert in, that is have skills in, all practical domains; indeed have perfect knowledge of the future; occupy all perspectives; and in various other ways cease to be human. But real virtuous agents have limited perspective being among other things gendered, old or young, historically and culturally located, and are by no means expert in all fields. We are left then with a dilemma: either we idealize and achieve extensional adequacy (there is extensional equivalence between the choices of virtuous agents and right action) in which case the choices determining rightness are not those of virtuous humans, or the account of rightness is not extensionally adequate.7 What makes actions right then on the view proposed is not that they would be chosen by virtuous agents, but that the actions themselves are favoured by v-reasons. But how do we understand the v-reasons? Explicating such reasons requires an account of what I have called the ‘target centered’ virtue ethical view of right action.8 According to Aristotle (Nicomachean Ethics 1976: henceforth NE) in deliberation virtuous

6 Unfortunately ‘rightness’ can have different uses in the literature. Sometimes it is conceptually linked to rationality so that an agent does something right even if mistaken, because supported by evidence at her disposal. This is not the use at issue here. Unfortunately too the waters are muddied by the use of the notion “moral” rightness (a notion not at home in virtue ethics and not used in connection with rightness here) which seems often to be associated with rationality or motive-centered conceptions. (For a rejection of the moral (in a narrow sense)/ non-moral distinction in the context of virtue ethics see Swanton 2014). 7 See further on problems of idealizing, Enoch (2005). 8 Originally in Swanton (2001).

A Virtue Ethics

43

agents aim at ‘practical truth’ or correctness in action.9 This aim is realized by attaining the “target” of relevant virtues, what Aristotle calls the ‘mean’ (relative to action) of those virtues. The mean itself is a multidimensional notion including action towards the right objects, in the right manner, for the right reason, to the right extent, with the right instruments, at the right times and so on, relative to the virtue in question.10 The v-reasons which contribute to making an action A right11 on a particular occasion—the v-reasons in favour of A—are constituted by the mean of relevant virtues in relation to action: a mean at which virtuous agents aim (NE 1106b16–24) and who characteristically exhibit wisdom in prosecuting that aim. Right-makers are thus conceptualized in terms of virtue concepts as applied to action, but as we shall see, the relation between such concepts and v-reasons in favour of a particular action is not straightforward. In particular, I shall argue, an act’s kindness, for example, is a default reason for an action which may be “switched off ”: it is not always a reason in favour of an act. V-reasons favouring actions on this account are constituted by certain kinds of facts—the attainment of a target of a virtue on relevant dimensions, for example an act’s being generous in an appropriate way, to appropriate people and so forth. Such acts hit at least some aspect of the mean of relevant virtue(s) in action. But what in more detail is it to attain the mean? First, attaining the mean in action is not identical to expressing virtue, which as Aristotle also points out requires further features, notably acting from a settled disposition including a settled disposition of the right affective states (NE Book 2 (iv)). For an act to be not merely right, but also fine (kalos) or excellent, it must express virtue. However in assessing or deliberating about practical correctness in action, we are not 9 Indeed in Crisp’s (2000) translation of the Nicomachean Ethics the notion of ‘correct’ is used as in the generous person ‘will give of his resources for the sake of the fine and correctly’ (e.g. in the right way to the right people at the right time) ‘with the other qualifications that attach to correct giving’ (1120a24–9). (See Annas 2014.) 10 Notice that this is a framework only, but a quite substantive one, since it is very easy to ignore or get wrong such things as appropriate manner (e.g. of caring in the case of the terminally ill); appropriate objects (e.g. to whom it is appropriate or required to disclose one’s reasons), appropriate instruments (e.g. when it is appropriate to bypass protocol). Reasoning about such matters is sophisticated and difficult, so it is wrong to claim that such an Aristotelian approach is ‘trivial’ or ‘viciously circular’ even if in its thin description being right means being in accord with (the targets of) virtue (Das 2003; Adams 2006: 10). 11 I say ‘contribute’ since reasons in favour may be strengthened by other reasons in favour, or may be countered in some way by reasons against (see Dancy 2004: 15–16).

44

Christine Swanton

characteristically focused on all the features of what makes an act kalos, but on contextually relevant or salient targets of the relevant virtues in relation to action. Unfortunately in recent literature, including that in virtue ethics, rightness has come to include features proper to what makes acts kalos, such as praiseworthiness and general goodness. I distinguish rightness as practical correctness from praiseworthiness, which is neither necessary nor sufficient for correctness in action. Second, to understand what it is to attain the mean in action we have to understand the many faces of “rightness”. Rightness as correctness is an umbrella notion where rightness can take several forms, discussion of which (and of their relationships) would take us too far afield. These include actions that are merely permissible, actions that are not only permissible but desirable, actions that are obligatory or required, actions that are admirable though perhaps not required. In all these cases rightness as correctness on a virtue ethical account is constituted by attaining the “target” of relevant virtues (what Aristotle calls the mean in relation to action). However what counts as hitting the target of a virtue will depend on the aspect of rightness in question. A merely permissible act of generosity may be minimally decent but not admirable. The donation does not hit the target of generosity in a stellar way, but at least it is not stingy. Avoiding stinginess will be required: a stellar act of philanthropy may be admirable but not required. Third, as already noted, the mean is multidimensional. Correct actions are understood as actions that hit the targets of relevant virtues, on contextually relevant or salient dimensions of the mean. Which dimensions of the mean are salient depends both on context and on the nature of the virtue. Some contexts, e.g. those in which we are interested in saving a life, may downplay the relevance of motive altogether (Das 2003) whereas in other contexts motive may be highly salient for rightness. Some virtues such as caring and kindness focus particularly on motive and manner—a focus which is transmitted to caring and kindness predicated of actions—whereas others such as beneficence are more focused on results-oriented external targets. Fourth, the targets of many virtues may be implicated in the rightness of an action. The targets of these virtues may be in conflict with each other, generating virtue ethical dilemmas. How these are to be conceived is briefly explored below in the context of “tragic” dilemmas (section 8).

A Virtue Ethics

45

Fifth, given that phronesis is fallible, and although virtuous agents aim at the mean and are more reliable in attaining it than the non-virtuous, virtuous agents may miss the targets of virtues and be wrong about v-reasons (See Curzer 2005). My account of v-rules as described so far can be summarized thus: (1) The v-rules codify (in a suitable sense) v-reasons specified by the thick v-concepts, applied to action. (2) A v-reason R (being kind, say) favours v-act A in context C insofar as R in C specifies a respect or respects in which A attains some relevant aspect of the mean in relation to action, of a relevant virtue (kindness, say). What counts as ‘relevant’ depends on the aspect of rightness in question; which virtues are in play in C; and which aspects of the mean of the relevant virtues are salient in C.

4. THE THICK CONCEPTS, V-REASONS, AND V-ACTS

V-rules enjoin v-acts, and what makes v-acts right are the v-reasons which favour them. The previous section gave an account of what it is for a v-reason to favour an act. However it need not always be the case that a v-reason (an act’s being kind for example) always favours that act. If v-reasons are particularist, then it would seem the v-property of a v-act need not make or help make an act right overall; indeed it may even have negative valence. But does not the “virtuousness” of an act guarantee its rightness, at least pro tanto? We need then to understand the nature of the v-acts that are enjoined by the v-rules. To understand the nature of a v-act requires an account of the relation between the features that make an act kind for example (its resultance base)12 and the act being kind. The properties forming the resultance base of properties specified by the thick concepts have been called by both Dancy and Little ‘naturally shapeless’, in the following sense: ( . . . ) the items grouped together under a moral classification such as ‘cruel’ do not form a kind recognizable as such at the natural level ( . . . ) of the infinitely

12 For discussion of the notion of resultance base, and the distinction between resultance and supervenience, see Dancy (1993: 73–9).

46

Christine Swanton

many ways of being cruel—kicking a dog, teasing a sensitive person, and forgetting to invite someone to a party might qualify—there is no saying what they have in common ( . . . ) except by helping oneself to the moral concept of cruelty. (Little 2000: 279)

The applicability of the thick virtue concepts (and vice concepts for that matter) depends on a (shapeless) descriptive resultance base, but they are also normative. Their normativity does not depend on a special nonnatural property which is also part of the resultance base. Rather the normativity of a thick concept is determined by its evaluative point as Williams (1995: 206) points out: ( . . . ) to understand how such a concept can be applied to a new sort of situation it is likely that one will have to grasp its evaluative point, the outlook of the people who use it.

Dancy (1993: 85) also notes that neither the ‘natural grouping’ (the resultance base) of such a concept, nor its evaluative point or ‘concern,’ are ‘independently intelligible’: The shape of the resultance bases is invisible to those who lack the concept of the relevant concern. But the concern also has not got enough determinate content to explain the choices that are made.13

In the light of this account, and using honest act as an example, we understand a v-act as follows: (a) A v-act (e.g. honest act) is of a type that (characteristically but not infallibly) exhibits the evaluative point or function of e.g. the virtue of honesty in relation to action, namely meeting the (contextually relevant or salient) target(s) of e.g. the virtue of honesty in relation to action. (b) A v-act (e.g. honest act) has ‘descriptive constraints’ (such as is not a lie) necessary for the thick concept (e.g. honest) to be properly applicable to actions.

13 What determines ‘relevant concern’ is controversial, but I do not have space to investigate this interesting issue here.

A Virtue Ethics

47

(c) It is sometimes the case that these constraints are such that a virtuous agent acting rightly may do something that is subsumable under a vice concept as applied to acts: she may do something unjust or dishonest. Given these features of the thick concepts we cannot say that telling lies is a kind of honesty just because a virtuous agent rightly tells a lie. Indeed following the v-rules may point one away from right action.14 For example consider cases of dishonest acts (lies) of the kind described in The No. 1 Ladies’ Detective Agency (Smith 1998). Assume that in the mouths of the heroine Mma. Ramotse, and of her helper in resolving a serious case, Mr. J.L.B. Matekoni (Chapter 18 ‘A Lot of Lies’), certain of these lies are right, an assumption which given the plot, is plausible. Paradoxically, honest acts by contrast would miss the contextually salient target, in relation to action, of the virtue of honesty: if one were honest one would be honest at the wrong time to the wrong people. This seems paradoxical because ‘honesty’ is the character term used to refer to excellence in fidelity regarding the dissemination of information. Here is another example. The v-rule ‘Be generous’ enjoins generous acts, but a generous act may be wrong, precisely because it misses the target of generosity. It misses contextually relevant dimensions of the mean. For example what is for a wealthy person a generous present may not be recommended, may indeed be wrong because the present has gone to the wrong people at the wrong time. Poor relatives at a family Christmas in a public present opening ceremony, who themselves could only afford a cheap gift and who cannot repay the generosity at subsequent Christmases, are humiliated and embarrassed by the generous gift. We would not say that because the act misses the target of generosity and is thereby wrong in this case it ceases to be generous. Rather the v-rule ‘Be generous’ (in the relevant context of a wealthy person giving a present here and now at Xmas) should here be violated. She should not be generous with her presents at this time and in this context.15 14 For this reason subsuming acts under v-concepts cannot guarantee their rightness, and one cannot appeal to the doctrine of the unity of the virtues (as dispositions of character) to underwrite that guarantee (an appeal made by Crisp 2000, and Annas 2011). See further on this issue, relating to Annas, Irwin (2013). 15 Note that the boundaries between being generous but inappropriately so, and not being generous at all are vague and contested (as is the case with the thick concepts as applied to acts in general).

48

Christine Swanton

In summary, given that the v-rules are not an infallible guide to the target(s) of the relevant virtues or indeed to rightness in general we should distinguish a v-act from an act which is virtuous in respect V (i.e. one which hits the (contextually salient) target of V).16 Nonetheless, the thick virtue concepts enable us to articulate reasons by evaluatively loaded description: they can substantively add to the stock of reasons. Acts for example do not merely promote welfare; they may do so kindly or unkindly, in a caring or cold manner, generously or ungenerously, graciously or with resentment. This point is made by Dancy (2004:17): ( . . . ) the applicability of a thick concept is capable of altering, or adding to, the reasons thrown up from below. That an action is obscene makes a difference to how we should act (though not always the same difference) beyond any made by the features that make the action obscene.

In short through re-description and addition to the stock of reasons, the thick concepts make salient what matters. 5. V-RULES CODIFY DEFAULT REASONS

On the view defended in the remainder of this chapter v-rules codify default reasons; they are “true for the most part” in a sense to be explained. As such they guide action and allow virtue ethics to be weakly codifiable. The action guidingness objection to virtue ethics can be overcome, but in a way compatible with particularism about reasons for action. Or so I shall argue. A default rule is understood as a defeasible generalization. For example the v-rule ‘Be honest’ expresses a default that lying is wrong, and the v-rule ‘Be generous’ expresses a default that generous acts are right, unless ‘certain complicating factors intervene’ (Horty 2012: 154). There are three broad kinds of ‘complicating factors’ which defeat defaults. (a) Rebutting defeat. The wrongness of lying for example is defeated by an overriding reason that (for example) the consequences of not lying are devastating. 16

This is the terminology I deployed in Swanton (2003: ch. 11).

A Virtue Ethics

49

(b) Exclusionary defeat. An excluded reason is not overridden by an outweighing reason, but despite the excluded reason’s weight or merit (such as that the generosity would do a lot of good) one refrains from acting for that reason. For example I have made a promise to my spouse not to be generous to this man. In such a case a promise is regarded as an exclusionary reason. (c) “Switching off ” defeat. This kind of defeat occurs when in a particular case the characteristic positive valence of kindness for example has been ‘switched off ’ (Dancy 2004:113), and in that particular case kindness no longer has positive valence. In an example of Dancy’s (2013: 12) the characteristic positive valence of considerateness is alleged to be switched off in the case of the considerate act of wiping the torturer’s brow. Here (according to Dancy), the considerateness of the act ‘hardly functions as a reason’ to wipe the sweat off.17 Note that (c) differs from (b) in that where R (generosity say) has been excluded from consideration as a reason, it is not assumed that the positive default status of generosity has been switched off. The overall merits of generosity in the circumstances may remain intact. Note also that exclusionary reasons are also understood as defaults: that is, such reasons could themselves be defeated by rebutting reasons of sufficient strength. As defaults, exclusionary reasons need not be regarded as absolute. As Raz claims, an exclusionary reason is not simply ‘a reason to avoid thinking, considering, or attending to certain matters.’18 Having considered the strength of a reason one may decide to ‘cancel’ the exclusionary reason. This is not the same as rebutting defeat. Assuming that exclusionary defeats are not reducible to rebutting defeats (an issue not discussed in this chapter)19 we can interpret some v-rules as characteristically expressing exclusionary default reasons such as ‘Be obedient’ understood as ‘Obey (legitimate) authority’ while others characteristically express rebuttable defaults such as ‘Be generous’. Particularist virtue ethics supposes that (at least most) v-reasons are susceptible to switching off defeat. Such a reason is ‘set up to be a reason 17

Of course people differ in their responses to examples including this one. Raz (1999: 184, “Postscript to the Second Edition: Rethinking exclusionary reasons”). My thanks to Garrett Cullity for alerting me to this passage and for discussion. 19 But see Horty (2012) for a thorough discussion and what I regard as successful arguments for the claim that they are not so reducible. 18

50

Christine Swanton

in advance’ (Dancy 2004: 112): they ‘arrive switched on’ (113) and operate as reasons unless ‘switched off if the circumstances so conspire’ (113). Criticism has focused on the particularists’ apparent inability to give an adequate account of the notion of a reason “arriving switched on”, even though it is clearly important to do so to avoid the “flattening of the moral landscape” objection cited above. Indeed Dancy himself describes his account of a default reason as ‘a bit thin’ (113). However in his (2007) Dancy offers a suggestion that seems to me to be on the right track. A default reason does not require what Dancy calls ‘some form of positive intervention’ (2007: 92) to make it a reason, though like all reasons it requires ‘enablers’ (conditions for the presence of reasons not themselves part of a reason) to permit the reason to operate as a reason, such as the ‘general enabler’ (Dancy 2004: 40) that an agent has the capacity to act on that reason. To show the difference between reasons that require intervening factors to make them reasons from those that do not, Dancy offers this example. That John asks Tim not to do it is ‘made into a reason’ (2007: 92) for Tim to do it by some intervening consideration such as a Parlour Game. In general the proposal allows us to distinguish Little’s colour of shoelace example from the inflicting pain example as reasons of different kinds when these considerations operate as reasons. Dancy concludes that ‘some features do not need to be turned into reasons by the presence of some other feature, and these are the default reasons’ (92). If so then we can make the following claim. From the fact that default reasons do not need to be turned into reasons, that is made into reasons by intervening factors, we cannot infer that nothing makes them possess default status as reasons. It is not intervening factors, but standing though not necessarily universally present features that make default reasons arrive already switched on. Default status needs to be explained by appeal to such “standing” conditions. Part of the problem in understanding this idea may lie in the fact that Dancy appears to confine the need for explanation to the intervening factors. He says: ‘( . . . )where the justness of an action counts in favour of doing it, there is nothing to explain; in that sense this is what we should expect to happen’ (2004: 113). Nietzsche by contrast thinks that the nature of justice, that ‘rarest of all virtues’, ‘an impossible virtue’,20 is systematically wrongly understood

20

Friedrich Nietzsche (1983) Untimely Meditations II, 6.

A Virtue Ethics

51

in a resentment-filled and punitive culture such as ours (Swanton 2015). What Dancy rightly objects to is that the standing features which make v-reasons arrive already switched on, whatever they are, must survive as a “trace” when the default reason is defeated. He asks: ‘What is the sense in which there is a trace of the non-exceptional case in the exceptional one?’ (2004: 115). Let us illustrate the problem with helpfulness. Consider a case where one greatly helps someone’s creative efforts to the point where the work scarcely expresses anymore his own genuine creative effort. The purpose of the help is to strengthen his chance of winning an artistic competition by making his work more congenial to the conservatives on the panel. The strategy works. Now what makes a disposition to help a virtue, and thereby also makes helping a reason for action that arrives already switched on, is that it promotes welfare in a way that expresses and strengthens bonds between people. However in this case the help undermines the originality and integrity of the developing work, not only dirtying the prize won but, let us say, starting a process of character corruption as the “artist’s” career unfolds.21 In this case the helping contaminates creativity. As a result, just as there need be no trace of goodness in the lemon juice when it has been contaminated by a poison, there need be nothing good about the helpfulness. How do we determine what makes v-reasons arrive switched on? What in general are the “standing conditions”? Dancy himself suggests where we should look for an answer to these questions: in the very nature of the feature which provides the reason, be it courage, kindness, or pleasure. In other words the explanation for the switched on status of reasons has to be uncovered through normative ethics: by investigating the evaluative point of courage in situations of danger, the point of justice in situations of scarcity, the purpose of help in the face of genuine neediness, the nature of pleasure as pleasant. For Aristotle, being pleasurable thus arrives switched on as a reason although it is switched off when contaminated by the sadism of the pleasure (absent perhaps special contexts). Pleasure is not good ‘without qualification.’ In cases of contamination, the pleasantness of the sadistic pleasure to the sadist does not

21

Compare the case of Keating in Ayn Rand (1952).

52

Christine Swanton

survive as a ‘trace’ imparting positive valence to the pleasurableness as a reason favouring the sadism. Most virtue ethicists at least require a substantive account of the reason-giving status of the reasons codified in the v-rules. Such accounts explain why those reasons arrive switched on by elucidating the point and function of the various virtues in relation to the goodness of a life to which it is proper to aspire in the type of world in which one finds oneself. That elucidation allows us to give normative “shape” to the heterogeneous natural features judged to be kind and so forth. Various Aristotelian, Humean, Confucian, Nietzschean styles of virtue ethics will of course provide differing accounts both of human nature, the type of world we live in and to what we can aspire, and the nature of virtue in such worlds. Naturally it is contentious what the v-rules having positive default are, and how we are to understand them. ‘Be meek’, ‘Be humble’, are controversial examples of (putative) v-rules: indeed understanding the nature of meekness as a virtue (or vice) for example, occupies books and articles in substantive virtue theory (e.g. Pettigrove 2012). Even if we are clear that a particular v-rule has positive default it may be questionable how strong that default is and why. Providing accounts of the nature of v-reasons is not only controversial but massively complicated. Take the putative v-rule ‘Be obedient.’ In this autonomy-obsessed society many would find the very idea of such a rule distasteful even repugnant. However the (putative) rule is described at such a high level of generality and abstraction that it is indeterminate whether being obedient has negative or positive default. To determine its default status we need to uncover the point, function, and contexts of obedience in a good life. This is a complex matter for the nature of a virtue is contoured by several broad contexts including roles, cultural and historical contexts, and stage of life. This is particularly true of a disposition of appropriate obedience where that is understood as a virtue—a disposition of excellence in relation to the field of obeying authoritative edicts. In professional role ethics obedience has positive default status: roles are structured by authority relations, such as rules, protocols, bosses’ instructions, whose points are themselves structured by the function of the institution in which the roles are embedded, be it medicine, university education, business, law. Justified whistle-blowing provides an example where the status of obedience as providing exclusionary reasons has been switched off or overridden. In children,

A Virtue Ethics

53

obedience is also a reason having positive default, but is switched off in the case of “stranger danger”. Here is another complication: the nature of the world in relation to which “standing” conditions are understood. In particular, default status is not necessarily indexed to normality. In what Vayrynen (2004) calls ‘Nasty Worlds’ which are assumed to be abnormal (unlike this world which is assumed to be good enough to be called normal)22 the default status of many v-rules may switch from positive to negative; in other words the (current) v-rule would no longer be a v-rule. Nastiness (of various sorts) would no longer be intervening factors turning features into reasons but standing conditions making e.g. being mistrusting and being insensitive “arrive switched on” as reasons, and thereby having positive default.23 Notice finally that the v-rules do not merely codify reasons having positive default status: the reasons codified may vary considerably in both importance and stringency. Some are both highly stringent and highly important such as ‘Be just’. Others are neither important (relatively speaking) nor stringent such as ‘Be witty’.

6. THE V-RULES, PARTICULARISM, AND CODIFIABILITY

My suggestion is that the v-rules codify default reasons that are capable of switching off defeat. Understanding the v-rules in the way suggested above produces a normative ethics that is both codifiable and Dancystyle particularist. However the very idea that Dancy’s particularism is compatible with any sense of codifiability may seem astonishing. It is thus important to show that the “uncodifiability” reading is incorrect. Discussions of his position are complicated by crucial ambiguities in particular the notions of a (moral) principle and ‘generalism’. What is common to the several types of ‘moral principle’ distinguished by Dancy in his (2004: 3ff.) is that the reasons they identify have invariant status. As Dancy claims (2004: 76): 22 This is of course a highly problematic assumption. For an account of how problematic see Tessman (2005). 23 Nicholas Everitt (2007: 281, n.16) gives the following example: ‘The American comedian Richard Pryor recalling the tough area in which he grew up remarked crisply “If you were sensitive in my neighbourhood you were food”.’

54

Christine Swanton

Moral principles, however we conceive of them, seem all to be in the business of specifying features as general [i.e. invariant] reasons. The principle that it is wrong to lie, for instance, presumably claims that mendacity is always a wrongmaking feature wherever it occurs ( . . . )

Dancy then defines particularism as the view that ‘the possibility of moral thought and action does not depend on the provision of a suitable supply of such principles’ (2004: 73).24 This is of course compatible with the actual existence of a suitable supply of such principles. Does such a supply actually exist? It depends on what one means by ‘principle’ in a sense relevant to ethics. Dancy is clear on this: principles in that sense must be norms or rules that ‘we could hope to grasp and operate’ (2004:12), one of whose main purposes is ‘telling us why we should do the action’ (2004: 127). The sort of “principle” which Frank Jackson has in mind: ‘the sort of ‘principle’ that has an entire world-description on its left hand side ( . . . ) is blatantly incapable of serving the purposes for which principles are usually required (e.g. guidance, or the specification of a reason for doing something)’ (Dancy 2004: 126). One might think that a suitable supply of principles does exist: those specified by the v-rules. However Dancy denies that in general the thick concepts can provide moral principles for they do not have invariant valence: ‘My own view is that almost all the standard thick concepts, such as those of fidelity, gratitude, reparation ( . . . ) are of variant valence’ (2004: 121), though they do ‘give a certain shape to what is thrown up from below ( . . . ) and in some way prepare things for overall judgment’ (122). This ‘shape’ is understood through investigations of the evaluative point of fidelity and so forth, and as we saw above, this is what justifies the status of v-reasons as arriving already “switched on”. However as alluded to above, he does not himself offer a normative theory such as virtue ethics to assuage the critics: those who claim that he cannot account for ‘moral pluses and moral minuses’. Appeal to v-rules within a particularist virtue ethical framework, I have suggested, can do the job that is needed.

24 Of course ‘suitable supply’ is a vague and contested idea: what is presumably meant here is some kind of codifiability of at least a large and non-peripheral part of ethics in terms of invariant principles.

A Virtue Ethics

55

This position may be described thus: Virtue Ethical Particularism (VEP): Though there is no ‘suitable supply’ of moral principles codifying invariant reasons for action, there is a suitable supply of rules codifying reasons as v-reasons, and at least most v-reasons have variant valence (though arriving already switched on as argued above).

The “default generalism” of VEP must be contrasted with: Invariant Generalism (IG) There is a suitable supply of principles codifying invariant reasons for action.

VEP must also be distinguished from three other particularist positions, as follows. Extreme Particularism (EP): Moral reasons are not expressible in any kind of (usable) rule or principle.25 Weak Particularism (WP): The normative valence of some reasons varies. (See Cullity 2002)

WP simply denies that ‘a consideration cannot count in favour of any action unless it counts in favour of every action’ (Cullity 2002: 169, Abstract). This position is weaker than VEP since the existence of one reason with variant valence is compatible with there being a ‘suitable supply’ of moral principles. Strong Particularism (SP) There are no normative reasons of invariant valence.

VEP is weaker than SP. For example VEP is consistent with the view that it is analytic that being just or being one’s duty always has positive valence (Cullity 2002). The attraction of VEP is this. The view that morality does not depend on the truth of IG is compatible with the truth of a claim one might deem essential for resolving the “flattening of the moral landscape” 25 A potent reason for the apparently widespread assumption that Dancy subscribes to (EP) is that the concept of codifiability is assumed to apply only to invariant principles. See for example Little (2000: 276).

56

Christine Swanton

problem, namely that morality does depend on a suitable supply of normative generalizations having positive default, such as those capturable in v-rules. Unfortunately there may be a serious problem with VEP. Is there ‘a suitable supply’ of v-rules? That is, are the v-reasons codified by such rules sufficiently pervasive? Call a negative response to these questions the Non–Pervasiveness Objection to VEP. There are two aspects to the Non–Pervasiveness Objection (NPO). First many normative reasons are what Dancy calls ‘enticing’ (as opposed to peremptory) and these may be deemed to be “non-moral”. Second many peremptory reasons arguably do not map onto v-rules. In the case of both of these types of reasons there are allegedly no targets of virtues to which v-rules guide. According to the first aspect of the NPO, for many types of reasons, there are no virtues whose study may offer grounds for the claim that they arrive already switched on. According to the second aspect of the NPO there are no v-rules, rules generated by virtue concepts, codifying many types of reasons. Consider first the enticing reasons. According to Dancy: Enticing reasons are to do with what would be fun, amusing, attractive, exciting, pleasant, and so on. (2004: 21)

One interpretation of this passage causes difficulty for my view. The difficulty is this. On that interpretation, virtue reasons are “moral” reasons and such reasons have to do with the peremptory: what virtue demands of us.26 Furthermore what would be fun, amusing, and so on are not within the domain of the moral and do not ‘take us to an ought’ (2004: 21) or a demand of virtue. However in reply we may note that Aristotle and others such as Hume do not confine virtue to the “moral” which has no clear sense. Not only are there virtues and vices associated with these alleged domains of the enticing, but they may make demands on us. Those who think that the purpose of life is to have fun are liable to go wrong in the domain of feeling and action in relation to “being fun”, as indeed are those joyless individuals in the grip of Hume’s ‘monkish virtues’. As for the amusing, both Aristotle and Hume recognize virtue and vice in this area. For Hume:

26

See further Annas (2014) and Crisp (2014).

A Virtue Ethics

57

Cheerfulness could scarce admit of blame from its excess, were it not that dissolute mirth, without a proper cause or subject, is a sure symptom and characteristic of folly, and on that account disgustful (E para. 208, 258 n. 1).

Kant itemizes a disposition to mocking banter as a vice of disrespect; Aristotle contrasts wittiness as a virtue with for example boorishness. Peremptory reasons apply in all these cases: for example the appropriate virtue may require in various circumstances that one not have fun in the wrong way (through sarcasm) or at the wrong time (when solemnity is required) and so forth. The amusing, the funny, the exciting, may give rise to both enticing and peremptory reasons in all sorts of different contexts. The second aspect of the NPO is this. Many peremptory reasons, it may be claimed are not capturable in v-rules. As Dancy himself puts it, to assess the normative force of features such as telling jokes, making certain remarks, making certain gestures, one needs to bring to them some ‘sort of practical shape’ (2004: 84) and this we do via the thick concepts such as the tasteful or distasteful. But then, according to the objection, for a very large class of these the shape is not provided by virtue/vice concepts (including being distasteful).27 I by contrast think that being distasteful is a vice concept, a concept that can and does generate a v-rule. Some people are distasteful as matter of disposition and may even glory in it. Here it is a vice of exhibitionism. Furthermore it is clear that ‘Don’t be distasteful’ is a rather stringent default rule even if it is not an important rule by comparison with e.g. ‘Be just’. Notice that the above view is not committed to the thesis that all v-reasons are expressible in v-rules. The view that there is a “suitable supply” of such rules, and that this supply is necessary to avoid the “flattening of the moral landscape” objection, does not imply that all v-reasons need be understood in terms of v-rules. There may be no v-rule ‘Be sentimental’, for example, and no v-rule ‘Don’t be sentimental’. In relation to sentimentality it is controversial where the default lies, if indeed there is a default. If I am correct, reasons for actions should not be analyzed or defined as features expressible in default rules, but are simply features in favour of actions as many (such as Dancy and Scanlon) believe. 27

A view expressed to me by Dancy in conversation.

58

Christine Swanton 7. THE BURDEN OF PROOF

I have shown what a virtue ethical particularism could look like. But why is there so much resistance to particularism, despite the well-recognized complexity of the ethical realm? Consider a Rossian and a Particularist arguing about whether kindness always has positive valence. The context is Jerry’s just presented mediocre paper and Jim’s kind praise of the paper. PARTICULARIST:

We do not want kind praise when reading an academic paper: we want the praise to be wrung out by the sheer excellence of the paper.28 ROSSIAN: Yes: one should not praise mediocre academic papers out of kindness in academic contexts, but the kindness of the praise is at least one good thing about the situation. PARTICULARIST: No: the kindness of the praise is a negative feature. Jim’s action inherently undermines the integrity of the academic purpose. ROSSIAN: No you can’t say that: there are plenty of occasions where kind praise of mediocre academic papers is called for: where a speaker is unusually anxious; where the speaker is a graduate student giving her first paper in a terrifying context, and so on. PARTICULARIST: I was not generalizing to those kinds of cases: I was just talking about Jim’s reaction to Jerry’s paper today. In fact let me put in a few more details about Jerry’s performance, and Jim’s behaviour. Jerry’s paper was not only mediocre but he was particularly arrogant and nasty to his opponents. He did not deserve kindness in relation to his paper. Furthermore, as we know, Jim is constantly self-sacrificing, but in staying behind to bestow kindness on Jerry he missed the opera. I would say that Jim’s act of kindness here is not even pro tanto appropriate. Kindness here is expressive of a vice of self-abnegation and this is a potent reason why the kindness of the praise in this particular case has negative valence. None of this of course suggests that kind helpfulness on Jim’s part, were James to collapse of heat exhaustion in the middle of yet another sneering remark, is out of order, even were he to miss the opera, especially since Jim, amongst all the impractical academics on show, is the only one who has taken a first aid course. 28 I owe this example to Jonathan Dancy (in conversation) but he is not responsible for the embellishments that follow.

A Virtue Ethics

59

As I have discovered, there is a drive to defend IG in the face of examples such as these. The reason for the difficulty is that the use of examples designed by the particularist to persuade one out of the invariance “prejudice”, fail to persuade when one is in the grip of that prejudice. This is, of course, a tendentious way of putting it. So let us say that the reason for the failure of what, for the particularist, are persuasive examples, is that the non-particularist is against particularism for serious theoretical reasons. Particularism can’t be right—it is after all ‘wrong and bad’ as Hooker (2000) says. The burden of proof is therefore seen to be on the particularist, and it is hard to discharge that burden through examples. That is why it seems axiomatic to many that the v-rules express invariant moral principles (‘thick moral features ( . . . ) are guaranteed of carrying a given valence of moral significance’ (Little 2000: 289). So the particularist needs to reduce the force of the theoretical objections. I hope to have shown that a virtue ethics conforming to VEP can supply rules which account for moral pluses and minuses, that the notion of a switched off default in the case of v-reasons can be explained by substantive accounts of the evaluative points and grounds of virtues—accounts which explain why v-reasons arrive already switched on—and that the ‘flattening of the moral landscape’ objection can thereby be disarmed. What about the scepticism concerning moral education of the young? The v-rules are action guiding, since they help locate what matters normatively speaking (the targets of the virtues) and (unless switched off, excluded, or overridden) guide one thereby to right action. What Hursthouse describes are what might be called basic v-rules; that is, rules that provide anchors for our moral thought and can be taught to children, but are too abstract and insufficiently sophisticated to guide us reliably in, for example, role contexts. Here we need accounts of the targets of what I have called ‘differentiated’ virtue (2007) such as virtue differentiated by role. Differentiated virtue allows for the nuances of, for example, role differentiation, cultural and historical location, and the narrative particularity of our lives (Swanton 2007). A virtue ethics of roles could say systematic things about the relation between role virtue and the basic virtues that generate the v-rules; a narrative virtue ethics differentiates basic virtue according to norms of narrative;29 a virtue

29

For discussion of cultural and narrative dimensions see Walker (1998).

60

Christine Swanton

ethics sensitive to the aspirational quality of virtue as something developed over time can speak of norms of cultivation;30 a virtue ethics sensitive to cultural location or the “burdened” quality of virtue31 will also add to normative complexity. Forms of differentiation add exponentially to the complexity of the moral realm, and with some notable exceptions virtue ethics has barely touched on this complexity.

8. CONCLUSION

This paper has argued that there could be a codifiable Dancian particularism applied to virtue ethics. The trouble with discussion of particularism so far has been that it has been conducted at a meta-ethical level so that objections such as unpredictability and so forth appear to loom large. What needs to happen and what I have attempted here is to show how particularism can be applied to a normative ethics: one containing a ‘suitable supply’ of v-rules as opposed to moral principles in the invariant sense. The v-rules guide by objective standards reflecting the evaluative point of the virtues, as explored through various types of virtue theory. Though the v-reasons characteristically codified by the v-rules as applied to action are susceptible to switched off default, some such as reasons of justice may nonetheless be near absolute. Their default status is highly stringent (as well as important). Finally can the v-rules guide us even in tragic dilemmas, understood as dilemmas in which all alternatives are terrible? Indeed they can; in fact I would go as far as to say that v-rules are particularly suitable in identifying a large number of potentially relevant features in such a dilemma. A mistake in the analysis of tragic dilemmas is to focus on only a very few v- rules such as the requirement to be beneficent, and since one cannot act in accordance with those rules it is assumed that one cannot but act wrongly in such a dilemma. However in tragic dilemmas there are many v-rules in play such as ‘Be morally serious’, ‘Be caring’, ‘Be resolute’, ‘Be courageous’, ‘Be non-maleficent’, ‘Act with integrity’. To be morally serious is to be disposed to appreciate the gravity of situations, and that virtue too has targets. To act with integrity (in the

30 For an account of virtue sensitive to the developmental aspirational quality of virtue see Annas (2011). 31 For the idea of burdened virtue see Tessman (2005).

A Virtue Ethics

61

example given below) involves not cooperating with evil. Sophie of Sophie’s Choice (Styron 1975) could neither act beneficently nor nonmaleficently because her acts could not possess the descriptive features characteristic of such acts. Perhaps her actions could conform to the rules ‘Be morally serious’, ‘Be courageous’, ‘Be resolute’, ‘Act with integrity’, and even ‘Be caring’ (insofar as that v-rule tends to target mainly the motives of action). She conformed to or could have conformed to at least the first of these rules but not the rules of resoluteness and integrity. However one might argue that the positive default status of ‘Be resolute’ and ‘Act with integrity’ (understood in the present way) are here switched off, since resoluteness and integrity in this kind of situation would seriously detract from Sophie’s virtue as a mother.32 On this view it is not that the v-rule ‘Be resolute’ (for example) is overridden by the claims of a conflicting v-rule: it is rather that being resolute, in this particular context is thought to have negative valence. Given all this, we might say that Sophie acted (or could have acted) rightly. It does not follow from this claim that a v-rule (be courageous or be morally serious say) directs Sophie to just one of the alternatives: what might be required is that one act on either alternative in a certain way. However for some, such as Hursthouse, an act may be too terrible overall to be called right (though on her view one can make a right decision). For others such as myself rightness as correctness is relative to situation: targets of virtues are relative to situation, and the contextually determined targets of some virtues may be met even in terrible situations. Whichever view one adopts, even in tragic dilemmas, some v-rules may guide one to targets of virtue, but not of course infallibly. Acknowledgments I am indebted to the participants of the Arizona Normative Ethics Workshop, and to Garrett Cullity who commented on previous drafts. I am also grateful to Jonathan Dancy and two referees for comments.

32 Slote (2011) claims that ‘she can’t in all conscience refuse to cooperate with evil in a case like this’ (76), but does not go so far as claiming that the v-reason of integrity (understood as ‘Don’t cooperate with evil’) here has negative valence.

62

Christine Swanton REFERENCES

Adams, Robert Merrihew (2006). A Theory of Virtue: Excellence in Being for the Good (Oxford: Clarendon). Annas, Julia (2011). Intelligent Virtue (Oxford: Oxford University Press). Annas, Julia (2014). ‘Why Virtue Ethics does not have a Problem with Right Action’ in Mark Timmons ed. Oxford Studies in Normative Ethics Vol 4. (Oxford: Oxford University Press). Aristotle (1976). Nicomachean Ethics trans. J. A. K. Thomson, revised H. Tredennick (Harmondsworth: Penguin). Aristotle (2000). Nicomachean Ethics trans. Roger Crisp (Cambridge: Cambridge University Press). Crisp, Roger (2000). ‘Particularizing Particularism’ in Brad Hooker and Margaret Little eds. Moral Particularism (Oxford: Clarendon Press), 23–47. Crisp, Roger (2014). ‘Supererogation and Virtue’ in Mark Timmons ed. Oxford Studies in Normative Ethics Vol 4 (Oxford: Oxford University Press). Cullity, Garrett (2002). ‘Particularism and Presumptive Reasons’ Proceedings of the Aristotelian Society Supp. Vol. 76: 169–90. Curzer, Howard (2005). ‘How Good People Can Do Bad Things: Aristotle on the Misdeeds of the Virtuous’ in David Sedley ed. Oxford Studies in Ancient Philosophy XXVIII (Oxford: Oxford University Press), 233–56. Dancy, Jonathan (1993). Moral Reasons (Oxford: Blackwell). Dancy, Jonathan (2000). Practical Reality (Oxford: Oxford University Press). Dancy, Jonathan (2004). Ethics Without Principles (Oxford: Clarendon Press). Dancy, Jonathan (2007). ‘Defending the Right’ Journal of Moral Philosophy 4: 85–98. Dancy, Jonathan (2013). ‘Moral Particularism’, The Stanford Encyclopedia of Philosophy (Fall 2013 Edition), Edward N. Zalta ed., URL = . Das, Ramon (2003). ‘Virtue Ethics and Right Action’ Australasian Journal of Philosophy 81: 330–4. Enoch, David (2005). ‘Why Idealize?’ Ethics 115, 4: 759–87. Everitt, Nicholas (2007). ‘Problems with Virtue Theory’ Philosophy 82: 275–99. Hooker, Brad (2000). ‘Moral Particularism –Wrong and Bad’ in Hooker and Little eds. 1–23. Hooker, Brad and Margaret Little eds. (2000). Moral Particularism (Oxford: Clarendon Press). Horty, John F. (2012). Reasons as Defaults (Oxford: Oxford University Press). Hudson, Stephen (1986). Human Character and Morality (Boston: Routledge and Kegan Paul). Hume, David (1975). Enquiries Concerning Human Understanding and Concerning the Principles of Morals, 3rd edition, ed. P. H. Nidditch (Oxford: Clarendon Press). Hursthouse, Rosalind (1999). On Virtue Ethics (Oxford: Oxford University Press). Hursthouse, Rosalind (2006). ‘Practical Wisdom: A Mundane Account’, Proceedings of The Aristotelian Society 106: 285–309. Irwin, Terence (2013). ‘Review: Julia Annas Intelligent Virtue’ Ethics 123: 549–56. Lance, M. and Little, M. (2006). ‘Particularism and Anti Theory’ in D. Copp ed. The Oxford Handbook of Ethical Theory Oxford: Oxford University Press, 20–1; cited in McKeever and Ridge, 2007. Little, Margaret Olivia (2000). ‘Moral Generalities Revisited’ in Hooker and Little eds. 276–304. McDowell, John (1979). ‘Virtue and Reason’ Monist 62: 331–50. McKeever, Sean and Michael Ridge (2006). Principled Ethics: Generalism as a Regulative Ideal (Oxford: Oxford University Press).

A Virtue Ethics

63

McKeever and Michael Ridge (2007). ‘Turning on Default Reasons’ Journal of Moral Philosophy 4, 1: 55–76. McNaughton, David and Piers Rawling (2000). ‘Unprincipled Ethics’ in Hooker and Little eds. 256–75. Nietzsche, Friedrich (1983). Untimely Meditations, Essay II ‘On the Uses and Disadvantages of History for Life’ trans. R. J. Hollingdale (Cambridge: Cambridge University Press). Pettigrove, Glen (2011). ‘Is Virtue Ethics Self-Effacing?’ Journal of Ethics 15, 3: 191–207. Pettigrove, Glen (2012). ‘Meekness and Moral Anger’ Ethics 122, 2: 341–70. Rand, Ayn (1952). The Fountainhead (New York: Penguin). Raz, Joseph (1999). Practical Reason and Norms 2nd ed. (Oxford: Oxford University Press). Scanlon, T. M. (2014). Being Realistic About Reasons (Oxford: Oxford University Press). Slote, Michael (2011). The Impossibility of Perfection: Aristotle, Feminism, and the Complexities of Ethics (Oxford: Oxford University Press). Smith, Alexander McCall (1998). The No. 1 Ladies’ Detective Agency (New York: Anchor Books) Styron, William (1975). Sophie’s Choice (New York: Random House). Swanton, Christine (2001). ‘A Virtue Ethical Account of Right Action’ Ethics 112: 32–52. Swanton, Christine (2003). Virtue Ethics: A Pluralistic View (Oxford: Oxford University Press). Swanton, Christine (2007). ‘Virtue Ethics, Role Ethics, and Business Ethics,’ in Rebecca L. Walker and Philip J. Ivanhoe eds. Working Virtue: Virtue Ethics and Contemporary Moral Problems (Oxford: Clarendon Press), 207–24. Swanton, Christine (2014). ‘The Notion of the Moral: The Relation between Virtue Ethics and Virtue Epistemology’ Philosophical Studies (Special Volume Selected Papers APA Pacific Meetings 2013) Vol. 171, 1: 121–34. Swanton, Christine (2015). The Virtue Ethics of Hume and Nietzsche (Oxford: Wiley Blackwell). Tessman, Lisa (2005). Burdened Virtues: Virtue Ethics for Liberatory Struggles (Oxford: Oxford University Press). van Zyl, Liezl (2014). ‘Right Action and the Targets of Virtue’ in Stan van Hooft ed. The Handbook of Virtue Ethics (Durham: Acumen), 118–29. Vayrynen, Pekka (2004). ‘Particularism and Default Reasons’ Ethical Theory and Moral Practice 7: 53–79. Walker, Margaret Urban (1998). Moral Understandings: A Feminist Study in Ethics (New York: Routledge). Williams, Bernard (1995). ‘Replies’ in J.E.J. Altham and Ross Harrison eds. World, Mind, and Ethics (Cambridge: Cambridge University Press), 185–225.

3 My Welfare and Yours MARK LEBAR

Human beings are ineluctably social creatures. There is no story of our emergence in history in which we are not living with one another. So if we want to understand what is good for us—what is the nature of our welfare or wellbeing—it would be strange indeed if we could do so without considering not only how others impinge on us, but how their welfare interacts with ours. Not all philosophical accounts of welfare can manage that task gracefully. My aim in this paper is to explore how one such account does so. In recent work I have developed a picture of rationality modeled along the lines of ancient eudaimonist thought (mostly Aristotle’s).1 I call it “Virtue Eudaimonism” (VE), and—as might be expected of a view heavily influenced by Aristotle—it maintains (inter alia) that living well2 is an Ultimate End for us (in a sense I shall explain), and that we live well by being wise and virtuous. More to the point: we live well in the company of others, and our good lives interact with the good lives of others in interesting ways. In this essay I will explore some of the ways the welfares of individuals interact and inform each other, focusing on the interactions with intimate others—those with whom our welfares become most intricately involved. My aim is both to shed light on these philosophically-underexplored relationships and to exhibit the power of VE to shed that light. My vehicle for this exploration is the character of Dr. Josef Breuer, from Irvin D. Yalom’s novel, When Nietzsche Wept.3 In this fictionalized account of a relationship the real Dr. Breuer and the real Friedrich

1

The Value of Living Well (OUP, 2013). For present purposes I treat “welfare”, “well-being”, “living well”, and “happiness” as interchangeable terms. 3 Harper Perennial, 1992. Page citations in text are to this book. 2

My Welfare and Yours

65

Nietzsche might have had, we can consider how individuals living well interact within a shared life, in particular the life Dr. Breuer shares with his wife, Mathilde. Breuer’s case is interesting because (with only modest embellishment) it is a nice example not only of how the goods of Josef and Mathilde interact but of what is at stake when they threaten to pull apart. I will explore how VE’s conception of what it means for us to be social rational animals illuminates all-too-familiar practical problems such as Breuer’s, and thus how my welfare and yours can be mutually informing.4 I

Dr. Breuer is a wealthy and successful late 19th-century Viennese physician. As Yalom has it, Breuer has pioneered a method of treating “hysteria” that we would now recognize as a precursor of modern psychotherapy. In this light he acts as a mentor to a younger colleague, Sigmund Freud, and is inveigled into a plan to treat the philosopher Nietzsche for his despair. Breuer engages in what will come to be known as Freud’s “talking cure” to help Nietzsche, but finds that successful treatment of such a deep and profound spirit as Nietzsche’s requires on his own part a considerable amount of self-scrutiny and, eventually, selfrevelation. Breuer’s self-scrutiny reveals his own deep unhappiness at a life that is, while by any external measure an enormous success, mostly empty to him. He feels trapped in his marriage, his family, and his career. Under the pressure of Nietzsche’s challenge to consider the source of his anxiety and sense of “crushing confinement” (197), his dread and longing to escape (223), Breuer contrives a plan to experience what it would be like for him to leave his life behind and begin again—a life now fully and consciously chosen, but involving his separation and divorce from Mathilde. Spoiler alert: Yalom does not have Breuer actually undertake this choice. Instead, he has Freud hypnotize Breuer and—via hypnotic suggestion and guided visualization—walk him through what that experience would be like. My aim is in no sense “therapeutic”; though their marriage is threatened with failure, I have no pretensions to insights as to how to save it. Indeed, it seems to me possible that dissolution might in the end have been the right course of action (although Yalom avoids this conclusion). 4

66

Mark LeBar

I part company with Yalom’s imagined narrative, to consider the decision Breuer faces as to his marriage to his wife Mathilde. As Breuer experiences his life and confronts his anxiety, his marriage and what it brings in its train (relationships and commitments) is a significant part of the problem. He exclaims to Freud: How can I be free!? It’s not the same as you, a young single man giving up a suffocating university career. It’s too late for me! I have a family, employees, patients, students. It’s far too late! We can talk forever, but I cannot change my life—it is woven too tight with the thread of other lives. (249)

So when he decides to break free, the first step in his (imagined) abandonment of his old life is to announce to Mathilde that he is leaving his family and his practice: his life as he has known it. How does Mathilde respond? Much as we might expect, and as (no doubt) Breuer himself expected, she responds with shock and anger: “Josef, this is madness!” Mathilde’s voice rose. Her eyes grew wide with fear. “What has happened to you? Since when is there a your life and a my life? We share a life; we made a covenant to combine our lives.” (256) “You speak of freedom. What a cruel joke to me! I wish I had your freedom—the freedom of a man to obtain an education, to choose a profession. ( . . . ) You want to leave? You want to make new life choices? Have you forgotten the choices you have already made? You chose to marry me. And do you truly not understand that you chose to commit yourself—to me, to us? What is choice if you refuse to honor it? ( . . . ) Don’t you understand that you cannot enter a covenant with me and then suddenly say, ‘No, I take it back, I’m not sure after all.’ That’s immoral. Evil.” (257–8)

Mathilde brings home the argument that Josef is not merely making decisions about his own life, but about hers as well. Moreover, these are not independent lives, but have, morally and in practice, been joined together. Breuer’s response is revealing. He does not deny either that he loves Mathilde or that he has a duty to her, but he claims that he has a “higher duty” to himself, and it is this duty that requires him to renounce his life and seek a freedom of a sort he does not yet feel. It would be a mistake to deny that there are real pressures on both sides of a decision such as the one Breuer faces. They are all too readily

My Welfare and Yours

67

imaginable, and faced by all too many people in one form or another. They are in that way quotidian. They may also be tragic: there may be no outcome in which one party or the other does not suffer, and suffer significantly. Yalom’s Breuer has the luxury of exploring these possibilities in hypnotic imagination, but most of us do not. I will consider how Breuer’s deliberation might be structured by the account VE gives of our good or welfare, and the way that good informs and is informed by those with whom we entangle our lives, as Josef and Mathilde have entangled theirs. I begin by working through VE’s view of what Breuer’s good looks like taken independently of Mathilde’s (or indeed, of other such commitments). Then I consider how such an independent good is affected when two have this sort of commitment and involvement with each other, but do not have conflicting deep interests. I conclude by considering what we can learn from such mutually informing goods when they threaten to come apart. II

On VE, Breuer’s good, like that of all rational social animals of the sort we are, is realized by fulfilling what we might think of as his “two natures”.5 Breuer’s first nature consists in his animality, including (to some degree) his sociality (since sociality is not found merely in humans). As Aristotle observes (NE I.7), we share many qualities with other living things, and what is good for them in virtue of those qualities is likewise by analogy generally good for us. Obvious examples would be nutrition, growth, and health. But unlike other animals, we also have what we can call a “second nature”. This second nature consists, roughly, of our rationality and its effects on our first nature. This is only rough because these effects are not one-way: though having reason changes our animality and sociality in important ways, it is also conditioned and structured by them.6 These mutual ramifications merit discussion.

5 In the idea of “two natures” I follow John McDowell, “Two Sorts of Naturalism”, in Hursthouse, Lawrence, and Quinn (eds.) Virtues and Reasons: Philippa Foot and Moral Theory (Oxford: Clarendon Press, 1995), 149–79. 6 This explains why the goodness of the properties we, like other animals, have is only “by analogy” good. While a discussion of the metaphysics of value involved in VE is beyond the ambit of this paper, the work of reason plays a key role in establishing or constituting value

68

Mark LeBar

VE follows Aristotle in supposing that we best think of our practical rationality in terms of our end-seeking nature. Ends are, roughly, objects of action or intention we take ourselves to have reason to seek.7 Some ends we seek for their own sake (these we call final ends); others we seek for the sake of further ends. Most of our final ends we seek both for themselves and for the sake of further ends—including and especially happiness (eudaimonia). Thus, for example, Breuer tries to understand hysteria both because he seeks to heal his patients and for the sake of medical understanding. Where do our ends come from? One important source is the impulses we inherit from our first natures. What in other animals are merely the urges and drives imposed by instinct are or can be “taken up” when aimed at objects we take ourselves to have reason to seek, and thus become ends. But our very capacities that do this “taking up” are shaped by our animal natures; thus Aristotle says that “choice” (prohairesis) is “either desiderative thought or deliberative desire” (NE VI.2). Overall, because we seek our ends not only for their own sake but for the sake of happiness, the idea of happiness is what affords an impetus to new ends, a basis for discarding potential ends or existing ends, and a framework within which many of our indeterminate ends (“I want to be a better physician”) take on shape and structure. Happiness (living well) is itself one of these indeterminate ends; it takes on shape and structure as we live our lives. Understanding what happiness consists in is something we can get better at, as we gain greater understanding of what is good for us and become more integrated in serving that good in our first and second natures. Thus we come to the crucial role of virtue in our welfare. VE follows Aristotle in maintaining that living well is impossible without virtue, and that living well is the point of virtue. (As with other dimensions of our lives, virtue gives us ends of its own, while we also seek them for the sake of happiness.)

properties, and this role is different in our own case from the case of other animals. See LeBar 2013, Part II. I thank an anonymous referee for raising this point. 7 “Seek” here is a placeholder for a wide variety of attitudes or actions that ends may give us reason to undertake, including “being constrained by”, as we shall see in §IV. For the constraining and other work of ends in our practical rationality, see LeBar 2013, ch. 1.

My Welfare and Yours

69

III

With that overview in hand, let us look at a concrete case in which Josef Breuer—considered for the moment merely as an unconnected individual—must choose how to advance his own good. As Yalom paints the picture, as a scientist and a Jew Breuer early on faced a fork in his career path in late 19th century Vienna. His best opportunity for advancement at the University of Vienna slipped away when his mentor died of typhus and Breuer was passed over for his replacement. Breuer chose not to try to press ahead in the University in the face of rising anti-Semitism, instead moving his research into his home and continuing it there, as well as undertaking a robust medical practice. Suppose we consider this decision as one Breuer undertook in light of his own welfare or well-being, disregarding the implications of his choice on others (as indeed they are disregarded in the novel). How would the ideal of welfare espoused by VE guide such a choice? We can begin with the guiding normative aim of Breuer’s life, the end of living well. This is his Ultimate End, the end for the sake of which he does all that he does, including pursuing other ends for their own sake. That is a normative or justificatory claim, not a claim about deliberation or motivation. This instance might, perhaps, be one in which it enters explicitly into Breuer’s deliberations, but it need not be. To the extent Breuer succeeds in his deliberations, however—to the extent he is wise— his Ultimate End of living well will do its work in constraining some considerations and giving greater weight to others whether he is conscious of these effects or not.8 There are two important features of this Ultimate End for our purposes. First, this end is indeterminate. Neither Breuer nor anyone else knows just what form it will take. It would be quite wrong to suppose that according to VE Breuer must have a Roycean “life plan”. A much better model is Tal Brewer’s notion of “dialectical activity”, which he describes thus: Some ( . . . ) activities have a self-unveiling character, in the sense that each successive engagement yields a further stretch of understanding of the

8 It may be that these effects must be “available” in the sense that in some justificatory contexts—where Breuer is pressed by others or by his own experience to defend his decisions— they become possible subjects of conscious thought.

Mark LeBar

70

goods internal to the activity, hence of what would count as a proper engagement in it.9

Part of what Breuer is doing, then, in choosing a response to this unwelcome development in his academic career is choosing a course of discovery about what life is best for him—embedded of course in a history and set of conditions that are themselves in part the products of his previous choices and constraints imposed upon him. Second, we should bear in mind the structure of ends Aristotle suggests. The fact that Breuer chooses to leave the academy and pursue private practice for the sake of living well, does not entail that the ends he seeks in doing so are merely instrumental—that he cannot pursue them for their own sake. The natural way to understand his interest in the issues in medical science that most engage him, for example, is as final ends. In fact, in Breuer’s clinical work as a forerunner for modern psychiatry, we can see even more of final ends at work. In the case of his treatment for the “hysteria” of Bertha Pappenheim (“Anna O.”), a charitable construal of his aims would be that he was treating her both for her own sake and for the sake of a better understanding of the therapeutic methods appropriate to such cases. Such multiplicities of final ends are commonplaces, not rarities. We ought not to be alarmed by the suggestion that he is engaging in private practice for its own sake as well as for the sake of living well. Moreover, as the treatment case suggests, the end of private practice will entrain new and further final ends. Formally, then, Breuer is developing a hierarchy of ends, along the lines suggested by Aristotle. By “hierarchy” here I mean to suggest (without making formal or precise) a collection of ends arranged in a complex structure, with a discernible order of significance in justification. That is, some ends matter more than others in justifying judgments and actions. Those higher in the hierarchy (such as the Ultimate End of living well) provide both justificatory support for and justificatory

9

Brewer continues:

If the activity's constitutive goods are complex and elusive enough, this dialectical process can be reiterated indefinitely, with each successive engagement yielding a clearer grasp of the activity’s proper form and preparing the way for a still more adequate and hence more revealing engagement in it. (The Retrieval of Ethics (Oxford, 2009), p. 37.)

My Welfare and Yours

71

constraint on the profusion of ends lower in the hierarchy. But this is a network, not merely a rank ordering, because our ends inform each other and our pursuit of them in various ways that make modeling them along a single dimension inappropriate. How do we get an arborescent framework of this sort? The engine postulated by VE is one of fittingness: our indeterminate Ultimate End takes shape and form by our making judgments as to what sorts of projects, commitments, ambitions—in short, what sorts of other final ends—fit into a good life. This is a form of Brewer’s “dialectical activity”, because in it we both discover and create what that life consists in. In making this suggestion I have in mind something similar in many respects to the “matches” Richard Kraut posits as at the center of his account of what is good for living creatures. Kraut says, “When we assess whether G is good for S, we must assess both G and S to see whether they are properly matched to each other ( . . . ) the ‘for’ in ‘G is good for S’ is best taken to indicate that G has a certain kind of suitability to S: their properties are so matched to each other that G serves S well.”10 VE takes the kind of “suitability” or “matching” Kraut is describing—what I am calling “fittingness”—to be a normative primitive. That is, apart from the idea that there is a relation of fittingness between our choices, plans, and so on, and our aim of living well by taking up those choices, plans, and so on, there is no further analysis available of the normative mechanism at work. That much I take to be in the spirit, if not the letter, of Kraut’s project. Other elements of VE’s story differ from Kraut’s. One important one, for example, is that on VE the correctness of these judgments, when we get them right, is not a matter of our apprehending and responding correctly to normative facts that are somehow “out there” awaiting our discovery, as Kraut suggests.11 Instead, the standards for correctness in these judgments (as all others) apply to their actual contribution to lives we recognize to be good. Of course, that recognition is itself a function of judgment, so obviously there is a kind of circularity to the process so described, but it need not be vicious.12 In fact, an inescapable upshot of

10

Kraut, What is Good and Why (Harvard, 2007), §21. Kraut, p. 8. 12 The resulting view is a form of constructivism about the standards for success in judgment and action. See LeBar 2013, chs. 5 and 6. On the circularity that is not (as I claim here) vicious, see ch. 11. 11

72

Mark LeBar

the thought that finding a way of living that can (to use Hume’s phrase) “bear its own survey” may be the combination of burden and promise of our “second natures” as creatures who negotiate their ways through the world using reason. But the reason we have is structured: it is public in its nature and the canons of success in judgment reflect a need for rational consistency, both with our own judgments and with the judgments of others.13 All that is a characterization of what Breuer ought to do, in choosing to live his life in the characteristic human way. But so far it is an isolated life. What does VE say about what happens when Breuer forms a close attachment with someone such as his wife, Mathilde? IV

As a general rule, humans do not live well alone. We live in company with others, and it is good for us that we do so. This is not just because of the advantages that accrue to social animals (e.g. the benefits of hunting in a pack). It is also because our sociality shapes, informs, and is shaped by our second natures—our capacities to set and pursue ends. This feature of us is reflected in VE in four ways, and we can consider Breuer for examples of these. First, others can be final ends for us (that is, we can seek others for their sake), just as most any object (concrete or abstract) may become a final end for us. Breuer may in this sense do any number of things for the sake of other people. One source of reasons for what he does in his practice as a physician, as we have seen, is the sake of his patients. Similarly, he may make arrangements for those in his employ in part for their sakes. He may do things for Mathilde or his children for their sake. All of these are perfectly ordinary manifestations of beneficence. VE understands these as reflections of ways in which Breuer promotes the final ends which others have become for him. But there are further ways in which our social natures are important in our hierarchies of ends. We get a clear statement of two of these from Kant. First, and most famously, Kant uses the language of ends to express perhaps the most intuitively forceful formulation of his

13

On the publicity of the standards for success in judgment, see LeBar 2013, ch. 8.

My Welfare and Yours

73

Categorical Imperative. “Rational nature,” he tells us, exists as an “end in itself ”; hence we are to “Act in such a way that you treat humanity, whether in your own person or in the person of another, always at the same time as an end and never simply as a means”.14 Though just what this means—or why it means that—may be somewhat obscure on Kant’s view, the idea has a natural home in VE’s framework. If we treat the “humanity” in ourselves and others as referring, loosely, to what we have here called “second nature”, we can take Kant to be marking an ordinary feature of final ends, namely, that they constrain, in a way that applies to our second natures. Having an end means (among other things) limiting our pursuit of other, subordinate ends. If we treat the idea of an “end in itself ” as a final end (one we seek for its own sake), we approach the idea that others by their nature impose constraints on the ways we may pursue our other ends. One way Breuer, for example, reflects this kind of constraint is in his respect for his patients’ privacy; even though it would advance his interests as a physician and researcher to treat their histories and treatment records as public, he does not do so, and goes to some lengths to distance the scholarly treatment of disorders from the individuals to whom those disorders belong. We need a bit more here, because we don’t quite yet have the idea that the ends others constitute for us are high enough in the hierarchy of ends such that our other ends are subordinate to them, rather than vice versa. Kant gets there, as I understand him, by insisting on the “objectivity” of the “endiness” of others, as opposed to other ends that arise from our inclinations and hence are merely “subjective”.15 This is not VE’s approach. Instead, the priority of these constraints on our pursuit of other ends is established by the same pressures that organize our ends into hierarchies in the first place: roughly, some ways of doing things make our lives go better than others. Here I will not take time to make that argument; suffice it to say that our seeing each other in ways that reflect this kind of priority is key to most of the ways our relationships 14 Groundwork of the Metaphysics of Morals, Ak. p. 429, Ellington translation (Indianapolis: Hackett, 1981). Kant’s language obscures the issue of whether it is the “rational nature” in others that is the final end, or some property (perhaps the welfare) of others that is that end, or whether it is the person himself or herself. On my view it is the latter, not any features of them or states they may be in, though there is not space here to make that argument. I thank an anonymous referee for pressing that question. 15 Groundwork, Ak. p. 428.

74

Mark LeBar

with others make our lives recognizably human, let alone good human lives.16 So a second way in which our social natures influence our hierarchies of ends is that others as final ends for us constrain our pursuit of other ends. But a third way also is found in Kant. Kant says that mere constraint “would harmonize only negatively and not positively with humanity as an end in itself ”. The “positive” part of seeing others as final ends involves something more: “the ends of any subject who is an end in himself must as far as possible be my ends also”.17 The idea that we can take on others’ ends as our own fits perfectly within VE’s framework.18 Aristotle treats “end-sharing” as an especially important element in the best kinds of relationships of philia—friendship or love, or more generally (we might say) of commitment.19 This is an important feature of Breuer’s relationship with his wife, and his love for her: if his love for her is sound, he has to a considerable degree made her ends his own. Consider an even more fictionalized example than Yalom’s. Let us suppose that Mathilde feels stifled in her life, and determines as best she can that she needs a creative outlet. She has always loved writing, has seen some of the success of emerging women novelists, and decides she wants to put serious energy into writing short stories and novels for publication. Now, Josef, insofar as he not merely loves her, but loves her as her husband and life-partner, takes Mathilde’s end as his own. He adopts the end of advancing Mathilde’s writing career, and he does so for her sake. The fact that he is helping her with her writing for her sake has implications for how he understands the end he is advancing. The “for her sake” flag here indicates the larger framework of ends within which this end is situated, and that has implications for how Breuer understands the end he seeks as her husband.

16 I indicate how I think this argument goes in “Virtue Ethics and Deontic Constraints”, Ethics 119 (2009): 642–71, and LeBar 2013, ch. 12. 17 Groundwork, Ak. p. 430. 18 I take this point to be distinct from the idea that others are final ends for us in that here it is their particular ends—what they care about—that we take to be reason-giving for us, rather than more narrowly their good or benefit. 19 I attribute this view to Aristotle in view of his ideas that we see our capacities for practical reasoning (phronesis—what we are here treating as “second nature”) as indeed ourselves, and friends we see as “second ourselves”, with whom we share lives in “discussion and thought” (NE IX.4, IX.9).

My Welfare and Yours

75

Unproblematically, for example, he might relieve Mathilde of some of the burden of household management to free up some of her time. But what about contributing to her end more directly? Suppose he thinks, fine as her writing is, that she simply has no ear for titling. He thinks her titles can be improved in such a way as to make her stories both more marketable and more coherent as an artistic whole. One way he might do this, since he is the one who is conveying the actual manuscripts to and from prospective publishers, is simply to make the emendations himself. If he had the aim of supporting her writing for the sake of some different end (say, simply to enhance family income), this might be defensible. But that he is doing it for her sake entails that it is not at all appropriate. It is her writing, an expression of her creative insight and imagination. That feature situates the end where it belongs, as part of Mathilde’s good, as part of a life that is good for Mathilde as a rational social animal, and as the particular rational social animal she is. So if he wants to help in this, he must find another way to do so. It is easy enough to see where things go wrong if this feature of Breuer’s end goes missing: Mathilde may well end up being treated as a “mere means” in the causal nexus that realizes Breuer’s conception of what would be good writing for Mathilde to undertake. That is obviously quite a different aim than the one that Breuer will have if he loves Mathilde and if her aim takes its proper place in his hierarchy of ends, for her sake. The point is that Mathilde’s end has been situated within Breuer’s hierarchy of ends because of the place it has in her hierarchy of ends. That the writing plays the role it does (or can) in Mathilde’s life is what gives it the determinate content it has for him. But if VE is right in its eudaimonism, even as Breuer pursues the end of advancing Mathilde’s writing for her sake, he also does so for his sake. That is, he pursues that end for her sake for his sake. That is not a conjunction (“for her sake and for his sake”); it is an embedding,20 and it reflects something important about our sociality and our rational endseeking natures.21 Breuer is so wired that, like most of us, it is good for 20 For an excellent discussion of this compound formulation, see Toni Rnnow-Rasmussen, Personal Value (Oxford University Press, 2011), chs. 5 and 6. 21 Might it represent a conjunction? In an imperfectly-formed love, yes. In fact, we might think that the process of acquiring virtue as Aristotle thinks of it (doing what virtue requires as the virtuous person requires it—NE II.4) involves a transition in the nature of the reasons we have for doing as we do. Only as we get better in our love, or virtue, do we fully appreciate and find reason-giving the sake of the other. I thank Mike Bishop for discussion of this point.

76

Mark LeBar

him to pursue some ends for the sake of others. It is good for him to have others in his life for whom he cares in this way, and who provide ends for him of this nature. Both of the “for the sake of ” elements here matter for the purpose of understanding how his love for Mathilde contributes to the ends he seeks as a practically rational social animal. The constraining nature of ends—in particular his Ultimate End of living well—is once again at work in the element of this story that reflects Breuer’s seeking this end for his own sake. It is because it is true of Breuer that entering into and honoring relationships that involve mutual end-advancement of the sort we are discussing are good for him, that he has reason to do so. This is easiest to see if we look at cases in which this constraining nature of one’s Ultimate End is not operative. Consider cases of obsession or servility, in which one person denigrates his or her own value for the sake of another.22 Something is wrong in such cases, and a good diagnosis of what that something might be is precisely that the end of advancing the interests or meeting the needs of others has been wrenched out of its natural home in which one sees one’s own good as having final value, indeed (as on the eudaimonist picture) ultimate value, in just the sense Aristotle says: it is that for the sake of which we do all we do. One feature of this constraining work may strike us as odd. It seems to make the possibility of Josef sacrificing for Mathilde impossible (at least a rational impossibility). Suppose Josef sees that if he gives up part of his research program, he can more effectively release Mathilde to do her writing, and he does so. Ordinarily, we’d like to say he sacrificed part of his research for her writing career. But on the view on offer, this seems inapt. If he had reason to do so, it was because the end of advancing Mathilde’s writing career figured more highly or significantly in his hierarchy of ends than did the marginal element of research that he has now given up. So it is no sacrifice. In effect, VE assimilates such cases of sacrifice to opportunity cost. But we think some opportunity costs are tantamount to sacrifice, even in the straight-forwardly intrapersonal case. I know, for example, that in choosing (with my wife) to have a family, I gave up time and energy that otherwise might have gone into my work as a philosopher. Given the 22 Jean Hampton worries very usefully about such cases in “Selflessness and Loss of Self”, Social Philosophy and Policy 10 (1993), 135–65.

My Welfare and Yours

77

significance to me of my work as a philosopher, that was a not an inconsiderable sacrifice. But there is nothing mysterious about it for that reason. The opportunity cost of that extra work was, I judged, higher than the opportunity cost I would have to accept to have a family. But it seems perfectly acceptable nonetheless to think of it as a sacrifice. Much the same is true here. To a great extent, our notion of sacrifice may be an artifact of unduly non-social and non-moralized conceptions of our interests. That makes it easy to see forgoing elements in the set of such interests as a “sacrifice”, when in fact it is very often a matter of tradeoffs for things we value more, and for which (importantly) we see ourselves as better off just for that valuing.23 While the language of sacrifice may yet be appropriate, it need not entail that one has really chosen to forego what is good for one to realize. I said there were four ways our social natures influence the lives of practical agency we live, as understood by VE. The fourth we can approach by example. There is a distinctive way in which Mathilde shapes the hierarchy of ends through which Josef pursues his agency. She can alter the relations of fit Josef sees in various of the opportunities available to him to pursue his ends. Suppose their son Robert is approaching the end of his home life. Josef and Mathilde (and Robert) agree that the next step is university. Josef thinks, in light of Robert’s aptitudes, that Robert would do best pursuing his studies in Salzburg. Left to his own devices, Josef would advance his ends as a father by setting Robert up for admission to Salzburg. This he does for his son’s sake but also for his own, since being a good father is a final end that has obvious salience for living well. Part of his task in ascertaining what would best suit Robert is to find a fit of conditions to Robert’s interests (in both senses), aptitudes, and so on, and in doing so Josef is also fitting his responsiveness to the needs of his son into his own life in a way that contributes to his own living well. This is part of the dialectical activity which his living well consists in. Now it is open to Mathilde to intervene, and by doing so to alter the relations of fit that Josef sees as giving him reasons for acting in various ways. Perhaps she has reason to believe that Robert would, for various reasons, do better in Geneva than in Salzburg, and makes that case to Josef. Now the place Mathilde has in Josef ’s life and his hierarchy of ends 23 This last point forecloses the argument that these cases are a matter of tradeoffs of welfare for other non-welfare values.

78

Mark LeBar

means that her input here has force of a sort that the input of others might not. Others might advise, and if Josef is wise, he will attend to good advice and adjust his thinking accordingly. But Mathilde’s input is not merely advice: it is to some degree authoritative over Josef ’s ends, and certainly to elements of their shared life, such as their parenting of Robert.24 How authoritative? It is difficult to say much about that question in the abstract, because we are confronted with a conflict of judgments. Part of practical wisdom in such relationships is getting a sense for when to defer and when to stand one’s ground. On the other hand, part of the commitment Josef and Mathilda enter into in forging their marriage is to grant authority where otherwise only advice can be given. To some extent, that is just what the commitment is: an agreement to, on some (indeterminate) occasions and to some (indeterminate) degree, subordinate one’s 24 Much more needs to be said about our relationships with our children; here I can only gesture at lines of thought that might be worth further investigation. Children begin in a condition in which they entirely lack the capacity to set and adjust ends, and end in adulthood with that capacity in full (or as full as we humans get, anyway). One way to think about our responsibilities to our children is to consider ourselves as something like trustees for the persons they will become. In effect that means that we are responsible as parents for forging their Ultimate End in the early stages of their lives, while shepherding their development and coaching them in the work of developing and adjusting their own economies of ends. (This is the crucial work of moral education, emphasized in Plato and Aristotle.) As time goes on, our children acquire their own voices and capacities for development, and the structure of ends they develop becomes increasing independent of those of their parents, in a way that is not the case for ongoing committed relationships such as marriage. But the story is complicated by the fact that, unlike in relationships with equals, with children parents are also in the modeling business. An important way in which they teach their children about adulthood and agency is through modeling adult decision-making, and this induces complications. Josef, for example, must realize that his children learn from the way that he chooses to live his life. He knows both that his children need him in their lives, and that he is experiencing the enormous centrifugal force of the oppression of his married and family life. So he is subject to a colossal collision of reasons. Staying would keep him present in his children’s lives but also (at least potentially) model a kind of abandonment of the authority of his Ultimate End of living well, in the extreme case subordinating his interests to theirs in a way that is itself quite unhealthy, not only for him but (especially given this modeling dimension) for them as well. Yet a further complication is the interaction of these elements with love. Part of what practical love consists in involves reciprocal or mutual authority and adjustment of ends. We want and need to love and be loved, but it is also crucial that our love be acknowledged. So an important part of what we want to do with children is acknowledge the force of their love for us. This may occur through registering the changes in the hierarchy of ends that they, through their love for us, impose upon us. It is partly by giving them that authority over us, and having them learn gradually what it is to be in relationships of reciprocal love and authority, that both they and we grow into relationships that contribute to living well.

My Welfare and Yours

79

own judgment to the judgment of one’s partner.25 We return to this question below. But this feature of the interaction between lives and agency—between the welfares of distinct but mutually committed individuals—is the locus of the fireworks when things begin to go bad, as they do in Breuer’s case; thus we get perhaps the most distinctive illumination of our interacting welfare at that point. V

It is the possibility that Josef ’s and Mathilde’s goods might pull apart that Yalom has Josef confront in his novel. Josef is in despair: Though I care for my wife and my children, I don’t love them! In fact, I resent being imprisoned by them. I lack courage: the courage either to change my life or to continue living it. I have lost sight of why I live—the point of it all. (p. 138)

Breuer dreams of the ground liquifying beneath his feet and his falling to land on a slab 40 feet underground. In a radical move, he engages Nietzsche in a trade, originally conceived as therapy for Nietzsche, but over time gaining urgency in the opposite direction: Nietzsche will treat Breuer’s despair, while Breuer treats Nietzsche’s despair and physical symptoms (e.g. migraines). Nietzsche’s prescription rests on the idea that “This moment exists forever, and you, alone, are your only audience” (p. 251). What this

25 Locke makes a related point about the social contract: if one retained one’s own natural authority even while under the contract—not deferring to the decision-making mechanisms of the commonwealth—the contract would “signify nothing” (Second Treatise of Government, §§96–7). Of course, in the political case, we have a relatively clear and unambiguous but asymmetrically authoritative way of resolving conflicts in judgment: the individual must defer to the majority. It is just the symmetry and equality of authority in committed relationships like marriage that make it difficult to specify how such conflicts can be resolved. I thank an anonymous referee for pressing this point. On the other hand, Locke also affords us a nice example of how not to understand this equality of authority. Locke maintains both that husbands have no more right over their wives than their wives have over them (§82), and that the mother “share power” with the father over their children, and also that when they have “different wills” over their “common concern” with the raising of their offspring, it is the “man’s share” to rule (§82). That is plainly a non-starter as an understanding of equal and reciprocal authority.

80

Mark LeBar

entails is that Breuer must seize his life in a way he has been reluctant to do: “I hate it,” Breuer almost shouted. “To live forever with the sense that I have not lived, have not tasted freedom—the idea fills me with horror.” “Then,” Nietzsche exhorted, “live in such a way that you love the idea!”

What stops Breuer, what constrains him, is precisely his duty to his wife and family. This, Nietzsche insists, must go: “Can duty take precedence over your love for yourself and for your own quest for unconditional freedom? If you have not attained yourself, then ‘duty’ is merely a euphemism for using others for your own enlargement.” Breuer summoned the energy for one further rebuttal. “There is such a thing as a duty to others, and I have been faithful to that duty. There, at least, I have the courage of my convictions.” “Better, Josef, far better, to have the courage to change your convictions. Duty and faithfulness are shams, curtains to hide behind. Self-liberation means a sacred no, even to duty.” ( . . . ) “But I cannot be free,” Breuer implored. “I have made sacred vows. I have a duty to my children, my students, my patients.” (p. 252)

A significant benefit of virtue eudaimonism is how much more insight we may gain into both happiness and our relations to others by considering them from the perspective of ends. Clearly something is wrong with Breuer’s notion of “duty” to others. It is something he feels he can fulfill by “caring for them” while at the same time “not loving” them. It is as though what matters is the fact that he has made vows, rather than what that fact entails for others for whom he cares. And (fictional) Nietzsche’s remonstrance that this is a matter of “changing convictions” rings hollow as well. It is not (merely) Breuer’s “convictions” that must change, but his life. He must understand differently what living his life means for him. What he needs is a completely different way of both thinking and feeling about what he is doing in living his life. That is natural terrain for the notion of ends in the framework of virtue that VE advocates. Though the emphasis of VE as an account of practical rationality is on the intellectual virtue of wisdom, the full picture of virtue includes the virtues of character which help to shape how we undertake advancing our ends and certainly our affective and conative

My Welfare and Yours

81

lives as we do so. The way in which our ends are related hierarchically is a matter not only of judgment but of appreciation for how they fit together—their significance in a single life which requires tradeoffs and satisficing among them. That is the work of virtue entire.26 There is a disconnect between the things that really matter for Breuer—what has meaning for him at a level sufficient to affect his dreams—and the framework of “duties” he has taken on, and feels bound to observe. A plausible account of virtue, situated within a eudaimonist framework, will ascribe, if not folly and vice, then lack of wisdom and virtue, to the choices Breuer has made about what to make of himself and how to live his life. Rectifying those shortcomings includes inter alia rethinking the ends Breuer sees himself as living for, and reshaping his character accordingly. However, there is a further problem which this lack of wisdom and virtue can obscure, and that is the potential conflict between a fuller love for Mathilde and his love for himself. So suppose we ascribe to Josef a fuller love for Mathilde—a genuine case of his seeing her as a final end for him of great importance, situated very highly in his hierarchy of ends, with all that we have seen that entails for his thought and action. If nevertheless he feels imprisoned by his life, including his life with her, how is he to think about what to do? In the book, Yalom has Breuer simulate what it would be like to leave Mathilde, with Freud’s help in a hypnotic state, a luxury of counterfactual experience most of us do not enjoy. Breuer decides after the hypnosis that he has good reason to reject the apparent attractions of freedom. In so doing he chooses his life in a way he has not previously done, and in turn is free in the life he has chosen in a way he has not previously been. If we take Freud’s hypnotic help out of the picture, what can we say about the practical reasons that bear on Josef ’s decision when he is confronted with such a choice? Mathilde is a final end for Josef, in virtue of his love for her and commitment to her. But the significance of Josef ’s Ultimate End of living well is that it regulates the pursuit of this final end, as others. If his subordinate end is genuinely in conflict with that Ultimate End, he must abandon it. It is possible, that is, that Nietzsche is right and Josef 26 Cf. Aristotle: “choice cannot exist either without thought and intellect or without a moral state; for good action and its opposite cannot exist without a combination of intellect and character” (NE VI.2: 1139a33–5).

82

Mark LeBar

must leave Mathilde and his life and practice in Vienna. Certainly we can think of cases in which remaining in relationships with others, however deeply loved, will be life-destroying, and if VE is right, we have reason to leave such relationships if we cannot eliminate or contain their toxicity. We do not have reason to see that sort of toxicity in the Breuer’s case. Still, in this respect Mathilde-as-Josef ’s-final-end is just like any other final end he has. That end exists within his hierarchical network of final ends, the competition between which for time, attention, resources— more generally priority in what he does—must be adjudicated by what fits best with Josef ’s Ultimate End of living well. But, as we have seen, Mathilde-as-Josef ’s-final-end is quite different from any other final end that is not a person, and different again from his other (personal) final ends with whom he does not stand in the intimate and committed relationship that he does with Mathilde. The particular kind of final end that Mathilde is, is one to whom Josef is accountable.27 One important element of VE’s understanding of accountability is that those to whom we are accountable partially shape (as we have seen) our hierarchy of ends: they participate in the part of our rational agency that consists in ordering, determining, and pursuing our various final ends. We might say: they are with us participants in the dialectical unfolding of those ends. This is part of what we can agree to do through contract with others,28 and of course Mathilde and Josef are bound by a marriage contract. What kind of effect does Mathilde have on Josef ’s hierarchy of ends, and what sort of weight does her judgment bear? The operating assumption of VE is that the Ultimate End is not subject to having its authority being overruled by the authority of another end, as we have seen; that is just what is meant by saying that it is Ultimate. So that presents a “top end” to Mathilde’s authority: she may not require of him that he do something that is inimical to his living well. But neither can we construe her input as minimal, if we are to understand their relationship as having the kind of character- and life-forming nature such relationships often For discussion of how VE fits accountability relations into a eudaimonist framework, see LeBar 2013, chapter 12. 28 In this way VE offers us a distinctive way of thinking about the authority we have and convey through contract and promise. 27

My Welfare and Yours

83

have. Where between these extremes does her authority fall? This question is important enough that it bears careful scrutiny. We have already seen that abstract and specific responses are unlikely to be in the offing here. We will have to settle for suggestions as to an outline for that scrutiny. First, the authority she has will be partly determined by the kind of understanding Josef and Mathilde had about the relationship they have entered into in marrying, and the dialectical unfolding of that understanding as they have lived together through the years. This understanding consists in part (but only in part) of their own attitudes towards their commitment. Like other practices, there may be facts about how relationships such as marriage can contribute to the good lives of those who marry, even if those facts are not recognized by the participants. Such facts may have some culture- and even individual-relativity, but they are not limited to the attitudes of the participants. Second, it seems plausible to suppose that in healthy relationships (those actually contributing to the Ultimate End of living well for both participants) their mutual authority is constrained by reciprocity. That is, neither has a degree of authority over the ends of the other that is not reciprocated.29 On any given occasion, of course, unless there is concurrence only one judgment can win out, so the reciprocity of authority constitutive of such a commitment must inhere in a relationship over time. Successful committed relationships consist in part of an evolving give and take in the assessment of shared ends and the means to them. However, this leaves open a further range of individual relativity. There may well be some relationships in which the mutual and reciprocal authority is quite limited—people committed to each other yet highly independent in their own autonomy—and others in which the setting and adjusting of ends is a far more collaborative undertaking, with much more mutual authority.

29 Harry Frankfurt writes, “Love does not necessarily include a desire for union of any other kind. It does not entail any interest in reciprocity or symmetry in the relationship between lover and beloved. Moreover, because the beloved may be entirely unaware of the love, and may be entirely unaffected by it, loving entails no special obligation to the beloved” (Taking Ourselves Seriously & Getting it Right (Stanford Univ. Press, 2006), p. 41). This may be true of the attitude or emotion of love. But it is not true of love in the kind of loving and committed relationship of the sort that is under consideration here.

84

Mark LeBar

This reciprocity may be realized in another way. David Owens distinguishes between the effects of demand and command in personal relationships.30 The difference is in the locus of authority to reorder the ends of the other. Owens understands a demand as a request that is situated within, and draws on the normative structure of, the relationship itself. Thus, a request for help in moving works somewhat differently, normatively, coming from a good friend, than coming from a stranger or acquaintance. The friend adverts (implicitly) to the reciprocity inherent in friendship in making what would otherwise be a mere request into a demand. A command, on the other hand, “imposes an obligation ( . . . ) by communicating the intention of hereby imposing that obligation.”31 It is by a unilateral act of will of the commanding agent that authority is supposed to be exercised, and it is hard to see how such authority can possibly be reciprocal. Thus the reciprocity involved in healthy relationships allows for demands but not command. Third, we may need to understand authority in a particular sense to grasp its role here. It is natural to understand authority as a kind of status or standing. For example, ordinarily if Meg orders me to do pushups, that means nothing: it gives me no reason at all to do pushups. The matter is different, however, if Meg is a sergeant and I am a private in her platoon. Her order now is reason-giving, and the explanation for the difference lies in her authority to give orders to me in light of her status and mine in our respective roles.32 Part of the picture I have been developing is that people in intimate relations, such as the Breuers, have standing in virtue of that relationship to effect changes in each other’s hierarchy of ends. What we are after now is a grasp of what that status comes to, and another appeal to standing won’t do the trick. Instead, something like a practical or actual understanding of authority is needed here. In this sense, Mathilde has the relevant sort of authority just in case her input into Josef ’s deliberations actually has the desired effect. This is not quite a causal point, but it is like a causal point. It is not a causal point because the relevant question is her capacity 30 David Owens, Shaping the Normative Landscape (Oxford Univ. Press, 2013), p. 100. Owens’ example is in the context of friendship, but his point generalizes. However, the way I understand the distinction goes beyond anything Owens says. 31 Owens, p. 101. 32 Cf. Stephen Darwall, The Second Person Standpoint (Harvard University Press, 2006), p. 259.

My Welfare and Yours

85

to shift the norms governing Josef ’s decision-making; that is a rational or normative matter, not a causal one. But it is like a causal power in the sense that either she realizes the effects she aims for in Josef ’s hierarchy of ends, or she does not. There is no lingering dispositional effect to be accounted for here. The decision Josef faces, in these terms, is how much authority Mathilde will have, and that question, understood in what I am calling this practical sense, is determined by whether or not he actually accedes to her wishes. Finally, then, we can understand the intimate moral relationship they share as one in which there is a meeting of the minds on just what her status is, or should be. Because of the significance of this decision, the failure to have a meeting of the minds would really mark the dissolution of their marriage. A disagreement on a point this fundamental would render other points of agreement trivial. The marriage is literally made or broken on what Josef decides, given Mathilde’s aspirations for her place in their relationship. This is why disagreements are so revealing about relationships, and thinking about this disagreement in this light reveals what is at stake. The full story there is of course only partly philosophical and is substantially sociological and psychological. Where does all this leave us? VE certainly does not settle what Breuer should do. It is unlikely that any philosophical theory could do so, however, and it is probably unwise to hope that any could. At best what it provides is a kind of deliberative framework which, if VE is correct, underwrites (or should underwrite) not just Breuer’s decision-making in this case, but in all cases. On VE our welfare is substantially a matter of our choosing, and because we are essentially social beings, the welfare of others must affect and be affected by the choices of those with whom we are close. My welfare and yours may not readily be disentangled. Acknowledgments I thank Darryl Freeland, audiences at the departments of Philosophy at the University of Texas and Florida State University, at the George Mason University Dept. of Economics, and at the Workshop in Normative Ethics at the University of Arizona, and anonymous referees, for comments on earlier versions of this paper.

4 Moral Worth and Normative Ethics NOMY ARPALY

According to a number of theorists (Arpaly 2002, 2003; Markovits 2010; Arpaly and Schroeder 2014), a morally right action has moral worth if and only if it is performed for the right reasons, which are the reasons for which it is right, or the right-making features of the action. I have referred to morally worthy actions as “praiseworthy actions”, though, as we will see, perhaps “esteem-worthy actions” would be more precise, if one were to use Kantian terminology.1 The idea central to these theories, of acting for right reasons through acting on the right-making features of actions, deserves particular attention. My purpose in this paper will not be to defend the view of moral worth in question, which I have already done in a paper and two books, but to argue that it has some implications for normative ethics, especially that it provides support for a pluralistic, rather than a Kantian or utilitarian, view. However, it would be useful to provide a few reminders of the nature of the view before exploring its implications, and I shall do so briefly in this section. Right reasons for action do not need to be reasons that the agent believes to be the right reasons for action.2 Huckleberry Finn’s action of helping Jim has moral worth if his reasons for helping Jim are the rightmaking features of helping Jim (that it treats Jim with respect, perhaps), however wrong he takes his action to be. Likewise, a person who performs the right action because she thinks it is right might not be acting for the right reasons at all. A person might think, for example, that the moral thing to do is to promote the interests of Aryan people over

1

It should be added that what holds for right action also holds for supererogatory action, so really “morally desirable” might be more accurate than “right”. For ease of reference, though, I’ll talk about morally right actions. 2 Arpaly (2003) defends this thesis at length.

Moral Worth and Normative Ethics

87

people of other races. If this person hires the better job candidate, not because she is better, but because she is of the “right” race, she acts rightly and acts because she wants to do the right thing—but still for the wrong reasons. She performs the right action, but not for its rightmaking features, and thus her action has no moral worth. Another way to put this point, found in Michael Smith (1994), is to hold that acting out of a commitment to the right de dicto, as opposed to de re, is a kind of moral fetishism. On my view, being committed to the morally right whatever it is (when “maximize the beauty of ducks” is a possible candidate for the right moral theory in the agent’s mind) is not praiseworthy in the least. More precisely, a morally worthy action stems from a commitment to the right and the good correctly conceptualized. If utilitarianism has the right account of the features that make actions right then the agent performing a morally worthy action conceives of her action as maximizing utility, and is committed to maximizing utility so conceived; if Kantianism is correct she conceives of her action as respecting persons or as acting in a universalizable fashion, and so on. Whether she also conceives of her action as “the right action” is immaterial. Very few people act on exactly the right reasons, however. Consider a person who does the right thing from a commitment to respecting persons but does not respect women, homosexuals, or blacks. This agent is not worse than most people in history, including Kant himself. If respect for persons is what acting rightly is about, it would be very harsh to hold that on those occasions in which Immanuel Kant worked hard to keep his promises his promise-keeping had no moral worth at all, and neither did the right actions of any other 18th century European male. Such a person, who is not committed to the complete moral reasons, might be committed to at least some partial, pro tanto moral reasons, and act upon them while doing the right thing—and that, as far as moral worth goes, should count for something. Kant thought of moral worth as an all-or-nothing affair, but my view is open to the possibility of degrees of moral worth (here Markovits differs from me—she denies that there are such degrees).3

3 Following some of her arguments, I no longer accept my old (2003) view of degrees of moral worth. For the purpose of this paper I am not committed to what Timothy Schroeder and I say about degrees of moral worth, either (2014).

88

Nomy Arpaly

In Arpaly (2002 and 2003) and Arpaly and Schroeder (2014), I have also said that blame and vice involve a sort of “negative moral worth”—a failure to respond to the moral reasons one has for taking a certain course of action. Sometimes they also involve being motivated to perform an action by its “wrong-making features”. Julia Markovits and I hold that there is an important connection between moral worth and moral reasons. If one grants that the morally worthy action is done for its right-making features, then one can make inferences from claims about moral worth to claims about what the rightmaking features of actions are. From the latter claims one can then make inferences about which normative ethical theories might be the correct ones. Markovits (2010) has already applied the idea to utilitarianism in arguing that, if utilitarianism were correct then people who keep promises and are not motivated by considerations of utility would act with no moral worth. One way to put it is to say that she would act with no more moral worth than Kant’s prudent grocer. Kant’s prudent grocer, recall, acts honestly because he reasons that acting honestly will bring him money. He is a paradigm of a person whose right action has no moral worth as his motive is not based on the right-making features of his action. But despite what one would expect if utilitarianism were true, it is implausible to think that everyone who keeps promises for nonutilitarian reasons is equivalent to the grocer in the moral worth of his or her actions. Yet, if utilitarianism were correct, it would follow that these people, just like the grocer, act for morally irrelevant reasons and not for the right-making features of their actions, and so act without moral worth. This, Markovits argues, is a reason to reject utilitarianism. It is of course open to the utilitarian to reject the intuition that people act with moral worth for non-utilitarian reasons, but that would be a costly choice. So would the choice to hold that any motive, regardless of its connection to utility, grants moral worth if it increases utility (i.e. if an invisible hand were to cause greed to result in prosperity for all, actions motivated by pure greed could be morally worthy). I cannot explore all of Markovits’ replies to objections here, but her line of argument strikes me as plausible. I do not wish to argue that in every case in which a normative claim (such as utilitarianism) clashes with an intuition about moral worth we should follow the latter, but sometimes the price of conflicting with an intuition about moral worth can be too high for a normative theory (and here it is worth noting that Kant starts the main

Moral Worth and Normative Ethics

89

text of the Groundwork with intuitions about moral worth—the good will—before getting to normative ethics—the categorical imperative). In the next section I will discuss some general implications that facts about moral worth have on normative ethics if my view of moral worth is true. In the following section I will argue at some length for one such particular implication: That there is more than one kind of moral reason, or, in other words, more than one right-making feature for actions. It is possible for two right actions to have different right-making features. This claim, I will argue, follows from facts about the moral worth of altruistic action—a topic on which a well-known view, usually identified with Kantianism, is wrong.

GENERAL IMPLICATIONS

What are the right reasons for a moral action? When asking this question, I will assume as a starting point that there are actions that have at least some moral worth. Kant doubts that any morally worthy action had ever been performed, but I see no need to share his brand of pessimism. An important thing to remember about agents who have acted in a morally worthy manner is that most of them are not ethicists. I refuse to refer to these agents as ordinary people, as some of them are quite extraordinary (see, for example, people who saved Jewish people from the Nazis). I will call them “philosophically unsophisticated”. It is unlikely that when acting well they follow, with any precision, the categorical imperative or the principle of utility (or Peter Railton’s specific consequentialism,4 or . . . ). That alone is not enough to rule out either the categorical imperative or the principle of utility as moral truths. This is because it does not rule out the possibility that most praiseworthy moral actions are committed for pro tanto moral reasons rather than for the complete moral reasons, and the complete moral reasons are captured by the categorical imperative or the principle of utility. However, given my view of moral worth, the correct moral theory implies that the reasons for which the philosophically unsophisticated morally worthy agents act are either the complete moral reasons, or 4

See, e.g., Railton (1988).

90

Nomy Arpaly

pro tanto moral reasons, or at the very least be trivially derivable from such reasons. Commonsense morality is not necessarily the right moral theory, but some commonsense moral reasons—the ones acted upon by praiseworthy agents—are at least pro tanto moral reasons. What moral reasons do philosophically unsophisticated praiseworthy agents, whether ordinary agents or philosophically unsophisticated saints and heroes, act from? I will not try to provide an exhaustive list of reasons, but it is easy to think of some. “I promised ” is one such reason: Quite often people explain why they kept a promise simply by appeal to the fact that they have promised. The fact that someone is in need is cited just as often as a simple reason for acts of benevolence. In all likelihood these are reasons that genuinely motivate many praiseworthy acts of promise-keeping and of benevolence, respectively, and thus a true normative theory must account for their status either as complete moral reasons or pro tanto moral reasons for action, or as at least derivable instances of such reasons. The Arpaly-Markovits thesis, again in conjunction with the assumption that there exist morally worthy agents who are philosophically unsophisticated, has another implication for the search for the rightmaking features of actions. Acting on the reasons behind uncomplicated right actions should not require an amount of reflection few are capable of or cognitive abilities that only a few possess. When I say “uncomplicated” moral action I am allowing for the fact that some moral actions, such as voting for the right economic legislation, do require reflection and intelligence due to their very nature. Some, however, do not. For lack of better words, it is an advantage for one’s theory of the right not to be too lofty. One has to be careful not to underestimate the amount of complexity of which people are capable while at the same time being unable to articulate this complexity. People are capable of learning complicated dances without being able to provide purely verbal instructions as to how to perform them, and they are capable of learning complicated languages without being able to articulate their grammars (but this special ability to act for grammatical reasons, an ability learned in such a special way, is probably not representative of our abilities to act for complex reasons in general). However, there is something suspicious about a moral theory that justifies uncomplicated right action in a manner that is foreign and hard

Moral Worth and Normative Ethics

91

to explain to anyone but the brightest university students. For example, explaining to all but these students what “contradiction in conception” means is remarkably hard, and so I hold in suspicion the view that people who are not the brightest students routinely act to avoid such a contradiction (pending a report from cognitive science that we have a special contradiction-in-conception detection ability—we who contradict ourselves so often!). One must note quickly that this disadvantage of the “contradiction in conception” view is far from a big problem for Kantianism in general. After all, even children ask “what would happen if everyone did that?” and “what would you say if someone did it to you?”, and these considerations are at least Kantian in spirit. I do not wish to argue that one should expect all unsophisticated agents to fully understand one’s moral theory. For one, there is more to a moral theory than specifying the right-making features of actions. There is, for example, explicating conceptual connections between different features (e.g. a Kantian might ask what the relationship is between universalizability and respectfulness and a virtue ethicist might ask whether there is, in fact, a unity of the virtues). There is also, of course, the task of figuring out the implications of basic normative truths for the right thing to do in deeply controversial or complicated situations. However, it speaks well for one’s theory if a university-level course is not required to understand one’s description of the reasons for which moral agents act. ALTRUISM

In this section I would like to defend the view that some actions have moral worth when they are motivated by non-instrumental concern for the wellbeing of others. I will also argue that, if we assume my view of moral worth, it follows that there are at least two kinds of moral reasons, altruistic reasons being one such kind. In other words, there are at least two different right-making features that different actions can have, one of which is the fact that an action protects wellbeing (or, if you wish, prevents or alleviates ill-being). The distinction between alleviation of illbeing and increasing wellbeing is blurry, but I will talk mostly about protecting wellbeing and alleviating or preventing ill-being since it is plausible that it is sometimes morally required to protect wellbeing, or prevent or alleviate ill-being (and, more controversially, that it is ever

92

Nomy Arpaly

morally required to increase the wellbeing of people who are already doing well). There is no precise word in English for concern to protect wellbeing, though it is related to compassion, kindness and sympathy. As I don’t think it is precisely identical to any of these, I will refer to it, following Lawrence Blum (2009) as “altruism”. For the purpose of this paper, when I am talking about altruism, I am talking about genuine non-instrumental concern for the wellbeing of one’s fellow humans, and also concern for a specific person that is a “realizer” of such a non-instrumental concern. Let me explain what I am not talking about, which will be equally important for what follows. I am not talking about the emotion that Kant talks about when he says that “We love everything over which we have a decisive superiority, so we can toy with it, while it has a pleasant cheerfulness about it: little dogs, birds, grandchildren”.5 This statement, offensive to grandparents everywhere, is not about non-instrumental concern for another’s wellbeing, which has nothing to do with the joy of toying with the weak. I am also not talking about what Bennett (1974) talks about when he talks of “sympathy” as something that would motivate one to avoid taking a child to the doctor because one’s heart is melted by the child’s crying. This is not concern for the child’s wellbeing, unless, to invoke Mill for a moment, one assumes idiocy along with it. True altruism is also distinguished from what psychologists currently call “agreeableness”, which is sometimes all we refer to when we say casually that a person is “nice”. Agreeable people “get along” with people around them, but absence of aggression and aversion to causing conflict does not amount to caring non-instrumentally about anyone’s wellbeing. As I am using altruism as a term of art, I would like to stipulate the following: when a person is motivated by altruism, her reason for action is “this person needs help”. Her motivation does not include the extra consideration sometimes known as the avoidance of a contradiction in one’s will. Christine Korsgaard (1996a, 60) rephrases that consideration when she says that the morally worthy helper, according to Kant, has a motive to help that is deeper than that of the sympathetic helper because he has a “further stretch of motivating thought.” That further stretch “concerns the sort of world that this would be if no one helped, or better

5

Reflections on Anthropology, 15:490. Cited in Wood (1999, 272).

Moral Worth and Normative Ethics

93

still if no one perceived the need for help as a reason to help ( . . . ) Such a world would be unacceptable because we regard our own needs as reasons why we should be helped.” I would like to argue that the view summed up by Korsgaard (whether or not it is exactly Kant’s view) is false. One can be a morally worthy helper without having a “further stretch of motivating thought” of this sort. First, on pain of loftiness, it is fairly safe to say that one can be a morally worthy helper without having exactly the thought Korsgaard describes. I doubt that many philosophically unsophisticated people wonder what the world would be like if no one “perceived the need for help as a reason for help”! However, there is nothing unusual about wondering what things would be like if everyone acted (or failed to act) in a certain way, and more specifically, there is nothing unusual about a person wondering how everyone acting (or failing to act) a certain way would affect her. I am adding this specification because it is arguably possible to wonder what would happen if “everyone did that” or if “no one did that” in a way that is more consequentialist or ruleconsequentialist than Kantian, which happens when you simply consider the effects of hypothetical universalized action on the world. However, it is also common to wonder how some universalized course of action would affect you. A simple example would be the statement “I must pay my taxes, because if no one paid taxes there would be no state, and without a state, where would I be? It’s only fair”. This is a Kantian “stretch of thought”. It is extra-Kantian if the implicit answer to the rhetorical question is something like “I would not be able to have an income” or “I would not have the comfortable life I want” (the latter desire being the basis of your wish to keep your money). But consider a person—let’s call her Jennifer—who helps another individual—let’s call him John—who has serious difficulties writing his dissertation. John is not Jennifer’s student, nor is he a close friend, nor has he done her a favor in the past, nor is he likely to do her a favor in the future—there is no obvious issue of fairness in the background. It is simply that Jennifer notices John’s ill-being and is in a good position to help him. She helps him, perhaps at some cost to herself. Her reason? A human being needs help. It never occurs to her to ask herself how she would feel in a world in which nobody helped. No thought of a hypothetical need for help, expecting to be helped, doing her fair share etc. plays a role in her motive as she helps John.

Nomy Arpaly

94

We might even imagine that she does not take her needs as reasons to help her. Perhaps Jennifer is really tough and values independence so deeply that she does not want to be helped—not in general and certainly not for free—or think she should be helped. We can also imagine Jennifer as a super-hero who is rationally confident that no difficulties she might encounter are such that she cannot surmount them without help. But the Jennifer I have in mind is a real human who simply is not motivated by a belief about her hypothetical needs as she helps John with his. Jennifer’s motivation, it is worth remembering, is not hedonistic— she cares about other people’s wellbeing non-instrumentally rather than for the sake of some pleasure it might bring her. Neither is her motivation narcissistic. It is not the case that she helps John because of an inference along the lines of “do whatever Jennifer wants, Jennifer wants to help John, therefore I’ll help John”, but rather she helps John because John needs help. Some Kantians I have talked to seem to assume that a motivation other than the categorical imperative is always narcissistic, but people who act purely out of love for their children are surely not always motivated by the categorical imperative when they do, and are often far from narcissistic. If Robert cares about his daughter’s wellbeing and helps her as a result, it need not be the case that Robert is motivated, narcissistically, by the thought “Robert cares about Martha’s wellbeing”. Robert might be a self-hater who does not care very much about himself or the fact that he cares about something. Luckily, selfhelp books are wrong when they tell you that you first have to love yourself in order to be able to love another. Robert the self-hater can be motivated by the thought “Martha needs help”—his not caring about himself and his own cares does not mean that he does not care about Martha.6 Jennifer cares about John’s wellbeing in a different way—she cares about it as an instance of human wellbeing—but the same reasoning applies to her. To recapitulate: Jennifer helps John because, as stereotypical liberals like to say, she cares. She does not think, at any level, whether in lofty or pedestrian terms, about whether or not she would take her own needs as reasons for people to help her. Despite this fact, agents like her often

6

See Pettit and Smith (1990).

Moral Worth and Normative Ethics

95

receive praise, gratitude, esteem, and, in dramatic cases, even admiration for their actions. Some would have high regard for Jennifer especially because, when she helps, the thought of her own needs does not occupy her mind at all, not even hypothetically. Even if one thinks that a healthy or rational person has some concern for her needs and accepts help at least occasionally, one can still admire a helping action that is not motivated by any thought associated with one’s needs (in the case of the super-hero it is hard to see why it would be irrational of her to reject help). One can admire such an action in a basically healthy and virtuous person who thinks of her needs when they arise, but not when helping others, and even in a person whom one would criticize as suffering from significant irrationality because she refuses help. The “tough” version of Jennifer, the one who values her independence above all, may seem silly—like the stereotypical man who does not ask for directions—but that does not lower our regard for her helping John. Even if we imagine a Jennifer who is clinically depressed and who quite irrationally fails to take her needs as reasons for others to help her, we imagine a person who would normally be esteemed for helping John, as long as she does it because of genuine concern with wellbeing and not as a desperate gambit for attention, say. It is possible for a person to have a disorder and also perform morally worthy actions directed at others (and Kantian intuitions tell us that there is something noble about a depressed person doing just that). Similar things are true for a Jennifer who, morally speaking, does not respect herself enough. It is possible for a person to have a vice (lack of self respect) and still perform morally worthy actions. Robert Louis Stevenson said: “It is the history of our kindness that alone makes this world tolerable. If it were not for that—for the effect of kind words, kind looks, kind letters ( . . . ) I would be inclined to think our life a practical jest in the worst possible spirit”.7 Korsgaard’s morally worthy helper, the one who attends to another’s needs because she would want others to take her needs as reasons to help her, is arguably acting not out of kindness in the ordinary sense but out of a type of fairmindedness. Now, not everyone values kindness quite as much as

7 This quote is used in a moving way by Kay Jamison in her autobiography An Unquiet Mind.

96

Nomy Arpaly

Stevenson does, but Stevenson’s sentiment is not unusual, and many have some version of it. It seems that, if one has Korsgaard’s view, one is committed to thinking that Jennifer, who acts out of a type of kindness that is not related to fairness, is someone whose actions have no more moral worth than those of the prudent grocer when he acts for profit. Does that mean that the advocate of this Kantian view must alienate herself from an intuition that, for many, runs deep? Some contemporary Kantians—I am thinking of Wood (2014) and Johnson (1996), would say that’s not necessarily so. They would invoke Kant’s concept of moral merit, which they interpret as distinct from his concept of moral worth. Kant’s concept of merit is quite complex, but for the moment, it suffices to say that it is plausible that Jennifer’s action has merit. If one’s actions have merit, one deserves praise and encouragement for one’s action, even if one does not deserve esteem (the latter being reserved to morally worthy actors). Wood and Johnson point out that for Kant, actions performed out of pure self-interest do not deserve praise and encouragement, but actions performed out of compassion can deserve them under some conditions, as well as actions performed out of concern for honor. Thus, the Kantian does not need to lump Jennifer’s case together with the case of the prudent grocer. Can a Kantian who holds that even though Jennifer’s action has no moral worth, though she still deserves praise and encouragement, accommodate the common intuition that there is something valuable about Jennifer’s action and motivation when she helps? To answer this, we need to ask the question: What is the difference between deserving esteem for one’s action, which seems to be equivalent to one’s action having moral worth, and simply deserving praise? One clue is provided by the juxtaposition of praise and encouragement. These are both actions, whereas esteeming is not an action (the question “what are you doing?” can be answered with “praising and encouraging my child” but not with “esteeming Mozart”8). Usually, encouragement is deserved the way a reward is deserved, whereas esteem is deserved the way an evaluation is deserved. The word ‘praise’ in English refers to something that can be deserved in both ways.

8

I esteem James Dreier for coming up with this approximate test of action-ness.

Moral Worth and Normative Ethics

97

To see this, consider the following conversation: LOUIS: CLARA: LOUIS: CLARA: LOUIS: CLARA:

I think he deserves a lot of praise. Why? Because he is a very good painter. I don’t think he deserves to be praised as much as he is praised these days. Why? Because he never, ever, praises anyone else.

Louis and Clara are talking about different ways in which one can deserve praise. Louis is talking about whether praise is warranted in the case of the man they are talking about, which is really asking whether admiration for him is justified. Whether admiration for a person’s ability as a painter is justified is a question to which it is simply irrelevant whether or not he praises other painters: Bringing this up would be appealing to the “wrong kind of reasons”. Clara, on the other hand, is talking about a practical question: whether one should perform an act consisting in giving the man praise. She wants to know if he deserves the benefit of hearing people praise him, in the same way in which one can ask whether someone deserves the money that people pay for his paintings. Whether or not the painter she is discussing praises other painters may or may not be relevant to the question, but it is not patently absurd to suspect that it is. Talk of deserving praise can go both ways, then. Sometimes we say a person deserves praise and mean to say that admiration and esteem for her are warranted. At other times we say that a person deserves praise and mean that it would be good to say a few good words about him, just like a person can “deserve encouragement” in the sense that it is good to encourage him. To say that Jennifer deserves “praise and encouragement” but not “esteem” is to say that praising her is desirable, the way encouraging her is. This might be true, and it may well be one difference between her case and that of the prudent grocer. However, those who think Korsgaard is wrong about people like Jennifer do not think that the disagreement is about how we should act towards Jennifer, but rather exactly about how admirable or esteem-worthy she and her action are. We might agree that it is desirable to praise the honor seeker, at least in a society in which the right things are honored, perhaps because such honor-seeking, unlike the grocer’s greed, is likely to bring forth good

98

Nomy Arpaly

actions. Yet in the case of Jennifer we think there is something that “shines like a jewel” about the agent herself and the way she has acted— something beyond the desirability of praising the agent (or the desirability of encouraging her or of any other behavior towards her). To change the subject a little bit, can a broadly Kantian thinker be more accommodating towards Jennifer if he rejects Kant’s view of the emotions? Korsgaard points out what she sees as Kant’s idea of emotions and feelings as “stupid”, which would be a great adjective to describe sympathy if it in fact prevented otherwise smart people from taking children to the doctor (or if it lead otherwise smart people to help an art thief who is struggling with a heavy load, as Herman 1993 suggests). Korsgaard (1996b) raises the possibility of a generally Kantian view that does not include a Kantian view of emotion and feeling but is instead coupled with a more sophisticated, Aristotle-style view of these things. Indeed, there seems to be nothing to prevent us from being roughly Kantian and still holding a view according to which emotions and feelings can embody value judgments or respond to reasons. This could imply that some actions that seem guided by emotion and feeling, as opposed to some kind of explicit deliberation, have moral worth. I agree that it is open to Korsgaard, and to many who hold any number of mainstream Kantian normative views, to hold that an agent who seems to act out of simple sympathy can be responding in a sophisticated way to reasons or values. However, the challenging thing about Jennifer’s case for such a philosopher is not that her motive might be “emotional” but that her motive seems to have nothing to do with universalizability: She does not have that additional “stretch of thought”. As I have described her, Jennifer’s motive does not even have to be “emotional” in the ordinary sense: Perhaps, when she helps John, she is too tired to have warm feelings of compassion towards him, but she tells herself coldly and explicitly that John needs help, and that cold and explicit thought motivates her—a thought that has nothing to do with how she would have felt about a world in which nobody considers helping her. This version of Jennifer’s case would still have to be regarded by Korsgaard as involving no moral worth. I should mention that it seems perfectly possible to me for someone to act on a “visceral” motive and still be responding to reasons involving universalizability: Take, for example, a person who quickly declines a proposition involving tax evasion (or even tax avoidance) because he

Moral Worth and Normative Ethics

99

recoils, as viscerally as can be, at the thought of being a “free rider”. This person does not consciously deliberate about universalizability, and yet his feeling of revulsion at the thought of making an exception for himself could betray a sensitivity to universalizability-based moral reasons. By hypothesis, though, Jennifer is different.

MISGUIDED ALTRUISTIC ACTIONS AND TYPES OF REASONS

On my view, the esteem-worthiness of motives is not about reliability or “modal robustness”. The “non-accidentality”, as Markovits calls it, of the connection between the moral worth-granting motive and the right action done from it is in the fact that the right action is done from the reasons for which it is right. To be more precise, the non-accidental connection is not so much between motive and action but between motive and morality. If universalizability makes actions right and a person does the right action because it is universalizable, the connection between his motive and rightness goes beyond matters of probability. However, a fact that needs to be explained by anyone who holds that altruism grants moral worth is the fact that altruism sometimes leads to a wrong action. I am not talking about cases in which a person helps another not knowing that the other is a thief, or fails to take a crying child to the doctor because of not knowing the elementary fact that it would be in his best interest to go there. I am referring to cases in which there is no complication in the form of lacking information or false factual beliefs, but the altruistic agent does the wrong thing anyway. If altruism is a moral motive, how can it lead, without factual ignorance, to even one wrong action? Also in need of explanation is the fact that even though people who are not philosophers know full well that there is such a thing as a wrong action motivated by altruism, they often resist the conclusion that right actions motivated by altruism have no moral worth. Consider again Jennifer, who helps John with his dissertation in order to alleviate ill-being. Imagine that after getting to know Jennifer a little better you discover that she has at some point lied to her roommate. Jennifer, it turns out, faced the situation described by Thomas Hill in “Autonomy and Benevolent Lies.”9 Her roommate had barely recovered 9

Hill (1991).

100

Nomy Arpaly

from an extremely painful love affair with a person with whom she was obviously incompatible. One night she asked Jennifer whether she thought her ex-boyfriend would be open to getting back together, and Jennifer said “no” despite knowing that the ex-boyfriend had, in fact, wondered aloud about the rightness of the breakup and expressed a desire to get back together. Suppose Jennifer said “no” to her roommate for the same reason for which she helps John with his dissertation: her roommate risks utter misery and she wants to protect her wellbeing. Let us assume that Hill is correct that this action is wrong. Thus it appears that if my view of moral worth is true it should be expected that if we were to learn about Jennifer and her roommate, we would reevaluate her action vis á vis John and his dissertation and regard it as morally worthless. We would feel the same kind of disappointment we feel when we discover that a person’s good deeds were in fact motivated by money or by honor or by a need to see himself as a good person. But while Jennifer’s lie to her roommate is a sign of an imperfect character, many of us would keep having some esteem for her and for her action of helping John. We can imagine John, upon discovering Jennifer’s paternalistic lie, saying something like “Look, Jennifer, you are basically a good person and I really appreciate what you have done for me. But you’ve got to get it into your head that sometimes you have to let people make their own decisions.” Our attitude to Jennifer is not pure disillusionment but rather ambivalence. To account for this complex attitude, I propose the following. There is more than one type of moral reason—more than one right-making feature that different actions can have—and it is possible to act in a way that reveals a commitment to only part of morality. One right-making feature some actions have has to do with wellbeing, but it is not the only one. Altruism fails to deliver right actions every time because considerations that have to do with wellbeing present a moral reason to help that is sometimes overridden by other moral reasons, reasons to do something else. Accordingly, Jennifer’s helping John might not have quite as much moral worth as such an action could have—perhaps the right-making features of helping John, specified more fully, include not just “alleviates a person’s ill-being” but also, say, “does not violate a person’s autonomy” or “is universalizable”, in which case Jennifer was only moved by a part of that which makes her action right. Her doing the right thing is not completely accidental—her reasons for action are some of the right

Moral Worth and Normative Ethics

101

reasons. It is, however, partially accidental—it is lucky that in the case of helping John no overriding reason exists to refrain from helping. However, even if this is true, I see no reason to insist, with Kant, that moral worth is an all-or-nothing affair. When a writer of short stories who is good at character development but not at creating page-turning plots writes a story in which the former is essential and absence of the latter does not really matter, she shows a real aesthetic sensibility even though there are writerly tasks in which she would fail. Moral sensibility can be similar. More questions are raised by what I have said than I have space to answer, but I shall tackle two of them. I have argued that wellbeing provides a moral reason for action that is not based on universalizability. A question that one might raise is the following: we have an idea of what an agent is like who is motivated to protect wellbeing but is missing at least one additional consideration, thus doing wrong. What would an agent be like who is motivated by universalizability, but who is missing at least one additional consideration, thus acting wrongly? A full answer to the question requires another paper. It would require presenting a putative example of such an agent and defending the claim that he is in fact such an agent. Here, I will only provide the putative example and the basic argument. Strictly speaking, I will argue not that such an agent can exist, but that such an agent can exist if my view of moral worth is true. The relevant part of my theory is the part that states that a person acts viciously only if her course of action manifests some indifference to one or more moral reason. Imagine the following two agents: 1) The Tough Guy: Akhil thinks of himself as a tough guy. He is a person who does not want anyone to help him, and will not accept free help under any circumstances. He values his self-reliance to the point that offers of free help are perceived by him as a nuisance, and he is rather vain about the subject. One day Akhil sees Jardish, whom he knows from work, in the parking lot of a store, struggling with a heavy box that is he trying to carry the short distance to his car. Jardish has to take frequent breaks to catch his breath, but Akhil, of course, has spent a lot of time lifting weights. It would be remarkably easy for Akhil to help Jardish, but he does not offer help.

102

Nomy Arpaly

He, Akhil, would never have wanted anyone to do him such a favor, even if the box was heavy enough to give him trouble. With a lack of imagination typical of human beings, Akhil thus assumes that surely Jardish feels the same. “Maybe not everyone is quite as tough as me”, Akhil would say if asked why he did not help, “but surely anyone, especially a man, would be insulted if offered help with that situation. How could I offend Jardish by implying that he can’t find a solution himself?” In truth, Jardish, who is different from Akhil in values and temperament, would have gratefully accepted the offer. 2) The Other Tough Guy: Kabeer also considers himself a “tough guy”, develops his physical strength and values self-reliance in the same extreme way that Akhil does. He sees Jardish struggling with the box. He knows Jardish does not have macho values or macho vanity and guesses correctly that he wants help. He does not help, even though it would have been easy for him. If asked for his motives, he would say, “Why should I help? I don’t want help from anyone, so it’s only fair that I help no one”. He would say that honestly and sincerely. He always avoided, and will avoid, courses of actions that he would, in fact, regard as non-universalizable. There is a difference between our two tough guys: Kabeer is a lot worse than Akhil. In contemporary parlance, Kabeer is behaving and thinking like an asshole, or at the very least a jerk, whereas Akhil displays no such tendencies, though he might show himself to be more naïve than is desirable. The difference between vicious behavior and wellintentioned behavior gone wrong due to naiveté is, on my view, the following: Naïve behavior goes wrong due to a factual mistake, whereas vicious behavior originates in a failure to respond to a moral reason. If morality were all about universalizability, and Kabeer’s only crime was misapplying the universalizability criterion, we would perceive him as regrettably naïve, bumbling, or simply as someone who made an error in calculation—we would see him, in other words, as we see Akhil. However, this is not how we see him: We see him as vicious. There is a moral reason to which he is failing to respond when he shrugs off Jardish’s need for help. On to a different question. The Kantian view I have criticized in this paper is a view based on the idea of universalizability and thus rests on a common way to interpret the universal law formulation of the categorical

Moral Worth and Normative Ethics

103

imperative. There are, however, other formulae. Can there be a plausible view that anchors both considerations of wellbeing and fairness-style considerations in the popular Kantian idea of refraining from treating one’s fellow human as mere means and always treating her as an end? Perhaps there is a duty, grounded in the formula of humanity, to take on other people’s ends as one’s own, and Jennifer’s action of helping John seems esteem-worthy because she is acting on this duty. It is implausible, prima facie, that Jennifer acts from the formula of humanity. The duty to further the ends of other persons, at least as normally conceived by Kantians, is strictly incompatible with paternalism, which is an imposition on the other person of ends set for him by the agent. Though ordinary English makes it sound strange, Kantians regard paternalism of the sort expressed in benevolent lies as a paradigmatic way of treating persons merely as a means, or at the very least as a way of disrespecting the human ability to set ends as present in the victim of the paternalistic action. Jennifer, according to the most natural interpretation of her story, acts from the same motive when she helps John and when she lies to her roommate, and a motive that leads, without factual errors, to paternalism cannot be the formula of humanity. There is no reason to assume that Jennifer’s decision to lie to her roommate is based on a factual error, such as literally not knowing that the roommate is neither psychotic nor a child, but rather a competent, rational-enough human capable of setting her own ends, if not always very wisely. It is extremely unlikely that every case of paternalism involves such a factual error (though unfortunately such errors do happen, for example when the victim of paternalism is very old or very unusual). Instead, it seems that often the paternalistic agent sees the victim’s ability to make autonomous decisions, but disrespects that ability, or, in more old-fashioned terms, sees the humanity in the victim but treats it merely as a means—and it is hard to use something as a means when you do not know it is there. If we assume that Jennifer’s case is one of these common cases, we have to conclude that she is not acting on the formula of humanity when she lies to her roommate. By hypothesis, the motive for Jennifer’s action is the same when she lies to her roommate and when she helps John. An advocate of the formula of humanity may, of course, deny that is possible and insist that Jennifer helps John from a radically different motive from the one that leads her to lie to her roommate. She can hold that all virtuous help is motivated

104

Nomy Arpaly

by one thing—the formula of humanity—whereas all misguidedly paternalistic help is motivated by another—perhaps simply concern for wellbeing. That would be costly, as rather often the agent does not experience things that way, nor do the people around him—who think, for example, that the same good intentions that he always has were sometimes appropriate and sometimes misguided. In addition, whether one perceives the different formulae of the categorical imperative as versions of the same directive or as related in a more complex way, it is an awkward position for a Kantian view to hold that it can be said, of the same agent, that she follows one formula but not the other. Jennifer, by hypothesis, is not motivated by the thought of universalizability, and so, if the moral law is about univeralizability, it seems that she is not motivated by the moral law. If we say that she is motivated by the formula of humanity, we imply that she is, in fact, morally motivated. Usually, one would think that for a Kantian theory, either one is morally motivated or one is not—and the “eitheror” is an exclusive one. The question under discussion here is not whether the correct normative theory can be rightly called “Kantian”. This is partially a question of Kant interpretation and I will leave that to Kant scholars. The question is whether the correct normative theory involves one rightmaking feature or more than one. Hopefully I have shown reasons to suspect that there is more than one right-making feature that actions can have. It would take more to vindicate the suspicion, which I hope to attempt on another day. REFERENCES

Arpaly, N. (2002). “Moral Worth.” Journal of Philosophy 99, 223–45. Arpaly, N. (2003). Unprincipled Virtue. New York: Oxford University Press. Arpaly, N. and Schroeder, T. (2014). In Praise of Desire New York: Oxford University Press. Bennett, Jonathan. (1974). “The Conscience of Huckleberry Finn” Philosophy 49(188), 123–34. Blum, Lawrence. (2009). Friendship, Altruism, and Morality. London: Routledge and Kegan Paul. Herman, B. (1993). The Practice of Moral Judgment. Cambridge: Harvard University Press. Hill, Thomas E. (1991). “Servility and Self-Respect”. In Autonomy and Self-Respect. Cambridge: Cambridge University Press. Johnson, Robert. (1996). “Kant’s Conception of Merit” Pacific Philosophical Quarterly 77: 310–34. Korsgaard, C. (1996a). Creating the Kingdom of Ends. Cambridge: Cambridge University Press.

Moral Worth and Normative Ethics

105

Korsgaard, C. (1996b). “From Duty and For The Sake of The Noble”. In Stephen Engstrom & Jennifer Whiting eds., Aristotle, Kant, and the Stoics: Rethinking Happiness and Duty. Cambridge: Cambridge University Press. Markovits, J. (2010). “Acting for the Right Reasons.” Philosophical Review 119, 201–42. Pettit, P. and Smith, M. (1990). “Backgrounding Desire.” Philosophical Review 99, 565–92. Railton, P. (1988). “Alienation, Consequentialism, and the Demands of Morality.” In Samuel Scheffler ed. Consequentialism and Its Critics. Oxford: Oxford University Press, 93–133. Smith, Michael. (1994). The Moral Problem. Oxford: Blackwell. Wood, Allen. (1999). Kant’s Ethical Thought. Cambridge: Cambridge University Press. Wood, Allen. (2014). “Moral Worth, Merit, and Acting from Duty.” In The Free Development of Each. Oxford: Oxford University Press, 13–40.

5 Ideas of the Good in Moral and Political Philosophy T . M . SCANLON

INTRODUCTION

My topic is the relation between ideas of the good that a person should use in assessing his or her own life and ideas of the good that figure in moral and political philosophy. Moral Philosophy, as I will understand it, is concerned with principles regulating our conduct toward one another. It is concerned both with the content of these principles— with what morality requires—and with the question of acceptance— why and to what degree we should take them seriously as guides to action. Political philosophy, as I will understand it, is concerned with standards for assessing large scale social institutions that we participate in and expect others to participate in. Here again there are questions of content—what justice requires—and questions of acceptance—why we should care about justice. Answers to both questions of content and questions of acceptance seem to depend on claims about individual good, that is to say, claims about what individuals have reason to want. But it is important to distinguish between two different kinds of claims of this kind. Claims of the first kind are claims about what is good from an individual’s point of view, that is to say, about what a person has reason to want in general. Claims of the second kind are claims about what a person has reason to want taking into account only effects on his or her life. Claims of the latter kind might be called claims about what is good for a person, and I will sometimes adopt this terminology.1 1 Stephen Darwall draws a similar distinction, between “what an individual values, prefers, or takes an interest in” and the person’s good, or what is to be “valued for his sake.” See Welfare and Rational Care (Princeton: Princeton University Press, 2002), pp. 1–3. But by what is good

Ideas of the Good

107

The normative standpoint defined by the question of what, in general, an individual has reason to want is in one way an objective standpoint since it takes into account reasons provided considerations that go beyond the ways that person’s life may be affected, such as the good of others. But this point of view remains subjective in another sense, since it is defined by the question of what that person has reason to want and to do, and the answer to this question will depend on that person’s particular situation, aims, relationships, and so on. The content of moral principles and principles of justice clearly depends on facts about what is good for individuals. Justifications of particular principles of right and wrong appeal to how individuals’ lives would be affected if people generally behaved in ways that those principles require, and how they would be affected if people behaved in ways that the principles forbid. Such justifications appeal, for example, to the reasons individuals have to want not to be subject to threats and violence, to be able to rely on agreements they have made, and so on. Utilitarian theories justify principles by appeal to such effects, but so do non-utilitarian theories such as my contractualism, and Rawls’ theory of justice, although they do this in different ways, and adopt different understandings of what individuals have reason to want for themselves. What alternative is there to this way of justifying moral and political principles? Appeal to brute facts about right and wrong, such as facts about rights, is one possibility. But it seems possible to give reasons for our judgments about moral right and wrong, and these reasons depend in large part on what it would be like to live under such principles, and what things would be like if accepted standards were different. What else could they appeal to? Social institutions and the content of moral requirements might be justified by direct appeal to some good other than the good of those to whom these requirements apply. Indeed, in the case of some institutions this seems the appropriate mode of justification. For example, my rights from a person’s point of view, I have in mind what that person has reason to want or do, not just what he or she in fact values or prefers. Darwall also identifies what he calls a person’s good with his or her welfare. It may be that, given how Darwall understands welfare, I would agree. But for present purposes I will leave open exactly how this idea is understood. All that matters is that there is a distinction between a person’s good and the broader notion of what is good from a person’s point of view. I thank Connie Rosati for prompting me to make this clearer.

108

T. M. Scanlon

and duties as a Harvard faculty member, such as the right to speak and vote at departmental meetings, are justified not by what is in my interest but by what is necessary in order for the University to operate in a way that serves certain goals. These include promoting the good of other individuals by providing education, and useful research, but also producing advancements in knowledge and understanding that are good in themselves, apart any from any practical benefits they might lead to. In the case of large-scale social institutions such as the state, however, justifications of the latter sort, which appeal to impersonal values, have something of a bad name. “Perfectionist” is one thing that such views are sometimes called. In contrast to such views, Rawls says at the outset of A Theory of Justice that he is viewing a society as “a cooperative venture for mutual advantage.” (TJ, 4–5.) This may seem an unexceptionable remark, and he offers no justification for it. But it is a substantive assumption, in favor of what might be called individualist answers at least to questions of content. My inclination is to accept this assumption, and count myself an individualist in this sense—that is, as someone who believes that in both moral and political philosophy answers to questions of content must be based on claims about what individuals have reason to want because of the way their own lives are affected. Consider now questions of acceptance—of what reason we have to accept moral principles as binding on us, or to care about principles of justice. Answers to these questions will need at least to take into account claims about what is good for us—that is to say, how our lives will be affected by the acceptance of moral principles or principles of justice. Answers to questions of acceptance cannot, I would say, appeal only to claims about individual welfare. Our reason for accepting moral principles as binding on us, and for wanting to live under just institutions, is not simply that living in these ways makes us better off. Insofar as living in accord with morality and with justice are things we have reason to want, it is because they constitute relations with others that we have reason to want to have. This is a matter of how our lives are affected, even if it is not a matter of welfare in the usual sense. A full answer to questions of acceptance, however, will have to go beyond claims about how our lives are affected, and consider whether living in accord with morality and justice is compatible with values that go beyond our own lives. Such an answer will therefore involve claims about what individuals have reason to want in general.

Ideas of the Good

109

In the remainder of this paper I will, as I have said, be concerned with how the conceptions of what individuals have reason to want that are relevant to answering questions of content and acceptance are related to the conceptions that individuals rightly employ in making decisions about their own lives. It might seem that there should be conceptions of what individuals have reason to want that can serve all three of these functions: as the basis for individuals’ own personal decisions about their lives, as a basis for arguments about the content of moral principles and principles of justice, and as the basis for arguments about why individuals should care about such principles. It seems to me, however, that this is not the case: the conception of what individuals have reason to want that figures in arguments about the content of moral requirements differs from the conceptions that individuals rightly appeal to in guiding their own lives. If this is so, it makes the question of acceptance more difficult to answer. It is more difficult to show how individuals can be justified in accepting the requirements of morality and justice if the content of these requirements takes account of how their lives are affected only in ways that are in some tension with their own conceptions of what is good. I will consider how this question arises and is dealt with by a number of apparently diverse thinkers: Henry Sidgwick, Karl Marx, John Rawls, and Bernard Williams. SIDGWICK’S DUALISM OF PRACTICAL REASON

Consider first the way in which this question arises for Sidgwick. At the end of The Methods of Ethics, Sidgwick distinguishes two rational points of view: two ways in which we can arrive at conclusions about what we ought to do.2 From one point of view we consider only how our own life is affected. From the other we take into account effects on the lives of all sentient beings. We may refer to the first of these as the point of view of what is good for a person, in the terms I used above, and refer to the latter as the point of view of what is impartially good. So described, it may seem as if conclusions reached from these points of view cannot conflict—they are conclusions about two different things: about what is good for me and 2 The Methods of Ethics, Seventh Edition (Chicago: University of Chicago Press, 1962), pp. 496–509.

110

T. M. Scanlon

what is impartially good. But as I have said, Sidgwick understands these as two ways in which we can arrive at conclusions about what we ought to do. Since what is best for me is not always the same as what would be impartially best, conclusions about what we ought to do reached from these two points of view can conflict. When they do, practical reasoning leads to conflicting directives. In such cases Sidgwick believed that, our practical reason is “divided against itself.”3 He saw this as a crisis, which he referred to as the Dualism of Practical Reason. Although it is not essential to this crisis, which could arise under different understandings of what is good for me and what is impartially good, it will be helpful for what follows to say more about how Sidgwick understands these two notions. First, he believes that what is of ultimate value in both cases is agreeable consciousness. In the case of my good, it is my own agreeable consciousness. In the case of impartial good, it is agreeable consciousness had by any sentient being. Each of these forms of good is what he calls a mathematical whole composed of occurrences of agreeable consciousness, and considering these wholes rationally requires a form of impartiality. Insofar as one is concerned with either type of good, it is irrational to prefer a unit of agreeable consciousness at one time to an otherwise identical unit occurring at some other time, and insofar as one is concerned with what is impartially good, it is irrational to prefer a unit of agreeable consciousness occurring in one life to an otherwise identical unit occurring in some other life. Although the ultimate value at stake in these two kinds of judgment is in one sense the same—agreeable consciousness—there is an important difference in the way it is valued. An additional unit of agreeable consciousness in my conscious life is good for me. That is to say, I have reason to want it because of its contribution to my life. An additional unit of agreeable consciousness in someone’s life is impartially good: it is something anyone has reason to want because of its contribution to someone’s life. To put the matter in the terms used by Thomas Nagel in The Possibility of Altruism, reasons of self-interest are subjective reasons, requiring a reference to the person in question, whereas reasons to promote the good generally are what Nagel called objective reasons: the

3

The Methods of Ethics, p. 508.

Ideas of the Good

111

facts that are such reasons can be described, in a way that captures their normative significance, without reference to the person for whom they are reasons.4 The conflict that troubled Sidgwick depends on the independence of these two forms of value. This conflict would not arise if a unit of agreeable consciousness were good for me only if, and because, it is good impartially (as Sidgwick says, good “from the point of view of the universe”) and happens to occur within my life. If this were so, then conclusions about my own good would simply be partial conclusions about what is good impersonally (they would be conclusions about what is good in one part of the universe, so to speak.) Such conclusions would therefore have no normative weight against conflicting conclusions about what is impersonally good, on balance. They would simply be subsumed by judgments of the latter kind, and the problem Sidgwick faces would not arise.5 I will call the claim that the good for an individual does depend on impartial good in this way “the Moorean Thesis,” after G. E. Moore, who espoused something like this view.6 There are many cases in which what is good for me does depend on what is good in a sense that does not concern only effects on my life. For example, success in my main aims makes my life better only if these aims are worth pursuing, and this can be so apart from any benefit to me. Success in finding a cure for cancer, or proving a deep theorem would make my life better. Success in counting blades of grass would not. It may be maintained that curing cancer and proving theorems of mathematical importance are things worth doing only if, and because, they lead to results that make other people’s lives better. On this view, the idea of what is good for individuals is, ultimately, the only thing that matters. I doubt that this is true.7 Even if it is, however, the present point 4

Nagel, The Possibility of Altruism (Oxford: Oxford University Press, 1970), pp. 90ff. Sidgwick recognizes this and believes that the Dualism of Practical Reason cannot be avoided in this way. See pp. 497–8 and, earlier 420–1. 6 G. E. Moore, Principia Ethica (Cambridge: Cambridge University Press, 1903), sections 59, 60. This Moorean Thesis is defended by Donald Regan in “Why Am I my Brother’s Keeper?” in Wallace, Pettit, Scheffler and Smith, eds., Reason and Value: Themes from the Moral Philosophy of Joseph Raz (Oxford: Oxford University Press, 2004), pp. 202–30. For criticism of Moore’s claim that good for is unintelligible, see Richard Kraut, Against Absolute Goodness (Oxford: Oxford University Press, 2011), Chapter 12. 7 For a thoughtful defense, see Richard Kraut, What Is Good and Why (Cambridge, MA: Harvard University Press, 2007), esp. Chapter 4 and pp. 268–9. In assessing this issue it is 5

112

T. M. Scanlon

remains: that the degree to which success in one’s pursuits makes one’s life better depends on whether these aims are worth pursuing for reasons that do not depend solely on effects on one’s own life. This does not mean that something’s being good for a person always depends on its being impartially good, in the way that the Moorean Thesis claims. The latter thesis seems particularly implausible in the case of the kind of good Sidgwick took to be basic: agreeable consciousness. A unit of my agreeable consciousness is not good for me only because, as a unit of someone’s agreeable consciousness, it is good impersonally that it occur. Rejection of the Moorean Thesis may or may not be compatible with what Thomas Nagel argued in The Possibility of Altruism. Nagel’s claim was that we must be able to see what is good for us as also good impersonally, and hence to see, and to be motivated by the fact that, similar things in other people’s lives are also good in the same way. So far, this is compatible with what I have just said, because it does not mean that we have to see something as good for us only because it is an occurrence in our lives of something that is impersonally good. But Nagel does also say that any subjective reason has to be statable in objective form with the same motivational content.8 So the compatibility of his view with the claim I have just made is not entirely clear. (It should also be said that this claim of Nagel’s is one of the most controversial in his book.) If the Moorean Thesis is mistaken, then there can be conflicts between the answers to Sidgwick’s two questions. Answers to the question about what is good for me are naturally understood as conclusions about what is in my self-interest. Answers to the question of what is impartially good can be described in either of two ways, both of which Sidgwick employs: simply as conclusions about what we have most reason to want, from an impartial point of view, or as conclusions about what is required by Utilitarian Duty. When these conclusions are understood in the latter way, the conflict is between morality and self-interest. important to bear in mind the distinction made above between what is good for an individual (makes his or her life better) and the good for an individual (what he or she has reason to want.) It seems obvious that something is good only if it is part of the good for some individuals in the latter sense. The question is whether it must also be good for some individuals in the former sense. 8

The Possibility of Altruism, pp. 110–15.

Ideas of the Good

113

The question of what to do when faced with a conflict of this kind would be resolved by an answer to the question of acceptance: such an answer would tell us what reason we have to care about what is morally right, and how much weight we should give to the fact something is morally required when this conflicts with what is in our own interest. As I said earlier, this is a question about what a person has, in the broadest sense, most reason to do. It seems obvious that in some cases of such conflict what I have most reason to do is to do what morality requires. If using a certain industrial process would cause cancer in people living near the plant, the fact that I could make more money by using this process does not make it the case that this is what I have most reason to do. It is easy to think of many more examples of this kind. So it is puzzling why Sidgwick says that whenever duty and self-interest conflict, practical reason is “divided against itself.” I believe that the best explanation of Sidgwick’s view about the Dualism of Practical Reason lies in his assumption that there could be a rational answer about what to do when conclusions about duty and self-interest conflict only if there were some more ultimate normative standard on which both of these conclusions depend for their normative force, and in terms of which their force can be compared, yielding a judgment about what one ought to do. This more ultimate standard might, for example, be an idea of the good such that something is good for me, or good impartially, only because it is good in this deeper sense. Or it might be a substantive idea of what one ought to do such that one ought to do what is in one’s self-interest, and one ought to do what is impartially best, only because one ought, in this more ultimate sense, to do these things. Sidgwick, very plausibly, believes that there is no such deeper substantive normative idea. This amounts to a generalized rejection of the Moorean Thesis: just as it is not true that something is good for me only if and because it is impartially good (and occurs in my life), there is no other idea of good such that something is good for me, or good impartially, only if and because it is good in this deeper sense. (And the same is true put in terms of ‘ought.’ The ‘oughts’ of self-interest and impartial rationality do not derive their normative force from some more general substantive ‘ought.’) Matters change if we put things in terms of reasons, and take the most authoritative practical question to be what we have most reason to do all

114

T. M. Scanlon

things considered. This question takes into account, within one normative frame, so to speak, what is good for us (what we have reason to want because of how we are affected by it) and what is good for others (what we have reason to want because of how they are affected by it), along with any other reasons we may have. Conclusions reached by considering this question have rational authority, since being rational involves being guided by conclusions of this kind. It would be irrational not to be guided by these conclusions. The idea that the question of what I have most reason to do is more basic than the question of what is in my self-interest or what is impartially best, and that the strengths of our reasons to do what is best in these ways can be compared, is not implausible in the way that the Moorean Thesis is implausible. It is not implausible to say that something is in my self-interest only if it involves my life being affected in a way that I have reason to want; nor is it implausible to say that something is impartially good only if it affects the lives of sentient creatures generally in ways that one has reason to want. Given that this is so, when there is a conflict between duty and self-interest (or between self-interest and what would be best from an impartial point of view) we can meaningfully ask which of these conflicting reasons is the stronger. It may be that this question does not have a determinate answer in every case. But it has such an answer in many cases, as the examples I have referred to show. So it is not true, as Sidgwick thought, that practical reason is divided against itself whenever there is a conflict between duty and self-interest. This solution to Sidgwick’s Dualism would not be available if the strengths of different reasons could be compared only if these strengths can be explained in terms of some common reason-providing factor, such as pleasure, or some other unified conception of the good. Although this is sometimes maintained, it seems to me clearly false.9 An additional 9 The importance for Sidgwick of such a basis is indicated in one of the reasons he gives for taking pleasure to be the sole ultimate good. “If we are not to systematize human activities by taking Universal Happiness as their common end, on what other principles are we to systematize them? It should be observed that these principles must not only enable us to compare among themselves the values of the various non-hedonistic ends which we have been considering, but must also provide a common standard for comparing these values with that of Happiness; unless we are prepared to adopt the paradoxical position of rejecting happiness as absolutely valueless. For we have a practical need of determining not only whether we should pursue Truth rather than Beauty, of Freedom of some ideal constitution of society rather than

Ideas of the Good

115

reason for rejecting this idea is provided by the fact we have just seen: if it were true, it would lead either to an implausible version of the Moorean Thesis or to the implausible conclusion that whenever duty and selfinterest conflict there is no answer to the question of what we have most reason to do because practical reason is divided against itself. I have been considering Sidgwick’s Dualism of Practical Reason as a problem about what I have called the acceptance of morality. Such an explanation will, I have said, be a claim about what an individual has most reason to want or to do, all things considered. Such a claim must take into account what is good for that individual, along with the requirements of morality and other considerations. It will therefore depend on the way in which both morality and the good of an individual are understood. I have argued that Sidgwick’s problem arose in part from the falsity of the Moorean Thesis. That is to say, it arose because the way in which an individual’s good is taken into account in arriving at conclusions about duty (as something impartially good) differs from the way in which an individual sees his or her own good. (As valuable because it is his or hers.)

MARX, RAWLS, AND WILLIAMS ON ACCEPTANCE

I turn now to problems about the acceptance of morality and justice confronted by Karl Marx, John Rawls, and Bernard Williams. Unlike Sidgwick, they all believe that the question of acceptance—whether to accept the requirements of morality or justice when these conflict with one’s personal aims—can be answered, positively or negatively. But I will argue that the problems that underlie their answers arise, as in Sidgwick’s case, from the falsity of some version of the Moorean Thesis—from differences between the way in which a person conceives of his or her

either, or perhaps desert all of these for the life of worship and religious contemplation; but also how far we should follow any of these lines of endeavor, when we foresee among its consequences the pains of human or other sentient beings, or even the loss of pleasures that might otherwise have been enjoyed by them.” Methods of Ethics, p. 406. For arguments against the need for such a common standard see Rawls, A Theory of Justice (Cambridge, MA: Harvard University Press, 1971), pp. 548–60 (Second Edition, pp. 486–91) and Henry Richardson, Practical Reasoning about Final Ends (Cambridge: Cambridge University Press, 1997), Chapter VI.

116

T. M. Scanlon

own good and the way this good is taken into account in determining the content of morality, or justice. I begin with the problem of toleration as discussed by Marx. In his famous 1843 essay, “On the Jewish Question,” Marx was responding to Bruno Bauer, who had claimed that it was incoherent for Jews to ask for religious toleration.10 Bauer held that the outlook of religious toleration involves seeing various religions as of equal standing. Jews could therefore not demand, and accept, religious toleration without seeing Judaism as simply one religion among others, thereby, as he put it, “ceasing to be Jewish.” (The same would be true of Catholics, or adherents of any other religion, but Bauer discussed the case of Jews since it was their claim to toleration that was in question.) Marx replied that accepting religious toleration did not involve equating different religions, but only accepting that the difference between them was politically and legally irrelevant. This did not involve abolishing religion any more than eliminating the property qualification for voting involves abolishing property. Religious toleration is thus a coherent view. It involves distinguishing between two points of view, which Marx called “the point of view of man” and “the point of view of a citizen.” The point of view of man is relevant for making decisions about one’s own life. The point of view of citizen is the point of view one should adopt when making decisions in political life, such as about how to vote, or what kind of constitution to support. Putting this in the terms I have used above, the content of the principle of religious toleration is determined by appeal to the value, for any person, of living according to the tenets of his or her religion (whatever this religion may be). But a Jew, or any other religious person, follows his or her religion because of what he or she sees as its unique importance and value, not as an instance of the more general value of “living in accord with one’s religion.” Bauer thus sees the outlook of religious toleration as involving an analog of Moore’s Thesis: It involves thinking that living by the tenets of Judaism (or Roman Catholicism) is valuable only as an instance of the value of living in accord with one’s religion, which applies to everyone.

10 Karl Marx, “On the Jewish Question,” in Robert C. Tucker, ed., The Marx-Engels Reader, 2nd Edition (New York: W. W. Norton, 1978), pp. 26–52.

Ideas of the Good

117

This thesis, Bauer correctly thinks, is something no religious person should accept. Indeed, the concept of “religion,” which is central to the idea of religious toleration, is not important in the personal thinking of a religious person, since it lumps together views that such a person regards as important and valuable and other views that he or she sees as false and even pernicious. Marx points out that religious toleration does not involve an analog of Moore’s Thesis. What it involves, he said, is a distinction between two kinds of questions: political questions about what kinds of laws and policies to support, and personal questions about how to live one’s private life. Insofar as these are different questions, it may seem that answers to them do not conflict, in contrast to the case of Sidgwick’s Dualism. But even if answers to the questions Marx distinguishes do not directly conflict with one another, the separation between them— granting the authority of an impersonal, religiously neutral, form of reasoning in the political sphere, and restricting one’s religious values to private life—generates two kinds of tension. The first is what I will call empirical tension: one may wind up living and raising one’s children in an environment that is antithetical to one’s deepest values, because it is shaped in part by others who live according to different values. But even if one is lucky, and lives in a community in which most people share one’s values, one has to live with the normative tension, within oneself, of accepting that one’s deepest values do not apply to political institutions, but must be limited to one’s “private sphere.” In contrast to Sidgwick, who thought that Reason gave no answer about what to do when answers to his two questions conflicted, Marx had definite views about how people should view the tension between the questions he distinguished. He thought that under present circumstances we should regard what he called “political emancipation”—the establishment of religious toleration (and the elimination of the property qualification for voting)—as an important form of progress, even though accepting it involved these tensions. But he saw what I have called normative tension as a bad thing for a person—a form of alienation. So the thought we should strive to move beyond mere political emancipation to what he called human emancipation, which would eliminate this split within the person by abolishing both religion and property.

118

T. M. Scanlon

These views of Marx’s are (differing) answers to the question of acceptance—whether one should accept the principle of religious toleration. As such, they are claims about what an individual has most reason to want, all things considered, claims which incorporate, but are not exhausted by, claims about what is good for an individual. The problem that Marx sees, and his response to it, are an example of the phenomenon I mentioned earlier: if the way an individual’s good is represented in arguments about the content of principles of justice differs from the way in which his or her good is represented in his or her private thinking about what to do, this makes the acceptance of principles of justice more difficult to justify. In his later writings, John Rawls was concerned with this same tension. In his terms, it is a tension between an individual’s Comprehensive View (a view about the deepest questions of meaning and values of human life) and his or her sense of justice (the standards with respect to which he or she believes that social institutions and claims against them should be assessed). Rawls’ question is why a person should affirm a “liberal” sense of justice (including, for example, principles of religious toleration) when this is in tension with his or her comprehensive view in the two ways I have described. Unlike Marx, Rawls does not believe that this tension could be avoided by any feasible and morally defensible redesign of social institutions. In any free society, he believes, individuals will develop varying comprehensive views, which will give rise to this tension. This acceptance of empirical and normative tension is one thing that makes Rawls a liberal theorist. Rawls recognizes that comprehensive views can, and normally do, include not only provisions about how individuals should live their personal lives, but also provisions about how they should treat others, including those who may not share their views. In other words, comprehensive views commonly offer answers to questions of both of the kinds that Marx distinguished. This makes Marx’s proposed solution—of merely separating the two questions—less satisfactory. But it opens up the possibility of a different kind of resolution, through recognizing that the degree of normative tension will depend on the content of an individual’s comprehensive view, and may not be severe, or even arise at all, for views with a certain kind of content.

Ideas of the Good

119

Rawls argues, in Political Liberalism, that reasonable citizens who hold what he calls a reasonable comprehensive view will have good reason, consistent with and even provided by their own comprehensive views, to accept principles of justice including religious toleration.11 Such citizens may still be subject to what I called empirical tension, but they will not feel normative tension, because provisions of their own comprehensive view limits the applicability to political questions of some of its features. Rawls does not say that reasonable comprehensive views are the only ones that individuals have good reason to hold, or that the only comprehensive views held in a free society will be reasonable ones. Since his aim is only to show that a well-ordered society based on a liberal conception of justice could be stable, and reproduce itself over time, under favorable conditions, he does not need to claim either of these things. But in order to defend a positive answer to the question of acceptance for liberal principles of justice, one may need to make something very much like the former claim: that (at least under conditions relevant for us) individuals do not have good reason to hold views about what is good from their point of view which are “unreasonable” in Rawls’ sense. If they could have good reason for holding such views, then they could have good reason for rejecting liberal principles of justice, which conflict with them. So in order to claim that individuals have good reason to accept such principles we need to deny that they could have good reason to hold unreasonable comprehensive views. (I will return to this question later.)

11 It is a complex interpretive question exactly how to understand Rawls’ idea of a reasonable comprehensive view and the argument he bases on it. His initial description of the idea of a reasonable comprehensive view is mainly epistemic: such views (1) involve the exercise of theoretical reason to understand the world; (2) involve the use of practical reason to identify important values and balance these when they conflict; and (3) evolve over time in response to their adherents’ judgments of what they take to be good and sufficient reasons (Political Liberalism, p. 59). It does not seem to follow directly from these features that a reasonable comprehensive view gives its adherents reason to accept a liberal conception of justice. But Rawls goes on to say that a reasonable comprehensive view is one that can be held by a reasonable person. Reasonable persons, he says, will recognize “the burdens of judgment” and therefore recognize that there are many reasonable comprehensive views that it can be reasonable for individuals to hold. Thus he says that “Reasonable persons will think it unreasonable to use political power, should they possess it, to repress comprehensive views that are not unreasonable, though different from their own.” (p. 60). For helpful discussion of these points, see Samuel Freeman, Rawls (London: Routledge, 2007) pp. 345–51.

120

T. M. Scanlon

A similar internal tension is involved in my final example, from moral philosophy, Bernard Williams’ famous remark about “one thought too many.”12 Williams was discussing the views of Charles Fried, who in his book, Right and Wrong, considered a case in which a man in a burning building had to choose between saving his wife and saving a stranger. Fried was discussing this question within the context of what might be called a neo-Kantian moral theory. He said, plausibly, that the equality of persons that such a morality involved did not imply that it would be impermissible for the man to save his wife, rather than saving a stranger or deciding between them by flipping a coin. The reason, Fried said, was that morality must allow people to give special concern to their loved ones. Williams’ characteristically sharp retort was that this description of the matter gives the man “one thought too many.” As Harry Frankfurt later put it, “there is something fishy about the whole notion that when his wife is drowning, the man needs to rely upon some general rule from which a reason that justifies his decision to save her can be derived.”13 Williams’ objection may seem to miss the mark in a way that is similar to Bruno Bauer’s. In determining the content of moral principles, we need to take account of special reasons individuals have to help, and to favor, their family members and others close to them. Because individuals have these reasons, it would be reasonable for them to reject principles that required one always to give equal weight to the lives of loved ones and the lives of strangers. But, as Williams and Frankfurt correctly point out, reasons of this abstract sort—reasons to help and favor “close friends and family members”—are different from the special reasons individuals actually have for giving special attention to their particular friends and loved ones. Loving these people involves seeing helping them as overridingly important. Stating reasons in an abstract form, such as “special reasons for helping close friends and family members,” is a way of recognizing, and equating for moral purposes, the particular reasons of particular individuals to save particular people close to them. Like the category, religion, this abstract formulation is intended precisely to play this 12 Williams, “Persons, Character and Morality,” in Moral Luck (Cambridge: Cambridge University Press, 1981), p. 18. 13 Frankfurt, The Reasons of Love (Princeton: Princeton University Press, 2004), p. 36.

Ideas of the Good

121

equating role, for the purposes of moral justification. But this is not to suggest that any individual, performing an action recognized by this category, would generally be moved by such an abstract reason. To claim that an individual’s reason to save his or her spouse rather than a stranger is that one has reason to favor those who are close to one would be another analog of the Moorean Thesis. Following Marx, we should distinguish here between the personal question, “What do I have reason to do?” and the question of moral justification, “What is a person in my situation permitted to do, and why?” The abstract categories just mentioned are relevant only to this latter question. Their relevance is moral, not personal. Williams is not blind to this distinction. He is making the deeper claim that we do not have good reason, from our personal points of view, to give the categories of moral thinking the weight normally claimed for them. The idea of “one thought too many” may seem to suggest that the person in Fried’s example is supposed to accept an analog of Moorean Thesis: that his reasons for helping his friends and loved ones derive from a more abstract reason that everyone has, to help those to whom he or she is related in these ways. As I have said, this seems false and Fried is not asserting it. But Williams, like Marx in the later part of “On the Jewish Question,” holds not only that a person’s special reasons for aiding those near and dear are not based on the abstract reasons employed in moral thinking, but also that a person who takes these personal reasons to be limited by abstract moral justification—who is prepared to sacrifice his loved one and to set aside his “ground projects” when impersonal morality demands this—is lacking in genuine commitment to the things he supposedly loves.14 This is a version of what I called the normative tension: The cost of accepting the constraints of morality does not lie only in the sacrifices it might require on particular occasions—on occasions when we are not permitted to save our spouse because the cost in other lives is too great. 14 As Frankfurt (2004) puts it, Williams suggests “that the man already goes wrong if he thinks it is incumbent on him even to look for a principle from which he could infer that, in circumstances like those in which he finds himself, it is permissible to save one’s wife.” One should separate two questions here, whether the question of permissibility is a serious question, and whether it is what the man should first think about in an emergency situation. Williams seems to be making a (negative) claim about the former question as well as about the latter, and it is only the former question that I am interested in here.

122

T. M. Scanlon

Whether or not such occasions ever arise, accepting this as a possibility— accepting morality as a limitation of our commitment to our loved ones and our “ground projects”—is incompatible with whole-hearted versions of these commitments. This conclusion is a negative answer to the question of the acceptance of morality: Williams is claiming that we cannot give a positive answer to that question about a morality that involves this kind of alienation from our personal relations and personal projects.15

ANSWERING THE QUESTION OF ACCEPTANCE

I have so far been emphasizing parallels between the problems raised by Sidgwick, Marx, Rawls, and Williams. But there is an important difference between the forms that these problems take for Sidgwick and Williams on the one hand, and for Marx and Rawls on the other. Sidgwick and Williams interpret answers to their two questions as claims about what individuals in certain circumstances in fact have most reason to do. In Marx’s case, things are different. Bruno Bauer’s claim was that a person who sees his or her religion as uniquely correct and important cannot coherently endorse religious toleration. This is not a claim about what such a person has most reason to do, but only about the reasons such a person must see him or herself as having insofar as he or she has certain religious views. (Bauer may have thought that what such a person actually had most reason to do was to abandon his or her religious beliefs.) Marx’s response, similarly, was a proposal about how such attitudes can be coherent, not about whether a person had reason to hold these attitudes. (As Marx went on to say, “true emancipation” would involve abolishing religion.) Rawls’ concern in the writings I have been discussing was more complex, and shifts between his early writings and later ones. His concern throughout was with the stability of liberal institutions—with 15 Susan Wolf interprets Williams in this way, and offers a nuanced account of grounds for finding unconditional commitment to the authority of abstract moral justification unappealing in “ ‘One Thought Too Many’: Love, Morality, and the Ordering of Commitment” in Ulrike Heuer and Gerald Lang, eds., Luck, Value, and Commitment: Themes from the Ethics of Bernard Williams (New York: Oxford University Press, 2012), pp. 71–92.

Ideas of the Good

123

whether individuals who acquired a liberal sense of justice by growing up in a well-ordered liberal society would see themselves as having reason to affirm that sense of justice. In A Theory of Justice, his concern was with whether such individuals would have good reason to affirm the sense of justice that their society led them to form—whether affirming this sense of justice was compatible with their good. He based his affirmative answer to this question, the question of “the good of justice,” on several substantive claims about individual good, including the claim that this sense of justice was integral to the kind of relations with others that they have reason to want16 and the Kantian thesis that this sense of justice expresses their nature as moral beings.17 He later came to regard this answer as too dependent on controversial claims about the good. So in Political Liberalism what he aims to show is that citizens of a well-ordered society would see themselves as having reasons grounded in their particular comprehensive views, to affirm a liberal sense of justice, leaving aside the question of what comprehensive views individuals had good reason to adopt. In this paper I myself am, like Sidgwick and Williams, primarily concerned with the question of acceptance of morality and justice—of what reasons individuals have to take justice and moral requirements seriously. An answer to this question must take a stand on what individuals have reason to want and to do insofar as effects on their only lives are concerned, and what they have reason to want and to do, all things considered, not merely about what they should see themselves as having reason to if they hold certain other views.18 I have so far considered only one substantive account of the good for an individual, Sidgwick’s view that it is a matter simply of the amount of agreeable consciousness in a person’s life. Let me now sketch an alternative view. I believe that any plausible account of the good for an individual will be pluralistic, including factors of the following three kinds.

16

A Theory of Justice (Second Edition), pp. 499–500. A Theory of Justice (Second Edition), p. 501. 18 My answer to the question of acceptance is thus like Rawls’ account of the good of justice in A Theory of Justice in that it appeals to substantive claims about the good, but different in that it avoids appeal to the Kantian idea of expressing one’s moral nature. 17

124

T. M. Scanlon

Experiential factors, such as enjoyments, excitement, and the absence of states such as pain and fear. Valuable relationships, such as friendships, good relations with one’s family and with others, such as those with whom one engages in cooperative activities. Success in one’s main aims and projects, insofar as these are worth pursuing.

I do not have a systematic argument that things of these three kinds are good for a person and make his or her life better.19 But it seems to me clear that they do make a person’s life more choice worthy, and that any account of what is good for a person that did not include them would be flawed and incomplete. Moreover, it seems clear to me that these three forms of good are independent: none of them is fully reducible to any of the others: the goods of achievement and relationships are not, for example, explicable simply in terms of the experiences they involve or foster. Monistic experientialism does not seem to me at all plausible. Nor does it seem to me that there is likely to be a plausible monistic account of any other kind, that is to say, any other good not on this list, such that all of these things are good for a person only because they lead to or are required by this further good. What I want to do now is first to draw out from these assumptions some conclusions about the relation between what is good for an individual and what an individual has reason to want all things considered, and then to consider the implications of these conclusions for the question of the acceptance of morality and justice. As I said earlier, I reject the Moorean Thesis, according to which something is good for a person only if and because it is something that is good impersonally (a good thing to occur in the world) and occurs in that person’s life. But any conception of individual good that includes the three elements I have just listed will make what is good for a person depend, in more complex ways, on reasons that person has that are, at least in part, independent of effects on that person’s life. 19 For discussion see Derek Parfit, Reasons and Persons (Oxford: Oxford University Press, 1984), Appendix I; Joseph Raz, The Morality of Freedom (Oxford: Oxford University Press, 1986), Chapter 12; T. M. Scanlon, What We Owe to Each Other (Cambridge, MA: Harvard University Press, 1998), Chapter 3; Stephen Darwall, Welfare and Rational Care. Richard Kraut offers an account of individual good that is also pluralistic in its content, but unified within an overall idea of flourishing. See his What Is Good and Why, Chapter 3.

Ideas of the Good

125

This is most obviously true of the factors in my third category. Success in a pursuit makes a person’s life better only if it was something worth pursuing. Otherwise it is a case of wasted effort. As I said earlier, the reasons that make a pursuit worthwhile will in most cases have to do with factors other than effects on the life of the person in question, such as the effects on the lives of others (as in the case of searching for a cure for cancer) or the value of the skill or knowledge that one is striving to develop (as in intellectual and artistic pursuits). Something similar, but slightly different, is true of some goods in my second category, of valuable relationships. A relationship makes a person’s life better only if it is one worth having (as opposed, say, to relationships of servility and domination). It is not that some relationships are worth having because it is good, impartially, that such relationships occur in the world. But there are basic normative truths about which relationships are worth having and which are not. I believe that the best answer to the question of the acceptance of morality lies in this area.20 Individual morality, as I understand it is, most fundamentally, a matter of being concerned to act only in ways that are justifiable to others. This is something one has reason to do because of the kind of things other persons are. As creatures capable of understanding and responding to reasons, they are creatures we have reason to want to be related to on a basis that is justifiable to them. Since this is a valuable relationship, achieving it also makes our lives better, even though our reasons for wanting to achieve it derive in part from the kind of things they are, not simply from considerations of our individual good. The problems that I discussed arising for Marx, Rawls, and Williams had to do in part with the abstractness of the considerations that go into determining the content of morality and justice, as compared with the more concrete and specific reasons that properly guide individuals’ personal choices. It is plausible to think that this abstractness arises from the specific demands of moral justification—from the need to equate the interests of different individuals by, for example, treating different religions as equally important.

20 Here I sketch a view described more fully in What We Owe to Each Other, pp. 103–7, 147–88.

126

T. M. Scanlon

Like the moral views and conceptions of tolerance that I discussed above, this sketch of individual good deals with abstract categories: pursuits, relationships, etc. This is not only because it is just a sketch. Any general account of individual good will need to be filled in by “subjective factors,” because there are many worthwhile pursuits and many relationships worth having. What a given person will have reason to want insofar as his of her own life is concerned will be success in his or her particular aims, the quality of his or her actual relationships with particular people, and on what happens to these particular people. The appeal to abstract categories is thus not due simply to the demands of “the moral point of view” that treats people’s interests equally. It is, rather, a feature of any general account of individual good. Even Williams’ conception of individual good involves such categories, such as the idea of an individual’s “ground projects.” Moral argument and political justification also employ these abstract categories because they must be general, in order to provide justification applicable to all. These abstract categories are, as I have said, less important from an individual’s own point of view. This is not to say that they are never relevant: sometimes one acts from reasons of obligation deriving from such a relationship. There are occasions on which one might say, for example, “I had to help him. After all, he’s my brother.” Williams’ point, however, is that this remark is most likely to occur when one’s relation with one’s brother is in some way impaired. Ideally, one loves one’s brother. To love someone involves caring about them and seeing what happens to them as in itself providing one with reasons, not dependent on the objective frame of family or other obligations. These objective factors become relevant, however, when we are justifying claims about the content of morality. The facts relevant to such justification are in the first instance facts about what individuals in certain circumstances have reason to want, and to do. But not just any reason is relevant to determining what is owed to a person. What is relevant to determining what we owe to each other are reasons grounded in the way individuals in certain circumstances would be affected by given principles. Relations such as “friend,” “spouse,” or “family member” specify, in unavoidably general terms, important ways in which a person can be affected, and therefore play a role in the justification of moral principles. This works in several ways.

Ideas of the Good

127

Since there are many different relationships worth valuing, and different individuals to whom one may stand in these relationships, the “subjective” aspect of this component of individual good means that acceptable moral principles must require that individuals not be interfered with in their choices among these alternatives. And as we have seen, acceptable principles must also allow individuals to favor those to whom they stand in these relationships, in the ways that those relationships require. This does not mean that one’s friends and family members must be recognized as counting more, morally speaking, but only that one must be given space to give them preference. On their own, they count morally just the same as everyone else, quite apart from their relationship to one. Of course, we can imagine forms of relationship that involve holding that those to whom one stands in this relationship (members of one’s race or nation, for example) are of greater moral significance than other people. A defensible morality need not allow space to fulfill the requirements of these relationships, however, because the relationships that would demand this are not ones that we have reason to value. So individuals do not have good reason, all things considered, to want to fulfill their requirements. Elements in my third category (valuable pursuits, among which I will include the fulfillment of Comprehensive Views) play a similar role in shaping justifiable moral principles and principles of justice, with one important difference. Because of the variety of valuable pursuits and, I will assume, the variety of comprehensive views, worthy of endorsement, defensible principles must protect individuals’ ability to make their own choices among these. Defensible principles must allow individuals opportunities to pursue projects and values they have chosen, insofar as this is compatible with others having similar opportunities. But, and here is the difference I mentioned, one’s pursuits, even one’s ground projects and one’s comprehensive view, have no moral standing themselves. Unlike the individuals to whom one may stand in special relationships such as marriage or friendship, they make no moral demands on others on their own. Any moral claims regarding them are entirely dependent on the claims of those who hold them. This is what it means for a moral or political theory to be individualistic in the sense that I mentioned earlier in this lecture. The fact that pursuits and comprehensive views do not have moral significance on their own does not mean, however, that they do not have

128

T. M. Scanlon

any kind of significance. As Bruno Bauer pointed out, religions in particular typically maintain that their demands have significance and authority that is superior to those of other religions. And Bernard Williams held that individuals may rationally attach similarly unique importance to their “ground projects,” thus bringing them into conflict with the requirements of impersonal morality. This raises again the question of acceptance of principles of morality and justice: the question of why one should adhere to these principles when they conflict with other considerations. In this case the considerations in question are not considerations of self-interest but rather other things that individuals may have good reason to care about. An answer to this question of acceptance will be a claim about what individuals have most reason to want, all things considered, taking into account reasons to care about their relations of justifiability with others, and reasons to care about these other aims and projects. The most aggressive answer would be to hold that insofar as particular pursuits, comprehensive views, or other aims are incompatible with relations of justifiability to others they are not worthy of adherence. (This would be parallel to what I said earlier about similarly demanding forms of personal relationship.) This response is not, I should emphasize, itself a moral claim, but a claim of value, about the reasons supporting morality and those supporting other relationships and pursuits. This is the response that I find most plausible. It amounts, in Rawls’ terms, to saying that unreasonable comprehensive views are not worth holding. Rawls himself, of course, did not claim this as part of his political conception of justice, whatever may have been his own personal view. Perhaps there are other, more conciliatory responses that one should consider, but I do not at present see what they might be. Acknowledgments This paper was first presented as the Routledge Lecture at the University of Cambridge. I am grateful to members of the audience on that occasion for their comments, and to members of the audiences on subsequent occasions, including the 2014 Arizona Workshop in Normative Ethics. I am also grateful to Samuel Freeman, Frances Kamm, Christine Korsgaard, Daniel Star, to two anonymous reviewers for this volume, and especially to Derek Parfit for helpful comments and discussion.

6 Scrupulous Judgments JESSE S . SUMMERS AND WALTER SINNOTT - ARMSTRONG

THE ISSUE

Moral agents seem more praiseworthy when they care more about other people and are more careful about what they are morally required to do. We praise agents for paying close attention to moral requirements. Someone who is careful not to deceive others even on small matters seems to have the moral virtue of honesty to a high degree. Conversely, we condemn agents for being thoughtless in morally loaded situations. People who do not even try to remember what they promised to do lack the moral virtue of trustworthiness. Even when we disagree with someone’s particular moral judgments or concerns, there is something admirable about the fact that they are morally concerned. It seems better to try hard to figure out the morally right act, even when one figures incorrectly. It is surprising, then, that some people seem to pay too much attention to moral considerations. Here are two real cases of this phenomenon: Bridget: At a restaurant where Bridget worked as a waitress she had a recurring fear of accidentally poisoning her customers. Accordingly, she checked the containers of cleaning solvents stored in the kitchen cabinets to ensure she did not inadvertently powder the food she served with a lethal garnishing of chemicals. Despite the fact that her position never involved contact with the solvents, making her involvement in such a fatal faux pas a virtual impossibility, she was tormented by the idea and correspondingly checked the containers at almost every order. ( . . . ) Another behavior occurred upon the passing of ambulances. When hearing the sound of ambulance sirens she dropped whatever she was doing, crossed herself several times—sometimes a specific number of times—or looked to the sky. If driving, this often required pulling over to the side of the road either to cross or to pray with clasped hands for the

130

Jesse S. Summers and Walter Sinnott-Armstrong

ambulance passenger. If she was unable to perform these behaviors, she felt anxious and guilty, fearing that the condition of the afflicted passenger would worsen, or that the person would even die, and that she would then be morally responsible. In the same vein, she often had intrusive thoughts about her younger brother dying from illness or getting into a fatal car accident. Consequently, she would cross herself and pray ritualistically when the thoughts randomly arrived and in response to antecedents such as her brother catching a cold, or when he left the house with his friends on weekends. These fears sank roots deeply into her sense of self, with her fearing that if she failed to avert harm this meant she was a selfish, uncaring, and even evil person (Garcia, 2008). Benjamin: Thirty-three-year-old married man with three children. Following his marriage when 20, he became preoccupied with the thought that he had not had sufficient concentration during the wedding ceremony, rendering the marriage invalid. He went to see the rabbis about it, until a second ceremony was arranged. After this, he became preoccupied with his wife’s menstrual cleanliness. The thoughts filled his mind all day, although he tried to dismiss them. Although he was aware that this preoccupation was excessive, he would question his wife many times a day as to whether she had become menstrually unclean. If she dismissed his question, he would decide that she was not taking the matter seriously and he would remain tense. He consulted rabbis very frequently as to whether— given his doubts—he was permitted to be with his wife, and they always permitted it. Nevertheless, he avoided touching his wife whenever possible, and intercourse was very infrequent. He only agreed to intercourse because of the commandment ‘Be fruitful and multiply’, but on occasion his wife ‘forced’ intercourse on him. In addition, he took a long time to complete the portions of daily prayer that are considered to be especially important and requiring particular devotion, every such line adding an extra 10 minutes to his prayer (Greenberg, Witztum, & Pisante, 1987).

These two cases illustrate how morally-focused judgments are not always praiseworthy. Moral judgment and concern about morality can go too far. When people like Bridget and Benjamin go too far in this way often enough, they fit the diagnosis of scrupulosity, a form of ObsessiveCompulsive Disorder focused on morality or religion. Our goal in this paper is to use the traits that typify scrupulosity to understand the pathological way in which people with scrupulosity form moral judgments: they form their peculiar moral judgments in response to anxiety

Scrupulous Judgments

131

about wrongdoing. This pathological source of their moral judgments distinguishes scrupulous moral judgments from ordinary moral judgments. This distinction then illuminates the cases and the ways in which concern with morality is praiseworthy. Before we can learn these lessons, however, we need to clarify scrupulosity as a form of ObsessiveCompulsive Disorder (or OCD).

WHAT OBSESSIVE-COMPULSIVE DISORDER IS

Obsessive-Compulsive Disorder in general is an anxiety disorder in which a person has persistent, intrusive, and unjustified anxiety-evoking thoughts or “obsessions” that she resists or reduces by repeatedly performing some behavior or “compulsion.”1 For example, a patient might have recurring thoughts of a break-in at her house, and those obsessions might lead her to repeatedly and compulsively check the locks to doors of her house. But which thoughts count as obsessive and which behaviors count as compulsive? In order for obsessive thoughts to support a diagnosis of OCD, they must not be based solely on accurate assessments of real dangers—such as a higher-than-average likelihood of a break-in (as in a crime-ridden community) or to worse-than-average losses if a break-in were to occur (as with unusually valuable possessions).2 Nonetheless, the originating thought needn’t be totally inaccurate: there is some risk of a break-in, after all. (In some cases, of course, the intrusive thought may be entirely ungrounded.) Such thoughts might be exaggerated, but they needn’t be completely off-the-wall. Such thoughts are intrusive to the extent that they are unwelcome and inconsistent with many other beliefs of the person. Intrusive thoughts, so understood, may be present in as much as 90% of the population (Rachman & de Silva, 1978; Salkovskis & Harrison, 1984; Rassin

1 This and the following section are taken, with some revision, from the authors’ previous work on scrupulosity (Summers & Sinnott-Armstrong, forthcoming). 2 There are subtleties here: a person may have anxieties justified by accurate judgments (e.g., about the likelihood of a break-in), but the judgments may form in response to one’s having deliberately sought out data in order to legitimate prior anxiety (e.g., trying to prove to oneself that an apparently safe area is in fact dangerous), data that one otherwise would have no reason to suspect even exists. The judgments are then accurate, but the justification is still compatible with the diagnosis of OCD. This point will return below.

132

Jesse S. Summers and Walter Sinnott-Armstrong

& Muris, 2007). Most of the population reacts to such intrusive thoughts by checking the evidence, shifting attention to something else, trying to analyze why the thought occurred, or discussing the thought with other people (Wells & Davies, 1994). In contrast, those with OCD are more likely to react in two other ways: they worry about the thought or they punish themselves in some way for having the thought (Wells & Davies, 1994; Moore & Abramowitz, 2007). Why would anyone react in these harmful ways to an intrusive thought? Perhaps the person’s reaction is stronger because the thoughts seem to the person more serious, more worrisome, and therefore harder to ignore or dismiss. But why would they seem so serious? Three possibilities suggest themselves. (A) People with OCD might have thoughts that reflect or build on false beliefs or even delusions. In our example, they might believe that the risk of burglary is much higher than it really is and much higher than normal people believe it is. Or, they might believe that a burglary would cause much more harm than it really would or than normal people think it would. In short, they might overestimate the probability or the degree of harm. (B) The emotions of people with OCD might be stronger or more persistent than normal. In our example, they might have normal beliefs about probabilities and harms, but the perceived risk might cause more anxiety in them than in normal people. Belief in a 1% chance of a certain harm causes more anxiety in some people than in others, so maybe people with OCD feel extreme anxiety in the face of normal risks and normal beliefs about risks. (C) People with OCD might have less ability to control their mental states, perhaps because they cannot shift their attention to other things. In our example, they might have normal beliefs and normal initial anxiety reactions but lack the ability to overcome or control the same anxieties that normal people manage to overcome or control. Of course, OCD patients might display varying degrees of all three features interacting in complex ways at different times. How do such thoughts and anxieties lead to compulsive behaviors? When an obsessive thought of a break-in evokes a great deal of anxiety, the patient might discover that checking the lock temporarily reassures her and soothes the anxiety. Unfortunately, it reassures her only briefly: her underlying anxiety about the house’s security—or perhaps a more general anxiety—returns after, or even persists through, the checking. The thought’s repetition and the return of the anxiety leads to more

Scrupulous Judgments

133

compulsive behavior in a desperate attempt to soothe the anxiety. This circle is vicious. Compulsive actions brought about by obsessive thoughts are typically more than a response to a stimulus, especially as the condition develops over time. To complicate our lock-checking example, the patient might perform the compulsive lock checking as a way of staving off anxiety before any anxiety is felt (Salkovskis, 1999). When she walks near the front door, she checks its lock repeatedly just in case, though not in response to any felt anxiety about a break-in. Or, instead of just checking the lock, she may engage in other compulsions, such as strenuously avoiding vulnerable parts of the house, or seeking excessive reassurance from locksmiths or security companies that the house is secure from break-in. Or, perhaps to make the increasingly routine compulsion feel more reassuring, her compulsions may develop into rituals that are less closely responsive to the initial anxiety: she develops a particular way of checking the lock that she finds more reassuring. She checks it a certain number of times or while fixating on a certain neutralizing thought. Or, she develops the belief that the house will remain secure if she always leads with the right foot when stepping into the house or never touches the doorknob with her left hand. These more complex compulsions turn the checking, which initially responded (albeit excessively) to a specific anxiety about the vulnerability of the house, into a ritual that is more than just a temporary reassurance that the lock remains locked. The ritual takes on a life of its own, and sometimes it takes over the life of the patient.

WHAT SCRUPULOSITY IS

Scrupulosity is a form of OCD (Nelson et al., 2006; Tek & Ulug, 2001) with moral or religious obsessions and/or compulsions and the same cluster of features as other forms of OCD. Although scrupulosity is not a common form of OCD (Miller & Hedges, 2008), it is perhaps the earliest identified form. The earliest known cases in Judaism are in the 11th century and in Catholicism in the 16th century (cf. Mora, 1969; Greenberg et al., 1987; Greenberg & Huppert, 2010), and the earliest reference at all is in Plutarch (Plutarch, 1951: 373, 5–6, 9; cf. ).

134

Jesse S. Summers and Walter Sinnott-Armstrong

The English cleric Jeremy Taylor wrote in 1660 of those persons who “dare not eat for fear of gluttony; when they are married they are afraid to do their duty, for fear it be secretly an indulgence to the flesh ( . . . ) and yet they dare not omit it for fear they should be unjust.” A generation later Bishop John Moore referred to what we now call “scrupulosity” as “religious melancholy.” He describes the scrupulous as having a “fear, that what they do, is so defective and unfit to be presented unto God, that He will not accept it ( . . . ) [They experience] naughty, and sometimes Blasphemous Thoughts ( . . . ) [which] start in their Minds, while they are exercised in the Worship of God ( . . . ) [despite] all their endeavors to stifle and suppress them. ( . . . ) The more they struggle with them, the more they increase.” Interestingly, Moore noted of the scrupulous that “They are mostly good People ( . . . ) for bad men ( . . . ) rarely know anything of these kind of Thoughts” (Moore, 1692). Informally, scrupulosity is characterized as ‘‘seeing sin [or immorality] where there is none’’ or “focus[ing] on minor details of the person’s religion, to the exclusion of more important areas” (Nelson et al., 2006: 1072). In fact, the condition is more complicated and more closely connected to OCD than this characterization would suggest. The scrupulous patient has unwanted, anxiety-inducing intrusive thoughts that may be either secular (like Bridget’s thoughts of causing injury or harm) or religious (like Benjamin’s thought that his wife is “unclean” by Biblical standards, or like blasphemous or sacrilegious thoughts, doubts about dogma or about one’s sinfulness, or fears of damnation). The scrupulous patient responds to those obsessive thoughts with compulsions that are often religious in nature, such as excessive praying, disproportionate attention to religious minutiae, and repeated reassurance-seeking from religious authorities (Nelson et al., 2006; Abramowitz, 2008). To understand how scrupulosity is a form of OCD, consider how scrupulosity differs from extreme devotion to a religion (Taylor, 2002). Someone who is extremely devoted to a religion—a “fanatic,” to respect that word’s etymology—will care deeply about the features of her religion: practicing its rituals, studying its dogma, learning about the lives of its exemplars. She may be a religious leader, teacher, or maybe even a saint. A scrupulous patient, on the other hand, will have religious anxieties and religiously-flavored compulsions, but might not care about the features of that religion per se. For example, the extremely devoted

Scrupulous Judgments

135

Catholic may regularly pray the rosary in order to imitate his favorite saints or to carry out the penance prescribed by a priest during confession. In contrast, the scrupulous Catholic prays the rosary as a way of keeping bad thoughts or events from happening (Greenberg, 1984; Abramowitz, 2002; cf. Traig, 2004). One may be both extremely devoted to a religion and also scrupulous, just as one may be both extremely concerned with home security and also have a lock-checking form of OCD, but scrupulosity is more than simply extreme devotion to a religion. Nor, on the other hand, is scrupulosity just OCD as expressed in a religious context, since one could have a compulsion to check the lock on the door of a church or to line up the prayer books (cf. Tek & Ulug, 2001). Instead, scrupulosity both has a distinctively moral or religious presentation and characteristically displays a particular constellation of OCD features. Scrupulosity patients have many features that can be found in other cases of OCD. Scrupulosity shares with other forms of OCD obsessions and beliefs about responsibility for controlling intrusive thoughts. It also shares with them contamination fears and disgust sensitivity (especially towards sex and death), which a fuller discussion of OCD would consider (Olatunji et al., 2005). In addition to those common features, scrupulosity also has certain characteristic features: (a) moral perfectionism, (b) moral thought-action fusion, and (c) chronic moral doubt (Nelson et al., 2006; cf. Frost & Steketee, 2002; Olatunji et al., 2007). These traits are not necessary—the diagnostic categories that psychologists currently use rarely allow for necessary conditions—but we can still consider these traits to be typical or characteristic of scrupulosity patients. Perfectionism involves extremely stringent, rigid, and high moral or religious standards for oneself (and maybe also for others) and, correspondingly, a heightened sensitivity to anything that falls short of such standards. (Perfectionism in this sense is not to be confused with perfectionism as a theory of value or politics; cf. Hurka, 1993.) This kind of perfectionism—though understudied—seems to promote and maintain various psychopathologies (Shafran & Mansell, 2001; Egan, Wade, & Shafran, 2011). This is not surprising: having high standards and being sensitive to whether they are achieved can make one evervigilant and anxious about possible failures. Bridget had perfectionist standards regarding risk of harm to others. Restaurant workers should

136

Jesse S. Summers and Walter Sinnott-Armstrong

make sure that they do not poison customers, but Bridget took this far beyond any normal standards when she constantly checked the solvents containers. Perfectionism becomes even more extreme when mere thoughts about bad acts or consequences are judged to be just as morally bad as the acts or consequences themselves. This surprising characteristic of scrupulosity is called moral thought-action fusion (Berle & Starcevic, 2005; Shafran et al., 1996). There are in fact two forms of thoughtaction fusion: likelihood thought-action fusion and moral thoughtaction fusion. In both cases, one ascribes inordinate significance to her thoughts and other mental states. In likelihood thought-action fusion, the sufferer believes that her thoughts, merely by occurring, make some outcome more likely. A scrupulosity patient need not have likelihood thought-action fusion, so he might not believe that the thought itself makes the content of the thought more likely to occur (cf. Tolin et al., 2001; Rassin & Koster, 2003; Nelson et al., 2006; Berman et al., 2010; Berman et al., in press).3 In moral thought-action fusion, the scrupulous patient fails to distinguish doing something immoral from thinking about doing something immoral and, hence, treats thoughts as equivalent (or close) in moral status to actions. She sees having unacceptable thoughts, even if they are uncontrollable and intrusive thoughts, as (almost) as bad as having an immoral intention, making an immoral attempt, or even actually performing an immoral action (Abramowitz et al., 2002). The perceived wrongness of having the thought is therefore simply in entertaining its content, however unwillingly. As one scrupulous patient puts it (though the description suggests that she may have both types of thought-action fusion): I have trouble with wishing evil on people. When I was a girl, I wished my neighbor would get sick, and he did. I prayed hard for him, and he got better. One day I got upset with someone and wished her to have cancer. Now she has cancer, and I feel it is my fault. I pray for her a lot. Sometimes I wish that bad

A complication worth noting is that “belief ” is ambiguous between what is attributable based on one’s actions and what is attributable based on one’s speech: if I sincerely say that I believe that my mother’s life is safe regardless of how often I check the stove, yet I continue to check the stove repeatedly “just in case,” it’s unclear what exactly I believe. 3

Scrupulous Judgments

137

accidents would happen and that people would get hurt. I really don’t want these terrible things to happen, but these thoughts keep coming anyway. Am I committing sin in having all these evil thoughts? (Santa, 2007)

Although moral thought-action fusion is not as evident as perfectionism and chronic doubt in our examples, there are some hints even in those examples. Benjamin’s worries about the validity of his marriage ceremony given his lack of intention may suggest moral thought-action fusion. So does Bridget’s need to engage in rituals to prevent harm when she “had intrusive thoughts about her younger brother dying from illness or getting into a fatal car accident.” Her therapist at the time asserts that “underlying these rituals was the irrational belief that she could ritualistically prevent his death” (Garcia, 2008: 18), that having the thoughts without performing the rituals would therefore lead to harm. Perfectionism and thought-action fusion naturally lead to the third characteristic feature of scrupulosity: chronic doubt and intolerance of uncertainty about the moral or religious status of one’s acts and oneself. If one is antecedently anxious about one’s religious or moral life, one is on the lookout for slight or even imaginary failures, whose importance may be magnified. Bridget found uncertainty about the solvents intolerable because her doubts led to “a recurring fear of accidentally poisoning her customers.” All together, the scrupulous person sets high moral standards for her actions as well as her thoughts and then experiences uncertainty and doubt that she meets those standards adequately or often enough (Abramowitz et al., 2002; Ciarrocchi, 1995). Of course, these three characteristic features are not independent. Judging thoughts to be equivalent to actions can be seen as one kind of extreme standard that constitutes perfectionism. Perfectionism might produce chronic doubt, because it is more difficult to meet perfectionist standards. Thoughtaction fusion can also feed uncertainty and doubt when one begins to wonder whether one’s thought that someone might be harmed is really a wish that the person will be harmed. And uncertainty about which moral standards are correct can lead one to adopt extreme standards in order to make sure that one did not overlook any immorality. These interconnections show how the characteristic features of scrupulosity fuel one another and produce a coherent syndrome.

138

Jesse S. Summers and Walter Sinnott-Armstrong ABNORMALITY

How exactly do the three features that typify scrupulosity—perfectionism, moral thought-action fusion, and chronic doubt—differ from normal morality and religion? And how are they connected to normal morality and religion? One connection lies in shared content. Perfectionism in scrupulosity patients requires very high standards, but those standards are exaggerations of normal standards. Bridget is concerned about harming people. So are we. Benjamin is concerned about violating what he sees as moral rules. So are we. Non-religious people do not share Benjamin’s concern with religious rules, but a lot of normal religious people do. So the content is not what separates the moral judgments of those who have scrupulosity and those who do not. Thought-action fusion might seem to be an exception, because most normal moral judgments are about actions rather than mere thoughts. However, moral judgments of thoughts are not that rare. The Christian Bible, for example, emphasizes the need to keep one’s thoughts pure: “You have heard that it was said, ‘You shall not commit adultery’; but I say to you that everyone who looks at a woman with lust for her has already committed adultery with her in his heart.” (Matthew 5: 27–8). Consider also Kant’s opening to the Groundwork (1997), which signals similar concerns about the importance of thoughts: “Nothing can possibly be conceived in the world, or even out of it, which can be called good without qualification, except a good will.” That, even if its effects are thwarted, the good will, “like a jewel, ( . . . ) would still shine by its own light, as a thing which has its whole value in itself. Its usefulness or fruitfulness can neither add nor take away anything from this value.” Even if we ignore Kant exegesis or Christian theology, most people consider it immoral to think racist thoughts even without acting on them (cf. Baron, 2002). Many of us seem to think that there is something morally valuable about good thoughts and morally disvaluable about bad thoughts, regardless of their consequences. Thus, thoughtaction fusion shares its content with many normal moral judgments. Of course, moral judgments by scrupulosity sufferers still exaggerate common sense morality. But this exaggeration is made possible by features of common sense morality itself. To see this, just imagine that a hundred people are drowning in a lake and you save ten before you

Scrupulous Judgments

139

collapse from exhaustion. Most people would feel bad about not saving more. Those people share the perfectionist standard that we should help them all, not feeling content as long as it is obvious that things could be morally much better. Similarly, imagine that you are driving legally and carefully, but you still run over a child who darts in front of your car without warning. Most people would feel bad (“agent regret”) about killing the child, either because they could have driven even more slowly and carefully or just because they played some role—albeit blameless— in the death. These people share the perfectionist standard that we should minimize risk of killing (or harming) other people. Thus, even though moral judgments that characterize scrupulosity are not normal, they are still based in—or at least superficially similar to—judgments and reactions that are very normal.

CHARITY

Consider an example that is not an immediately obvious example of scrupulosity but that will relate to the existing moral literature. I should give some non-trivial amount to charity, but, although I realize I should not give so much that I am then a recipient of charity myself, that doesn’t tell me how much to give. May I spend some extra money on myself to enjoy regular relaxation? How about spending money to support art in my community? May I spend some extra money to eat for my own pleasure, or enough to avoid health problems, or should I suffer some health problems, perhaps even a shorter life, in order to give more overall to those even worse off ? Many of us fail to draw any principled line at all in cases of giving to charity. Instead, we give an amount that seems like enough. We acknowledge that we might be wrong about whether we’re giving enough, and we feel at a loss as to how exactly we would tell which amount is correct, but we are still not particularly motivated to try to determine the precisely correct amount. If we try to draw a line at least for the sake of consistency from year to year or from person to person, we likely draw it in an admittedly arbitrary place—the amount a friend gives or the amount we gave last year anchors what we give this year—or just to stick with an amount that “feels right.” We reconsider that line only when prompted to do so by some external change.

140

Jesse S. Summers and Walter Sinnott-Armstrong

Contrast how this situation looks to a person with scrupulosity who cares about giving to charity. (People with scrupulosity don’t care equally about all moral issues, and issues concerning charity are not common in the literature.) Such a person with scrupulosity would suffer from chronic doubt about whether any amount that she considers giving is precisely the right amount. If she is particularly concerned with perfection in her own actions, then she will be unable to draw any arbitrary line for purely pragmatic purposes. Her anxiety will not respond to reassurance by authorities or to her own thoughts that the amount of her donation is “reasonable,” so she will continue to look for a way to determine a non-arbitrary amount to give, a principled way of resolving her doubt, even when no principle is available. To add to her anxiety about the amount she gives, she may even worry that, if she fails to give the right amount with the right intentions, then her giving is unacceptable—perhaps as bad as giving nothing at all. To resolve this problem, then, the scrupulous would embrace an expansive requirement. They will see themselves as required to give as much as possible, well beyond what most people see as extreme. Only by giving as much as possible (perhaps even at the expense of her own health) could she be sure that the principled and non-arbitrary amount she is giving is enough. If, by giving more, she could reduce the harmfulness of a situation even by a miniscule amount, then she blames herself for doing less than everything possible. If her spending money on coffee instead of donating it to charity could play any role, however slight, in a child’s death from lack of adequate nutrition, then she is (by her own lights) entirely responsible for those lethal consequences. Believing that she must give away almost everything (and then giving it all away) becomes the only way for this person with scrupulosity to ease her doubts. As should be clear at this point, a person with scrupulosity is not obviously wrong in her moral conclusion about what to do here: some utilitarians will agree that one (ideally) should give away almost everything, and some (but not all) utilitarians add that one should feel guilty or ashamed for giving less (Sinnott-Armstrong, 2005). The example is meant to be one in which the scrupulous notice features that are actually morally relevant (the needs of others and the usefulness of charity), just as Bridget was concerned with potential poisoning and Benjamin was concerned with religious rules that many of the non-scrupulous in his

Scrupulous Judgments

141

community would have defended as morally relevant features (regardless of whether they are correct). Given that the scrupulous are responding to many of the right features of a situation, who can blame them for looking for ever more detailed moral principles, or for worrying that their actions may unknowingly violate some unnoticed principle? If anyone should be praised for having a good will, they should be—even if their actions are disruptive to their own lives and to the lives of (some) others. ANXIOUS MORALITY

So far we have suggested that at least some people with scrupulosity respond to morally relevant facts and to slippery slopes that also bother normal people. The core difference between these groups then lies in the anxiety that plagues people who suffer from scrupulosity, both the amount of anxiety and the ways in which they respond to the anxiety. Space prevents us from discussing in any detail what anxiety is. Anxiety, in its paradigm cases, is a feeling of extreme discomfort or fear in perceived anticipation of a potentially negative occurrence. But paradigm examples won’t help us understand the more interesting cases that chronic anxiety disorders reveal. For example, one may have anxiety that motivates action or belief but that one didn’t even recognize at the time: this is a “feeling” that one did not notice that she felt. Moreover, what may be an extreme and acute discomfort in one person may be a constant background state—a mood, even—in an anxious person and therefore not consciously noticed most of the time, just as we may not consciously notice the feeling of the clothes that we have been wearing all day. Here, we can only signal some of these difficulties, appeal to paradigm cases of anxiety to establish the sort of thing we have in mind, and note that it is characteristic of such a state that we are generally disposed to reduce anxiety when and to the extent possible. We will emphasize that anxiety can be related to one’s moral judgments in two different ways: anxiety may result from moral judgments, or anxiety may lead to moral judgments. Of course, in genuine cases of scrupulosity, these two likely co-occur and feed back on each other, but we will consider them separately for simplicity. Call the first type “judgment-driven scrupulosity.” In this type of scrupulosity, the initial moral judgments of the scrupulous may respond to the same features that ordinary moral judgments respond to, but, as a

142

Jesse S. Summers and Walter Sinnott-Armstrong

result of those judgments, they suffer from increased anxiety. Perhaps they doubt the correctness of their judgments, hesitating to implement the judgment they’ve formed in case it is wrong, in case there is a better option, or in case they do not have the purity of mind or intention necessary to make the action moral. Or, perhaps the decision feels weighty, or one feels responsible for all of the possible consequences of the action and is worried that she may not have anticipated some of those consequences. There are many ways in which one’s judgment may cause such increased anxiety. What is wrong with judgment-driven scrupulosity? It cannot be the initial judgment, since we stipulated in this type of scrupulosity that those initial judgments are formed in response to the same features to which ordinary moral judgments respond. There may be something wrong in their hesitancy to implement the judgment, or their doubts, even once implemented, that it was the right action, or in trying to summon up the purest possible intention before acting, but these may all be present even in ordinary moral judgments, particularly in hard cases. When faced with a morally difficult case, an ordinary moral reasoner may have difficulty making up her mind what to do, may hesitate before implementing it, may continue to wonder if she’s done the right thing or if her intention is really as pure as it should be. These reflect the recognition that one faces a morally important case, one not to be taken lightly. Notice, though, that judgment-driven scrupulosity is marked by such anxiety even when the cases are not morally important, when the judgments should be easy to make and implement. The scrupulous have an exaggerated sense of responsibility for their actions (Nelson, et al., 2006), thinking many of their own judgments and actions each to be very important, which explains why what should be easy cases are much more difficult. On its own, though, this doesn’t tell us whether the scrupulous are wrong to take seriously so many more of their actions than most people do. Perhaps we should all be more scrupulous, feeling more anxiety when making each of our moral decisions. Perhaps. But there are many reasons to doubt that judgment-driven anxiety is conducive to making good moral decisions and living a moral life. Broadly, the problem that anxiety introduces into moral judgments is that the anxiety narrows one’s attention to, in part, what will soothe that anxiety. But soothing one’s anxiety isn’t itself a moral goal, and

Scrupulous Judgments

143

soothing one’s anxiety is likely to lead one away from pursuing other moral goals.4 Here are just three ways in which the presence of anxiety can lead an agent astray, even if she has made her initial judgment in the absence of any anxiety: (1) Anxiety narrows one’s attention to a small, self-focused range of those things that will best soothe one’s anxiety and makes one fail to notice other morally relevant factors. Bridget, for example, ignores her customers’ and employer’s needs to have timely service in order to check on solvents, which is to soothe her own anxiety about whether she should serve the food. Even though her decisions to serve the food (and not to harm her customers, and not to irritate her employer, etc.) are in themselves reasonable, her subsequent actions show that she has focused excessively on the checking that will soothe her own anxiety to the exclusion of the other morally relevant factors. (2) Anxiety may lead one to focus on features other than those that are morally relevant. If one’s anxiety does not respond to morally relevant reassurance—to evidence, for example, that no harm is likely to occur— then one may try to reassure the anxiety in morally irrelevant ways instead. Remember the example of the lock-checker above: when relocking the door does not reassure the person that the door is locked, then locking it twice with the left hand before touching it twice with the right may be more reassuring. Similarly, when the scrupulous person judges that she may harm someone inadvertently, and the attendant feeling of anxiety doesn’t respond to the evidence that she is unlikely to do so, she may respond to features that are reassuring but morally irrelevant, like her belief that pulling over to cross herself as the ambulance goes by will prevent the harm, though this is irrelevant to preventing harm (even by Bridget’s own religion’s standards). (3) Anxiety makes moral judgments inflexible. Imagine a normal case: I worry that my attempt at a joke came off badly and that I actually insulted my friend. I’m anxious as a result, so I talk to my friend who

4 Our claims about anxiety should not be controversial, though we focus on anxiety’s generally pernicious effects here. It is true that anxiety also plays a valuable role in our lives, and some features of anxiety can, in the right circumstances, even have morally advantageous effects. Nonetheless, we doubt that anxiety itself is morally advantageous in general (cf. Kurth, 2015).

144

Jesse S. Summers and Walter Sinnott-Armstrong

tells me that he had realized it was a joke and thought nothing of it. I am relieved of my anxiety, and I revise my judgment that I may have done something wrong. But imagine the abnormal way this could have gone. My friend tells me he thought nothing of it, but I’m still anxious. I make sense of the anxiety: my friend was wrong to say he thought nothing of it; he was in fact offended, which even he may not realize. My judgment that I offended my friend becomes even more secure, sustained by the continued anxiety (Chekov, 1927). If this could be true in cases like this example, cases in which reassurance is possible (even if the person’s anxiety does not respond to the reassurance), it is all the more likely in cases in which reassurance is unlikely or impossible. If I worry that the bump I felt while driving could have been my hitting someone who then quickly stumbled off out of my sight, I’m very unlikely to reencounter the stranger to reassure myself. Or, if Benjamin judges that he failed to say his prayers with the appropriate devotion the first time, and this makes him anxious, his continuing anxiety indicates to him that he continues not to have the appropriate devotion even on the third or fourth repetition, so he judges it necessary to continue to repeat the prayer. His rabbi cannot tell Benjamin what was going on in his mind while praying, and god does not reassure him that his prayer was fine the first time. And notice that Bridget worries about powdering food with solvents in such small doses that it’s not possible to reassure herself by looking at the food itself, but only by checking if the levels of the solvents in their containers are changing over time. It may not be surprising that anxieties tend to focus on such cases where reassurance is difficult or impossible, since this makes sense of the scrupulous person’s continued anxiety. These three features of anxiety are going to be all the more pernicious in cases in which the anxiety does not result from one’s judgments but plays a more fundamental role in some cases of scrupulosity. Call this second type “anxiety-driven scrupulosity.” Instead of their moral judgments causing their anxiety, their anxiety causes their moral judgments. These cases start with anxiety that the scrupulous cannot make sense of, which they then rationalize post hoc by exaggerating their moral standards so that their standards will imply that they really are immoral and, hence, should feel such anxiety. Their extreme moral judgments thereby help them justify or make sense of their antecedent anxiety.

Scrupulous Judgments

145

Now, on its own, the fact that anxiety directs one to make judgments needn’t make the judgments themselves any less praiseworthy. Sometimes anxiety may point us to the morally-relevant features of a situation (Kurth, 2015). Indeed, some such piques of anxiety may correspond to what some call the “conscience.” Even when their judgments are responding to morally relevant features of a situation, though, the anxiety-driven scrupulous are responsive to those features because of the underlying anxiety that pushes them to make the judgment. And the presence of that anxiety, particularly the presence of anxiety that is not correctly reassured, may have a distorting effect, even if some particular judgments are not distorted. The scrupulous exaggerate, distort, or even form their moral judgments in at least the three ways suggested above in order to soothe their anxiety rather than simply in order to determine what really is morally right or best. They narrow their attention to those issues that seem relevant to the anxiety, ignoring other issues that may be more important morally. Instead, they focus on other features that are not really morally relevant. And their anxiety makes those judgments inflexible even after reassurance. In the case of anxiety-driven judgments, the anxiety may even prompt the person to form judgments that she would not otherwise have formed. All of this hardly seems conducive to making reliable moral judgments. Their moral judgments are formed in a vain attempt to make sense of anxiety that may not itself represent any morally relevant features of the world, so the judgments are epistemically suspect. They are also morally suspect, because they focus on the person’s anxiety instead of morally salient features of the world, such as the welfare of others. When the scrupulous are motivated to legitimate or soothe their own anxiety, even if they also have a desire to know and do what really is morally right, their high standards and demanding judgments are less praiseworthy from a moral point of view. They deserve to be pitied— maybe even condemned for the distorted emphasis they place on certain requirements—rather than praised. Return to our examples at the start. We usually see agents as morally better when they are more careful to do what is morally required, such as when they try their best never to deceive or break promises. However, that is largely because they are responding to values that we see as morally important. As a result, we praise them even when they are so careful that

146

Jesse S. Summers and Walter Sinnott-Armstrong

they hurt other people, such as when they refuse to lie even though the truth hurts, because there is something valuable and accurate (even if exaggerated) about their high standards. They are motivated only by a desire to act morally. In contrast, these grounds for praise may not apply to either form of scrupulosity. The scrupulous’ moral judgments appear to be directed at moral truth, but they respond primarily to underlying anxiety. Scrupulous persons worry persistently about doing or having done something wrong, and this shapes their moral judgments both in content and in whether and when they form such judgments at all. The problems that they find to legitimate their worries might be genuinely moral, and their guilt feelings might be genuine and genuinely motivating, but they are ultimately moved by the need to soothe their anxiety, not a concern to do what is moral, which distorts their judgments in many—though not all—cases. Our diagnosis, then, of the difference between moral judgments by the scrupulous (certainly of the anxiety-driven kind, and to a lesser degree also of the judgment-driven kind) and equally strict moral judgments by non-scrupulous people is that the scrupulous judgments are based on anxiety which has a distorting effect on the judgments and subsequent actions, whereas normal judgments are based on a motive to figure out what really is morally required. Notice that this diagnosis predicts that the scrupulous person’s doubts will be strongest regarding distant harms or harms that are not physically observable, since these are easy to legitimate and hard to reassure oneself about, so such harms will do a better job at rationalizing persistent and otherwise inexplicable anxiety.5 A scrupulous person who is worried about harms that aren’t physically observable or easily demonstrated to be false would feel no rational pressure to revise such worries. Therefore, he would believe that his persistent doubts—which he would feel regardless—are and continue to be legitimate. Purely religious or entirely mental harms are both indemonstrable and are often the scrupulous patient’s chief concern: if a marriage ceremony has to be undertaken with the proper intention or else god won’t accept it as valid, then how can Benjamin be convinced that the 5 It may also predict that anxiety will be exceptionally strong in cases of moral dilemmas. (Thanks to Charlie Kurth.) This may include cases in which all of one’s options are bad as well as those in which all of one’s options are of equally indeterminate value.

Scrupulous Judgments

147

ceremony was indeed said with the proper intention? The scrupulous often dismiss the reassurances of priests and rabbis, suggesting instead that the religious authority, just in virtue of offering such a reassurance, shows that he must not have understood just how inadequate the scrupulous prayer was (Miller & Hedges, 2008). Likewise, if the poisons that Bridget worries may land in the customers’ food are imperceptible and cause harm only after the customer has left, then the ordinary ways of reassuring herself that she hasn’t caused any harm won’t be effective. Therefore, the anxiety underlying the scrupulous person’s reasoning leads her to worry especially about these harms that are distant and difficult to detect. CONCLUDING THOUGHTS

As mentioned earlier, we do not suggest what does underlie correct moral motivation, only what does not. Moral motivation by anxiety appears to distort genuinely praiseworthy moral reasoning. More provocatively, this is sometimes so even if the conclusions reached are genuinely moral conclusions and even if the features that the person picks out to justify those conclusions are the morally relevant features that really do justify that conclusion. We might be happy that the person has picked out the morally relevant features of a situation and responds to them correctly, but what motivated her search for such features casts doubt on the reliability, and perhaps even on the praiseworthiness of her moral judgments and of herself. That on its own is enough to suggest that there are important differences between normal people and people with scrupulosity. This suggests two further conclusions that we will not develop here. First, even non-scrupulous people are sometimes motivated by anxiety. This normal motivation might not be as systematic as scrupulous anxiety, but its occasional presence still might affect our praise of non-scrupulous agents. When a person’s doubts motivate the person to find something to legitimate those doubts, and when what the person finds and acts on genuinely does legitimate those doubts, then there is a risk that the subsequent judgments will be distorted. However, the extent of the risk depends largely on whether the anxiety itself is generally responsive to the moral features of a situation. For the scrupulous, their anxiety is not easily reassured and is not generally responsive to the moral features of a

148

Jesse S. Summers and Walter Sinnott-Armstrong

situation, at least in the anxiety-driven form. But nothing we say here would rule out that anxiety may have a more constructive, if limited, role in ordinary moral lives, even if the risk for distortion is still present. Secondly, the way in which people with scrupulosity soothe their anxiety is by looking for specific, principled answers to moral or practical questions. This recalls Aristotle’s prescription that we should demand precision only as far as is appropriate for a subject (NE, Book I, 1094. b24). Morality seems not to admit of very precise prescriptions: we should be “reasonably” careful not to hurt others, give a “reasonable” amount to charity, and so on. One feature of scrupulous moral thoughts might be that they seek precision beyond what is appropriate for the subject, which is a mark that they are soothing anxiety but are not primarily interested in being moral. This in turn might make us question whether seeking precise moral answers in all practical cases is a pursuit demanded by morality or whether it is instead motivated by other attitudes that are, perhaps, less morally praiseworthy.

REFERENCES

Abramowitz, J. S. (2002). Treatment of scrupulous obsessions and compulsions using exposure and response prevention: A case report. Cognitive and Behavioral Practice, 8(1), 79–85. Abramowitz, J. S. (2008). Scrupulosity. In Clinical handbook of obsessive compulsive disorder and related conditions (pp. 156–72). Baltimore: Johns Hopkins University Press. Abramowitz, J. S., D Huppert, J., Cohen, A. B., Tolin, D. F., & Cahill, S. P. (2002). Religious obsessions and compulsions in a non-clinical sample: the Penn Inventory of Scrupulosity (PIOS). Behaviour Research and Therapy, 40(7), 825–38. Baron, M. (2002). Character, immorality, and punishment. In Rationality, Rules, and Ideals; Critical Essays on Bernard Gert’s Moral Theory with Reply, edited by W. Sinnott-Armstrong and R. Audi (Lanham, MD: Rowman and Littlefield). Berle, D., & Starcevic, V. (2005). Thought–action fusion: Review of the literature and future directions. Clinical Psychology Review, 25(3), 263–84. Berman, N. C., Abramowitz, J. S., Pardue, C. M., & Wheaton, M. G. (2010). The relationship between religion and thought–action fusion: Use of an in vivo paradigm. Behaviour research and therapy, 48(7), 670–4. Berman, N. C., Wheaton, M. G., & Abramowitz, J. S. (in press). Rigid rules of conduct and duty: Prediction of thought–action fusion. Journal of Cognitive Psychotherapy. Chekhov, A. (1927). Death of a government clerk (C. Garnet, Trans.). In S. Foote (Ed.), Anton Chekhov: Early Short Stories 1883–1883. New York: Modern Library. Ciarrocchi, J. W. (1995). The doubting disease: Help for scrupulosity and religious compulsions. Paulist Pr. Egan, S. J., Wade, T. D., & Shafran, R. (2011). Perfectionism as a transdiagnostic process: a clinical review. Clinical Psychology Review, 31(2), 203–12.

Scrupulous Judgments

149

Frost, R. O., & Steketee, G. (2002). Cognitive approaches to obsessions and compulsions: Theory, assessment, and treatment. Pergamon Press. Garcia, H. A. (2008). Targeting Catholic rituals as symptoms of Obsessive Compulsive Disorder: a Cognitive-Behavioral and Psychodynamic, Assimilative Integrationist approach. Pragmatic Case Studies in Psychotherapy; Vol 4, No 2 (2008). Greenberg, D. (1984). Are religious compulsions religious or compulsive: a phenomenological study. American journal of psychotherapy, 38(4), 524. Greenberg, D., & Huppert, J. (2010). Scrupulosity: A Unique Subtype of ObsessiveCompulsive Disorder. Current Psychiatry Reports, 12(4), 282–9. Greenberg, D., Witztum, E., & Pisante, J. (1987). Scrupulosity: Religious attitudes and clinical presentations. British Journal of Medical Psychology, 60(1), 29–37. Hurka, T. (1993). Perfectionism. Oxford: Oxford University Press. Kant, I. (1997). Groundwork of the Metaphysics of Morals (M. Gregor, Trans.). Cambridge: Cambridge University Press. Kurth, C. (2015). Moral Anxiety and Moral Agency. In Oxford Studies in Normative Ethics, vol 5. Oxford: Oxford University Press. Miller, C. H., & Hedges, D. W. (2008). Scrupulosity disorder: An overview and introductory analysis. Journal of Anxiety Disorders, 22(6), 1042–58. Moore, E. L., & Abramowitz, J. S. (2007). The cognitive mediation of thought-control strategies. Behaviour Research and Therapy, 45(8), 1949–55. Moore, John. (1692). Of Religious Melancholy: a sermon preach’d before the Queen at WhiteHall, March the 6th, 1692. In R. Hunter & I. Macalpine (Eds.), Three Hundred Years of Psychiatry, 1535–1860. London: Oxford University Press, 1963. Mora, G. (1969). The scrupulosity syndrome. International psychiatry clinics, 5(4), 163. Nelson, E. A., Abramowitz, J. S., Whiteside, S. P., & Deacon, B. J. (2006). Scrupulosity in patients with obsessive-compulsive disorder: relationship to clinical and cognitive phenomena. Journal of anxiety disorders, 20(8), 1071–86. Olatunji, B. O., Abramowitz, J. S., Williams, N. L., Connolly, K. M., & Lohr, J. M. (2007). Scrupulosity and obsessive-compulsive symptoms: Confirmatory factor analysis and validity of the Penn Inventory of Scrupulosity. Journal of anxiety disorders, 21(6), 771–87. Olatunji, B. O., Tolin, D. F., Huppert, J. D., & Lohr, J. M. (2005). The relation between fearfulness, disgust sensitivity and religious obsessions in a non-clinical sample. Personality and Individual Differences, 38(4), 891–902. Plutarch. (1951). Plutarch: Selected Lives and Essays (L. R. Loomis, Trans.). Roslyn, NY: Walter J. Black. Rachman, S., & de Silva, P. (1978). Abnormal and normal obsessions. Behaviour research and therapy, 16(4), 233–48. Rassin, E., & Koster, E. (2003). The correlation between thought-action fusion and religiosity in a normal sample. Behaviour Research and Therapy, 41(3), 361–8. Rassin, E., & Muris, P. (2007). Abnormal and normal obsessions: A reconsideration. Behaviour research and therapy, 45(5), 1065–70. Salkovskis, P. M. (1999). Understanding and treating Obsessive-Compulsive Disorder. Behavior Research and Therapy, 37(supplement), s29–52. Salkovskis, P. M., & Harrison, J. (1984). Abnormal and normal obsessions—a replication. Behaviour Research and Therapy, 22(5), 549–52. Santa, T. M. (2007). Understanding Scrupulosity: Questions, Helps, and Encouragement. Liguori, Missouri: Liguori/Triumph Publishers. Shafran, R., & Mansell, W. (2001). Perfectionism and psychopathology: a review of research and treatment. Clinical Psychology Review, 21(6), 879–906.

150

Jesse S. Summers and Walter Sinnott-Armstrong

Shafran, R., Thordarson, D. S., & Rachman, S. (1996). Thought-action fusion in obsessive compulsive disorder. Journal of Anxiety Disorders, 10(5), 379–91. Sinnott-Armstrong, W. (2005). You ought to be ashamed of yourself (when you violate an imperfect moral obligation), Philosophical Issues 15, Normativity, 193–208. Summers, J. S. & Sinnott-Armstrong, W. (forthcoming). Scrupulous Agents. Philosophical Psychology. Taylor, C. Z. (2002). Religious addiction: Obsession with spirituality. Pastoral Psychology, 50 (4), 291–315. Taylor, J. (1660). The rule of conscience. In D. Hunter & I. Macalpine (Eds.), Three Hundred Years of Psychiatry: 1535–1860. Cambridge: Cambridge University Press, 1963. Tek, C., & Ulug, B. (2001). Religiosity and religious obsessions in obsessive-compulsive disorder. Psychiatry Research, 104(2), 99–108. Tolin, D. F., Abramowitz, J. S., Kozak, M. J., & Foa, E. B. (2001). Fixity of belief, perceptual aberration, and magical ideation in obsessive–compulsive disorder. Journal of Anxiety Disorders, 15(6), 501–10. Traig, J. (2004). Devil in the details: Scenes from an obsessive girlhood. Little, Brown. Wells, A., & Davies, M. I. (1994). The Thought Control Questionnaire: A measure of individual differences in the control of unwanted thoughts. Behaviour Research and Therapy, 32(8), 871–8.

7 Obligations of Gratitude and Correlative Rights TONY MANELA

When one person performs an act of benevolence for another, the latter owes the former gratitude. I take it that gratitude, as the proper response to benevolence, includes certain behaviors—such as returning a favor— which aim at advancing the wellbeing of the benefactor.1 Some of these grateful behaviors are referred to as duties or obligations2 of gratitude; and about such obligations of gratitude, three platitudes are widely accepted: 1) The Obligation Platitude: gratitude entails certain strict obligations— things a beneficiary must do; 2) The Directionality Platitude: these obligations are owed to the benefactor; and 3) The No-Right Platitude: the benefactor does not have a right to the fulfillment of these obligations. These three platitudes present a tension. A longstanding tradition in law and moral philosophy holds that directed obligations always correlate to rights held by the person to whom the obligation is due.3 My strict 1 Roslyn Weiss denies this, arguing that gratitude is merely a feeling, something like a wellwishing for a benefactor (Roslyn Weiss, “The Moral and Social Dimensions of Gratitude,” The Journal of Southern Philosophy XXIII, no. 4 (1985): 491–501). But as others as far back as Aristotle have noted, one can have feelings that fail to motivate; and a beneficiary who feels goodwill for his benefactor but is unmoved to help her (say, to acquiesce to her request for a reasonable favor) falls short of gratitude. (Aristotle, “Nicomachean Ethics,” in The Complete Works of Aristotle, ed. Jonathan Barnes (Princeton: Princeton University Press, 1995), 1166b30–67a20.) I will proceed, then, on the premise that gratitude calls for certain behaviors. 2 I will treat the terms “duty” and “obligation” as synonymous. 3 The origins of this tradition are often attributed to Wesley Hohfeld, Fundamental Legal Conceptions (New Haven: Yale University Press, 1923). In reality, references to the correlativity of obligations and rights antedate Hohfeld’s paper by nearly two centuries. See, for instance,

152

Tony Manela

obligations to you to keep a promise, to tell the truth, or to refrain from harming you all correlate to rights you hold against me. If the three platitudes are true, then certain obligations of gratitude appear to be exceptions to the correlativity thesis, at least as it is applied to moral rights and obligations. My goal in this paper is to investigate the platitudes about gratitude, and the correlativity thesis, in order to dispel this tension. I will proceed as follows. In §1, I defend the claim that there are indeed certain strict obligations of gratitude—things a beneficiary morally must do, that admit of no latitude. In §2, I argue that these obligations are directed from the beneficiary to the benefactor. In §3, I argue that while benefactors do not have standing to demand gratitude, they do have special standing to remonstrate with beneficiaries in its absence. This special standing amounts to what I call an imperfect right. After defending all three platitudes, I argue in §4 for a modified version of the correlativity thesis: moral obligations entail correlative rights, but those rights may be perfect or imperfect. I conclude by discussing several ramifications this modification of the correlativity thesis has for other debates in moral and political philosophy.

1. THE OBLIGATION PLATITUDE

I take it as a given that beneficiaries morally ought to do what gratitude calls for—and that failure to perform at least certain acts of gratitude when gratitude is called for constitutes a moral shortcoming. The distinctively moral register in which we appraise or evaluate ingratitude leaves little doubt of this. Hume, for instance, wrote that “Of all crimes that human creatures are capable of committing, the most horrid and

Samuel Pufendorf, Two Books of the Elements of Universal Jurisprudence (Indianapolis: Liberty Fund, 2009): Definitions VIII.2, XII.3; Thomas Reid, “Of Morals,” in Essays on the Active Powers of Man (Cambridge: Cambridge University Press, 2011): chapter III; Thomas Rutherforth, Institutes of Natural Law: Being the Substance of a Course of Lectures on Grotius De Jure Belli Et Pacis (Baltimore: W. and J. Neal, 1832): II.V; William Paley, The Principles of Moral and Political Philosophy (Indianapolis: Liberty Fund, 2003): II.9; Johann Gottlieb Heineccius, A Methodical System of Universal Law: Or, the Laws of Nature and Nations, with Supplements and a Discourse by George Turnbull (Indianapolis: Liberty Fund, 2008): I.I.7. In this paper, I will be concerned primarily with the one-way entailment claim that directed obligations imply rights, and not the wider thesis that directed obligations and rights entail each other.

Obligations of Gratitude and Rights

153

unnatural is ingratitude ( . . . )”.4 Kant wrote that ingratitude was (along with envy and malice) “the essence of vileness and wickedness”.5 And Seneca ranked ingrates below thieves, rapists and adulterers.6 While philosophers agree that ingratitude warrants negative moral appraisal, however, they disagree about whether grateful behavior is morally obligatory, the way promise-keeping is. Skeptics in this debate note that behavior might merit moral condemnation without constituting the breach of a moral obligation.7 For example, a single rider on a crowded train morally ought to yield her seat at the request of a couple who wish to sit together, and to fail to do so is morally lousy; but the lone rider has no obligation to do so.8 Generally, the obligatory is distinguished from what merely ought to be done in that the obligatory is binding. It must be done. It admits of no latitude. Failure to do what is obligatory is not just bad or lousy, but prima facie wrong. Given that failure to perform some act (ç) is morally condemnable, I will take it that whether someone has a moral obligation to ç can be determined by considering the way the reason to ç presents itself to a conscientious agent in his practical deliberation.9 In contrast to other reasons (even other moral ones), the conscientious obligee should grant obligations a high “deliberative priority” vis-à-vis other considerations in deliberation.10 In particular, moral obligations should have a “peremptory

4 David Hume, A Treatise of Human Nature, ed. David Fate Norton (Oxford: Clarendon, 2007), III.1.1. 5 Immanuel Kant, Lectures on Ethics, trans. Louis Infield (Cambridge: Hackett, 1979), 218. 6 Lucius Annaeus Seneca, On Benefits, trans. Miriam Griffin and Brad Inwood (Chicago: University of Chicago Press, 2010), I.10.4. 7 Christopher Heath Wellman, “Gratitude as a Virtue,” Pacific Philosophical Quarterly 80 (1999). Other “obligation skeptics” about gratitude include A. D. M. Walker, “Gratefulness and Gratitude,” Proceedings of the Aristotelian Society 81 (1980–81): 52; Joel Feinberg, “Duties, Rights, and Claims,” American Philosophical Quarterly 3, no. 2 (1966); Bernard Williams, Ethics and the Limits of Philosophy (Cambridge, MA: Harvard University Press, 1985), 179. 8 This example is inspired by Julia Driver, “The Suberogatory,” Australasian Journal of Philosophy 70, no. 3 (1992): 286–95. 9 In looking to the deliberative “behavior” of the conscientious agent as one factor that helps distinguish the morally obligatory, I follow, inter alia, Joseph Raz, Practical Reason and Norms (Oxford: Oxford University Press, 1999); David Owens, Shaping the Normative Landscape (Oxford: Oxford University Press, 2012); Thomas Pink, “Moral Obligation,” Royal Institute of Philosophy Supplement 54 (2004): 159–85. 10 Williams, Ethics and the Limits of Philosophy, 183. Williams did not think that feature by itself was sufficient to show that some deliberating agent had a moral obligation.

154

Tony Manela

character”; they are, in the words of R. Jay Wallace and Samuel Scheffler, “presumptively decisive” in practical deliberations. The fact that one has made a promise to ç, for instance, should generally, for the conscientious agent at least, settle by itself the question of whether to ç. I will follow several philosophers in fleshing out this peremptory character of obligations in terms of “exclusionary reasons”.11 As Joseph Raz puts it, an obligation to ç is a first-order reason in favor of ç-ing, combined with a second-order exclusionary reason that militates against acting on various other first-order reasons that count against ç-ing.12 For the conscientious promisor, for instance, the judgment that he has an obligation to fulfill a promise will not only incline him to do what he promised; he will also be disposed not to act on certain other reasons that might tempt him to do otherwise. He may find it inconvenient to do what he promised; but far from weighing that inconvenience against his first-order reasons for doing what he promised, he will refuse to act on the basis of this inconvenience. This, together with the fact that promisebreaking warrants a distinctively moral negative appraisal, makes it the case that promise-keeping is a moral obligation.13 Turning to gratitude, we can now ask whether any grateful behaviors are obligatory in this sense. In addressing this question, philosophers have sometimes failed to cast the net widely enough over situations in which gratitude might be called for. Christopher Wellman, for instance, takes as paradigm cases of acts of gratitude the giving of thank-you gifts. Selma, in a genuine act of benevolence, buys Marge a lovely antique clock. Marge, several months later, has an opportunity to buy a quilt she knows Selma will love. Wellman claims that while Marge ought to buy the quilt for Selma, Marge has no strict obligation to do so. I think Wellman is right about this. But it would be too hasty to infer, as Wellman does, there are no strict obligations of gratitude.14 Wellman does not consider acts of gratitude that seem more paradigmatically obligatory—for instance, acts that would not just benefit a benefactor, but save her from suffering. Consider what I call grateful

11

Raz, Practical Reason and Norms; Owens, Shaping the Normative Landscape. Raz, Practical Reason and Norms, 178–82. 13 As opposed to a requirement in some other normative domain—like the aesthetic, the epistemic or the prudential. 14 Wellman, “Gratitude as a Virtue.” 12

Obligations of Gratitude and Rights

155

nonmaleficence: abstaining from harming a benefactor. This class of grateful conduct is largely ignored in the literature. One exception is A.D.M. Walker, who notes that a beneficiary may have obligations to refrain from harming benefactors, but only in virtue of his obligation to refrain from harming people generally. It would thus be improper, Walker claims, to call such obligations “obligations of gratitude.”15 But there are situations when it is permissible (if not morally ideal) to harm someone; and there are situations when it might be permissible to harm a stranger, but impermissible to harm a benefactor. Consider a case of fair business competition. Yancy16 is a businessman contemplating opening a new store in a certain area. He has both selfish reasons (he wants more money to buy a new Maserati) and altruistic reasons (he wants to give a managerial position to his dedicated employee Peter) for making the move. He is also aware that if he opens a store in the location he is considering, he will drive out of business a competitor who owns a small store nearby. The move seems, if not saintly, at least permissible. Now imagine Yancy discovers that the owner of the nearby store is Rachel—a woman who, in a genuine act of benevolence, saved his life ten years earlier. I want to claim that this detail makes the move impermissible. To open the store would be wrong. The fact that the move would harm his benefactor should change the way Yancy deliberates about what he ought to do. Insofar as Yancy is a conscientious beneficiary, we would expect him not simply to weigh Rachel’s wellbeing against the joy he would get from his new Maserati; we would expect him no longer to consider the joy of the new Maserati at all in making his decision. It may pain him to forgo such benefits, but we would expect him to act against his preferences here. Yancy’s discovery that Rachel is his competitor should also change the way he deliberates about his other moral obligations. Let’s say Yancy had promised his employee Peter a managerial position at the new location. It seems plausible to say that before Yancy knew the competitor was Rachel, his

15 Walker, “Gratefulness and Gratitude.” Walker seems to reverse his stance on this in a later article: “Political Obligation and the Argument from Gratitude,” Philosophy & Public Affairs 17, no. 3 (1988): 203–4. 16 For the sake of clarity, my examples in this paper will use paradigmatically masculine names beginning with the letter Y to refer to a beneficiary, and paradigmatically feminine names beginning with the letter R to refer to a benefactor.

156

Tony Manela

promise to Peter should trump his concern for the nameless competitor. But when Yancy discovers it would be Rachel he would put out of business, we would expect his deliberation to change. It seems plausible that his promise to Peter no longer excludes from his deliberation the fact that his competitor will suffer. In fact, depending on the particulars of the case, Yancy might now decide that refraining from opening the store is more important than keeping his promise to Peter.17 It might be objected here that a truly grateful beneficiary would not find himself deliberating about what to do, as Yancy does; nor should he have to muster any effort of will to do it. Gratitude may indeed call for certain behaviors, but as Christopher Wellman and Paul Camenisch claim, acts of gratitude should be done spontaneously, from natural inclination.18 But as my example shows, sometimes what gratitude calls for is burdensome enough to conflict with other legitimate desires and commitments a beneficiary might have. And since even the most grateful and virtuous of beneficiaries will have other commitments (even moral ones), we would expect even the most grateful beneficiary sometimes to have to muster an effort of will, to go against his inclinations, to do what gratitude calls for. His acts of gratitude are no less genuine for that.19 When we consider certain behaviors called for as part of the grateful response, we find good reason to believe that some such called-for behaviors are strict obligations: their absence warrants negative moral appraisal, and a conscientious beneficiary’s reasons for performing them should take a certain high priority in his practical deliberation. I turn now to the Directionality Platitude—the claim that these obligations are owed to the benefactor.

17 This is assuming, of course, that Yancy was unable to both respect his promise to Peter and spare Rachel (and other innocent parties) from substantial harm. 18 Wellman, “Gratitude as a Virtue,” 287; Paul F. Camenisch, “Gift and Gratitude in Ethics,” The Journal of Religious Ethics 9, no. 1 (1981): 4. 19 I’ll note parenthetically that this tells against conceptualizing gratitude, the called for response to a genuine act of benevolence, as an emotion. While emotions may indeed entail first-order inclinations to act in certain ways, a fully grateful beneficiary should also have additional second-order reasons that come into play when such first-order motivations are overwhelmed. A beneficiary who merely has the emotion of gratitude, without second-order reasons, is not showing fullest gratitude, but “fair-weather” gratitude.

Obligations of Gratitude and Rights

157

2. THE DIRECTIONALITY PLATITUDE

Many find it obvious that grateful behaviors, like grateful nonmaleficence, are obligations to a benefactor.20 While grateful behaviors certainly concern a benefactor, however, a skeptic could highlight the possibility that an obligatory action might concern some party without being owed to that party. Kant, for instance, claimed that we have obligations concerning animals, but not to them; and perhaps obligations of gratitude, if they exist, are the same sort of beast. If obligations of gratitude were directionless, owed to no one in particular, then gratitude would not present a counterexample to the correlativity thesis. A little reflection, however, shows that certain obligations of gratitude, like obligations of nonmaleficence, are owed to benefactors by the lights of the two most prominent theories of “directionality”: the interest theory and the will theory. According to interest theorists, obligations are owed to the person whose interests justify holding the obligee to be under an obligation.21 On this view, my obligation to keep the promise I made you is owed to you because your interest, qua promisee, in having the promise kept justifies my being held to have such a duty. On this conception of directionality, duties of grateful nonmaleficence would be owed to the benefactor, since it is the benefactor’s interests that ground a beneficiary’s duties.22 Interest theorists should be inclined to grant that Yancy owes it to Rachel not to open the store. See, for instance, Wellman, “Gratitude as a Virtue,” 289. Interest theorists include John Stuart Mill, Joseph Raz, and Matthew Kramer. Many interest theorists, like Raz, take themselves to be putting forward accounts of rights, rather than accounts of the direction of a duty; but I will take it that their theories can be interpreted as accounts of how to determine the “counterparty” of a duty—the person to whom it is owed. In this, I follow (among others) Gopal Sreenivasan, “Duties and Their Direction,” Ethics 120, no. 3 (2010): 465–94. 22 It is an interesting question whether all grateful behavior (and not just grateful nonmaleficence) is grounded in the interests of the benefactor. Certain gratitude-expressing behaviors, like doing a favor for a stranger (sometimes called “paying it forward”) when one receives an anonymous benefaction, or taking care of a deceased benefactor’s grave, don’t seem to fit this mold. On reflection, however, such behaviors could be described as advancing the ulterior (or indirect) interests of such benefactors. It seems plausible to me that a benefactor might have an interest in having her grave taken care of after she dies, or in seeing her beneficiary benefit strangers the way she benefited him. Ultimately, all I am claiming in this section is that certain obligations of gratitude, like Yancy’s obligation of grateful nonmaleficence, are rooted in the interests of the benefactor. This will be enough to show that there are strict directed moral obligations that do not correspond to moral rights, and, therefore, to show that there is a tension between certain obligations of gratitude and the correlativity thesis. 20 21

158

Tony Manela

Will theorists, by contrast, hold that my duties are duties to you insofar as you hold some degree of normative control over the force of those duties in my deliberation.23 For a will theorist, if I make a promise to you, my obligation to keep my promise is owed to you because you are the one who can insist that I fulfill it, release me from my promissory obligation, set the terms of compensation if I do not live up to it, or forgive me altogether for failing in it. It seems that by this conception of directionality, obligations of grateful nonmaleficence, like Yancy’s, are indeed obligations to the benefactor. Imagine Rachel discovers that Yancy has decided, after realizing she was his benefactor, not to open the new location. She writes to him, and gives him her blessing for opening the location anyway. “I realize this will set my interests back,” she tells him, “but I don’t want you to feel constrained on my account. You’ve worked hard to build your business, and I know you’ve promised your employee Peter a managerial position.” This ought to change the way Yancy deliberates about what to do. He should still be averse to harming Rachel, of course; but his promise to Peter may now regain priority in his deliberation. And this suggests that Rachel has some power to release Yancy (or at least reduce the weight) of the reason her saving Yancy’s life gives him. Rachel could also change the way Yancy ought to deliberate by issuing a request to him. She might write to Yancy and offer her blessing for opening the new location, but request a managerial position at the new location, so she doesn’t suffer terribly when her store closes. While requests do not necessarily create new obligations, they do (paradigmatically) aim at providing new reasons for doing something.24 It seems like Rachel’s request here might provide Yancy with something like a peremptory reason. Yancy might now find himself willing to open the store, but still unwilling to keep his promise to Peter, insofar as that would conflict with Rachel’s request, which he might see as taking precedence. We can see this as analogous to an obligor’s setting the terms of compensation for the breach of an obligation.

23

Will theorists include H. L. A. Hart, Carl Wellman, and Hillel Steiner. Indeed, this is what distinguishes requesting from informing or communicating a wish. See Geoffrey Cupit, “How Requests (and Promises) Create Obligations,” Philosophical Quarterly 44, no. 177 (1994): 446. 24

Obligations of Gratitude and Rights

159

Lastly, as I suggested above, Rachel has the power to forgive (or withhold forgiveness from) Yancy. Let’s say Yancy realizes it is Rachel he will drive out of business, but opens the new location anyway, because he just cannot stop thinking about his new Maserati. Shortly after opening the store, he feels ashamed of himself, seeks Rachel out and apologizes to her. While she may not demand an apology, it would seem false for Rachel to say, when Yancy presents his apology, that he had nothing to apologize for—that he did nothing wrong. It wouldn’t be presumptuous for Rachel to tell Yancy that she forgives him. And when she does, Yancy no longer needs do anything to make it up to her (though he may, of course, be called on to perform other acts of gratitude in the future).25 There are, then, certain strict obligations of gratitude, which are owed to the benefactor by the lights of two prominent theories of directionality: it is the benefactor’s interests that justify the beneficiary’s peremptory reasons; and the benefactor has special normative control over the beneficiary’s reasons for abstaining from harming her.26 I turn now to the No-Right Platitude—the claim that nobody, not even benefactors, has a right to the fulfillment of such obligations.

3. THE NO-RIGHT PLATITUDE

Some philosophers understand the term “right” in such a way that to have a right is simply to have a duty owed to you. This is the way Wesley 25 Again, it is an interesting question whether all grateful behaviors are such that a beneficiary’s reasons for performing them are under the normative control of benefactors. The same apparently problematic cases I mentioned in fn. 22 seem to tell against this possibility. Still, however, on reflection, it seems possible that a deceased benefactor might exercise normative power over a living beneficiary from beyond the grave, as it were, through some living executor to whom she transferred that power, or via the reading and execution of a will she left behind. In that will, she might, for instance, ask her beneficiary to dedicate the time and energy he would have been inclined to spend visiting and maintaining her grave to volunteering for a local soup kitchen. Such an act of will, it seems to me, should change the way the beneficiary ought to prioritize these activities in his practical deliberation. I am grateful to an anonymous referee for pressing me to engage this example. 26 It is worth noting that there are other accounts of directionality in the literature—such as Gopal Sreenivasan, “A Hybrid Theory of Claim-Rights” in Oxford Journal of Legal Studies 25, no. 2 (2005): 257–74; Leif Wenar, “The Nature of Claim-Rights,” Ethics 123, no. 2 (2013): 202–29. Though I do not have space to address them here, I suspect that each of these theories—and, indeed, any plausible theory of directionality—will allow that beneficiaries’ obligations of gratitude are owed to benefactors.

160

Tony Manela

Hohfeld proposed we use the term “claim-right”. For those who understand “rights” in this way, the falsity of the No-Right Platitude should follow with necessity from my claims so far. Yet intuitively, the possibility of a right to gratitude strikes us as absurd. And no philosopher writing about gratitude—even those who think there are strict directed obligations of gratitude—has asserted that benefactors have a right to a grateful response from a beneficiary.27 This suggests some understanding of “right” other than the one proposed by Hohfeld. Presumably Hohfeld had this understanding in mind when he decided to use the term “claimright” as the correlative position of a duty. What does it mean to have a right, in this pre-Hohfeldian sense? Remarks from Hohfeld’s famous essay suggest that before he defined the term as the correlate of a directed duty, he understood “claim-right” to mean something like entitlement to enforce a claim.28 Hohfeld, of course, was concerned with rights in the legal domain. In the moral realm, though, we can still make sense of a right as a standing to enforce— roughly, a permission to use or threaten force, or otherwise impermissible sanctions, to induce compliance. If you refuse to return property I have leant you, for instance, I might threaten to refuse to return something I borrowed from you, or threaten to violate your own property by entering your home to take mine back.29 If you break a promise to me, I might threaten to renege on a promise I made to you. Such measures, of course, may not always be justified—especially in cases where what is at stake in having a right respected is trivial. A pre-Hohfeldian right, though, will always imply the permissibility of a minimum level of enforcement, in the form of demanding.30 A full account of how demands amount to enforcement is beyond the scope of this paper; but I take it that one thing 27 Not all philosophers of gratitude deny that there can be rights to gratitude. At least one philosopher remains agnostic on the question. See Terrance McConnell, Gratitude (Philadelphia, PA: Temple University Press, 1993), 243 n. 63. 28 See Margaret Gilbert, “Giving Claim-Rights Their Due,” in Rights: Concepts and Contexts, ed. B. H. Bix and H. Spector (Farnham: Ashgate, 2012), 307–8. 29 At least in a state of nature. In a state of civil society, we surrender our property rights to our governments, trading the moral permission to enforce for the power to press charges or file civil suits. 30 I mean “demanding” to refer to a type of speech-act. I take it that in order to count as enforcement in a meaningful sense, demands must be able to influence other people. They thus must be voiced or uttered, and not “performed” silently, in one’s mind. Coleen Macnamara, “Taking Demands out of Blame,” in Blame: Its Nature and Norms, ed. D. J. Coates and N. A. Tognazzini (New York: Oxford University Press, 2013).

Obligations of Gratitude and Rights

161

demands do is to imply that force or sanctions might be forthcoming.31 This “proto-threat” quality is a feature shared by all demands—from the legitimate demands of a promisee to the “naked” demands of a pirate— and marks the minimum degree of enforcement to which a rightholder is entitled to resort. To have a pre-Hohfeldian right, then, is to have special standing to demand compliance, and to permissibly threaten and carry out unpleasant consequences if compliance is not forthcoming.32 Intuitively it seems that acts of gratitude, despite how beneficiaries should deliberate about them, may never be demanded. Christopher Wellman is right to note that this is true in the case of Marge’s thank-you gift for Selma; but it is also true in cases where gratitude calls on the beneficiary to prevent harm to his benefactor. No matter what is at stake for Rachel, and no matter Yancy’s reasons for doing so, Rachel simply may not demand that Yancy refrain from opening a new location in her neighborhood; and while she can accept an apology he freely offers, it seems she may not demand one. These intuitions are backed by a number of arguments in the literature on gratitude,33 the most cogent of which is suggested by Fred Berger. According to Berger, “While we have no hesitation in saying there is an obligation to show gratitude for help or for a gift, we do not feel at ease in saying it is something owed the grantor in the sense that he has a right to demand it. Such a demand shows the help or gift to be something less than a show of benevolence; it appears to be something done in order to gain favor.”34 Insofar as that is the case, the original act 31

John Skorupski, The Domain of Reasons (New York: Oxford University Press, 2010), 310. It might be objected here that many conventionally recognized moral rights, like my right not to be physically harmed, do not imply special standing on my part. Anyone, it seems, can permissibly “interfere” with a person who tries to harm me in order to stop her. This would imply that rights do not necessarily entail special standing. In response to this objection, it could be suggested that our intuitions in such cases rely on the assumption that when a person is being attacked (or stolen from, or having some other right violated), she is often not in a position to enforce her own rights in the moment. In such cases, we see ourselves as prima facie justified in “borrowing” a right-holder’s permission to interfere with a right-violator. For a longer reply to this sort of objection, see Carl Wellman, A Theory of Rights: Persons under Laws, Institutions and Morals (Totowa, NJ: Rowman & Allanheld, 1985), 137–9. 33 See, for instance, Barbara Herman, “Being Helped and Being Grateful: Imperfect Duties, the Ethics of Possession, and the Unity of Morality,” The Journal of Philosophy CIX, no. 5/6 (2012): 408; Margaret Visser, The Gift of Thanks (Boston: Houghton Mifflin Harcourt, 2009), 324; F. Hutcheson, An Inquiry into the Original of Our Ideas of Beauty and Virtue: In Two Treatises (J. and J. Knapton, 1729), 161. 34 Fred Berger, “Gratitude,” Ethics 85, no. 4 (1975): 300. For those who put forward similar arguments, see Weiss, “The Moral and Social Dimensions of Gratitude,” 492; Seneca, On 32

Tony Manela

162

of “benevolence” would not have been the sort of genuine benevolence that calls for gratitude. Thus to demand is to indicate that the grounds of the demand are not actually grounds for gratitude at all. I agree that if a benefactor had bestowed some benefit for the sole purpose of exacting a similar benefit in return later on, this would seem to render the original act of benevolence less than gratitude-worthy. But a benefactor’s demanding an act of gratitude does not necessarily entail that her primary intention in conferring the benefit was to gain standing to demand a return favor in the future. As Terrance McConnell points out, even if a benefactor “recognizes that her beneficiary will incur a debt of gratitude, it does not follow that the purpose, intent, or aim for which she undertook the act was to create such a debt( . . . )”35 Be that as it may, a benefactor’s belief that she could (permissibly) demand a return in the future does undermine her standing to demand an act of gratitude, whatever her original intentions were. Imagine a case in which a donor gives a recipient some benefit without the intention of claiming a return in the future, but with the belief (justified or unjustified) that she might permissibly demand a return if she changes her mind. Would we still condemn the absence of gratitude in a beneficiary in such a case? I think the answer is no. The case I described actually seems tantamount to an odd sort of lending, in which the donor presents her benefit to the recipient as a loan, repayment on which she does not intend (at the time of delivery) ever to demand. Should she ever change her mind and demand repayment, though, she reserves the right, as we say, to collect. A debtor has an obligation to repay a loan, and the creditor has standing to demand it. But what a creditor is owed is not gratitude. When a benefactor demands fulfillment of an obligation of gratitude, then, what is problematic is not necessarily the fact that the demand suggests some initial ulterior motive, but that it presupposes and entails (insofar as such a demand is put forward as legitimate) a belief on the part of the benefactor that in conferring her original benefit, she became entitled to demand a return. This presupposition undermines her desert Benefits, I.2; Gilbert Meilaender, The Theory and Practice of Virtue (Notre Dame, Ind.: University of Notre Dame Press, 1984), 170; Claudia Card, “Gratitude and Obligation,” American Philosophical Quarterly 25, no. 2 (1988): 120. 35

McConnell, Gratitude, 25.

Obligations of Gratitude and Rights

163

of gratitude, renders gratitude no longer owed.36 In demanding acts of gratitude, the benefactor thus makes it the case that a beneficiary who neither feels gratitude nor gives his benefactor’s interests special priority in his deliberations is no longer deserving of moral condemnation. And if obligations of gratitude may not be legitimately demanded, then there is no pre-Hohfeldian right to the fulfillment of obligations of gratitude, the way there is such a right to have a promise kept. It would be misleading to leave things there, however. While benefactors lack standing to demand or threaten, we need not conclude that benefactors must simply “lie down” and suffer ingratitude with silent disappointment. Claudia Card suggests as much when she says that a benefactor’s complaining of ingratitude comes close to complaining of her own poor judgment in selecting a beneficiary.37 But it would seem absurd for Rachel, should she discover Yancy’s ingratitude, to blame herself for saving his life. The person to blame is Yancy. Benefactors can, and perhaps sometimes should, hold ungrateful beneficiaries accountable. Though a benefactor may not do this by demanding, there are other speech-acts in her arsenal whose use becomes appropriate in the face of ingratitude. Imagine Rachel finds out that Yancy knows she will be put out of business by his decision to open a new location, and he proceeds anyway—not because of his promise to Peter, but simply because he wants his new Maserati. Though she may not demand that he desist, or threaten to sabotage the new store, she is entitled to resent him, and perhaps to express this resentment in a bid to get him to change his behavior. She seems to have a special standing to remonstrate with Yancy. “How could you do this to me?” she might ask. “Have you forgotten that I saved your life? How dare you drive me out of business just so you can have a new Maserati! All I ask is that you have the decency to leave my client base alone—and I will not forgive you if you don’t.” In remonstrating in such a way, Rachel expresses justified resentment, insists that Yancy account for his decision, and tries to influence Yancy by making certain normative reasons he already had motivationally salient to him. She thus influences Yancy in a way that Indeed, due to the sloppiness of the original “lending” act, it could be argued that nothing in the way of repayment is owed at all. 37 Card, “Gratitude and Obligation.” See also Weiss, “The Moral and Social Dimensions of Gratitude,” 497. 36

164

Tony Manela

falls short of the pushiness of demanding, while exerting more force than the request an unknown business competitor might have standing to make of Yancy under similar circumstances.38 This suggests a picture of gratitude as providing some kind of standing for a benefactor—not license to demand or enforce, but license to resent and remonstrate in the face of non-compliance. Reviving a term from the natural lawyers, I propose to call such standing an imperfect right. In addition to describing the standing of a benefactor, imperfect rights 38 Some might object to the possibility that there is such a thing as special standing to resent or to remonstrate in the face of ingratitude. Any third party, it might be argued, may permissibly resent and remonstrate with Yancy as Rachel can, since there is a universal moral liberty to feel and express feelings like resentment (Raz, 2010). I agree that everyone is at liberty to feel and express feelings, in the sense that they may not be forcibly stopped from doing so. But standing, as I understand it, is a subtler form of permission—permission which, if lacking, might warrant at most only a certain kind of reproach or rebuke. Consider a case in which Rachel comes to believe (mistakenly) that, despite her request that he not do so, Yancy has decided to open his new store in her vicinity. She resents his decision, shows up at his office, and begins to remonstrate with him. Yancy explains that she has been misinformed. In response to her request, he had decided not to open a new store in her vicinity after all. He considered not opening a new store at all, but instead, he has decided to open one on the other side of town, far from Rachel’s store, but in the vicinity of Ronnie’s. Rachel happens to know that Ronnie too saved Yancy’s life in a genuine act of benevolence years earlier, and that Ronnie also stands to be driven out of business by Yancy’s decision. Yancy, she realizes, is still violating an obligation of gratitude; but the fact that his violated obligation is to Ronnie, rather than Rachel, seems to change things subtly. Rachel may still be indignant at Yancy’s decision, but we would expect her resentment to cool, and she no longer seems, intuitively, to have the standing to remonstrate as forcefully. If she continues to remonstrate with Yancy, he might invoke norms of noninterference, and explain that though he may still be in the wrong, it is no longer any of her business— that it is between him and Ronnie. This apparent change in what Rachel has standing to feel or to say might be chalked up to a change in her belief that her own interests will be set back. She came to his office indignant at his ingratitude and disappointed that she would be driven out of business; she left with just as much indignation, but without the disappointment. On this view, any especially intense emotional reaction Rachel might feel could be reduced to 1) the indignation any third party might feel upon witnessing ingratitude, and 2) disappointment because her interests were the ones set back by that ingratitude. But this does not seem to be the case. Rachel’s standing to resent and remonstrate seems to be more than these two elements. Consider a case in which Yancy plans to open his new store near Rachel’s, despite her request that he desist. Yancy’s decision will not only drive Rachel out of business; it will also have an equally harmful impact on Theo, another business competitor in the area. When Theo discovers Yancy’s history with Rachel, he is just as indignant as any third party would be. And since he has just as much to lose as Rachel, he is just as disappointed as she. But intuitively, it seems that Rachel may resent and remonstrate with Yancy in ways that Theo could be rebuked for attempting. While it may be incumbent on Yancy at the very least to hear Rachel out, he may not owe the same to Theo. Rachel does seem to have special permission to remonstrate, the justification of whose appropriateness cannot be reduced to indignation and disappointment. I am grateful to Kit Wellman for pressing me on this point.

Obligations of Gratitude and Rights

165

might also be an apt way to characterize the standing some party has when she (merely) deserves some good treatment from another person.39 An employee who earns a raise, or an offender who earns forgiveness by atoning for a wrong, might also be said to possess imperfect rights, insofar as they may not demand these things, but may uncriticizably resent or remonstrate their absence. Though I do not have space to do the concept justice in this essay, it is worth noting and refuting one reason that the concept of imperfect rights has received so little positive attention in recent moral philosophy. Because they do not amount to standing to claim or demand, it has been argued that imperfect rights fail to fulfill perhaps the most important function of rights: to help us guarantee respect for ourselves. As Feinberg puts it, “Having rights enables us to ‘stand up like men,’ to look others in the eye, and to feel in some fundamental way the equal of anyone. To think of oneself as the holder of rights is not to be unduly but properly proud, to have that minimal self-respect that is necessary to be worthy of the love and esteem of others.”40 He goes on to write that this aspect of rights argues “well their supreme moral importance.” Any right which we could not claim or demand, he concludes derisively, “would be a very ‘imperfect’ right indeed!” Pace Feinberg, however, imperfect rights can indeed play an important role in guaranteeing respect. Imagine a person whose perfect rights are meticulously respected, but who is nonetheless scrupulously denied everything to which she has an imperfect right. Despite her exceptional hard work, she is never offered promotions or bonuses that were not guaranteed to her by contract. She praises the do-gooders of her society, but she herself never receives praise for her praiseworthy deeds. She always thanks people when they do her favors, but when she performs genuine acts of benevolence, she is never shown gratitude. When others wrong her, she forgives (when self-respect permits); but when she atones for similar transgressions, she is never forgiven. If such a person remained silent and unresenting in all these cases, we would be hard pressed to call her properly proud. We might be inclined to say she suffers from a lack

39 Rutherforth, Institutes of Natural Law: Being the Substance of a Course of Lectures on Grotius De Jure Belli Et Pacis, 31. 40 Joel Feinberg, “The Nature and Value of Rights,” Journal of Value Inquiry 4 (1970): 252.

166

Tony Manela

of self-respect, from a sort of servility.41 Just as a self-respecting person will sometimes assert her perfect rights, a self-respecting person would sometimes “press” her imperfect rights as well, at the very least by resenting, and occasionally by remonstrating. Of course, even if one has standing to do so, remonstrating is often not a wise or helpful course of action (just as actually demanding, even if legitimate, may be unwise or unjustified). But still, the mere recognition of one’s imperfect rights may be an important and overlooked component of self-respect. At the very least, imperfect rights would help explain what is morally wrong with a benefactor who always remains silent in the face of ingratitude. Let us return to the No-Right Platitude: the claim that benefactors lack a right to the fulfillment of obligations of gratitude. We can say that there are three senses of “right”: Hohfeldian claim-rights, preHohfeldian perfect rights, and pre-Hohfeldian imperfect rights. In virtue of being owed certain strict directed obligations of gratitude, benefactors can indeed have rights to gratitude in the first sense. And in virtue of a benefactor’s special standing to resent and remonstrate, benefactors may sometimes have a right to gratitude in the third (pre-Hohfeldian imperfect) sense. As my development of Berger’s argument showed, however, benefactors seem to lack rights to the fulfillment of obligations of gratitude, in the pre-Hohfeldian perfect sense of standing to demand or enforce.

4. CORRELATIVITY RECONSIDERED

My example of grateful nonmaleficence gives us strong reason to endorse all three gratitude platitudes: there are strict moral obligations of gratitude, the way there are strict moral obligations of promise-keeping; these obligations are owed to someone in particular, in the ways promissory obligations are; yet there is no perfect right to the fulfillment of these obligations, the way there is a perfect right to have a promise kept. Rather than rejecting any of the platitudes, then, it becomes more 41 This would be a wider kind of servility than that described by Thomas Hill—on whose account servility is a failure to take one’s perfect rights seriously. Hill is skeptical about whether it is wrong to “demand less for oneself than one deserves”, where, I take it, he uses the term “demand” in a looser sense than the one I used above. Thomas Hill, “Servility and SelfRespect,” in Dignity, Character, and Self-Respect, ed. Robin S. Dillon (London: Routledge, 1995), 92.

Obligations of Gratitude and Rights

167

plausible ask whether the tension among them can be resolved by modifying the correlativity thesis. The thesis, as I stated it at the beginning of this paper, is that strict directed obligations entail rights. As my discussion in the previous section showed, there are three distinct senses of the term “rights”, each of which corresponds to a more specific version of the correlativity thesis. Consider first CT1: Strict directed duties entail Hohfeldian claim-rights.

As a premise in substantive moral reasoning, CT1 has a substantial drawback: as Matthew Kramer points out, CT1 is little more than a stipulative definition of the term “claim-right”.42 Given that we’ve identified some strict directed obligation to someone, all CT1 allows us to infer is that this person has a strict obligation owed to her. No inferences can be made about the permissible responses of a wronged second party. Now consider CT2: Strict directed duties entail pre-Hohfeldian perfect rights.

Unlike CT1, CT2 is more than a stipulative definition. It says that where there is a strict directed duty, there is standing to demand. This version of the correlativity thesis allows for substantive inferences in moral reasoning (from claims about obligations to claims about standing to demand or enforce); but, as my discussion of the gratitude platitudes showed, CT2 is inconsistent with certain obligations of gratitude (namely, obligations of grateful nonmaleficence). There are strict directed obligations of gratitude that do not correlate to perfect rights. Now consider CT3: Strict directed duties entail pre-Hohfeldian rights, which may be perfect or imperfect.

42 Matthew Kramer, “Rights without Trimmings,” in A Debate over Rights, ed. Matthew Kramer, N. E. Simmonds, and Hillel Steiner (Oxford: Oxford University Press, 1998), 23. Kramer does not necessarily see this as a drawback of CT1, which may still serve a useful heuristic function for legislators trying to decide which duties to impose and which rights to enshrine or codify. But for those of us engaged in discovering what moral obligations and rights there are, CT1 is less than helpful.

168

Tony Manela

Like CT2, CT3 is substantive, and does more than stipulate a definition, as CT1 does. Unlike CT2, though, CT3 allows that a strict directed duty need not entail standing to demand or enforce. If we think of strict obligations as perfect obligations, then we might say that CT2 is perfection-preserving: if the directed obligation in the antecedent of the entailment claim is perfect, then the right it entails must also be perfect. CT3, by contrast, is not perfection-preserving. The perfection of the obligation in the antecedent of the entailment claim does not entail the perfection of the right it correlates to. Because of this, CT3 is consistent with the case of grateful nonmaleficence, since there are imperfect rights to gratitude. CT3 is thus a version of the correlativity thesis that is both substantive and plausible, in that it is consistent with obligations of gratitude. Knowing which version of the correlativity thesis can be relied upon is essential to the success of a certain inference pattern in moral reasoning: a sort of modus tollens, inferring from the absence of a right to the absence of a strict obligation. Philosophers who rely on CT2 in such inferences sometimes take the lack of standing to demand some act to imply that that act cannot be strictly obligatory. But if CT2 is false, then such inferences are not valid. In what remains of this paper, I want to explore two ramifications this entails for arguments in political and moral philosophy. Consider first efforts to ground political obligations in gratitude to the state.43 There are many strong arguments against the claim that our obligations to the state are obligations of gratitude,44 but one of the strongest runs as follows: even if the state deserves gratitude from its subjects, no benefactor has standing to demand grateful acts. And granting CT2 (that our obligations to the state are only those acts the state may demand of us), it follows that we have no obligations of gratitude to the state.45 But if CT2 is false, then we may indeed have obligations of gratitude to the state that the state may not exact by force.46 Such obligations might include, for instance, the obligation of a skilled worker 43

Depending on how one interprets Plato’s Crito, this argument has an ancient pedigree. For a list of these arguments (and rebuttals to them), see Walker, “Political Obligation and the Argument from Gratitude.” 45 Christopher Heath Wellman, “Gratitude and Political Obligation,” APA Newsletters: Newsletter on Philosophy and Law 99 (1999): 180–208. 46 McConnell, Gratitude, 207. 44

Obligations of Gratitude and Rights

169

educated at the expense of his state to remain in that state, rather than emigrate to a country with a higher quality of life.47 It is worth noting that this conclusion applies to efforts to ground political obligation in moral phenomena besides gratitude, such as fair play. The shakiness of CT2 suggests that the mere fact that some action may not be demanded by the state is not enough to infer that such an action is not obligatory. The rejection of CT2 also has ramifications more squarely in the realm of moral philosophy. Consider the class of actions called the suberogatory—actions that are bad to do but not morally forbidden. Julia Driver gives the example of Roger, whose dying brother Bob asks him to donate a kidney. Driver notes (rightly) that Roger’s brother has no perfect right to the kidney—may not demand it—and seems to infer that Roger has no obligation to his brother to provide it.48 She then takes our profound moral disapproval of Roger’s refusal to give the kidney to be evidence of a new moral concept, distinct from moral obligation: the suberogatory. But if CT2 cannot be relied upon, then the inference from a lack of standing to demand to a lack of an obligation can be questioned. And indeed, there is good reason to think that Roger does have a strict obligation to give the kidney, independent of anyone’s standing to demand. We would expect Roger (assuming he is a conscientious person, and has a healthy relationship with his brother) to see his brother’s need as providing a protected reason to give his kidney. We would also allow that Roger’s brother has the power to release him from this obligation, to forgive him if he breaches it, to resent him if he withholds his kidney for superficial reasons, and perhaps even to remonstrate with him under those conditions too. To label Roger’s decision suberogatory fails to do justice to the peremptory nature of Roger’s reason, to the normative control his brother has over that reason, and to his brother’s standing to 47 See Kieran Oberman, “Can Brain Drain Justify Immigration Restrictions?,” Ethics 123, no. 3 (2013): 427–55. Oberman does not believe such obligations are obligations of gratitude; but regardless of what principle such obligations emerge from, the rejection of CT2 means that the mere fact that such obligations may not be demanded should not count as a strike against their existence. 48 Driver, “The Suberogatory,” 287. This is an inference she makes more explicitly in an earlier example of suberogation—the case of the lone train rider asked to give up his convenient seat so a couple, who arrived on the train later than he, might sit together. For the lone rider to fail to do so is blameworthy, she writes: “The problem is justifying the blame when the agent is acting within his rights. The people who want to sit together have no claim against the person ahead of them in line. Thus, he has no obligation to pass up the more convenient seat.”

170

Tony Manela

hold him responsible for making the wrong decision (even if this falls short of standing to enforce). This suggests that it may be more fruitful to characterize at least certain “suberogatory” acts as a subset of strict directed obligations: those that, like obligations of gratitude, may not be enforced or demanded. Acknowledgments I am deeply grateful to Maggie Little, Henry Richardson, and Karen Stohr for their insights on the ideas in this essay. I am also grateful to Kit Wellman, to audiences at the Rocky Mountain Ethics Congress and the Workshop in Normative Ethics, and to two anonymous reviewers for comments on earlier versions of this project.

8 Moral Anxiety and Moral Agency CHARLIE KURTH

Nothing in human affairs is worthy of great anxiety. – Plato

How much should I give to charity? Is it okay for me to break this promise? As an advocate of women’s rights, can I vote for the pro-life candidate? When we face difficult moral decisions like these, we feel a distinctive unease: we must make a choice but we are unsure what the correct thing to do is. Yet despite the pervasiveness of this phenomenon, surprisingly little work has been done to either characterize this emotion—this moral anxiety—or explain the role it plays in moral decision making. That’s a mistake. Given that moral anxiety is a pervasive feature of our lives, it is important that we understand what it is. Moreover, and more importantly, given the many ways in which emotions can inform and distort our reasoning, it is also important to understand the role that moral anxiety plays in moral decision making. In what follows, I will argue that moral anxiety is central to good moral decision making and agency—it’s an emotion that we ought to cultivate. This claim is striking on its own. But it also upends the familiar picture, one found among philosophers and folk alike, of anxiety as an inherently destructive emotion: the anxious person is someone consumed—paralyzed—by intense anxiety. What could be valuable in that? To make my case, I begin by developing a model of moral anxiety that builds from work on anxiety in general and social anxiety in particular (§§1–3). The resulting account reveals moral anxiety to be an emotion that (i) we experience when we are uncertain about the correctness of a moral decision that we are contemplating or have made, and that (ii) prompts epistemic behaviors like deliberation and information gathering that are aimed at resolving our underlying uncertainty. With this model in hand, I then argue that moral anxiety is an emotion that is particularly well-suited to engage the capacities that are essential to good moral decision making and agency—deliberation, reflection, and

172

Charlie Kurth

the like. Thus, it is something we ought to cultivate (§§4–5). The end result is a novel account of the moral psychology that underlies moral thought and action.

1. VARIETIES OF ANXIETY

Anxiety—as a general psychological phenomenon—is (a) an aversive emotional response to uncertainty about a possible threat or danger that (b) provokes distinctive cognitive and motivational tendencies that are aimed at addressing the uncertainty in question.1 While anxiety in general displays this core pair of features, we also find an interesting diversity in both the situations that provoke it, and the behaviors that subsequently result. For instance, uncertainty about how to act in an unfamiliar social situation (e.g., a fancy gala) can bring anxiety that tends to provoke deference to social authorities; knowing that your behavior broke the rules, but feeling uncertain about whether you will be punished for it, brings anxiety that prompts efforts to make up for what you did (e.g., a preemptive apology); uncertainty about existential matters (e.g., does God exist?) can bring anxiety that leads to epistemic behaviors (e.g., reflection, consultation of religious texts). Moreover, while we often focus on situations where anxiety manifests in chronic, debilitating ways, such cases are (for most of us) exceptions. In fact, it’s widely accepted among social and clinical psychologists that anxiety typically manifests as a moderate and generally beneficial emotion. As the psychologist David Barlow explains, [a]nxiety functions to warn of a potential danger situation and triggers the recruitment of internal ( . . . ) psychological defense mechanisms ( . . . ) [This] serves the adaptive purpose of protecting the integrity of the individual and allowing a higher and more mature level of functioning.2 1 The term ‘anxiety’ as used in ordinary speech, philosophy, and psychology refers to a range of phenomena that are unlikely to constitute a unified kind. Thus, in this essay, I follow others (e.g., Baumeister & Tice 1990, Öhman 2008) in focusing on an important dimension of what we refer to as ‘anxiety’—namely, anxiety that concerns uncertainty. There is also some tendency to use ‘anxiety’ to refer only to significant or clinical levels of unease/worry, rather than the broader psychological phenomenon that I will be focusing on. As we’ll see, there’s good reason to prefer the more encompassing usage I will be employing. 2 Barlow 2001: 8; see also Baumeister & Tice 1990, Marks & Nesse 1994, Leary & Kowalski 1995: 22ff.

Moral Anxiety and Moral Agency

173

To be clear, to claim that anxiety is generally beneficial is not to deny that it can lead to problems—intense bouts (or deficits) of anxiety can be disastrous. Rather, it’s meant to draw out that even though anxiety can sometimes go awry, it—like other dimensions of our emotional repertoire—typically manifests in a moderate and beneficial manner. In this way, anxiety is no different than, say, fear or anger: these emotions are good things to have even though they can at times manifest in unfortunate ways (e.g., phobias, rage).3 Taken together, these observations suggest that the label ‘anxiety’ picks out a family of emotional responses that share a common core (namely, (a)–(b) above), and that these responses take on distinct forms in order to help individuals address specific kinds of problems. So, for instance, social anxiety prompts things like deference and caution that help one manage uncertain social situations. Similarly, punishment anxiety brings efforts to make amends that can help reduce the likelihood of punishment.4 While this initial gloss on anxiety is instructive, it leaves two important questions unanswered. (1) How can we individuate particular varieties of anxiety in a way that allows us to see them as (in a sense) species of a common genus? (2) How can we explain the distinctive concern of a particular variety of anxiety in a way that allows us to both understand when such anxiety is appropriate (or rational), and make sense of the resulting behaviors as generally beneficial? In responding to these questions, it will be helpful to draw on tools from the philosophy of emotion. In particular, understanding emotions in terms of their ‘formal objects’ will allow us to develop a general answer that we can then use (in §§2-3) to understand moral anxiety. The formal object of an emotion is what that emotion is about; it’s the property that is implicitly ascribed by the emotion to its target. Thus, the formal object of fear is something like danger: to fear the dog is to see the dog as dangerous; the formal object of anger is, roughly, an offense 3 This suggests that the real issue is not whether anxiety (or anger or fear) is generally beneficial. Rather, it’s how to cultivate anxiety (or anger or fear) so as to draw out its benefits—a question I take up in §5. 4 See Marks & Nesse 1994; Barlow 2001. I’m not claiming that every type of anxiety is beneficial. One might think that existential anxiety—anxiety about (e.g.) God existence—is unlike social or punishment anxiety in that it is not a generally beneficial emotional response (or even a genuine kind of anxiety). Whether this is true is a matter for another time.

174

Charlie Kurth

against you (or your interests): to be angry at the comment is to see it as offensive.5 Moreover, and as these examples suggest, the properties that are constitutive of the formal objects of emotions are typically taken to be evaluative properties: they are things—like dangers and offenses— with normative significance for the individual experiencing the emotion.6 But if having an emotional experience involves attributing a formal object—an evaluative property—to the emotion’s target, then emotional experiences are, in a real sense, forms of evaluative awareness: they are experiences where one sees one’s situation in a normatively loaded way. Given the above renderings of the formal objects of fear and anger, our discussion suggests the following glosses on the distinctive kind of evaluative awareness associated with these emotions: To fear the dog is to see the dog as dangerous—as something to be avoided. To feel anger at someone’s comment is to see the comment as offensive—as something calling for a response.

Moreover, recognizing that experiences of fear and anger involve evaluative awareness of this sort helps explain both why they tend to bring the distinctive behaviors that they do, and why these emotions are things that can be beneficial. Given that fear is about danger, it makes sense that it tends to result in a fight/flight/freeze response; likewise, given that anger concerns an offense to you or yours, it makes sense that it tends to trigger things like aggression and outrage.7 5 See Deonna & Teroni 2012 for an overview of emotions and formal objects. D’Arms & Jacobson (2003) reject this way of understanding emotions. As they see it, what is typically taken to be a descriptive account of an emotion’s formal object is actually better understood as a normative gloss that expresses when it is appropriate to feel the emotion in question. I’m quite sympathetic to D’Arms & Jacobson’s proposal and what follows could be restated in a way that fits with it. But for ease of presentation, I will stick with the more familiar formal objects model. 6 Here I aim to be neutral on metaethical debates regarding the nature of evaluative properties. 7 The general claims in the text are widely accepted among philosophers and psychologists studying emotions—e.g., Solomon (1973), de Sousa (1987), Lazarus (1991), Ekman (1992), Nussbaum (2001), D’Arms & Jacobson (2003), Roberts (2003), and Prinz (2004). That said, this sketch needs fleshing out. For instance, with regard to the claim that emotional experiences are experiences where one sees one’s situation in a normatively loaded way, there is much controversy over both how we should understand both the nature of this awareness (e.g., do emotions provide perceptual awareness?), and what the underlying evaluative content of emotional experience is (e.g., is this content propositional or something more non-cognitive?). There are also important questions about how to characterize the evaluative dimension of

Moral Anxiety and Moral Agency

175

This appeal to the formal objects of emotions provides us with the beginnings of answers to questions (1)–(2) concerning what’s distinctive of various types of anxiety. We can start by asking what the formal object of anxiety is. The above discussion (feature (a) in particular) suggests that the formal object of anxiety is problematic uncertainty: to feel anxious about a situation is to see that situation as involving a threat or danger whose potential is unpredictable, uncontrollable, or otherwise open to question. But if the formal object of anxiety is problematic uncertainty, we should be able to distinguish different types of anxiety (social, punishment, moral, etc.) by saying more about the nature of the uncertainty in question. Moreover, getting clearer about the nature of the uncertainty involved in different types of anxiety should also help us understand both when it is appropriate to feel a particular form of anxiety and why anxiety of that sort can be beneficial. In what follows, I develop this picture by taking a closer look at social anxiety. Social anxiety is a nice starting place both because its phenomenology is familiar and because it has been studied extensively by clinical and social psychologists. Thus, social anxiety will provide us with a framework that we can use to develop a model of moral anxiety. Moreover, the general strategy will also give us an important methodological tool: recognizing the significant similarities between social and moral anxiety will allow us to flesh out our account of moral anxiety by drawing on empirical work concerning social anxiety. 2. SOCIAL ANXIETY AS TEMPLATE

As a first pass, we can understand social anxiety as an aversive emotional response to uncertainty about how others will judge one’s social worthiness—one worries that others will deem one as, in some way, not socially good enough. Moreover, this concern typically leads to a distinctive combination of thoughts and behaviors aimed at both addressing one’s uncertainty (e.g., efforts to determine if one has run afoul of the social conventions), and minimizing one’s chances of being viewed as socially ‘unworthy’ (e.g., avoidance, deference to social superiors). So we have the makings of an explanation for why social anxiety is a emotional awareness: is the normativity prudential, moral, emotion-specific? Fortunately, for the discussion that follows, we can stay largely neutral on these details.

176

Charlie Kurth

generally beneficial emotion: it helps one to recognize possible threats to one’s social standing and to respond appropriately to them. Again, this is not to deny that social anxiety can have very bad effects. In some cases (e.g., for individuals who engage in high levels of self-monitoring), anxiety can lead to the chronic self-esteem difficulties and withdrawal behavior that characterizes Social Anxiety Disorder.8 However, and in line with the remarks above (§1), such cases are exceptions: social anxiety is, for most of us, most of the time, a moderate and beneficial emotional response—a twinge that brings awareness and caution, not an intense dread that brings consuming worry. An example will help draw this out. You are at a dinner party and are seated next to Sam. He’s a new acquaintance, but someone with whom you have several common friends—thus you’re concerned to make a good impression. So far, your conversation has been going well. But the mood quickly shifts in a way that makes you anxious: the discussion becomes awkward and Sam makes eye contact less frequently. Sensing this, you begin to worry that you’ve said something inappropriate. Given your anxiety, you start trying to figure out what you might have said that could have triggered the change. Was it the story about the shenanigans you pulled back in college? The talk of the election? Will he mention this to your mutual friends? Will they care? Taking advantage of a lull in the festivities, you replay portions of your conversation in your head. You’re trying to understand what happened and what you could say to restore the earlier tenor of the discussion. There are also changes in your general demeanor: you become more cautious and deferential. You are, for instance, more hesitant to engage Sam in conversation again and, when you do, you are more deferential—more inclined to apologize when, say, you accidently interrupt him. You also seek to insert comments here and there with the hope of clarifying—mitigating—some of your earlier remarks about college and politics. Applying this example to the general framework developed in §1 indicates that we should understand the formal object of social anxiety as something like problematic uncertainty about how others will view your social worthiness. This in turn suggests the following gloss on the distinctive kind of evaluative awareness associated with social anxiety: 8 How exactly social anxiety manifests will, like any aspect of personality, be subject to both individual differences and situational factors. See Barlow 2001: chap. 13 for further discussion.

Moral Anxiety and Moral Agency

177

To feel socially anxious about (e.g.) one’s comment is to see the comment as making one a possible target of social criticism—as something to be addressed or corrected.

Recognizing that social anxiety is a form of social awareness helps us see that it has an important place in human psychology: because social anxiety helps one see that one’s comment (attire, behavior, etc.) might be deemed socially inappropriate, it provides one with a signal that one might be subject to social criticism and thus needs to tread lightly.9 This in turn helps explain both when social anxiety is appropriate, and why it tends to bring the distinctive behaviors that it does. Given that the formal object of social anxiety is problematic uncertainty about how others perceive our social worthiness, it is appropriate to feel socially anxious when there is uncertainty of this sort. Similarly, given that social anxiety concerns uncertainty about whether our behavior violates the social sensibilities of others, it makes sense that we tend to respond with the epistemic and risk avoidance behaviors that we do. More specifically, and as the dinner party example reveals, social anxiety tends to bring epistemic behaviors that are aimed at identifying and assessing the social error that one may have made so that one can repair the damage that one might have done. But if one cannot determine how one might have erred, one typically falls back on behaviors that will minimize the risks of social sanction—e.g., deference and avoidance behaviors.10

9 As suggested in note 7, I intend my account to be neutral with regard to debates in the philosophy of emotion. But debates between perceptual and appraisal/cognitive theories of emotion are a place where this is not entirely possible. Perceptual theories (e.g., Ekman 1992, de Sousa 1987) maintain that emotional experiences are, or necessarily involve, (something like) perceptions of value. Appraisal/cognitive theories (e.g., Lazarus 1991, Nussbaum 2001) hold that emotional experiences are, or necessarily involve, (something like) value judgments. However, while my account of anxiety as a form of social awareness can be made to fit with most perceptual and appraisal/cognitive theories, it is not compatible with extreme appraisal/ cognitive views that equate emotions with value judgments (e.g., Solomon 1973). This is because if emotions just are cognitive evaluative judgments, then they cannot play a role, as I maintain, in raising and shaping awareness prior to those judgments. Given that we have independent reason to reject these extreme proposals on, for instance, the grounds that they cannot make sense of cases where one feels fear but judges that there’s nothing to be afraid of (e.g., Roberts 2003, chap. 1), I do not see this restriction on the scope of my proposal as a cost. 10 For a nice discussion of both social anxiety’s signaling function and its role in engaging epistemic and risk minimization behaviors from the perspective of psychologists, see Leary & Kowalski 1995 and Miceli & Castelfranchi 2007.

178

Charlie Kurth

The upshot, then, is that when social anxiety is appropriate to the situation we are in, as (intuitively) it is in the dinner party case, it functions not just as a distinctive form of social awareness, but one that prompts generally useful behaviors that helps us monitor, assess, and regulate our interactions with others in social settings. It is, in short, a valuable emotion—one that enables us to better understand and navigate the complexities of social life. 3. MORAL ANXIETY: WHAT IT IS AND WHAT IT DOES

We now have a general account of ‘anxiety’ as a label for a family of emotional responses to distinct forms of uncertainty. We have also developed a more detailed account of social anxiety that has allowed us to both draw out its distinctive features, and understand how it can productively inform and shape social interaction. In this section, I will use this account of social anxiety as a template from which to develop a model of moral anxiety. This investigation is significant in two ways. First, it reveals that, though there are important differences between social and moral anxiety, they also share significant similarities in their basic structure—similarities that help support the claim that moral anxiety, like social anxiety, is a genuine and generally beneficial feature of human psychology (§3.1). Second, it allows us to distinguish moral anxiety from similar emotions like punishment anxiety and anticipatory guilt (§3.2). 3.1 The Basic Model The examples at the beginning of the paper suggest that moral anxiety is an aversive emotional response to uncertainty about the correctness of a moral decision one is contemplating (or has made).11 Moreover, when one feels morally anxious, one tends to do things that are aimed at both resolving one’s underlying uncertainty (e.g., information gathering, reflection, deliberation), and minimizing the risks that come with having to act in the face of such uncertainty (e.g., deferring to “moral authorities”). This suggests that, as with social anxiety, we can see moral anxiety as involving a distinctive form of awareness and motivation—one that 11 In addition to feeling morally anxious about what to do, one can also feel morally anxious about what to value. For brevity’s sake, I will focus only on the former.

Moral Anxiety and Moral Agency

179

tends to provoke behaviors that help one better understand what the morally correct thing to do is. Again, an example will help illustrate this. Your mother’s Alzheimer’s has advanced dramatically in the last year and you can no longer provide her with the care she needs. You’re inclined to follow the doctor’s recommendation and put her in a nursing home. But this decision makes you anxious.12 Given your anxiety, you start to reflect on the details of the situation that you now face: your mother was always so concerned about elderly care facilities—in fact, last year, before things turned for the worse, you promised her that you would never put her in such a place. Recalling your promise you now wonder—would she have thought that in a situation like this it would be inappropriate to put her in a nursing home? In exploring this question, you’re trying to sort out whether it’s permissible to break the promise you’ve made. Still unsure, you decide to ask a close friend for guidance and are prepared to refine—perhaps even change—your thinking about what to do based on your conversation. As with the dinner party case, we have an intuitive and familiar example that both works to highlight some of the distinctive features of moral anxiety, and helps us understand how it can contribute positively to moral thought and action. Building from this example and our initial gloss of moral anxiety, we can say that the formal object of moral anxiety is, roughly, problematic uncertainty about the correctness of a moral decision one is contemplating (or has made). This in turn suggests the following gloss on the distinctive kind of awareness associated with moral anxiety: To feel morally anxious about a decision that one is contemplating is to see one’s decision as possibly open to moral objection—as something calling for a cautious approach.

Thus, when one feels morally anxious, one sees one’s situation in the same kind of evaluatively loaded way that is associated with feelings of social anxiety. This means that the claim that moral anxiety is a distinctive form of moral awareness should be just as plausible as the claim that

12 Presumably, you’re also feeling other emotions—sadness, frustration, anger. Here I want to focus specifically on your anxiety.

180

Charlie Kurth

social anxiety is a distinctive form of social awareness: both serve as signals that one faces a particular kind of problematic uncertainty. Moreover, and again like social anxiety, the fact that the awareness here is evaluative awareness helps explain both when moral anxiety is appropriate and why it prompts the behaviors that it does. Given that the formal object of moral anxiety is problematic uncertainty about the correctness of one’s decision, it is appropriate to feel morally anxious to the extent that the decision one is contemplating might be open to moral objection. And given that moral anxiety concerns uncertainty about the correctness of one’s decision, it makes sense that it prompts epistemic behaviors—information gathering, reflection, deliberation, and the like—that are aimed at helping one resolve the uncertainty about whether the decision one is contemplating is morally justified.13 Stepping back, we can see that looking to social anxiety has allowed us to develop an account of what moral anxiety is and what it does. In short, it is a variety of anxiety that provides us with a distinct form of normatively-loaded awareness: it both signals to us that the decision we are considering may be open to moral objection, and prompts behaviors—especially epistemic behaviors—that are aimed at helping us address our underlying uncertainty. Moreover, given the broad functional and structural similarities that we’ve identified between social and moral anxiety, we can also see how moral anxiety can be valuable. When our moral anxiety is appropriate to the situation we are in, as (intuitively) it is in the case of your promise to your mother, we have an emotion that can help us monitor, assess, and regulate our moral thinking and doing. Such a capacity will be particularly useful in the early stages of moral development when we are more likely to face new and novel moral situations—situations that involve the problematic uncertainty that moral anxiety can help us recognize and work though. But since even morally mature individuals can face novel and difficult moral situations, 13 Here we see a contrast between social and moral anxiety. In the case of social anxiety, avoidance, deference, and other efforts to minimize risk play the dominant role—epistemic behaviors typically come in just to help identify the best risk minimization strategy (Miceli & Castelfranchi 2007). With moral anxiety, by contrast, we see more emphasis on epistemic behaviors—an emphasis that makes sense given moral anxiety’s formal object. That said, I do not deny that hard moral questions can provoke deference and avoidance behaviors. Clearly, they can: with regard to your promise to your mother, you might just defer to her doctors, or try to pass the decision off to your sister. But such behaviors are often a last resort. After all, you feel moral anxiety, in part, because you must come to a decision about what to do (more on this is §4.3).

Moral Anxiety and Moral Agency

181

moral anxiety will continue to be valuable even as we become wiser and more experienced.14 3.2 Moral Anxiety as Distinct The discussion so far allows us to distinguish moral anxiety from both social anxiety and anxiety about punishment in terms of their formal objects. In short, moral anxiety concerns uncertainty about the correctness of a moral decision that you are contemplating, while social and punishment anxiety concern, respectively, uncertainty about how others will view your social worthiness, and uncertainty about whether you will be punished for a wrong you have done. Moreover, and as we’ve seen, it is because these varieties of anxiety have different formal objects that we can explain why they prompt the distinct behaviors that they do. However, even after distinguishing between social, punishment, and moral anxiety, one still might be puzzled about how moral anxiety differs from other similar emotions—especially anticipatory guilt and what we might call ‘practical anxiety’: the anxiety you feel about the correctness of a practical decision that you are contemplating. We can start with anticipatory guilt. Anticipatory guilt is an emotion we experience in advance of acting in a certain way; it is (roughly) the concern that one will do wrong by X-ing. Thus, anticipatory guilt is different from moral anxiety in that, with anticipatory guilt, one’s concern is focused on doing something that one believes to be wrong. With moral anxiety, by contrast, one’s concern is focused on one’s uncertainty: is the decision to X open to challenge?15 Moreover, recognizing this difference in what anticipatory guilt and moral anxiety concern (i.e., differences in their formal objects) indicates that we should also expect there to be differences in the associated behaviors that they prompt. In particular, rather than the epistemic behaviors associated with moral anxiety, we should expect anticipatory guilt to prompt things like a motivation away from X-ing and a tendency to engage in preemptive, reparative behaviors

14 I say more about moral anxiety’s value in §4. In that discussion I also take up questions about the problems that moral anxiety can bring. 15 Velleman (2003) discusses a phenomenon that he terms ‘moral anxiety’ but he’s clearly looking at something akin to anticipatory guilt, not the uncertainty-based form of moral anxiety I’m exploring.

182

Charlie Kurth

(e.g., apology) when one suspects that such motivation will not be sufficient to keep one from X-ing.16 The more difficult case concerns distinguishing moral anxiety from practical anxiety—the anxiety that one feels about the correctness of a practical decision that one is contemplating.17 After all, both concern the correctness of a decision under consideration. How these emotions differ will turn on larger questions about how to distinguish the moral from the practical, and what the correct theory of emotion is. I want to remain neutral on these issues. That said, it’s still possible to illustrate how one might distinguish these varieties of anxiety.18 For instance, while the formal objects of moral and practical anxiety both appeal to concerns about the correctness of the decision at hand, one might characterize this ‘correctness’ in different ways. For instance, one might hold that for moral anxiety, correctness amounts to being justified from an impartial point of view; while for practical anxiety correctness amounts to being justified from a more agent-centered point of view. Moreover, given this way of articulating the difference in the formal objects of these emotions, we should expect to find corresponding differences in the behaviors they prompt. In particular, if moral anxiety concerns uncertainty about the correctness of a decision as viewed from an impartial point of view, then we should expect there to be epistemic behaviors (e.g., reflection, deliberation) that tend to be impartial in this way; and if practical anxiety concerns uncertainty about the correctness of a decision as viewed from an agent-centered point of view, then we should expect epistemic behaviors that tend to be agent-centered. The upshot, then, is that our general model of anxiety allows us to make principled and plausible distinctions between different varieties of anxiety. It warrants us in taking moral anxiety to be a genuine and important psychological phenomenon.

4. MORAL ANXIETY, DELIBERATIVE VIRTUE, AND AGENCY

The model of moral anxiety we now have provides a descriptive account of what moral anxiety is and what it does: moral anxiety is a response to 16

See Frank 1988 for a discussion of these features of anticipatory guilt. I discuss practical anxiety at length in Kurth (MS.1). 18 I do not believe much of substance turns on the possibility that, pace the discussion that follows, moral anxiety is just practical anxiety with moral content. 17

Moral Anxiety and Moral Agency

183

uncertainty about the correctness of a moral decision that prompts epistemic behaviors (e.g., deliberation, information gathering) aimed at resolving the uncertainty at hand. Having made these descriptive claims, we are in a position to defend a normative one: moral anxiety is something we ought to cultivate, for it is central to good moral decision making and agency.19 The core of my argument is as follows: P1) Our concepts of good moral decision making and agency presume that individuals have certain deliberative virtues—e.g., they’re appropriately attuned and receptive to reasons. P2) Having these deliberative virtues requires that individuals have distinctive metacognitive capacities. P3) Moral anxiety is a psychological mechanism that engages the needed metacognitive capacities. C1) So, moral anxiety is a mechanism that engages the very capacities that are essential to good moral decision making and agency. P4) We ought to cultivate capacities that make distinctive and important contributions to our ability to engage in good moral decision making and agency. P5) Moral anxiety makes distinctive and important contributions to our ability to engage in good moral decision making and agency. C2) So, moral anxiety is something that we ought to cultivate.

I take P1 to be an uncontroversial claim about the relationship between certain deliberative virtues and our concepts of good moral decision making and agency. More specifically, these concepts presume that individuals are appropriately sensitive to matters of moral justification: to deem a moral decision a good one, or to see an individual as exhibiting effective moral agency, is to (among other things) see the decision making/individual in question as being attuned and receptive to the relevant reasons in the right sort of way.20 I take P4 to be similarly 19 In making this claim, my focus is on the decision making and agency of actual humans, not ideal or perfect decision making and agency as such. 20 A couple points: (1) I take this gloss on the deliberative virtues associated with good moral decision making and agency to border on platitude. For advocates, see the references in footnote 22 later in this chapter. (2) I do not take reasons-responsiveness to exhaust the deliberative virtues that underlie good moral decision making and agency; things like curiosity, imagination, humility, and enthusiasm are also likely to play an important role.

184

Charlie Kurth

uncontroversial. So the real substance of the argument lies in P2, P3, and P5. In what follows, I flesh out and defend these premises and argue that, given their plausibility, we have reason to conclude that moral anxiety is an emotion we ought to cultivate.21 4.1 Premise 2: Moral Decision Making and Moral Metacognition Good moral decision making and agency presume that individuals are appropriately attuned and receptive to the reasons that are pertinent to the issue at hand. This is a claim about the psychology of good moral agents. So we should ask what—psychologically speaking—it involves. At minimum, the kind of attunement and receptivity implicit in these concepts involves a distinctive set of metacognitive capacities, three of which are noteworthy: (1) The ability to monitor one’s (moral) cognitions (e.g., beliefs, desires, emotions, and attitudes) as they relate to each other and to the features of the world that they purport to be about. (2) The ability to recognize and assess problems with one’s (moral) cognitions (e.g., inconsistency, falsity, insufficient justification). (3) The ability to bolster, revise, qualify, or even abandon one’s (moral) cognitions in light of one’s assessments of them.22 Here two points of elaboration will be helpful. First, the capacities described in (1)–(3) are metacognitive in the sense that they are forms of mental processing that—at the conscious or unconscious level— function to monitor, assess, and regulate other aspects of mental processing. Second, these capacities are moral in the sense that the abilities in

21 To be clear: the claim here is not that we should see moral anxiety itself as a virtue; rather, the claim is that moral anxiety is something to cultivate because it is central to our ability to exhibit the deliberative virtues—e.g., attunement and receptivity—that are essential to good moral decision making and agency. 22 While the capacities in (1)–(3) are not generally made explicit, they’re clearly presumed in a wide range of proposals in normative and metaethics. For instance, these metacognitive capacities underlie the distinctive form of self-consciousness that is central to Christine Korsgaard’s (2009) account of agency as self-constitution, the wants/interests mechanism that is key to Peter Railton’s (1986) accounts of moral and prudential judgment, Allan Gibbard (1990) and Philip Kitcher’s (2011) accounts of the role of normative discussion in practical decision making, and Valarie Tiberius’s (2008) virtue theoretic account of practical reflection.

Moral Anxiety and Moral Agency

185

question are tied to the engagement of, e.g., a distinctly moral perspective (c.f., §3). 4.2 Premise 3: Moral Anxiety and Moral Metacognition We’ve already seen that moral anxiety is implicated in moral metacognition (§3). But we can say more. In fact, recent work in social and clinical psychology allows us to flesh out the model of moral anxiety that we developed above. The result will be a theoretically and empirically grounded defense of P3. Support for the claim that moral anxiety performs (or is engaged in) the metacognitive monitoring and assessing function of (1)–(2) comes from the earlier observation that moral anxiety is a form of moral awareness. After all, in order for moral anxiety to function as a signal that one’s decision may be vulnerable to challenge, it must be (or be associated with) mechanisms that perform the monitoring and assessing functions noted above. Moreover, work in psychology and cognitive science gives empirical backing to this theoretical claim. In particular, this work suggests that anxiety is “the phenomenological correlate of ( . . . ) [cognitive] conflict” in the sense that we feel anxious when we recognize (perhaps only at the unconscious level) that our cognitions— e.g., our beliefs, desires, intentions, feelings—are in tension with one another.23 Importantly, this research doesn’t just focus on social anxiety or anxiety in general—some of it examines the anxiety that results from conflicts concerning moral matters. For instance, recent research indicates that recognizing that one’s decision conflicts with one’s conception of oneself as a moral person tends to produce the unease and stress that is characteristic of anxiety.24 This adds further support to the claim that moral anxiety involves a set of mechanisms that both monitor and assess one’s moral cognitions, and makes one aware of any tensions in them.

23 McClure et al. (2007): 221. See also, Barlow (2001): chap. 3, Miceli & Castelfranchi (2007): 310, and Öhman (2008). The quote in the text raises interesting questions about the differences between (moral) anxiety and cognitive dissonance. Briefly, these phenomena can be distinguished along at least two dimensions. First, (moral) anxiety concerns not just the conflicting cognitions that are characteristic of cognitive dissonance, but more specifically the ones that provoke uncertainty. Second, while resolution of cognitive dissonance tends to take the path of least resistance (Aronson 1999), this is not the case for (moral) anxiety. 24 See, e.g., Burroughs & Rindlfleish (2002): 351–5; Aquino et al. (2009).

186

Charlie Kurth

For evidence that moral anxiety is engaged in the regulatory metacognitive functions of (3), we can return to the claim from §3 that moral anxiety prompts distinctive epistemic behaviors. These behaviors function, as we saw, to help one resolve one’s underlying uncertainty about the correctness of the moral decision that one is considering. But for moral anxiety to engage such behaviors (e.g., reflection, deliberation, information gathering) just is for it to engage the metacognitive regulatory functions in (3). Again, empirical research supports this account of moral anxiety’s regulative role. In particular, this work indicates that anxiety about the correctness of a decision one is contemplating engages four broad patterns of regulating behavior. (i) Anxiety tends to bring more focused thoughts and reflections. When anxious, one will be more focused on the threatening features of one’s situation and more inclined toward reflection about them.25 (ii) Anxiety also prompts distinctive forms of reasoning. It tends to engage one in various forms of detailed, hypothetical reasoning as one tries to think through how best to respond to the situation one faces.26 (iii) Anxiety also tends to bring a shift in one’s deliberative perspective. As a result of anxiety, one will tend to seek out alternative viewpoints in an effort to enrich one’s understanding of the situation one faces. One might, for example, ask a friend for advice about what to do, or seek out more information about the details of one’s situation.27 Relatedly, (iv) anxiety tends to prompt more open-minded inquiry. When anxious, one becomes more receptive to new information and less likely to be dismissive of conflicting viewpoints.28 Taken together, this research nicely supports the theoretical picture of moral anxiety as a psychological mechanism that regulates moral judgment by informing and shaping our understanding of the moral decision we face.

25 On anxiety’s role in focusing attention and increasing vigilance, see Mathews 1990: 461, Faucher & Tappolet 2002, and Öhman 2008, 715–16. On increased reflection, see Messer 1970. 26 For discussion of the role that anxiety and other emotions concerned with uncertainty play in generating detail-oriented processing, see Mathews 1990: 456–7, Tiedens & Linton 2001, and Schwartz & Clore 2007. 27 MacKuen et. al. 2010 shows that anxiety prompted by challenges to one’s position on affirmative action policy leads one to seek out more information on affirmative action policy, and to seek information that both affirms and challenges one’s viewpoint. Anxiety’s role in shifting one to a more social perspective is a central theme of Baumeister & Tice 1990. 28 MacKuen et. al. 2010.

Moral Anxiety and Moral Agency

187

That is, it supports the claim that moral anxiety performs the metacognitive functions in (3).29 4.3 Premise 5: Moral Anxiety Contributes Productively The discussion so far indicates that our concepts of good moral decision making and agency presuppose that we have certain metacognitive capacities—capacities of the very sort that moral anxiety functions to engage. Thus, in moral anxiety, we have a psychological mechanism that can do the very things that we’re interested in. This just is conclusion C1 from above. But why think that moral anxiety plays a distinctive and important role in the metacognition that underlies moral decision making and agency such that it’s something we ought to cultivate? That is, why accept P5 and the resulting conclusion C2?30 To draw out the plausibility of P5—namely, the claim that moral anxiety plays a distinctive and important role in engaging metacognition—first notice that the discussion so far indicates that moral anxiety is an emotional response that is in the business of engaging 29 Objection: The discussion in the text maintains that moral anxiety causes epistemic behaviors like reflection and information gathering. But surely the causal process also runs in the other direction: reflection and information gathering can bring anxiety. If that’s so, then two worries arise. First, unless there’s reason to think that this alternative causal route is infrequent, we have a challenge to P3 (i.e., the claim that moral anxiety prompts important forms of metacognition). Moreover, rumination that leads to anxiety is a symptom of General Anxiety Disorder and other clinical conditions. So we also have reason to doubt that moral anxiety is generally beneficial. Reply: I grant that this alternative (and potentially pernicious) causal pattern occurs. However, there is good reason to think that it is atypical. As we’ve seen (§1), the consensus among psychologists is that anxiety is a generally beneficial response. More importantly, the empirical work cited above (notes 25–8) provides evidence that—among normals—anxiety causes epistemic behaviors like reflection and information gathering. Moreover, recent work also suggests that in cases where reflection brings anxiety, this alternative causal route is the upshot of specific (but atypical) triggers. For instance, clinical levels self-monitoring appear to engage feedback mechanisms that lead to anxiety-provoking levels of reflection (Barlow 2001, Mor & Winquist 2002). There are similar findings for individuals inclined to take an immersed, first-personal reflective perspective on their situations as opposed to a distance, third-personal one (Kross & Ayduk 2008). In light of all this, we can set the objector’s concerns aside. 30 The argument that follows is not meant to establish that moral anxiety is necessary for good moral decision making and agency. It may be, but I won’t (and needn’t) argue for that here. For present purposes, all I need to establish is that moral anxiety plays a distinct and important role in engaging the metacognition—i.e., the monitoring, assessing, and regulating functions described in (1)–(3) above—that is essential to good moral decision making and agency.

188

Charlie Kurth

the metacognitive functions that are essential to good moral decision making and agency. More specifically, moral anxiety is, as we’ve seen, a sensitivity to problematic uncertainty that we have developed in order to better negotiate the complexities of moral life (§3). So even if there are other ways to engage these metacognitive functions, these functions are things that moral anxiety is particularly well-placed to perform (more on this shortly). Moreover, the corresponding claim for social anxiety (namely, that it’s in the business of engaging the metacognitive functions that allow us to navigate the complexities of social interaction) is one that is well supported by both personal experience and work in psychology (§§1–2). So combining this claim about social anxiety with the structural similarities that we’ve identified between social and moral anxiety, gives further support to the claim that moral anxiety is important: just as social anxiety brings an awareness and sensitivity that is central to our ability to understand and navigate the complexities of social life, moral anxiety brings a corresponding awareness and sensitivity that is central to our ability to understand and navigate the complexities of moral life. One might be skeptical. One might allow that moral anxiety can do what we take to be essential to good moral decision making and agency, but maintain that it only makes a small contribution—after all, things like humility, open-mindedness, or just a general curiosity and desire for improvement are more plausible candidates for the kinds psychological mechanisms whose cultivation would matter for good moral decision making and agency. In fact, one might think that given (moral) anxiety’s tendency to bring pernicious forms of motivated reasoning, it will have a detrimental, not just a trivial, effect on decision making and agency. I suspect that much of what drives the thought that moral anxiety is unhelpful—even pernicious—comes from conflating it with related phenomena like social and punishment anxiety. As we’ve seen (§3), these varieties of anxiety are concerned with uncertainty about (respectively) how others will view one’s social worthiness and whether one will be punished for a wrong that one has done. As such, they are more likely to prompt reasoning that is motivated by self-interested concerns (e.g., a biased search for reasons aimed at presenting oneself in a favorable light). By contrast, moral anxiety is concerned with uncertainty about the correctness of a moral decision that one is contemplating. So it will

Moral Anxiety and Moral Agency

189

typically engage reasoning that’s motivated by a concern for accuracy.31 But once we recognize this difference, the thought that moral anxiety will make one particularly susceptible to pernicious forms of motivated reasoning fades—in fact, it’s likely to engage the very sort of constructive reasoning that we’re interested in. Moreover, recognizing these differences also helps explain something that would otherwise be puzzling— namely, why psychologists sometimes find “anxiety” leading to maladaptive behaviors, but other times ones that are quite helpful.32 But there’s still the thought that things like humility, openmindedness, and curiosity are more effective ways to secure the deliberative virtues that underlie good moral decision making and agency. I do not deny that character traits/skills like these are valuable. Humility and open-mindedness bring an important degree of caution and a willingness to consider new or conflicting evidence.33 Curiosity motivates us to work through challenges and to explore issues we find puzzling.34 Clearly these are valuable traits to have. But moral anxiety adds something further. As we’ve seen, moral anxiety functions as a distinctive kind of signal—namely, one that disrupts our current behavior and prompts reassessment.35 So while traits like humility and curiosity play an important epistemic role, they are unlike moral anxiety in that they are not inherently disruptive epistemic mechanisms. Rather, they are traits that take hold after we’ve come to see that we face a puzzle or problem. Thus, moral anxiety’s distinctive value lies in both (a) its ability to make us aware of the need to engage the deliberation and inquiry that humility, open-mindedness, and curiosity help guide and (b) its tendency to give us a motivational push in this direction.36 31 The claim that differences in concern (self-interest vs. accuracy) affect the motivations that drive one’s deliberation and inquiry in these ways has significant empirical support. See, e.g., Chen, Shechter & Chaiken 1996; Tiedens & Linton 2001; MacKuen et. al. 2010. 32 For a similar conclusion about the need to specify different kinds of anxiety in order to explain how “anxiety” can have both positive and negative consequences, see Matthews 1986. More specifically, this experimental work indicates that while forms of anxiety associated with worry, apprehensiveness, and self-reproach tend to negatively influence performance on creativity tasks, forms of anxiety associated with a tense, frustrated drive tend to enhance performance. 33 See, e.g., Arpaly & Schroeder 2014: 241–5. 34 For more, see Morton 2010. 35 For a similar point, see Baumeister & Tice 1990 and Öhman 2008. 36 MacKuen et. al. 2010 provides empirical support for this: it shows that anxiety prompted by challenges to one’s position on affirmative action policy not only leads one to seek out more

190

Charlie Kurth

To further draw this out, consider a modified version of the case of your promise to your Alzheimer’s stricken mother. Her doctors have just told you that it’s time to put her in a nursing home and you’re inclined to go along with their recommendation. But you feel anxious about this—an anxiety that gets you to recall your promise. This both disrupts your initial inclination to just follow the doctor’s advice, and prompts reassessment and reflection. So even if you ultimately decide to put your mother in a nursing home, moral anxiety has played an important role: it has helped you recognize the significance of the situation you face in a way that gets you to think through your decision.37 The upshot, then, is an affirmation of P5—the claim that moral anxiety matters. Though one might worry that moral anxiety is trivial or pernicious, such concerns are misplaced. Moreover, in comparison with things like humility, curiosity, and open-mindedness, moral anxiety has distinctive signaling and motivational features that give it a particularly important role to play in engaging the metacognitive capacities that we’re interested in when we’re interested in what makes for good moral decision making and agency. 4.4 The Pay-Off: Moral Anxiety as an Emotion to Cultivate Recognizing the importance of moral anxiety’s role in moral decision making and agency is significant. It gives us the evidence that we need to accept P5, and so (with C1 and P4) makes a case for the conclusion C2—namely, that moral anxiety is an emotion that we ought to cultivate. But once we see this, we also get a better understanding of why we ought to cultivate moral anxiety. In short, the capacity to feel moral anxiety stands to the virtues associated with good moral decision making and agency (e.g., attunement and receptivity) in the same way that the capacity to feel sympathy stands to the virtue of benevolence, or the capacity to feel anger stands to the virtue of moral outrage/ courage: we can exhibit these virtues in large part because we have the associated emotional capacities. Sympathy, anger, and anxiety bring the information (see note 27) but also prompts an increased willingness to explore new solutions to affirmative action issues—that is, it prompts a kind of open-mindedness and curiosity. 37

As noted in §3.1, while moral anxiety’s value as a disruptive signal is likely to be more beneficial in the earlier stages of moral development, it will still be an important part of the moral psychology of the mature moral agent—though difficult and novel moral decisions will become less frequent as one grows older and wiser, they’re not going to go away.

Moral Anxiety and Moral Agency

191

normatively loaded awareness and motivation that is distinctive of emotional engagement, and that facilitates the moral sensitivity central to virtuous agency.

5. THREE IMPLICATIONS

5.1 Cultivation Recognizing the above relationship between emotions like anxiety, sympathy, and anger on the one hand, and their associated virtues on the other, helps us understand what it means to say that moral anxiety should be cultivated. To say that moral anxiety (or sympathy or anger) is an emotion to cultivate is not to suggest that we want to just experience it more often or more intensely. Rather, it’s to say that moral anxiety (or sympathy or anger) is something we should learn to feel at the right times and in the right ways. In the case of moral anxiety, this will involve, among other things, leaning to recognize when our unease is the result of problematic uncertainty about what to do (as opposed to concerns about, say, punishment). It will also involve developing the ability to channel the motivational dimension of moral anxiety into the epistemic behaviors that will help us address our uncertainty about what to do. Clearly, more needs to be done to flesh this out. While I cannot do that here, two points are worth noting. First, providing an account of cultivation is something anyone who countenances talk of virtue must do—so there’s no special problem for moral anxiety. Second, with regard to moral anxiety, we’re likely in a better position to develop a plausible, substantive answer insofar as there is a wealth of work in clinical and social psychology on which we can draw. 5.2 The Significance of Psychological Conflict The conclusion that moral anxiety is something that should be cultivated comes with a novel, empirically grounded picture of the moral psychology of the virtuous agent. On this picture, good moral decision making and agency are a function of, among other things, one’s tendency to feel morally anxious in the face of hard moral choices. This in turn indicates that a good moral decision is (often) the product of psychological conflict—conflict that leads one to consciously explore, evaluate, and reject possible challenges to the moral decision that one is contemplating. This

192

Charlie Kurth

is not to deny that much of (good) moral decision making and agency is the product of automatic processes.38 But it is to insist that any account of moral decision making must acknowledge that moral decisions will often benefit from the moral anxiety that gets one to consciously explore and assess possible objections to the moral decision that one is contemplating. 5.3 A Challenge to Existing Accounts The picture of the moral psychology of the virtuous agent that comes out of my account of moral anxiety is significant in its own right. But it also challenges other prominent proposals. For instance, the claim that moral anxiety is something that we ought to cultivate contrasts with what we find in Hellenistic (and perhaps Platonic) accounts. The Stoic’s concern with passions, for instance, is (in part) that these emotions bring impulsive action. Thus, passions must be transformed into other affective states or motivational dispositions that will be more conducive to helping one secure the good life—e.g., a lustful passion becomes a desire for a loving friend.39 Applying this to moral anxiety suggests a picture on which it is valuable, if it is, only because it can be transformed into some other affective state or motivational disposition that will help one engage in good decision making. But given what we’ve learned here, this thought is mistaken. Moral anxiety itself can play a constructive role in moral life: it is useful both as a signal and as a motivator. My proposal also challenges the picture of the virtuous agent that we get from contemporary Aristotelians. While these neo-Aristotelians recognize that learning to be virtuous will bring frustration and psychological conflict, they insist that this will not be the case for virtuous individuals. For the virtuous, thought and action is “effortless” and “unimpeded by frustration and inner conflict.”40 Against this picture, our investigation of moral anxiety suggests that anxiety is part of what makes certain decisions—like your decision about your Alzheimer’sstricken mother—virtuous: your concerns about whether you’re doing the right thing, and the deliberation it brings, express an admirable moral sensitivity—an attunement that reveals your understanding of the 38

See, e.g., Railton 2009, Arpaly & Schroeder 2014, chaps 1–3. This sketch glosses over much richness in the Stoic position. See Brennan 2003 for details. 40 Annas 2011: 73. 39

Moral Anxiety and Moral Agency

193

complexity of moral life.41 If that’s right (and more needs to be done to show that it is), then we have reason to question these neo-Aristotelian accounts of the psychology of the virtuous agent.42

6. CONCLUSION

Moral anxiety is not just a familiar part of moral life. It is an emotion that has an important role to play in moral decision making and agency. In particular, we have seen that moral anxiety is like sympathy and anger in that each is crucial to virtuous thought and action. Moreover, the picture of good moral decision making and agency that results from our examination of moral anxiety is one that challenges other prominent accounts. On the picture developed here, the mark of many good moral decisions is that they are the product of psychological conflict—anxiety—that leads one to explore, evaluate, and reject possible challenges to the moral decision that one is contemplating. In a slogan: Much in human affairs is worthy of anxiety. Acknowledgments In working on this paper, I have benefited from conversations with Anne Margaret Baxley, David Brink, Eric Brown, Justin D’Arms, Julia Driver, Elizabeth Foreman, Lizzie Schechter, Walter Sinnott-Armstrong, Jesse Summers, Mike Tiboris, and Eric Wiland. Thanks as well to those in my Fall 2013 seminar on Morality and Emotion and to Mark Timmons and the two OUP reviewers for their valuable input.

REFERENCES

Annas, J. 2011. Intelligent Virtue. Oxford: Oxford University Press. Aronson, E. 1999. The Social Animal (8th ed.). New York: Worth Publishers/Freeman. Arpaly, N. & T. Schroeder. 2014. In Praise of Desire. Oxford: Oxford University Press. Aquino, K., D. Freeman, A. Reed II, V. K. G. Lim, & W. Felps. 2009. “Testing a SocialCognitive Model of Moral Behavior: The Interactive Influence of Situations and Moral Identity Centrality,” Journal of Personality and Social Psychology 97: 123–41. Barlow, D. 2001. Anxiety and Its Disorders, 2nd Edition. New York: Guilford Press. 41 This is a point that has been recognized by others—e.g., Arpaly & Schroeder 2014, 241–5, Stohr 2003. 42 I further develop this argument in Kurth (MS.2).

194

Charlie Kurth

Baumeister, R. & D. Tice. 1990. “Anxiety and Social Exclusion,” Journal of Social & Clinical Psychology 9: 165–95. Brennan, T. 2003. “Stoic Moral Psychology,” in Cambridge Companion to the Stoics, B. Inwood (ed.). Cambridge: Cambridge University Press, 233–56. Burroughs, J. & A. Rindlfleish. 2002. “Materialism and Well-Being: A Conflicting Values Approach,” Journal of Consumer Research 29: 348–70. Chen, S., D. Shechter, & S. Chaiken. 1996. “Getting at the Truth or Getting Along,” Journal of Personality and Social Psychology 71: 262–75. D’Arms, J. & D. Jacobson. 2003. “The Significance of Recalcitrant Emotions (or, AntiQuasijudgmentalism),” Philosophy 52: 127–45. de Sousa, R. 1987. The Rationality of Emotion. Cambridge, MA: MIT Press. Deonna, J. & F. Teroni. 2012. Emotions: A Philosophical Introduction. New York: Routledge. Ekman, P. 1992. “An Argument for Basic Emotions,” Cognition and Emotion 6: 169–200. Faucher, L. & C. Tappolet. 2002. “Fear and the Focus of Attention,” Consciousness & Emotion 3: 105–44. Frank, R. 1988. Passions within Reason. New York: Norton. Gibbard, A. 1990. Wise Choices, Apt Feelings. Cambridge, MA: Harvard University Press. Kitcher, P. 2011. The Ethical Project. Cambridge, MA: Harvard University Press. Korsgaard, C. 2009. Self-Constitution. Oxford: Oxford University Press. Kross, E. & O. Ayduk. 2008. “Facilitating Adaptive Emotional Analysis,” Personality and Social Psychology Bulletin 34: 924–38. Kurth, C. Manuscript 1. “Anxiety, Normative Uncertainty, and Social Regulation.” Kurth C. Manuscript 2. “Emotion, Deliberation, and the Skill Model of Agency.” Lazarus, R. 1991. Emotion and Adaptation. New York: Oxford University Press. Leary, M. & L. Kowalski. 1995. Social Anxiety. New York: Guilford Press. MacKuen, M., J. Wolak, L. Keele, & G. E. Marcus. 2010. “Civil Engagements: Resolute Partisanship or Reflective Deliberation,” American Journal of Political Science 54: 440–58. Marks, I. & R. Nesse. 1994. “Fear and Fitness: An Evolutionary Analysis of Anxiety Disorders,” Ethology and Sociobiology 15: 247–61. Mathews, A. 1990. “Why Worry? The Cognitive Function of Anxiety,” Behaviour Research & Therapy 28: 455–68. Matthews, G. 1986. “The Effects of Anxiety on Intellectual Performance: When and Why are They Found?” Journal of Research in Personality 20: 385–401. McClure, S. M., M. M. Botvinick, N. Yeung, J. D. Greene, J. D. Cohen. 2007. “Conflict Monitoring in Cognition-Emotion Competition,” in Handbook of Emotion Regulation, J. Gross (ed). New York: Gilford, 204–28. Messer, S. 1970. “The Effect of Anxiety over Intellectual Performance on ReflectionImpulsivity in Children,” Child Development 41: 723–35. Miceli, M. & C. Castelfranchi. 2007. “Anxiety as an ‘Epistemic’ Emotion,” Anxiety, Stress & Coping 18: 291–319. Mor, N. & J. Winquist. 2002. “Self-Focused Attention and Negative Affect: A Meta-Analysis,” Psychological Bulletin 128: 638–62. Morton, A. 2010. “Epistemic Emotions” in Oxford Handbook of Philosophy of Emotion. P. Goldie (ed.), Oxford: Oxford University Press, 385–400. Nussbaum, M. 2001. Upheavals of Thought. Cambridge: Cambridge University Press. Öhman, A. 2008. “Fear and Anxiety,” in Handbook of Emotions. M. Lewis, J. M. HavilandJones, and L. F. Barrett (eds.), New York: Gilford Press. Prinz, J. 2004. Gut Reactions. Oxford, Oxford University Press. Railton, P. 1986. “Moral Realism,” Philosophical Review 95: 163–97.

Moral Anxiety and Moral Agency

195

Railton, P. 2009. “Practical Competence and Fluent Agency,” in D. Sobel & S. Wall (eds.), Reasons for Action. New York: Cambridge University Press, 81–115. Roberts, R. 2003. Emotions: An Essay in Aid of Moral Psychology. Cambridge: Cambridge University Press. Schwartz, N. & G. Clore. 2007. “Feelings and Phenomenal Experiences,” in E. T. Higgins & A. Kruglanski (eds.), Social Psychology: Handbook of Basic Principles (2nd edition) New York: Gilford. Solomon, R. 1973. “Emotions and Choice,” Review of Metaphysics 27: 20–41. Stohr, K. 2003. “Moral Cacophony: When Continence is a Virtue,” Journal of Ethics 7: 339–63. Tiberius, V. 2008. The Reflective Life. New York: Oxford University Press. Tiedens, L. & S. Linton. 2001. “Judgment Under Emotional Certainty and Uncertainty,” Journal of Social and Personality Psychology 81: 973–88. Velleman, J. D. 2003. “Don’t Worry, Feel Guilty,” Royal Institute of Philosophy Supplement 52: 235–48.

9 The Narrative Calculus ANTTI KAUPPINEN

In this paper, I undertake the potentially quixotic task of quantifying the prudential value of a good life story (or life history, as some might prefer to call it). Whether or not the effort turns out to be successful, it forces us to think systematically about the good-making features of life stories, instead of merely relying on intuitions about cases. If we can identify independently plausible principles, they will also support particular narrativist intuitions. Let’s begin with a well-known story deriving originally from Michael Slote. Imagine two politicians, Uphill and Downhill. Uphill begins in political wilderness, isolated, bitter, and unhappy, and then in the middle of his career becomes successful and happy, finishing on a high note. Downhill begins with a meteoric rise to the top, and is just as successful and happy as Uphill ends up being. But in the middle of his career, Downhill succumbs to a scandal, and disappears in the political wilderness, isolated, bitter, and unhappy, ending on a low note. Even if the total sum of the kind of things that are intrinsically good for a person at a given moment considered in isolation—such as, perhaps, happiness, desire-satisfaction, exercise of human potential, or goal-achievement—is the same in these two lives, as it surely may be, many have the intuition that Uphill’s life is better. They’d rather lead that kind of life, and want it for people they care about. As Slote puts it, “Without hearing anything more, I think our natural, immediate reaction to these examples would be that the first man was the more fortunate” (Slote 1983, 23). If this reaction stands up to scrutiny, it seems that the shape of a life matters in addition to the sum of moment-by-moment value it contains. What best explains the apparent significance of a life’s shape—for short, the shape-of-a-life phenomenon? Perhaps the most popular account is what I call temporal distributivism. According to it, the order in which non-relationally good and bad things happen in someone’s life

The Narrative Calculus

197

affects its prudential value. I argue that this is false, but we may be mislead into thinking otherwise by the fact that certain temporal distributions are characteristically linked with prudentially good life stories. According to narrativism, in turn, it is the narrative (or narratable) relations among events and segments of one’s life that explain why Uphill’s life is better. While narrativism has had influential defenders, such as David Velleman (1991/2000), Jeff McMahan (2002), John Martin Fischer (2005), and Connie Rosati (2013), there are few if any theories about what makes for a good life story. In Section 2, I sketch such an account. I first turn briefly to narratology to argue that our lives are narratable because narratives concern the adventures of planning agents: events have narrative significance in virtue of their role in goal-directed pursuits. Contrary to some narrativists, I argue that it is not the stories that we actually tell about our lives that make a difference to its value, but rather the objective narratable relations that obtain between the events that constitute our lives. (This is why some might prefer to call the object of my inquiry life histories.1) I argue that the narrative value of events is a multiple of three main factors: their positive or negative causal contribution to the agent’s present or future goals, the value of those goals, and the degree to which success in achieving the goal is deserved in virtue of exercising agential capacities. This Narrative Calculus yields a prudential narrative value for each event in an arbitrary period (or at least provides a heuristic for arriving at one), and allows for adding up those values to arrive at the value of the whole story for an agent. In spite of being additivist, it is holistic in that momentary narrative values of events can be retrospectively changed by their instrumental contribution and prospectively changed by desert. Since the Narrative Calculus assigns a value to events on the basis of their contribution to giving a progressive shape to one’s life, I call my variant of narrativism progressivism. Progressivism cashes out value-relevant narrative characteristics in teleological terms. In Section 3, I argue that doing so suffices to capture and buttress the kind of intuitions that motivate the narrativist approach. Section 4 defends the view against criticisms leveled recently against similar accounts. In the final section, I make a few observations about the

1

This term was suggested to me by an anonymous referee for this volume.

198

Antti Kauppinen

deeper rationale of progressivism. The core idea is that narrative value is the measure of how we fare as temporally extended planning agents vulnerable to fate. We are not just subjects of better or worse experience at various moments, but active shapers of our stories over time. Considered under this aspect, our lives can go better or worse for us in virtue of having a more or less progressive shape. 1. FROM TEMPORAL DISTRIBUTION TO NARRATIVE SHAPE

To state the various views aiming to capture or debunk the shape-of-alife phenomenon, I need to introduce some terminology. I will call an individual’s level of well-being (the intrinsic value of life for her) at a moment the momentary value of life for her at that time. It could be that momentary value is determined exclusively by the intrinsic features of an individual’s life at the time, such as her hedonic level or perhaps level of actualizing human potential (I’ll try to remain neutral on these issues here). Call this thesis moment internalism, and the kind of momentary value that depends only on non-relational features of life at a time ‘atomic value’. Its denial, moment externalism, maintains that it is possible for momentary value to depend on what happens at other times, that is, the relational features of the individual’s life at the time. The simplest way to arrive at the value of a whole life (or a segment of a life) for an individual is adding up the momentary values at each moment during her life. Call this principle additivism. If additivism is combined with moment internalism, we get a thesis concerning the value of a whole life that I’ll call atomism. For a very clear example of an atomistic theory, consider Fred Feldman’s (2004) version of hedonism. According to it, the intrinsic value of someone’s life for her is simply the sum of hedonic values at all the times in her life. The shape-of-a-life phenomenon poses a challenge to this theory and other forms of atomism. If Uphill’s life is better than Downhill’s in spite of containing the same sum of pleasure, self-actualization, and other atomic goods, something has to give. This motivates adopting some form of holism. Since atomism is a combination of additivism and moment internalism, holism comes in two main varieties. Internalist holism keeps moment internalism, but rejects additivism. The value of a whole life is not the sum of momentary values at each time during it, but contains some additional factor. Velleman’s view is of this type. Externalist holism, in

The Narrative Calculus

199

turn, keeps additivism, but rejects moment internalism. It holds on to the idea that the value of a life is the sum of how well we fare at each moment, but denies that momentary value is determined only by what happens at the time. My view will be a version of externalist holism. With these terminological stipulations in place, I can state the most popular view that tries to capture the shape-of-a-life phenomenon: Temporal Distributivism The temporal arrangement of atomic values non-instrumentally matters to the overall welfare value of a life.2

The most common form of temporal distributivism holds that it is better to start out with low atomic well-being and end up with high atomic well-being than vice versa (e.g. Kamm 2003, 222). Call this type of view improvementism (Bradley 2011, 48). Other variants are possible: Slote (1983), for example, thinks that atomic benefits in the ‘prime of life’ count for more in overall evaluation than benefits early in life. But besides being ad hoc, Slote’s type of view doesn’t account for common intuitions about the preferability of improvement (in some sense) in more restricted periods. Joshua Glasgow offers a clear example: [I]magine that you can choose between living a year that begins unemployed and lonely and ends up with you in love and happily employed, or living a year that begins with you being involved in fulfilling work and head-over-heels in love and ends up out of work and alone. All else is equal: your achievements and failures even out (the love and work you find or lose is found or lost by sheer luck), your meals contain the same foods, and so on. ( . . . ) Despite other things remaining constant, the uphill year seems preferable to the downhill year. (Glasgow 2013, 668–9)

Why is the Uphill year preferable? The improvementist holds that “It is bad to be worse off than you used to be” (Glasgow 2013, 668). Importantly, according to improvementism, this is not the case because people are made unhappy by their life getting worse (this would mean that the sum total of atomic values would be different in the two lives), or because people happen to prefer their lives to get better in terms of 2

I draw here on Dale Dorsey’s (2015) formulation.

200

Antti Kauppinen

atomic benefits (this would leave it contingent which year it is better to live).3 It claims, instead, that improvement in atomic value is an additional intrinsic good in Uphill (Glasgow 2013, 670). I share the intuition that Uphill is a better year to live than Downhill. But I don’t believe that improvementism is the correct explanation. In fact, I believe that the temporal distribution of well-being, as such, doesn’t make any difference to the subject’s total welfare. It only misleadingly seems so to us. Why? It could be, as Fred Feldman (2004) suggests, that we tacitly attribute additional pleasure deriving from observing improvement in one’s life to Uphill, although we’re not supposed to. In that case, Uphill’s life will really be better, but this is straightforwardly explained by the atomist. Or alternatively, perhaps we find “a sort of excellence, or beauty, or appropriateness” (135) in Uphill’s life that is missing in Downhill’s. But this intuition, Feldman points out, does not target the prudential value of the life, its value for Uphill himself, but rather a kind of aesthetic value. This is an important challenge that it is good to bear in mind; I will return to it in Section 4. The two debunking explanations that Feldman offers may explain away intuitions about some cases with an Uphill/Downhill structure. But I believe that the actual Uphill/Downhill cases offered by shapists are more robust. However carefully I focus on prudential value and remind myself to regard the sum of atomic values as identical, Uphill still seems like the life I’d prefer for someone I care about. But it doesn’t follow that it is better because of the temporal arrangement of moment-internal values. Indeed, according to (strong) narrativists, an Uphill life is better than a Downhill life if (and only if) and because its story is prudentially better. The narrativist thus offers a kind of debunking account of why improvement seems to matter: it is typically, but not necessarily, associated with a better underlying story. The best-known, if partial defense of narrativism is offered by David Velleman (1991/2000). Velleman is in fact best classified as a hybrid theorist, since he also endorses temporal distributivism. Nevertheless, he influentially maintains that “The meaning of a benefit depends not only on whether it follows or precedes hardships but also on the specific narrative relation between the goods and evils involved” (Velleman

3

These points are nicely made in Dorsey (2015).

The Narrative Calculus

201

1991/2000, 63). Why think so? Because, first, two lives that have a different narrative structure can have a different prudential value, even if we hold fixed both the sum and the temporal distribution of atomic values. And second, two lives that do not differ narratively and contain the same total of atomic value will be equally good for someone, even if the temporal distribution of value is different. Here’s Velleman’s famous example of the first kind of scenario, with names added for convenience: Megan/Betty “In one life [Megan] your first ten years of marriage are troubled and end in divorce, but you immediately remarry happily; in another life [Betty] the troubled years of your first marriage lead to eventual happiness as the relationship matures. Both lives contain ten years of marital strife followed by contentment; but let us suppose that in the former, you regard your first ten years of marriage as a dead loss, whereas in the latter you regard them as the foundation of your happiness.” (Velleman 1991/2000, 65)

The Betty scenario seems better for you, because the first ten years are not a dead loss but the foundation of happiness (let’s assume that you’re correct in regarding them as such). As Velleman puts it, “The bad times are as bad in both lives, but in one they are cast off and in the other they are redeemed. Surely, these two decades can affect the value of your life differently, even if you are equally well off at each moment of their duration.” (1991/2000, 65–6). This is a very plausible explanation. Velleman also offers a number of other scenarios to the same effect, but no systematic theory of what makes a life story prudentially good. As Glasgow observes, temporal distributivists can accept Velleman’s claim if they modestly maintain that temporal distribution of atomic values is just one factor in determining whole life value. Yet there is reason to think that even this more modest thesis is false. When temporal distribution of goods and evils is dissociated from narrative progress— which is no easy task, since the two tend to go together—we can see that on its own, the former makes no difference to the value of the life. Dale Dorsey (2015) offers a number of examples, of which the simplest and perhaps most convincing involves two people in an Experience Machine. One has an Uphill experience and the other a Downhill experience. Arguably, there is no narrative difference in such cases: both are simply subject to an illusion, which is also equally pleasant overall. Arguably,

202

Antti Kauppinen

also, there is no difference in how good the lives are—meaningless gains and losses do not make a difference to welfare. If so, all that matters for how someone fares is the sum of atomic benefits and the shape of her story.

2. A NARRATIVE CALCULUS

But what kind of life story is it good for one to have? Can we go beyond examples and formulate some general principles? Let us begin by acknowledging that the common terminology is potentially misleading. A narrative or story in the most obvious sense is a representation of a life, or at a minimum a sequence of events. In my view, it is not intrinsically valuable that some kind of story can be told of our lives.4 After all, the many different true stories could be told of the same series of events. What is good for us is rather that certain narratable relations obtain between the events that constitute our lives. In deference to convention, I’ll continue to talk about narrative value, but it is worth bearing in mind that I am really referring to the intrinsic prudential value of narratable features of a life. What I say is therefore compatible with claiming that people who do not see their lives in narrative terms can lead perfectly good lives, as Galen Strawson (2004) famously maintains. Indeed, they can lead lives that are high in narrative value in my sense. What Is a Narrative? Still, it is worth thinking about narratives in order to understand what kind of relations are potentially narratable. I said that a narrative represents a sequence of events, but not all such representations are narratives. Consider the following: 4 To be sure, it can be instrumentally good for us if we tell ourselves a story of our life that makes us feel good about it, since it contributes to our level of happiness at the time of telling. I discuss the relationship between personal stories and happiness understood as a comprehensive affective condition in Kauppinen (2013). I thus partially agree with Rosati when she says that storytelling is good for us because it helps us “see ourselves and our lives in a way that supports our sense of our own worth, that stimulates us to move forward, that helps to secure our sense of who we are and a sense of direction, and that moves us forward as a matter of our own volition” (Rosati 2013, 45). This is only partial agreement, however, since on my view, such storytelling has only instrumental value for happiness. Rosati explicitly denies this (Rosati 2013, 48), but I suspect it is because she doesn’t think things like sense of worth or direction are components of happiness, while I argue they are.

The Narrative Calculus

203

9 am meeting with Arya. 9.30 am meeting with Benjen. 10 am meeting with Cersei. The extinction of dinosaurs was caused by a change in the global temperature, which resulted from a comet hit. Finely chop a carrot, two stalks of celery, and a large onion. Lightly fry them without colouring for at least 20 minutes in olive oil and butter. When they are soft, add a dash of salt and white wine.

The first is a mere list of events, perhaps a diary entry, the second a (dubious) causal explanation, and the third a recipe for a delicious soffritto. There has been a lot of debate and little agreement among narratologists about what minimally distinguishes narratives from such non-narrative representations of event sequences. But the various definitions and theories tend to point in the direction of goal-directed human agency extended over time. For example, Jerome Bruner says that the subject matter of narratives is constituted by “human or human-like intentions and the vicissitudes and consequences that mark their course” (Bruner 1986, 13). Kenneth and Mary Gergen observe that “An intelligible story is one in which events are selected that serve to make the goal more or less probable” (Gergen and Gergen 1988, 21). Finally, David Herman argues in the same vein that “the property of narrativehood attaches to sequences of states, events, and actions that involve an identifiable participant or set of participants equipped with certain beliefs about the world seeking to accomplish goal-directed plans” (Herman 2004, 90). Without getting any deeper into the debate among narratologists, I’m going to assume that narratives concern the exercise of planning agency over time and its consequences. Once we move from proto-narratives describing simple agent-involving transitions from one state to another to descriptions of incidents that occur in pursuit of long-term goals, further characteristic (and perhaps essential) features of stories emerge. Here is a sketch of the standard, basically Aristotelian model that one can find in screenwriting guides and narrative psychology (McAdams 1993, 25–30). First, there is some initiating event. The normal run of things is disrupted by something that calls for action. Maybe a prince and a princess are supposed to get married, but a dragon snatches him away. The princess adopts the goal of rescuing him. Here begins the rising

204

Antti Kauppinen

action or complication of the story: the protagonist engages in instrumental reasoning and makes an attempt to reach the goal. Maybe the princess builds a ladder to climb up to the dragon’s lair. But we don’t always get what we want so easily: we face obstacles and challenges. The dragon burns down the ladder. The princess has to rethink, try again, seek help, and change. Maybe she enlists a giant eagle to carry her up to the lair. Eventually, the story reaches a turning point: either the protagonist exhausts the available means and fails, or is guaranteed to succeed as long as nothing out of the ordinary happens. So in the dragon’s lair, the princess slays the dragon and rescues the prince. Resolution or denouement follows: they marry. But only in fairy tales do they live happily ever after. In real life, one episode or chapter of life is followed by another. As long as we live, we face new problems, and must seek new solutions. Contrary to what Alasdair McIntyre (1981) claims, life doesn’t really have a beginning, middle, and end in the Aristotelian fashion. It is not about solving a problem to restore equilibrium. Nonage isn’t a complication and dotage isn’t a resolution. Rather, life consists of a chain of chapters that can be linked in various ways. If the whole has a narrative arc, it has to be understood in different terms. I’ll say that a whole life story can be predominantly progressive, regressive, or stagnant, in spite of lacking the classical three-act structure. But how can we talk about progress across distinct projects, in the absence of a single ultimate telos? The answer must lie in the way earlier projects influence later ones or in the way later projects alter the meaning of earlier ones. I’ll discuss different possible relations below. When Is a Story Prudentially Good? Our aim in working out a narrative calculus is to give an account of the contribution of individual events to making a story prudentially good. Since stories concern the pursuit of goals, the most obvious way your story can go badly for you, other things being equal, is if you fail to reach your goal. In turn, the good kind of story arc will result in success, again other things being equal. If we focus on the various actions and mere happenings that take place in the course of your story, we see they will either promote or hinder that goal, or don’t form part of the story. (Goalrelated events are the basic atoms of the story.) If your goal is to rescue the mountain gorillas in the Congo, a successful fundraiser or the passage of

The Narrative Calculus

205

a new conservation law will move your story forward, while contracting a tropical disease will slow it down. The narrative value of these events hangs, in part, on just how much they contribute to reaching or frustrating the goal. This is what I’ll call the Instrumentality Factor (IF). Unfortunately, it is not easy to quantify the contribution of individual events to reaching a goal. It is tempting to think that IF hangs on the extent to which an event increases or decreases the odds of achieving the goal, but I’m not at all sure if that will get the right results. After all, there isn’t even a consensus on when one event is the cause of another, not to mention on the measure of one event’s causal contribution towards another. So I will have to rely on an intuitive understanding of promoting a goal. It will turn out to be important that what we do can contribute not only to the goals we pursue at the time (our concurrent goals), but also subsequent goals. I’ll return to this point below. Finally, it’s perhaps worth emphasizing that Instrumentality Factor is a component of intrinsic value, not instrumental value. It is intrinsically good for us that our efforts turn out to be instrumental in bringing about a goal. This may be most obvious when our goal is to benefit someone else. Suppose I succeed in such a project without ever coming to know it. In that case, the actions that bring about benefit for another are not instrumental in bringing about something that is intrinsically valuable for me. But I maintain it is still intrinsically good for me that I’ve spent my time and energy usefully—that is what contributes to my well-being. The second feature that makes a difference to how good it is for us to live a story is what I will call the Value Factor. It doesn’t just matter to the progress of our story how effective we are in pursuing our goals, but also what direction the story is heading—how valuable those goals are, non-narratively speaking. I take it to be quite obvious that the story of a life spent in the challenging and successful pursuit of counting blades of grass is not the story of the best kind of life for the protagonist. If our efforts can be wasted by being unsuccessful, they can just as surely be wasted by being misdirected. I emphasize that this is a matter of selfinterest: it is not, or not just, that the subject’s life is morally better, for example, if she pursues objectively valuable goals, but that it is better for her to invest her limited time in genuinely worthwhile activities. I’m not going to assume any particular view of which things are non-narratively valuable. Plausibly, one’s own happiness has some non-narrative value, and justice or beauty or understanding have more. The Value Factor

206

Antti Kauppinen

does not play a large role in the debate about narrativism, so I will not defend it further here. The final factor of narrative value I’ll discuss here, the Desert Factor, is somewhat complex. The rough idea is that we have more at stake in goals that we have acted to bring about than in goals we have done little to achieve. So if I spend day and night working on my rice paddy to have enough to eat for the winter, and you do hardly anything on yours, success or failure contributes less to the value of your story. There’s various ways people have proposed for capturing this element, but I think it’s best to talk about a variety of desert I’ll call agential. The idea is that in a familiar but non-moral sense, we deserve success in a pursuit when we exercise the capacities that make us the kind of agents we are. This is related, but not identical with the idea that effort is the desert base. I prefer to focus on exercise and development of our capacities, since mere effort is often stupid: you don’t do well as an agent if you expend a lot of physical effort instead of making use of intelligence and self-control, for example. This is the point at which considerations relating to responsibility and determinism interact with narrative value: if it were the case that we weren’t capable of the kind of agency in virtue of which we deserve success, but mere drifters in the flow of events, our lives would have little narrative value. I believe the best way to calculate the Desert Factor is allowing the degree of desert to intensify the narrative value of all events promoting or hindering the goal. The reason for thinking this way is that not only actions, but also non-actional events are better for our story, when achieving the goal they advance is more deserved. If I’ve exerted myself more than you to make the rice grow, the rain that saves both our crops means more to me than to you. So non-actional events that turn out your way can also positively contribute to the narrative value of your life for you. Nevertheless, the more success hangs on non-actional events, the lower the Desert Factor will be. So other things being equal, the less your success is explained by lucky happenings, and the more it is explained by the exercise of your capacities, the better your story is for you. If all you did to grow your rice was to put out a sack of grains, which were serendipitously spread on the paddy by ducks, fertilized by your neighbour’s cows, and watered by the rain while you were cavorting with your neighbour’s spouse, your agricultural success contributes little to the value of your story. Similarly, actions and mere happenings that

The Narrative Calculus

207

hinder success are the worse the more richly deserved the success is. If you exerted your intelligence to the utmost to be the first to solve a math problem, and failed just because a meteor hit your house as you were almost there, being hit by a meteor is worse for you than it would have been had you made only nominal effort. Before summing up these factors in a narrative calculus, I need to say a little more about narrative relations on a larger scale. It turns out to be crucial for explaining narrative intuitions that events can play a role in the pursuit of non-concurrent goals. There are at least three kinds of narrative progress among episodes or segments of one’s life. The first and simplest is if two episodes are unified by a higher-order goal. Maybe my higher-order aim is to bring about basic income. I first pursue it by campaigning for the Green Party, who has it in its platform, and when that fails, engage in a media campaign that persuades the Conservative leadership of the idea. I’ve made narrative progress in the sense of approaching realization of the same goal. What I’ve already said about Instrumentality and Desert suffices to capture this. However, there is room for large-scale progress even if there’s no unifying aim organizing distinct pursuits. Most importantly, it may be that earlier efforts make later pursuits more successful. First, they may transform or position the subject in such a way that she aims at more valuable goals than she would otherwise. This is most dramatic in the case of redemption stories, in which (say) an alcoholic has to hit rock bottom to come to realize that taking responsibility for one’s family is more important than chasing the fleeting high.5 But you can, of course, also gain insight from doing something that is successful on its own terms, such as a career in science. Second, earlier segments may be instrumental in making one’s later pursuits more successful than otherwise, either by way of improving the agent’s abilities or changing her environment, including her external resources. Successfully completing a PhD (somewhat) improves your odds of success in pursuing an 5 The psychologist Dan McAdams (2006) argues that redemption stories form a characteristically American genre of thinking about one’s life. People who think in redemptive terms will describe something bad that happened to them, but then say that it “ended up making me a better person,” or “taught me to love my family,” or “toughened me up for the challenges ahead.” (McAdams 2006, 24) Redemption stories contrast most sharply with contamination sequences, in which “the bad ruins, spoils, sullies, or contaminates the good that precedes it” (McAdams 2006, 211).

208

Antti Kauppinen

academic career, while it won’t help you with a career in bartending. And, again, growing up as the child of an alcoholic can have the partially redeeming feature of “toughening you up” for later pursuits. It is to capture this potential narrative role that the Instrumentality Factor needs to include a reference to contributing to non-concurrent goals. And finally, perhaps earlier efforts can make later success more deserved even if they weren’t directed toward the later goal. Here I’m thinking of someone who tries his hand at everything but always fails for reasons beyond his control, and then finally succeeds at something completely new, like being a cook. It isn’t outrageous to think that such a person’s eventual success is agentially deserved, and better for her than for someone who hasn’t experienced a similar odyssey. Since the narrative value of an event is the result of the interaction of all these factors, it is best to express it as a multiple of them: The Narrative Calculus (NC) The intrinsic prudential narrative value NV(e) of an event e = contribution of e toward the agent’s concurrent and non-concurrent goal(s)  objective nonnarrative value of the goals e contributes to  degree to which success in achieving the agent’s goal(s) is deserved in virtue of her actions.

Let me immediately grant that the narrative calculus is inferior to the hedonic calculus that inspires it in terms of precision (though we shouldn’t exaggerate the exactitude of the hedonic calculus either). There is no algorithm for assigning values for the different factors involved. In spite of the rather grandiose name, I only mean to provide a heuristic for the purposes of comparing the narrative values of options. Further yet, the three factors that I list are almost certainly not the only ones that matter. For example, not all of our goals are equally comprehensive or central to defining who we are, so it would make good sense to factor in the time-relative centrality of the goal that an event promotes or hinders. Our interest is rarely in the narrative value of a singular event, but rather in the narrative value of a sequence of events or a possible life. Since NV is defined for events, there is a straightforward way of calculating the narrative value of a sequence for an agent, namely adding together the NVs of all the events that comprise it:

The Narrative Calculus

209

Additivity of Narrative Value The prudential narrative value of an agent’s life during period from event e1 to event en, NV(e1–en) = the sum of NV(e1)...NV(en)

When the events between which narrative value is summed up are birth and death, we get narrative value for the whole life: Whole-Life Narrative Value The prudential narrative value of a whole life L, WLNV(L) = NV(birthx–deathx)

It bears emphasizing that although NV is additive, it is not atomistically additive: we can’t look at the value of what happens at each moment in isolation from what happens at other times, and sum the atomic values up to arrive at the value of the life for the agent. This is because the NV of an event depends on what happens at other times. Given the Instrumentality Factor, the extent of the contribution of an event to the agent’s concurrent goals depends on what happens at other times, and its contribution to the agent’s non-concurrent goals on what goals the agent will have and how her pursuit will be affected by the event. Given the Desert Factor, the NV of an event depends also on what the agent deserves in virtue of actions at other times during pursuit of the goal (and perhaps actions in pursuit of other goals, if the desert base is broader). This is why NC is holistic in spite of being additivist. Since NC aims to capture the contribution of events to narrative progress through one’s whole life, I will call the view that endorses it as the measure of holistic value progressivism. Progressivism vs. Achievementism The picture of narratable features of life that I sketched on the basis of narratological literature emphasizes the importance of goal-directed pursuits for narrative value. I will finish this section with a brief comparison between progressivism and achievementism. Achievementists maintain that the achievement of goals that one has invested in, or perhaps a valuable or rational subset of them, independently contributes to the value of one’s life. According to Thomas Keller’s (2004) unrestricted achievementism, the attainment of any goal through one’s own efforts is good for one (and the better the more effort one has made), while

210

Antti Kauppinen

Thomas Scanlon’s (1998) rational aim view counts only success in achieving rational goals. Achievementist views no doubt tap into some of the same intuitions as progressivism. But there are at least three major differences. The first has to do with bookkeeping. Goal-achievement is a matter of attaining a goal through one’s actions. NC factors out the value of goal-achievement to the value of the actions that promote the goal and the value of goalattainment itself. The latter may have little or no narrative value. What is narratively paramount is moving our story forward in the direction of a (valuable) goal in a way that will result in deserved success, not goalattainment as such.6 Second, achievementists think that the more one has already invested in the pursuit of a goal, the more valuable its achievement is, other things being equal. The notion of investment can be cashed out in terms of effort (Keller 2004) or self-sacrifice (Portmore 2007). Portmore argues convincingly that mere effort will not do as a measure of value. In one of his examples, Fred and Greg both spend the same time in training for the Olympics with the aim of winning a medal, making an equal effort, but while Fred loves nothing better than solitary practice, Greg doesn’t like it, and would rather be with his family. Portmore argues that winning a medal is more valuable for Greg, since he’s sacrificed more for the goal (2007, 20–1). Can progressivism, which says nothing about effort or sacrifice, capture this intuition? Yes. Progressivism will understand investment differently, again focusing on the NV of one’s actions rather than goal-achievement. If one has invested in a goal, then one has performed actions in order to promote it. Other things being equal, the more one has exercised one’s capacities, the more one has to lose in case it turns out the actions together fail to sufficiently promote the goal, whether it is because of some insurmountable obstacle or abandoning the goal for a different one. (In either case, there will be the possibility of redemption.) In the Fred/Greg case, both perform the same actions. But as I see it, Greg must exercise his capacities to a higher degree in overcoming his reluctance to train and his desire to be with his family. 6

So, for example, if you achieve your goal posthumously, it increases your level of welfare during those times of your life you exercised your capacities in its pursuit. It turns out you weren’t wasting your time. This has the advantage that there is no need to postulate a level of welfare while you’re dead.

The Narrative Calculus

211

His willpower works overtime. (Something analogous will be the case whenever you knowingly sacrifice something for the sake of something else.) So Greg is more deserving of success. That’s why his story is better than Fred’s, if they both win a medal. The third difference is that because achievementism doesn’t factor out the value of goal-promoting actions, it also doesn’t capture the significance of promoting future goals. As Portmore rightly observes, achievementism has no room for redemption of past efforts that were invested in a different goal (2007, 23–4). Focusing on individual aims misses the bigger picture, even if narrative meaning hangs in part on goal-achievement. In including the contribution to future goals, NC resembles Portmore’s Not-for-Naught view, according to which the redemption of one’s self-sacrifices in itself contributes to one’s welfare (Portmore 2007, 13). But as before, the notion of sacrifice plays no role in progressivism. 3. TESTING THE THEORY

The aim of the Narrative Calculus is to give a systematic account of what makes one life narratively better for the subject than another. But does progressivism capture the intuitions that motivate the narrativist approach in the first place? Let us start with the Uphill/Downhill case. I believe our judgment doesn’t result from the mere fact that the bad segment precedes the good in one and succeeds it in the other. It is rather that we assume that Uphill’s struggles in early life instrumentally contribute to his success in later life (giving their Instrumentality Factor a boost) and maybe teach him to pursue more valuable goals. Meanwhile, perhaps, Downhill’s early success has the opposite effect of contributing to later failure, as his pride goeth before the fall. We may also think that given his early tribulations, Uphill’s in itself equal success in pursuing his political goals is more deserved than Downhill’s, which comes about with relative ease. If this is the case, the Desert Factor must be based in part on non-concurrent goal-directed activities, as speculated above. What about the Megan/Betty case? From the perspective of progressivism, the features that make the Betty choice better turn out to be similar to the Uphill/Downhill scenario. Call the husband in the scenarios Don. Only in the Betty life, the events of the first marriage—the marriage counseling, the painful arguments—end up having a positive

212

Antti Kauppinen

instrumental contribution to later success, so they have a higher narrative value. That’s what makes the scenario a classic redemption story. It is of course a possibility in real life that struggles in one marriage pay off in a new marriage to someone else—but in the case as described, it is postulated that the success of the Megan scenario doesn’t in any way owe to the events of the first marriage. Further, Don’s later happiness is more deserved with Betty than with Megan, since it results directly from his intentional actions and so owes less to the sheer luck of meeting a compatible person. A third kind of case that narrativists often appeal to involves contrasting a windfall with an earned reward. For example, Simon Keller (2004, 29–30) contrasts two authors, Bill and Steve, who each aim to write and publish a novel. Only Bill succeeds, and his life improves in many non-narrative ways, such as increased self-esteem and enjoying the cathartic experience of being in the public eye. Steve, however, fails to complete his novel, but wins the lottery, and enjoys the same atomic goods as Bill, forgetting all about his failure to complete the novel. Keller maintains that Bill’s life is better of the two, because the story of this part of his life is “a story of hard work leading to a successful culmination” (Keller 2004, 30). If we take the case as Keller describes it, it is trivial for progressivism, as only Bill’s actions sufficiently promote his goal, while Steve fails. It is somewhat more interesting if we assume that both really aimed at self-esteem and fame, but while Bill achieved this through his own efforts, Steve achieved it by chance. In that case, the difference comes to the Desert Factor: while both reached an equally valuable goal, the narrative value of the events that promote Bill’s goal is enhanced by his deserving the success more as a result of bringing it about (to a greater degree) by the exercise of his own capacities. Fourth, NC helps us see why the persuasiveness of Glasgow’s Uphill/ Downhill year case after all hangs on its narrative structure. Recall that the comparison is between “living a year that begins unemployed and lonely and ends up with you in love and happily employed, or living a year that begins with you being involved in fulfilling work and headover-heels in love and ends up out of work and alone” (Glasgow 2013, 668). I believe we only have the intuition that Uphill is better because we assume that the characters aim at employment and love all along. For some reason we’re not given, Downhill’s pursuit of these goals eventually fails, while Uphill’s eventually succeeds. It is thus not credible for

The Narrative Calculus

213

Glasgow to postulate that there is no narrative difference between the cases.7

4. OBJECTIONS

In the previous sections, I have sketched an account of what makes narratives prudentially good, and used it to explain common narrativist intuitions. But what about objections to narrativism? One common objection is that stories are one thing and life another: we can tell many different true stories about a life, so how could a story determine how good someone’s life is for her? This objection fails to make the distinction between a narrative and the narratable features of a life that it is about. As I have emphasized, NC is exclusively concerned with the latter. While any actual narrative is inevitably selective, the features of a life that are relevant to NV include all potentially narratable events. Another common objection is that narrativists confuse prudential and aesthetic value—that life stories can be more or less entertaining or dramatic, but we shouldn’t think that aesthetically better stories are better for the protagonist.8 It is one motivation for developing NC to make clear that what is at issue is not aesthetic judgment. A peak-to-peak type of life story may be aesthetically boring, but that doesn’t reduce its prudential value—other things being equal, we are better off without episodes of failure and hardship. There are, however, more serious and specific challenges. I will focus here on potential obstacles for combining holism with moment externalism, as it is central to the way I have formulated NC that later events can alter the momentary value of earlier events. In this respect, it contrasts clearly with Velleman’s view, which does not allow for retrospective 7 Elizabeth Harman suggested a better comparison. Suppose one gay man lives through social change from exclusion to acceptance (and thus experiences a change from fear to joy), and another through social change from acceptance to exclusion. Importantly, the pattern of change in atomic value in no way owes to what the protagonists personally do. Other things being equal, wouldn’t we still prefer to lead the former life? That seems right. Here’s a speculative response to this kind of intriguing case. For the gay community as a whole, the first is a tale of narrative success and the other of failure (during the characters’ lifetime). Perhaps when people identify with a group, the story of their life becomes intertwined with that of the group, and gains or loses value accordingly, independently of their individual role. 8 Fischer (2005, 380) seems content with narrative value being a species of aesthetic value, but he is an exception in this regard.

214

Antti Kauppinen

change of momentary value—he says that “when subsequent developments alter the meaning of an event they can alter its contribution to the value of one’s life, but they cannot retroactively change the impact that it had on one’s well-being at the time” (1991/2000, 68). However, as Jeff McMahan points out, Velleman’s commitment to moment internalism is dialogically indefensible, since he does allow for retrospective change in the prudential value of life during a longer period, and there is no ontological difference between life at a moment and life during an extended period (McMahan 2002, 180). Narrativists, at least, lack sufficient reason for endorsing moment internalism. Ben Bradley raises several more serious challenges to moment externalism. His general claim is that only moment internalism is compatible with the supervenience thesis that the intrinsic value of something depends solely on its intrinsic properties (Bradley 2009, 19–20). After all, moment externalism maintains that the intrinsic value of someone’s life at a time can be affected by its relational, extrinsic properties. But why would this be a problem? Bradley claims that this would undermine the whole project of locating fundamental good- or bad-making properties (Bradley 2009, 19–20), that is, the properties whose instantiation completely determines what value there is. But this is a strange claim. What the moment externalist maintains is that the intrinsic value of S’s life at t depends on a) the intrinsic properties of S’s life at t and b) the relational (narrative) properties of S’s life at t. This is to say that the instantiation of the intrinsic and narrative properties of S’s life at t together completely determines what value there is. There is no reason to think fundamental good-making properties couldn’t be holistic. Bradley’s second charge is that allowing retrospective value change cannot account for prudential choice in certain situations. For concreteness, suppose that Don faces the following choice at t1: Betty Stay with Betty, with whom Don has been married for 20 years (since t0). Their marriage has had its ups and downs, but after their struggles, Don and Betty have found a way to live together that makes them both pretty happy. Megan+ Leave Betty and marry Megan+, a young and attractive secretary, who is head over heels in love with Don.

The Narrative Calculus

215

Let us stipulate that in this variant, life with Megan+ would be somewhat better for Don than life with Betty from t1 onwards. It doesn’t matter for the argument whether this is because he would be happier, or because the narrative value of his life from t1 would be higher, or both. Nevertheless, if Don’s choice of Betty at t1would sufficiently increase the value of his past life for him (for the reasons rehearsed above), the value of his life as a whole would be higher with Betty. (For this to be the case, it can’t be true that future with Megan+ would be very much better than future with Betty.) The question is: what should Don do, prudentially speaking?9 One simple answer from the narrativist perspective is that what is best for someone is always to maximize the value of her whole life. After all, we are not creatures living in this one moment, but temporally extended beings whose ‘past selves’ are really just past aspects of the one and the same self. Shouldn’t we be concerned with our lives as a whole? Some narrativists say so. Jeff McMahan says that what a cross-temporally psychologically unified individual “has most egoistic reason to want, at any given time, will be what would be best for his life as a whole, and this will be importantly affected by considerations of narrative unity” (McMahan 2002, 176). Bradley protests: I see nothing in the notion of prudence to indicate that one should focus primarily on one’s whole life. Why not care only about one’s future wellbeing? If someone chooses the best future available to her, it seems difficult to accuse her of being imprudent. (Bradley 2011, 59)

Perhaps, but perhaps not. Consider the case of Martin Luther Bling, whose life until around 1966 is the same as Martin Luther King’s. However, in 1966, MLB decides to cash in on his achievements, selling his name and likeness to advertisers, and starting a civil rights theme park with a name resembling “Negroland” and segregated facilities for black and white visitors, which contributes significantly to undermining the achievements of the civil rights movement. As a businessman, he is fairly happy and somewhat successful, until he dies of a heart attack in 1969, 9 Let’s assume we won’t settle for a satisficing view, on which it is prudentially permissible for Don to choose an option that is sufficiently good either from a whole-life or future perspective.

216

Antti Kauppinen

thus leading a life that is plausibly somewhat better during that last three years than MLK’s troubled years after the passage of the Civil Rights Acts. However, given that MLB’s post-1966 actions undermine the contribution of his pre-1966 activities to the objectively highly valuable goal of racial equality (while increasing their contribution to the less valuable goal of his personal wealth), it is also plausible that his life as a whole is prudentially worse than MLK’s. It doesn’t seem to me like a terribly tough bullet to bite to say that it would be good prudential advice to MLK in 1966 to implore him not to tarnish the worth of his past achievement by trading meaning for cash. Someone with strong narrativist intuitions might well accept that the prudential choice in each case is the one that maximizes whole life value. Nevertheless, I’m not going to rely on this assumption. I’ll grant that the value of future activities—for short, future value—can have prudential weight in addition to its contribution to whole life value. This is a kind of future bias in Parfit’s (1984) sense.10 Combined with retrospective value change, it threatens to result in what Bradley labels “dualism of prudence”: the impossibility of giving a single answer to what is in one’s self-interest when faced with the choice of a better life as a whole or a better future (Bradley 2011, 60–1). To avoid such implausible dualism, there must be some way of mapping whole-life and future values on a single scale. The suggestion I want to try out is that we can leave determining the relative weights of whole life value and future value to the individual subject (or, say, her ideal advisor). While it is not up to you what kind of life story is best for you or what kind of future is best for you, it is to some extent up to you how much relative weight they have in arriving at a unitary measure of prudential value, which could be just the weighted sum of WLNV (and the sum of atomic welfare values) and future NV (and the sum of future atomic values). But not entirely up to you: the stronger your psychological connectedness to your past selves is, the more weight the value of the past has in determining your self-interest. Let us say your weighting preferences are sensitive when they 10 Interestingly, Parfit observes that bias towards future “does not apply to events that give us either pride or shame: events that either gild or stain our picture of our lives” (1984, 160). While it is a relief to discover that a painful operation is in the past instead of ahead of us, it is no relief to discover we’ve done something shameful in the past. It may be a general truth that future bias is rational, if at all, only with respect to experiential goods and evils, but not with respect to non-experiential ones—such as narrative goods and evils.

The Narrative Calculus

217

reflect the degree of psychological unity. But although you must give more weight to whole life welfare when you’re more rather than less psychologically connected, it’s up to you just how much weight it will have, given some degree of connectedness. (Perhaps there should be a minimum threshold.) On this picture, it will be in someone’s self-interest to maximize whole life value even at the expense of future value, if they are strongly psychologically connected to their past selves and care about their life as a whole. Others, however, may be relatively disconnected and care little for anything but the future. Imagine that Don is such a future-oriented person through his life. This has the interesting consequence that what prudence recommends him to do changes with his temporal perspective. Bradley puts it as follows (substituting my names): If the value of [Megan+] after t1 is sufficiently greater than the value of [Betty] after t1, then it will turn out that prudence dictates choosing [Megan+] at t1. But imagine that [Don] faced a choice of [Betty] or [Megan+] at the start of the life, knowing in advance exactly how things would develop. Then the very same view would have [Don] choose [Betty] over [Megan+], since at the start of the life, the whole life and its future part are the same. But the only thing that changes between the start of the life and t1 is that some time goes by and some of the life gets lived in just the way [Don] knew it would. It cannot make sense to choose one way at the start, knowing full well what is to happen, and then change one’s mind only because some time has elapsed. (Bradley 2011, 60)

Bradley is right in drawing out the implication for the future-oriented: if the future segment of your life has value for you above and beyond its contribution to the value of your life as a whole, mere passage of time can change what it is prudential for you to do. But contra Bradley, this is a feature, not a bug. Mere passage of time does change one thing: what is past and what is future, your temporal perspective. If, like Don, you’re future-oriented, change in temporal perspective will change what is in your self-interest even though nothing else changes. (If you’re uncomfortable with this, maybe you need to be whole-life-oriented.) So I think it does make sense for Don to change his mind about what’s best for him when he arrives to t1. But that doesn’t mean it wouldn’t be prudentially rational for Don to precommit to Betty at t0 in full knowledge of the facts. For example, he should take steps to ensure

218

Antti Kauppinen

that he’ll never meet Megan. If it is not obvious why this is the case, imagine that at t0 Don meets an angel who shows him, It’s a Wonderful Life -style, how his life will be in the two options. He’ll see that after t1, he’ll be somewhat better off with Megan. But he’ll also see that the next 20 years of his life will be somewhat pointless, unless he binds himself to Betty now. Being concerned with what is the future segment of his life at t0, he will maximize his self-interest by precommitting. Of course, he knows that his future self will have different interests. But they’re not his interests at t0, and the psychological connections to his future self will be weak. So it is prudential for him at t0 to ensure he won’t ruin a good thing 20 years hence. 5. CONCLUSION

The life of an animal, with some possible exceptions at the more cognitively sophisticated end of the spectrum, consists of isolated pursuits: find food, find a mate when in season, protect the offspring until it can fend for itself, begin again. Such a life can go better or worse for the animal, but there is no reason to think that it is better for a dog, say, to have an unhappy puppyhood and a joyous adulthood than vice versa, as long as the two lives offer an equal sum of satisfactions. The shape of a life begins to matter when there is a story to be told about it. And there’s a story to be told whenever an agent adopts longrange goals, forms concrete plans about how to reach them, takes some means to her ends and thereby exposes herself to luck and obstacles, and meets with success or failure. Such narratable episodes add up to a whole life with a narrative shape when earlier goals, efforts, and incidents inform later pursuits, or fail to do so. I’ve argued above that the events that comprise the agent’s life gain in intrinsic value for the agent when they contribute to merited success in pursuit of valuable goals, whether the agent has the goals at the time of action or not—in short, when they contribute to giving the life a progressive shape. But why should we think it’s intrinsically good for us to lead a life with a progressive shape? Why should it matter to our self-interest whether a certain kind of true story could be told of our life? Here we approach the bedrock at which answers give out. We can just point to lives that differ narratively and say “Look, isn’t this better for a person?”. But it may be worth considering what kind of answer might be given to a

The Narrative Calculus

219

parallel question about pleasure or happiness. Nearly everyone agrees that they are among the things that are intrinsically good for us. But why? One kind of thing we can say begins with an appeal to the kind of beings we are. It is a deep fact about us that we are subjects of conscious experience. As a result of this, events can affect us in ways they can’t affect trees or stones: they can change the felt quality of that experience for the better or for the worse. This is a kind of explanation of why pleasure or happiness is good for us. It looks to our nature to locate a dimension along which our good can be directly affected. But we are not just subjects of experience. We are also the kind of beings who take charge of the world we live in, active agents who choose ends and means. How we fare in our pursuits matters to us beyond the associated experiences. That’s why we wouldn’t trade a successful career for a perfect illusion of one inside an Experience Machine. Imagine you’re on your deathbed after a life rich in narrative value. You have not wasted your brief time on this Earth—what you did made a difference for the better, objectively speaking, and made your own subsequent pursuits better, too. You had luck, both in having opportunities for pursuing valuable goals, and in things turning out your way. But your success wasn’t due to sheer luck, or it wouldn’t be as richly deserved as it is. It is fitting for you to look back on your life with a mix of pride and gratitude, and for others to look on it with admiration and even wonder.11 Now wouldn’t that be a life worth wanting, for your own sake? REFERENCES

Bradley, Ben (2009). Well-Being and Death. Oxford: Oxford University Press. Bradley, Ben (2011). ‘Narrativity, Freedom, and Redeeming the Past’, Social Theory and Practice, 37 (1): 47–62. Bruner, Jerome (1986). Actual Minds, Possible Worlds. Cambridge, MA: Harvard University Press. Dorsey, Dale (2015). ‘The Significance of a Life’s Shape’, Ethics, 125 (2): 303–30. Feldman, Fred (2004). Pleasure and the Good Life. Oxford: Oxford University Press. Fischer, John Martin (2005). ‘Free Will, Death, and Immortality: The Role of Narrative’, Philosophical Papers, 34 (3): 379–403. Gergen, Kenneth and Gergen, Mary (1988). ‘Narrative and the Self as Relationship’, Advances in Experimental Social Psychology, 21: 17–56. 11 I argue elsewhere (Kauppinen 2012) that the value such a life instantiates to a high degree is meaningfulness. But the argument of this paper does not hang on this controversial identification of meaning in life with having a good life history.

220

Antti Kauppinen

Glasgow, Joshua (2013). ‘The Shape of a Life and the Value of Loss and Gain’, Philosophical Studies 162: 665–82. Herman, David (2004). Story Logic. Lincoln: University of Nebraska Press. Kamm, Frances M. (2003). ‘Rescuing Ivan Ilyich: How We Live and How We Die’, Ethics, 113: 202–33. Kauppinen, Antti (2012). ‘Meaningfulness and Time’, Philosophical and Phenomenological Research, 84 (2): 345–77. Kauppinen, Antti (2013). ‘Meaning and Happiness’, Philosophical Topics, 41 (1): 161–85. Keller, Simon (2004). ‘Welfare and the Achievement of Goals’, Philosophical Studies, 121, 27–41. McAdams, Dan (1993). The Stories We Live By. New York: Guilford Press. McAdams, Dan (2006). The Redemptive Self. New York: Oxford University Press. McIntyre, Alasdair (1981). After Virtue. Notre Dame, IN: Notre Dame University Press. McMahan, Jeff (2002). The Ethics of Killing. Oxford: Oxford University Press. Parfit, Derek (1984). Reasons and Persons. Oxford: Oxford University Press. Portmore, Douglas (2007). ‘Welfare, Achievement, and Self-Sacrifice’, Journal of Ethics and Social Philosophy, 2 (2), 1–28. Rosati, Connie (2013). ‘The Story of a Life’, Social Philosophy and Policy, 30 (1–2): 21–50. Scanlon, Thomas (1998). What We Owe to Each Other. Cambridge, MA: Harvard University Press. Slote, Michael (1983). Goods and Virtues. Oxford: Clarendon Press. Strawson, Galen (2004). ‘Against Narrativity’, Ratio, 27: 428–52. Velleman, David (1991/2000). ‘Well-Being and Time’, reprinted in The Possibility of Practical Reason, pp. 56–84. Oxford: Oxford University Press.

10 Morality Within the Realm of the Morally Permissible ELIZABETH HARMAN

0. INTRODUCTION

It is common to distinguish among moral theories by comparing which behaviors are morally wrong according to each theory, and why these behaviors are morally wrong according to each theory. But there is another way that moral theories differ: they differ in their views of the realm of moral permissibility. I will argue that we can divide moral theories into four categories according to their views of the realm of the morally permissible. First, consider this question: 1. Can an agent face two morally permissible options, one of which is morally better than the other?

Views that answer “no” to this question hold that morality is very demanding, such that one is always morally required to behave in the morally best way available to one; or that morality is not very demanding because one often has options that do not differ morally at all. Such views do not acknowledge a non-trivial realm of moral permissibility. They hold that there are sometimes behaviors that are merely morally permissible (not also morally required), but they hold that whenever one has more than one morally permissible option, the options do not differ morally. Views that answer “no” to this question fall into the first category. (Such views are false, in my view, because they deny the existence of supererogatory actions.) Views in the other three categories answer “yes” to the first question. (It might be useful to refer to the chart below (“Four categories of Moral Theories”) while reading my mapping out of the four categories of views.)

222

Elizabeth Harman

Now consider this question: 2. Are there any cases in which an agent has more than one morally permissible option available to her, and one of those options is such that, all things considered, she should take that option?

Views that answer “no” to this question hold that whenever we have moral permissibility, we also have what we might call all-thingsconsidered permissibility. On these views, if an action is not morally wrong then it is never the case that one should not perform that action, all things considered. (Such views are false, in my view, because we often have options available to us that we should not take but that are not morally wrong.) Views that answer “yes” to question 1 but “no” to question 2 fall into the second category. Views in the remaining two categories answer “yes” to both questions 1 and 2. Now consider this question: 3. Given that there are some options one should take all things considered, but which are not morally required, are any of these such that one should perform them for purely moral reasons?

Views that answer “no” to this question hold that moral reasons cannot settle that one should do something without making it morally required. Views that answer “yes” to the first and second questions, but “no” to the third question, fall into the third category. Many ethicists are committed to a view in the third category, but I will argue that views in the third category are implausible for reasons not generally recognized. Once we grant that there is a robust realm of the permissible, in which one sometimes has permissible options that differ morally, and once we grant that sometimes one should take one of these options and should not take others of these options, it is strange to hold that it is only options that are partly favored by non-moral considerations that can win out to be all-things-considered required. Consider supererogatory options. Sometimes one should not perform a particular supererogatory action, all things considered (perhaps it would be too costly to oneself). Some moral views might hold in addition the following: Sometimes a supererogatory action is such that, all things considered, one should perform it; one’s reasons favor it; failing to perform

The Realm of the Morally Permissible

223

that action would be doing something that one all things considered should not do; but it would not be morally wrong. I will argue that indeed this is sometimes the case. If any supererogatory action is such that one should perform it, all things considered, because of the moral considerations that favor it, then failing to perform that action is what I will call a morally permissible moral mistake: an option that one should not take—for moral reasons—but that is not morally wrong. I will argue that there are some morally permissible moral mistakes; but the crucial claim I need for this paper is a weaker claim: that a moral theory can hold that there are some morally permissible moral mistakes. Such moral theories fall in the fourth category; they answer “yes” to all three questions. While I claim that the phenomenon of morally permissible moral mistakes is underappreciated by moral philosophers, I claim that recognition of the phenomenon is pervasive in ordinary moral thinking. Consider how natural it would be in ordinary conversation to say to a friend, “Well, you don’t have to do it. But it would be a nice thing to do. You should do it.” Moral theories that answer “yes” to all three questions hold that some options are morally permissible moral mistakes: they are actions that one should not take, all things considered—for moral reasons—but that are not morally wrong. These theories see moral reasons as relevant not just to which actions are morally required but also to which actions one should perform, all things considered. These theories see morality as playing an important role within the realm of the morally permissible. My goals in this paper are the following. I will introduce and develop the categorization of moral theories into the four categories I have mentioned; I think this categorization is interesting in its own right. I will argue that the true view is in the fourth category, thus arguing for the existence of what I call morally permissible moral mistakes. Meanwhile, I will also be arguing for a weaker claim: that a view in the fourth category is coherent. We should recognize that there can be moral theories in the fourth category. 1. IS THERE A NON-TRIVIAL REALM OF MORAL PERMISSIBILITY?

Consider the following question: Question 1: Can an agent face two morally permissible options, one of which is morally better than the other?

Elizabeth Harman

224

Four Categories of Moral Theories 1. Can an agent face two morally permissible options, one of which is morally better than the other?

No

Yes

There is no non-trivial realm of moral permissibility.

2. Are there any cases in which an agent has more than one morally permissible option available to her, and one of those options is such that, all things considered, she should take that option?

(Examples: Consequentialism, Simple Kantianism)

No

Yes

Moral permissibility coincides with all-things considered permissibility. (Weird)

3. Given that there are some options one should take, all things considered, but which are not morally required, are any of these such that one should perform them for purely moral reasons?

No

Yes

While non-moral reasons can sometimes win out within the realm of the morally permissible, moral reasons (though present) can never win out there.

Sometimes one should take an option for purely moral reasons, though it is not morally required. (For example, sometimes one should perform a supererogatory action.) Failing to take such an option is making a morally permissible moral mistake.

(Many moral theorists implicitly assume that the true moral theory is in this category.)

(I argue that the true view is in this category.)

Some ethical theories answer “no” to this question. Consider a standard consequentialist view: Consequentialism: An option is morally permissible just in case there is no alternative option with morally better consequences. Consequences are morally better or worse impartially (betterness is not relative to agents or times)1. Options themselves are morally better or worse only to the extent that their consequences are morally better or worse. 1 This sentence clarifies that the consequentialism in question is, as we might say, really consequentialism, and not a deontological view “consequentialized”.

The Realm of the Morally Permissible

225

On this view, if one option is morally better than another, that is because its consequences are better, but then it is morally required to choose that option over the other. So the answer to Question 1 is “no.” Consider also the following view: Simple Kantianism: An option taken for a particular reason is morally permissible just in case its agent can reasonably wish it to be a law of human nature that agents in general take options like that for reasons like that. Morality consists of this test of moral permissibility. An option is morally better than another option just in case the first option is morally permissible and the second option is morally impermissible.

On this view, the answer to question 1 is also “no.” (I’m not claiming that any Kantian view really has this form.) If morality is simply a test of moral permissibility for actions, then there is no room for one morally permissible option to be morally better than another morally permissible option. The correct answer to question 1 is “yes,” however. The existence of the supererogatory shows the answer to be “yes.” Consider the following two cases. Burning building: A passerby, Angela, sees a building on fire, and a child at the second-story window. Angela could keep her distance, or she could attempt to enter the building and save the child. This would involve risking her life to try to save the child’s life. Crying stranger: While staying at a hostel in Rome, Brian, an American college student, hears another guest crying in his room. When Brian sees the other man in the hallway, he has two options: say nothing, or ask if he is okay and wants to talk, thus opening himself up to being burdened by the man’s problems and to being delayed in his plans to explore the city.

I make the following claims about these cases. It is morally permissible for each agent to try to help, but it is also morally permissible to refrain from trying to help. Trying to help is morally better than not trying to help.

In both cases, trying to help would be supererogatory: it would be a morally good thing to do, but it is not morally required. I think that cases

226

Elizabeth Harman

like this show that the answer to question 1 is “yes.” But this is controversial. Some people deny that any actions are supererogatory. (One might hold that helping in these cases is morally required, as Consequentialism might hold. Or one might hold that helping is not morally required but also is not morally better, as Simple Kantianism might hold. Alternatively, either Consequentialism or Simple Kantianism might hold that it is not morally required for Angela to help, but it is morally required for Brian to help.2,3) Views that answer “no” to question 1 fall into the first category: these are views on which there is no non-trivial realm of moral permissibility. 2. WHAT IS THE RELATIONSHIP BETWEEN MORAL PERMISSIBILITY AND ALL-THINGS-CONSIDERED PERMISSIBILITY?

Let’s restrict our attention now to views that answer “yes” to Question 1. Consider the following question. Question 2. Are there any cases in which an agent has more than one morally permissible option available to her, and one of those options is such that, all things considered, she should take that option?

This question is less controversial than question 1. Of those people who answer “yes” to question 1, I think most will answer “yes” to question 2 as well. Consider the following cases. Pool: Carla is playing a friendly game of pool. No money is at stake, nor is either party likely to feel particularly happy or upset with a win or a loss. But Carla is doing her best to win. She contemplates a difficult shot at the number 4 ball; if she is successful, she’ll be well positioned for another shot. Alternatively, she could take an easy shot at the number 5 ball; but then she’d be unlikely to have any good options for the next shot. Given how her friend is playing, Carla has

2 Simple Kantianism might hold that we can reasonably wish for a world in which people do not constantly risk their lives to save others, but we cannot reasonably wish for a world in which people do not reach out to sad strangers. 3 What a consequentialist view would hold about the case may depend on what Angela should think about the likelihood that she can save the child and survive herself; if this is less than 50% likely, then a consequentialist view might hold that, given her information, she should not try to help.

The Realm of the Morally Permissible

227

little hope of winning unless she can sink several balls on this turn. Carla should go for the trickier shot. Teeth: Ernie is tired and feels like going to bed without brushing his teeth. Ernie should brush his teeth. Deportation: Fiona is on her way to a deportation hearing when she sees a man trip and fall down on a busy sidewalk. The man is not seriously hurt, but Fiona would normally stop to help him and make sure he is okay. If Fiona stops, this may result in her being deported. Fiona should not stop.

I claim that in the above three cases, each agent faces two morally permissible options, but should perform one of them rather than the other. I do not think my claims about these cases are controversial. As with most philosophers’ uses of example cases, my claim is that in the most natural version of the above cases, given my non-moral descriptions of them, the further claims I make about the cases (the normative claims) are true. I would like to introduce another example case, which I will use in a slightly more unusual way. Here is the case: Charity: George is deciding between working with two different charities. One addresses lung cancer and the other addresses famine. The work he could do to contribute to the famine charity would make more of a difference in the world: he is in more of a position to be helpful to them, and their work saves more lives. But George’s mother died of lung cancer recently. Working with the lung cancer charity would enable him to cope with the loss of his mother: it would be a statement of love and he would meet others who have suffered similar losses. It is morally permissible for George to volunteer with either charity. But all things considered, he should volunteer with the lung cancer charity.

I do not think it is obvious from the non-moral details I have given about this case that the normative claims are true of it. I claim the following: The Charity case is a possible case: it is possible for the normative claims to be true in a case in which these non-normative claims are true.

These four cases show that the answer to question 2 is “yes.” Although the view is unusual, we can consider a moral view that answers “yes” to question 1 but “no” to question 2. One might have the

228

Elizabeth Harman

view that the realm of the morally permissible coincides with the realm of all-things-considered permissibility. Within this realm, various considerations tell in favor of various options. These considerations make these options eligible as reasonable choices to make; but they do not settle that one should take one option rather than another. It is true that considerations sometimes make an option eligible as a reasonable choice to make without making it the option one should take. For example, suppose you are deciding whether to have Japanese food or Indian food for lunch. You are torn. You enjoy both foods, and both restaurants, for quite different reasons. The Indian food is yummy, filling, and less expensive, but the restaurant is further away. The Japanese food has a delicate sophistication which is more of a meaningful aesthetic experience for you; it is equally yummy, the restaurant is closer, but Japanese food is less filling and more expensive. You realize that even a slight improvement in one option or the other would not settle that you should take one option rather than the other. If you suddenly found a $1 off coupon to one restaurant, for example, that would not settle that you should go there. If you are in such a case, your options are on a par: it is not the case that you should take either one, nor that you should not take either one, nor do small sweetenings (or improvements) in either option make it the case that you should take that option. (By contrast, if the two options were exactly equally supported by your reasons, then a small sweetening of either option would settle that you should take that option.)4 In this case, it seems that your reasons to go to the Japanese restaurant make that option eligible as a reasonable option for you to take, without making it the option that you should take; and the same is true of your reasons to go to the Indian restaurant. While this does sometimes happen—sometimes one faces two significantly different options that are both supported by one’s reasons without either being favored by one’s reasons—I think this is unusual. A view that answers “yes” to Question 1 but “no” to Question 2 holds that this kind of situation is in fact incredibly common, and that whenever one finds

4 It is controversial whether it is possible for two options to be on a par. If this is possible (as I believe it is), then there are four possible relations that two available actions ç and ł can stand in: it may be that it is better to ç than ł; it may be that it is better to ł than ç; it may be that it is equally good to ç as to ł; or it may be that ç-ing and ł-ing are on a par. For an argument that two options can be on a par, see Ruth Chang, “The Possibility of Parity,” Ethics 2002.

The Realm of the Morally Permissible

229

oneself with more than one morally permissible option, these options are also all-things-considered permissible: neither is favored by one’s reasons. 3. MORALLY PERMISSIBLE MORAL MISTAKES

Now let’s confine our attention to views that answer “yes” to both question 1 and question 2. On these views, an agent can sometimes face at least two morally permissible options that themselves differ morally; and sometimes, among morally permissible options, an agent should take one of these options rather than the other. Now consider: Question 3. Given that there are some options one should take, all things considered, but which are not morally required, are any of these such that one should perform them for purely moral reasons?

Note that the examples we considered in section 2 do not help us to answer this question. In each of those cases, the explanation of why the agents should take the particular options involves some non-moral considerations, such as strategic considerations or prudential considerations. Let’s consider views that answer “no” to question 3. On these views, although moral reasons are present in the realm of the morally permissible, moral reasons cannot win out. When an agent faces at least two morally permissible options that differ morally, she might be in any of the following three situations: (a) she should take a morally worse option over a morally better option, because of some non-moral reasons, such as prudential reasons (b) she should take a morally better option over a morally worse option, because of some non-moral reasons, such as prudential reasons (c) neither option is such that she should take it, either because each option is equally supported by her reasons, or because the options are on a par We saw an example of a case in which (a) is true with the Deportation case: it is morally worse to rush to one’s own deportation case rather than stop to help someone who has fallen, but rushing on is the thing that

230

Elizabeth Harman

Fiona should do.5 We can see an example of a case in which (b) is true if we consider someone who is choosing between volunteering at two different charities; one of them does more good, so that volunteering there is the morally better choice, but it is also much more convenient for her. She should volunteer at the morally better, more convenient charity, but not for purely moral reasons; part of the reason she should do so is that it is convenient for her. (Indeed, we can imagine a version of the case in which the primary factor in the explanation of why she should choose that charity is the convenience.) I do believe that there are cases in which, between two options that differ morally, the reasons to take the morally better option are on a par with the reasons to take the morally worse option; but I do not think it is obvious what these cases look like. (I am not sure whether a morally good option can ever be equally supported by the agent’s reasons as a morally worse option is; this does not matter for our discussion.) Views that answer “yes” to questions 1 and 2, but “no” to question 3, recognize that cases in which an agent faces at least two morally permissible options that differ morally may be of the form (a), (b), or (c), but they leave out an important further possibility: (d) she should take a morally better option rather than a morally worse option because of the moral reasons that favor the better option

But there are cases in this category. Consider this case: Mechanic: Hannah is a car mechanic who owns her own small business. Irene, who is passing through town, brings her car in. Irene tells Hannah about her life. Irene has barely enough money to make it where she is heading with her three kids, starting a new life having escaped from domestic abuse. She can just cover the cost of the repairs and make her trip, but she’ll have to drive much longer days than she was planning because the repair will eat up the cost of one night in a motel. Hannah knows that it’s morally permissible for her to charge Irene for the repair. But she considers not charging her. Hannah concludes, “I shouldn’t charge her.” She’s right.

5 Many people facing deportation have moral duties to family members to avoid getting deported. Suppose that Fiona has no family who are relying on her to stay in the country, and that her reasons for staying are self-interested. Then hers is a case of type (a).

The Realm of the Morally Permissible

231

My claim about this case is: The Mechanic case is a possible case: it is possible for the normative claims to be true in a case in which these non-normative claims are true.6

I think my claim about this case is pretty uncontroversial, but that its import is generally underappreciated. We often have options available to us that are clearly supererogatory. Sometimes one takes a supererogatory option, and in doing so one is responding appropriately to one’s reasons. And sometimes it is not just that the supererogatory option was a reasonable option to take, but it was the option favored by one’s reasons. Sometimes an agent thinks, regarding a supererogatory action, “I should do this,” and she is right about that: given all of her reasons, that is the thing she should do. Consider Hannah’s deliberative process. She recognizes that it is morally permissible to charge Irene for the repair to her car. Of course it is! But still she considers helping Irene. That she considers helping her seems utterly reasonable, even though the only reasons to help Irene are moral reasons: that it would be good for Irene and her children, who are escaping a bad situation. Hannah’s choice to help Irene also seems reasonable. My claims about deliberation are these: In deliberating about what to do, an agent might come to know that a particular option is supererogatory, while continuing to consider taking that option. Furthermore, the agent might consider the possibility that she should take that option. She might conclude that she should take that option, and she might be right, even though throughout her deliberation, she knows that the option is supererogatory—it is morally good to do, but it is not morally required—and

6 My point here is that I haven’t filled in all the details of the Mechanic Case to make it uncontroversial and obvious that it is morally permissible for Hannah to charge Irene, and yet that she shouldn’t charge her. My claim is that there is some way of filling in those details so that those normative claims will be true. I think it should be fairly uncontroversial that there is some way of filling in the details so that those normative claims will be true, though I do not think there is a particular way of filling in the details such that it will be generally uncontroversial that that way of filling in the details makes the normative claims true. Note that I am not saying, regarding a complete non-moral description of the situation, that it is merely possible that the normative claims are true in that case. Rather, I am saying that it is possible for the partial description I gave to hold of a case in which the normative claims hold. (Thus, I do not mean to suggest that the moral might not supervene on the non-moral.)

232

Elizabeth Harman

she knows that the only considerations that favor doing it are moral considerations. This is true, but it is also reflected in the phenomenology of deliberation. It often seems to an agent that she knows that an option is supererogatory, even while deliberating and even while she concludes that she should take it because of the moral considerations that favor it. This seeming is often accurate.

These claims about deliberation and about the accuracy of the phenomenology of deliberating about whether to take a supererogatory option must be denied by any moral theory that answers “yes” to questions 1 and 2 but “no” to question 3.7 I conclude that the answer to question 3 is “yes”, because sometimes an agent should perform a supererogatory action; and in some of these cases, the only considerations that favor that action are moral considerations. Sometimes one should take an option, for moral reasons, though it is not morally required. I call the failures to take such options morally permissible moral mistakes. Here are some terminological stipulations: S’s ç-ing is a mistake = (def) S should not ç, all things considered. S’s ç-ing is a moral mistake = (def) S should not ç, all things considered, and the reasons against ç-ing that win out to make it the case that S should not ç are moral reasons. S’s ç-ing is a morally permissible moral mistake = (def) S’s ç-ing is a moral mistake but S’s ç-ing is not morally wrong.

In the Mechanic case, Hannah’s charging Irene for the repair would be a morally permissible moral mistake. She should not charge her; the reasons against charging her that win out to make it the case that he should not charge her are moral reasons; and yet it is not morally wrong to charge her. In sections 4 and 5, I will consider some objections to my argument in this section.

7 The importance of taking the phenomenology of deliberation seriously is also stressed by Horgan and Timmons in their “Untying a Knot from the Inside Out: Reflections on the ‘Paradox’ of Supererogation” Social Philosophy and Policy (2010) 27: 2: 29–63.

The Realm of the Morally Permissible

233

4. REJECTING MORALITY WITHIN THE REALM OF THE MORALLY PERMISSIBLE

I have argued that the answer to question 3 is “yes”, and that moral reasons in favor of an option sometimes win out to make it the case that an agent should take that option without making that option morally required. Restricting our attention to views that answer “yes” to questions 1 and 2, let’s consider more carefully views that answer “no” to question 3. How can they resist the argument I made in section 3? What alternative picture(s) can they offer? There is more than one way that such views might go. A view might hold: • What looks like a supererogatory action that should be performed in cases like Mechanic is actually a morally required action

Or: • What looks like a supererogatory action that should be performed in cases like Mechanic is actually an action made eligible by the agent’s reasons, but it is not an action the agent should perform.

Or: • What looks like a supererogatory action that should be performed that is supported by purely moral reasons is actually supported by some non-moral reasons, and that is a crucial part of why it should be performed.

Or, finally: • What looks like a supererogatory action that should be performed in cases like Mechanic is actually an action that the agent should not perform, in light of all of the agent’s reasons; this is part of what is morally admirable about the action.

And these responses might be combined to handle different cases differently. I will take each kind of response in turn. My reaction to each response will be to acknowledge that there are some cases with the structure described, but to deny that the Mechanic case, and others

234

Elizabeth Harman

like it, must have the structure described by my opponent rather than the structure I have described. Let’s consider first this response to cases like the Mechanic case: • What looks like a supererogatory action that should be performed in cases like Mechanic is actually a morally required action

My opponent points out that morality sometimes forbids doing what is ordinarily permissible. Sometimes it is morally wrong to charge someone for a service or a good, although it is normally morally permissible to do so. That’s certainly right. Consider the following case: Asthma: A woman and her daughter are fleeing an abuser. Their car breaks down, and it will take all of their cash to pay for the repair. They have somewhere to stay when they get where they are going, but they have no way to pay for the daughter’s asthma medicine, which they need to buy right away to prevent the possibility of an asthma attack, which could be lifethreatening. Julie, the mechanic, learns all of this. Her garage has been making more money recently, and she has not already made any plans for the extra money.

It may well be morally required for Julie to give the woman the repair for free. The health risk to the daughter makes Asthma a significantly different case from Mechanic. My opponent says this: The Asthma case shows that sometimes a mechanic may be morally obligated not to charge a customer, because of the burden on the customer and the mechanic’s ability to get by without the payment. If we consider the original Mechanic case, in which Hannah should give Irene the repair for free, this is simply a case in which it is morally required to do so. It’s not as morally wrong for Hannah to charge for the repair in Mechanic as it would be for Julie to do so in Asthma, but it is still morally wrong.

I am not moved by this. Given what is actually at stake for Irene and her children in Mechanic—driving longer hours on their trip, versus one more night in a motel and easier driving days—and even taking into account what they have recently been through, it simply could not be morally wrong for Hannah to charge them. Nevertheless, it is utterly reasonable for Hannah to consider giving them the repair for

The Realm of the Morally Permissible

235

free, and she might well conclude, correctly, that she should do so. This first opponent that we are considering wants to agree with me that there are versions of the Mechanic case in which Hannah should give them the repair for free, but disagrees with me that it is morally permissible to charge them in these versions of the case. My response is that there is simply no version of the Mechanic case in which it is morally wrong for Hannah to charge them (although it may be wrong for Julie to charge for the repair in Asthma, which is not a version of the Mechanic case). Note also that this first opponent must deny the claims I made about deliberation. The first opponent must hold that, while Hannah is correct that she should give Irene the repair for free, she is simply wrong in her belief that doing so is supererogatory. One cannot know first that an option is supererogatory while leaving it open whether one should perform that action, if it is only favored by moral considerations. Rather, if one opens the question of whether, all things considered, one should perform the action, then one is also opening the question of whether the action is supererogatory. These claims are implausible. Let’s turn now to an opponent who says this: • What looks like a supererogatory action that should be performed in cases like Mechanic is actually an action made eligible by the agent’s reasons, but it is not an action the agent should perform.

Recall the case of choosing whether to eat Indian food or Japanese food for lunch. Sometimes two significantly different options are both supported by one’s reasons, but neither is favored by one’s options. Sometimes they are not exactly equally supported by one’s reasons, as we can see by noting that small improvements (or sweetenings) in each option do not make either option the one the agent should take. In such cases, the two options are on a par. As I’ve already commented, I find it implausible that a supererogatory action is ever equally supported by the agent’s reasons as a non-supererogatory alternative, but that is not crucial for our discussion. I will restrict my attention, in discussing this opponent’s view, to the idea that sometimes a supererogatory option may be on a par with a non-supererogatory option. I certainly agree that this can happen.

236

Elizabeth Harman

This second opponent might furthermore say the following thing about the phenomenology of deliberation: An agent such as Hannah, who knows that one of her options is supererogatory, may nevertheless consider taking that option, yes. But Hannah does not ask herself the question “what should I do?” Rather, she asks herself, “what to do?” In deciding what to do, Hannah is asking a different question. She is asking what she chooses. This is a question that she must settle for herself, even if her reasons do not favor one option over another. Knowing the answer to “what should I do?” does not settle the answer to the question she’s really asking, “what to do?” in any cases in which her reasons do not favor one option over another.

It is certainly true that sometimes we must choose between options that we know to be on a par or equally favored by our reasons. I must choose one can of diet coke from the fridge, though I know it is not true that I should take the one on the left rather than the one on the right. We often know, when we are choosing between two very similar options, that neither option is favored by our reasons. But how often do we know this about choices that differ significantly? I think it is unusual to know this. Rather, I think the normal mode of deliberation when choosing between options that differ significantly involves implicitly taking seriously the possibility that one option is favored by one’s reasons and trying to figure out which one it is. How often do our reasons settle that we should do one thing rather than another, and how often do our reasons simply make more than one significantly different option eligible to us? I think the latter situation is rare. We should not mistakenly think it is more common than it is by confusing that situation with the following situations: (i) It is difficult for the agent to know how to weigh some very different kinds of factors that tell in favor of two different options. (ii) It is difficult for the agent to know what will happen if she chooses each of two different options. (iii) It is better for the agent to choose between two options now, rather than waiting to think about it more, even if waiting would increase the likelihood that the right choice would be made. (iv) The agent has two options that are both wonderful and that would lead to good outcomes.

The Realm of the Morally Permissible

237

(v) The agent is choosing between two different types of actions; it is not the case that, in general, agents choosing between these types of actions should always choose one particular type of action over the other. Each situation, (i) through (v), is compatible with its being the case that, in light of all of her reasons, the agent should pick one of her options rather than any of her other options. Sometimes an agent’s reasons favor an outcome, though it is very hard to see that that is true, as in (i). Sometimes there is a lot of information that is relevant that the agent does not have, which can increase how difficult a decision feels, as in (ii); but it does not mean that the choice is up for grabs; there are still facts about what credences are warranted in light of whatever evidence the agent does have, and there may be a single option that the agent should take in light of those credences. Sometimes there is a reason to choose now, even if doing so would involve choosing an option that is worse in light of all of one’s reasons (perhaps ignoring the reason to choose now), as in (iii); in such a case, it may be true that the agent should have chosen differently, in light of her reasons, even though it is also true that she should not have waited to figure out that that was the right choice to make. Sometimes we say to someone, “you can’t go wrong,” when what is really true is not that she cannot act as she should not, in light of all of her reasons, but that she has two options that are both wonderful, as in (iv). Sometimes an agent should make a particular choice, given the details of her situation, although a different choice would have been warranted if her situation had been a bit different; thus, it is not the case that agents in her situation should always choose in a certain way, as in (v). Usually one’s reasons support one choice over the others available to one. While there is a lot of moral permissibility in our lives, there is not very much all-things-considered permissibility in our lives. We often do things that we should not do. Often, it isn’t so terrible that we do things we should not do. There is a certain amount of weakness of will in our lives that isn’t terribly bad for us (eating that extra chocolate or skipping that day at the gym). Sometimes we choose to make a choice quickly rather than making the right choice; this may be the life policy that one should adopt, although it means that one sometimes does particular things one should not do. And sometimes we simply are wrong about

238

Elizabeth Harman

what we should do; but the wrong choice is not seriously worse than the right one. On the other hand, sometimes it is terrible that we do things we should not do; sometimes we make big mistakes. My third opponent says this: • What looks like a supererogatory action that should be performed that is supported by purely moral reasons is actually supported by some non-moral reasons, and that is a crucial part of why it should be performed.

The third opponent might point out that people are often made to feel good by helping others, and that this always counts as a reason in favor of helping others, so it is hard to find a morally good action that should be performed for purely moral reasons. In fact, the claim I have made is a bit more modest: that there are cases in which an agent has two morally permissible options, and should perform one of these options, where the reasons in favor of the option that win out to make it the case that she should take that option are moral reasons. One might hold that if Hannah should give Irene the repair without charging her, the reasons that win out to make it the case that she should do so are the considerations that it would be good for Irene and her children in a difficult time. Even if it would make Hannah happy, that isn’t part of what makes it what she should do. But alternatively, we can simply stipulate that she won’t be made happy, that somehow helping Irene would make Hannah feel emotionally involved in an unsettling way. So let’s set aside the thought that helping someone often makes one happy. The third opponent might instead revise a thought that the second opponent had: that often our reasons underdetermine what we should do. Consider, for example, Ruth Chang’s view, according to which one’s “given reasons” (the reasons we find ourselves with) often underdetermine what one should do, but then one makes a choice to endow a reason with extra strength, creating a new, coinciding “voluntarist reason” which settles that one should act in the way that reason supports.8 Chang’s motivating examples include cases like these:

See Ruth Chang, “Voluntarist Reasons and the Sources of Normativity” (in Reasons for Action, Sobel and Wall, eds. (New York: Cambridge University Press, 2009), pp. 243–71), which is part of a bigger project, encompassing additional papers and a book manuscript, Making it Matter. 8

The Realm of the Morally Permissible

239

Romance: Kyle and Louis are in a romantic and sexual relationship, but they have not defined it. So far it is casual. Kyle has no compelling reason to take care of Louis’s needs, for example by cooking him dinner if he is exhausted after a long day of work, or by picking up Louis’s laundry left on the floor of Louis’s apartment. It is not the case that Kyle should do these things. But if Kyle commits himself to Louis—a mental act—it thereby becomes true that Kyle should do these things. Career: Megan is trying to decide between two career directions. She will enter either a Masters of Fine Arts program in order to become an actor or a Masters of Social Work program in order to become a psychotherapist. Megan feels passion for acting, but the career path is uncertain and stressful. Megan has a real talent for both acting and therapy; her listening skills and empathy are an asset for both. Therapy is a safer path; she would feel that her work has meaning because she would be helping others; but that path involves less of an outlet for her creative energy. Megan’s reasons do not settle what she should do. Megan thinks it over and makes a commitment—a mental act—to pursue a career as a psychotherapist. It thereby becomes true that Megan should become a psychotherapist.

Chang clarifies that she does not see the relevant commitments as the formation of intentions, and the phenomenon she is discussing is not the phenomenon of how having formed an intention may influence how one should act. Also, the commitments are private mental acts. The phenomenon Chang is interested in is not the way that commitments we make to others impact how we should act. Chang’s view presents a challenge to my own view because if she is right, then it may be that whenever one should choose a supererogatory act over a non-supererogatory act, this is because one has privately committed to this endeavor, in thinking it over, thus creating a new “voluntarist reason” which is (perhaps) not properly thought of as a moral reason. So even in cases in which it looks like there are only moral reasons in favor of performing the supererogatory act, it turns out that there is a non-moral reason in favor of it. Perhaps something close to Chang’s description does happen sometimes. Perhaps sometimes an agent should perform a supererogatory action simply because she wants to, or because she feels like doing so. But there would still be many cases, such as Mechanic, in which the agent lacks that independent inclination, and is moved by the considerations that favor the supererogatory action. And it is not clear that the

240

Elizabeth Harman

mere fact of inclination or desire ever provides a reason. It may be that inclinations and desires indicate the presence of reasons, without themselves being reasons—one may have an inclination or desire for a reason, where that reason is typically also a reason to do the thing one wants to do.9 As I understand her views, Chang and I disagree about the extent to which a person’s given reasons underdetermine what she should do. The choices that Chang discusses—whether to be seriously involved with a particular person, what hobbies to have, what career to pursue, etc.—are difficult choices. But I think it is rare that in such cases a person faces two options that are equally favored by her reasons or are on a par. As I argued above, while situations (i) through (v) are common, none of them need be situations in which a person’s given reasons do not settle what she should do. My fourth opponent says: • What looks like a supererogatory action that should be performed in cases like Mechanic is actually an action that the agent should not perform, in light of all of the agent’s reasons; this is part of what is morally admirable about the action.

This opponent says that what is so admirable about supererogatory actions such as Hannah’s in Mechanic is Hannah’s willingness to act contrary to what her reasons support, in order to help another: it is precisely because she should not do it that the action is supererogatory, and praiseworthy. Unfortunately I do think that the idea that supererogatory actions are ones that agents should not perform, in light of all their reasons, is common.10 But the view is false. I will make three points about this. 9 There is an extensive literature on the question of whether a person’s desires or inclinations themselves provide reasons, or should be responsive to independently-existing reasons. I don’t want to take a stand on that issue in this paper, but I do have sympathy with the view that desires and inclinations do not provide reasons. 10 Some are puzzled by the existence of the supererogatory—how can an action be morally good to do but not morally required? One solution to this apparent puzzle holds that supererogatory actions are recommended by a perspective that includes only a subset of one’s reasons, and that these actions are not morally required because an agent has other reasons that tell against performing the actions, such that all things considered the agent should not perform the actions. An example of this view occurs in Jamie Dreier’s “Why Ethical Satisficing Makes Sense and Rational Satisficing Doesn’t” (in Satisficing and Maximing, Michael Byron, ed., Cambridge University Press, 2004, 131–54). The view also appears in Douglas Portmore’s

The Realm of the Morally Permissible

241

First, we should distinguish the question of whether it is in Hannah’s interest to help from the question of whether she should not help. Sure, it is contrary to her own interests to help. But that does not mean that she should not help. Morality often requires self-sacrifice of various kinds; and we should engage in that self-sacrifice. Second, while self-sacrifice is admirable, there is nothing particularly admirable about a willingness to do what one shouldn’t do, for someone else’s sake. Suppose that you were willing to break a promise to Joe in order to help Karen. The fact that you shouldn’t break the promise doesn’t make it admirable to do it anyway in order to help Karen. Third, it is true that sometimes one shouldn’t perform a supererogatory action, because of one’s self-regarding reasons. Sometimes a supererogatory action would be so personally costly that one’s reasons tell against doing it.11 When an agent performs such an action, there is definitely something special about it. It is praiseworthy, as any supererogatory action is, and perhaps it is praiseworthy in a unique way. A willingness to disregard the force of one’s own interests for another’s sake may warrant special praise.12 But not all supererogatory action involves a failure to appreciate the force of the reasons one has. Hannah may be seeing her reasons correctly and clearly when she concludes that she should give the woman the repair for free. 5. OTHER MORAL CATEGORIES

Someone might object to my discussion so far by claiming that “morally permissible moral mistake” is just a new term for an already-familiar phenomenon, so I have introduced unnecessary new terminology, which is therefore obfuscating. The objector holds not just that ordinary morality is committed to the existence of morally permissible moral

“Position-relative consequentialism, agent-centered options, and supererogation,” Ethics (2003) 113: 303–32. Portmore’s commitment to the view that all supererogatory action is a mistake is pointed out by Betsy Postow in “Supererogation Again,” Journal of Value Inquiry (2005) 39: 245–53. Portmore repudiates the view in “Are Moral Reasons Overriding?” Ethical Theory and Moral Practice (2008) 11: 369–88. 11

Raz claims that no supererogatory actions are such that they should not be performed. “Permissions and Supererogation” American Philosophical Quarterly (1975) 12. 12 Unless it involves a systematic inappropriate devaluing of oneself—too common among women in sexist cultures like our own.

242

Elizabeth Harman

mistakes (a claim with which I agree), but that moral philosophy already countenances the phenomenon. I will consider three different versions of this objection. The first objector holds that someone makes a morally permissible moral mistake just in case she fails to take her morally best option. This suggestion fails in both directions. First, it is not true that one’s morally best option is in general the option that one should take, all things considered. For example, it may be that the morally best thing I could do right now would be to go to the local hospital and offer up one of my kidneys and some of my liver. But that is not the thing that I should do right now, all things considered. Failing to do that is not making a morally permissible moral mistake, but it is failing to do the morally best thing I could do right now. The claim fails in the other direction as well: sometimes an agent ought to perform a supererogatory action, but it is not the morally best thing she could do right now. (The morally best thing she could do may be so personally costly that, all things considered, she should not do it.)13 The second objector holds that someone makes a morally permissible moral mistake just in case she fails to take an opportunity to fulfill an imperfect duty; after all, each action that would fulfill an imperfect duty has moral reasons in favor of it, but is not morally required. This suggestion fails because it is not true that whenever one has an opportunity to fulfill an imperfect duty, one should do so, all things considered. For example, suppose that you are on your way home from work and you can feel a migraine coming on. If you get home quickly, you can take a pill that would prevent a full-blown migraine that would last several 13 Paul McNamara has some work on the supererogatory in which it might seem that he makes one of my central claims. Both McNamara and I say things like this: “Sometimes one ought to do something that is not morally required, and the thing one ought to do is a supererogatory action.” But McNamara and I use “ought” in quite different ways, so he is not making the same claim that I am making. He uses “ought” to pick out an agent’s morally best actions. So when he says “sometimes an agent ought to do something that is not morally required” he simply means that sometimes an agent’s morally best option is not morally required. And when he says “sometimes an agent ought to perform a supererogatory action” he simply means that sometimes an agent’s morally best option is a supererogatory action. McNamara makes these kinds of claims in his “Supererogation, Inside and Out: Toward an Adequate Scheme for Common Sense Morality” (in Oxford Studies in Normative Ethics, ed. Mark Timmons, Oxford University Press (2011), 202–35) and in other papers. McNamara does not address the question of whether it is ever true that one ought to perform a supererogatory action, all things considered.

The Realm of the Morally Permissible

243

hours. If you suddenly see an opportunity to do something nice for someone—perhaps to help someone who is picking up trash in a park— it is not the case that you should take that opportunity, although helping others is something that you have an imperfect duty to do. Rather, you should hurry home and take your pill. Failing to stop to help the person picking up trash is not a morally permissible moral mistake, because it is not a failure to do something that you should do.14 Finally, the third objector holds that someone makes a morally permissible moral mistake just in case she does something suberogatory. The category of the suberogatory is a kind of inverse of the category of the supererogatory: while supererogatory actions are morally good to do, but not morally required, suberogatory actions are morally bad to do, but not morally wrong. It is controversial whether any actions are suberogatory, but I agree with the objector that some are. The objector’s suggestion fails because it is not true that every morally permissible moral mistake is suberogatory. For example, consider Hannah’s charging Irene for the repair in Mechanic. This would not be a morally bad thing to do; but she should not do it, for moral reasons. So the category of morally permissible moral mistakes is not identical to the category of suberogatory actions. Nevertheless, the objector may be right about something: every suberogatory action may be a morally permissible moral mistake: something that one should not do, for moral reasons, but that is not morally wrong.15,16

14 This case provides another counterexample to the claim that one should always take a morally better option over a morally worse option. It is morally better to stop and help the person picking up trash, rather than driving straight home; but you should not stop and help. 15 I see it as an open question whether all suberogatory actions are morally permissible moral mistakes. I discuss the suberogatory in my “Morally Permissible Moral Mistakes” (Ethics, forthcoming) For interesting discussion of the suberogatory (also called “offence”), see Roderick M. Chisholm, “Supererogation and Offence: A Conceptual Scheme for Ethics,” Ratio (1963) 5: 1–14; David Heyd, Supererogation, Cambridge: Cambridge University Press (1982); Gregory Mellema, Beyond the Call of Duty: Supererogation, Obligation, and Offence, State University of New York Press (1991); Julia Driver, “The Suberogatory,” Australasian Journal of Philosophy (1992) 70: 3: 286–95; Hallie Liberto, “Denying the Suberogatory,” Philosophia (2011) 40: 395–402; and Paul McNamara, “Supererogation, Inside and Out: Toward an Adequate Scheme for Common Sense Morality” (2011). 16 I develop and defend the notion of a morally permissible moral mistake in three other papers: “Morally Permissible Moral Mistakes” (Ethics, forthcoming), “Eating Meat as a Morally Permissible Moral Mistake” (forthcoming in Philosophy Comes to Dinner, Chignell, Cuneo, and Halteman, eds., Routledge) and “Gamete Donation as a Morally Permissible Moral Mistake” (manuscript).

244

Elizabeth Harman 6. CONCLUSION

I have argued that we can divide moral theories into four categories, according to how they view the realm of the morally permissible. The first category includes views on which it is not possible for an agent to have two morally permissible options, such that one option is morally better than the other; these views do not acknowledge a non-trivial realm of moral permissibility. The second category includes views on which there is a non-trivial realm of moral permissibility, but on which it is never the case that an agent should act in one morally permissible way rather than in another morally permissible way. The third category includes views on which there is a non-trivial realm of moral permissibility, and on which sometimes one should act in one morally permissible way rather than another; however, views in this category deny that sometimes one should take one morally permissible option rather than another for moral reasons. Views in the fourth category recognize the robust presence of morality within the realm of the morally permissible: on these views, sometimes one should take one morally permissible option rather than another, for moral reasons. These views hold that some options are morally permissible moral mistakes. I have argued that the true moral view acknowledges the existence of morally permissible moral mistakes. But I have also argued for the following weaker conclusions. It is coherent to hold that there are morally permissible moral mistakes. We should acknowledge that a moral theory could hold that there are morally permissible moral mistakes. Acknowledgments For helpful comments on drafts of this paper, I thank Alex Guerrero, Peter Graham, and audiences at the Massachusetts Institute of Technology and the 2014 Arizona Workshop in Normative Ethics.

11 Fairness, Participation, and the Real Problem of Collective Harm JULIA NEFSKY

1. INTRODUCTION

In a wide range of cases, people collectively cause a morally significant outcome but no individual act seems to make a difference. For example, while climate change is caused in part by millions of people acting in certain ways, any one such act does not seem to make a difference. Take a single driver off the road and the problem of climate change will surely be just as bad. Similarly, while large-scale consumer patterns can have a major impact on the lives of people across the globe, it’s hard to believe that a single purchase will itself make a difference. Will things be any better for anyone if, for instance, the cup of coffee I buy this morning is fair trade rather than conventional? Or take the case of voting in a national election, where—let us suppose—it matters a great deal for the welfare of the population which candidate wins. While our votes collectively determine who wins, a single vote won’t make a difference: the winner will be the same give or take any individual vote. The problem in cases of this sort is that it seems each person can argue, “it won’t make a difference whether or not I do X, so I have no reason to do it.” The challenge is to explain where this argument goes wrong. If we cannot, this would be very troubling; morality would be powerless in wide range of cases in which it should have force. I call this “the Problem of Collective Harm”, and the cases in which it arises “Collective Harm Cases”.1 This follows my terminology in Julia Nefsky, “Consequentialism and the Problem of Collective Harm: A Reply to Kagan.” Philosophy & Public Affairs 39 (2011): 364–95 and in my “How You Can Help, Without Making a Difference”, draft. I think these names, on balance, do a nice job of giving a sense of the relevant phenomenon, but they are in one way misleading. 1

246

Julia Nefsky

There are two ways of trying to solve the problem.2 The first is to try to deny the description of the cases: to argue that, contrary to appearances, one’s individual act actually does, or at least might, make a difference with respect to the morally significant outcome in question and that there is, thus, no problem after all. The second is to accept that there are cases in which one’s individual act cannot itself make a difference, but to argue that there is still reason for action—a reason that doesn’t have to do with the difference you make in outcome. Those who take this second type of approach, in other words, respond by rejecting the implication from ‘it won’t make a difference’ to ‘there is no reason to do it.’ In this paper I am interested in the possibility of taking the second kind of approach—a “Rejecting the Implication” approach.3 I centre the discussion around three such proposals. The first two—which I call “Weak Participation” and “Strong Participation”—both claim that we can solve the problem by attending to the fact that individuals in Collective Harm Cases are acting as part of a group that makes a difference. The third proposal—the “Fairness Approach”—attempts to solve the problem by appealing to considerations of fairness. After raising some initial questions about each proposal individually, I argue that there is a common, basic problem with all three. While they do identify considerations other than the difference you make, it’s not clear that these considerations can provide the reasons for action we are looking for as long as it is true that your act won’t make a difference. More specifically, I argue that unless your act could play a significant instrumental role in bringing about the outcome in question, the explanations these views are offering as to why you have reason to do it cannot get off the ground. And yet not playing any such role seems to be just what we are accepting when we accept that your act won’t make a While in many Collective Harm Cases it is harm that is at stake, this need not be so. In a voting case, for instance, it could matter a lot morally which candidate wins, but not because there will be more harm if one candidate wins than if the other does. This would still count as a “Collective Harm Case” on my definition. We can think of “Collective Harm Cases” as naming a wider class of cases after a subset, namely the subset in which what is at stake is harm. 2 This excludes biting the bullet and accepting that there is no reason for action in Collective Harm Cases. 3 The “Denying the Description” approach will briefly come back into the discussion at the end of this paper. I discuss the plausibility of such an approach elsewhere (in “Consequentialism and the Problem of Collective Harm”).

Fairness, Participation, and Collective Harm

247

difference. I bring out this problem through a discussion of the Fairness Approach in Section 6. In Section 7 we see how the same problem arises for the other two approaches. The problem, I then argue, extends beyond these three proposals. In particular, if we try to apply certain general non-consequentialist moral theories to Collective Harm Cases, we encounter the same basic problem. This clarifies why it is that the Problem of Collective Harm poses a genuine challenge not just for consequentialists but also for nonconsequentialists. None of this will be to say that notions of fairness or participation will not be important in understanding the morality of Collective Harm Cases, or that non-consequentialist theories cannot ultimately handle these cases effectively. Rather, the aim of the discussion is to obtain a sharper understanding of what the core challenge in the Problem of Collective Harm really is. This, in turn, helps us identify what the central task needs to be if we are to properly address it. 2. TWO EXAMPLES

I have given a few quick examples of real-world Collective Harm Cases, but I will largely focus on the following two imaginary cases. Both are from Derek Parfit. Drops of Water: Ten thousand men lie in the desert, suffering from intensely painful thirst. We are ten thousand people near the desert, and each of us has a pint of water. We cannot go into the desert ourselves, but we can pour our pints into a water cart. The cart will be driven into the desert, and any water in it will be evenly distributed amongst the men. If most of us add our pints, this will be enough to alleviate their suffering. But while our acts of adding our pints would collectively alleviate their suffering, any individual such act does not seem to make a difference. A single pint added to the cart only allows each man to drink an extra ten-thousandth of a pint of water. This is no more than a single drop. And, however much water they are receiving, one drop more or less is simply too miniscule an amount to make any difference to how they feel.4

4 This is adapted from Derek Parfit, Reasons and Persons (Oxford: Oxford University Press, 1984), p. 76 and follows my presentation in “How You Can Help, Without Making a Difference.”

248

Julia Nefsky

Harmless Torturers: A victim is hooked up to an electric shock machine that has a thousand settings, controlled by a dial. There are a thousand “torturers”, each of whom turns the dial up a single notch. At the first setting there is no electric current. At the second setting there is a tiny electric current, but it is too tiny to be perceived by the victim. In general, the settings increase the voltage by increments so tiny that the difference between any two adjacent settings is too miniscule to make a difference to how the victim feels. But, while any two adjacent settings are indistinguishable, after many increases the victim feels pain, and by the thousandth setting the victim is in excruciating pain. While the torturers’ dial-turns together result in a lot of pain for the victim, it seems that no individual makes a difference for the worse. Take any given torturer’s dial-turn away, and the pain would be just as bad.5

The Problem of Collective Harm arises in both cases. In Harmless Torturers, each torturer can argue, “my dial-turn did not make a difference; the victim would have been in just as much pain had I not done it. So I did not do anything wrong.” In Drops of Water, each of us can say, “it won’t make a difference whether or not I add my pint, so there’s no reason to add it.” True, the dehydrated men will not be relieved from their pain unless enough of us add our pints. But this does not change the fact that things will be the same for them give or take any single pint. 3. WEAK PARTICIPATION

Weak Participation and Strong Participation both claim that we can solve the Problem of Collective Harm by attending to the fact that, even if you do not yourself make a difference, you may be part of a group that does. We together make a great deal of difference in Collective Harm Cases, and this can explain why each of us has acted wrongly or rightly. Weak Participation essentially starts and stops with this claim. Parfit tries this simple appeal to what people together do in Reasons and Persons, giving the following principle, which he calls “C7”: (C7) Even if an act harms no one, this act may be wrong because it is one of a set of acts that together harm other people. Similarly, even if some act benefits 5 This is adapted from Reasons and Persons, p. 80 and follows my presentation in “Consequentialism and the Problem of Collective Harm”.

Fairness, Participation, and Collective Harm

249

no one, it can be what someone ought to do, because it is one of a set of acts that together benefit other people.6

In Harmless Torturers, for instance, while turning up the dial one notch makes no difference to the harm suffered by the victim, it is one of a set of acts that together cause him a lot of pain. The torturers together make a difference for the worse, and so each of them has acted wrongly. Does this response work? Not as it stands. After all, what determines who is in the group?7 Why can’t each individual say: my act makes no difference, and so it’s not part of the set of acts that together makes a difference for the worse? It is unclear, in other words, why I should count as part of the group that harms if what I do makes no difference with respect to that harm. The point is not (or not just) that we need some criteria that will tell us which acts are in the group and which are not. The point is that if we want to derive normative implications directly from my being part of a group, we need an understanding of what this sort of participation consists in that explains why it has these normative implications. Still, Weak Participation might seem to be pointing down the right path. Weak Participation says: even if you do not by yourself make a difference, if you together with others do, this can explain why you have acted wrongly or rightly. Strong Participation can be thought of as an attempt to substantiate this idea.8

6 Reasons and Persons, p. 70. Parfit intends this to be a consequentialist proposal. For a discussion of whether it should be accepted by consequentialists see Frank Jackson, “Which Effects,” Reading Parfit, ed. Jonathan Dancy (Oxford: Blackwell, 1997), 42–53, and Ben Eggleston, “Should Consequentialists Make Parfit’s Second Mistake: A Reply to Jackson,” Australasian Journal of Philosophy 78 (2000): 1–15. In this paper, though, I am interested in the proposal independently of whether it would be attractive to a consequentialist. 7 Parfit attempts to address this question. He sets out various principles that attempt to further elaborate on C7 and, if they succeed, would specify who is in the group. But, as he admits in a later paper, they do not succeed: the further principles exclude the very acts he is trying to capture, acts that cannot themselves make a difference. See his “Comments,” Ethics 96 (1986), pp. 846–8. 8 When substantiated in this way, though, it is clearly not consequentialist (i.e. See note 6 in this chapter).

250

Julia Nefsky 4. STRONG PARTICIPATION

Strong Participation, unlike Weak Participation, employs a robust conception of acting with others—of participating in a collective action. Using this conception of what it is to participate, it attempts to explain why mere participation, regardless of whether you can make a difference, has normative implications. Christopher Kutz develops such a proposal. On Kutz’s account, you participate in a collective action when you act with a “participatory intention”, an intention to do your part in a collective project.9 Suppose we are building a sandcastle. Your act of filling the bucket with water is part of our collective action because it is performed with the intention of contributing to our collective end, a sandcastle. You fill the bucket so that we can mix water with sand, which will allow us to sculpt the castle. If you had instead filled it with the intention of pouring it on my head, this would not have been part of our collective action. “Jointly acting groups,” Kutz writes, “consist of individuals who intend to contribute to a collective end.”10 Next Kutz argues that (1) you are accountable for actions that are ascribable to you, and for the consequences (whether intended or not) of those actions, and that (2) when you participate in a collective action, that action is ascribable to you. When you take part in a collective action, you are an “inclusive” author of that action.11 You are one of a number of people who can say, “We did it.” It is, in this sense, your action, and so you can be held accountable for it and its consequences.12 Kutz claims that this holds regardless of whether you individually make a difference. He gives examples like the following. Suppose that

9 Christopher Kutz, Complicity (Cambridge: Cambridge University Press, 2000), Chapter 3. There is a further condition, namely that your conception of the joint project sufficiently overlaps with that of the other members of the group. (Complicity, pp. 94–5.) 10 Complicity, p. 89. There is disagreement about how to understand the phenomena of collective action. Kutz’s account—as he explains—is minimalistic. Other accounts bring in stronger conditions, conditions requiring, for instance, mutual knowledge or mutual responsiveness. And they often invoke not just intentions to do your part but intentions that we do the group action. In part because Kutz’s account is already a minimalist one, it is not as though we could resolve the core problem that I will raise (in section 7) by substituting in another available account of collective action. 11 Where this is contrasted with being the “exclusive” author of an action. (Complicity, pp. 105–6.) 12 Complicity, pp. 137–9.

Fairness, Participation, and Collective Harm

251

you and I are having a picnic. While I am getting the drinks from the car, you put the blanket down on a bed of flowers and ruin it. Even though it is your individual act and not mine that causes this damage, you put the blanket down as part of our collective action of picnicking together. So, while none of my individual acts make a difference for the worse, the damage is a result of a collective action that is ascribable to me. Thus, I can be to some extent accountable for it.13 A similar thing can be said in Collective Harm Cases: even if one’s individual act makes no difference with respect to the outcome of concern, one can be accountable if one is a participant in a collective action that results in this outcome. One issue that arises in trying to solve the Problem of Collective Harm in this way is that there are, as Kutz points out, many Collective Harm Cases that simply do not involve collective action. These are cases of “unstructured collective harms”: cases in which harm results from many people acting in certain ways, without there being any joint project that they are all intentionally participating in.14 Much environmental damage is unstructured harm; drivers, for instance, are not intending to do their part in some collective project whenever they drive. Similarly, people who buy conventional rather than fair trade goods are not intending to do their part in a joint project. There might be, however, a simple way of extending the picture to cover unstructured cases. Typically when there is a risk of unstructured collective harm, there is also an opportunity to participate in a collective action of preventing that harm. In Drops of Water, if none of us add our pints and the men continue to suffer, we would not be intentionally participating in a collective project. So, this would be a case of unstructured collective harm. But if we do add our pints, it seems we would be engaged in a joint project of alleviating their suffering. Similarly, while drivers are not engaged in a collective project when they drive, they could participate in a collective action aimed at mitigating environmental damage if they each reduce the amount they drive with the aim of doing their part in such a project. If Strong Participation explains why there is reason to refrain from participating in harmful collective actions even if you make no difference, it should also be able to explain why

13

Complicity, p. 154.

14

Complicity, p. 166.

252

Julia Nefsky

there is reason to participate in beneficial collective actions even if you make no difference. The next question to ask, though, is: how exactly does Strong Participation explain why there is reason to do either? So far, what it has explained is why participants in a collective action can be accountable, even if they make no difference. But, at least on Kutz’s notion of accountability, asserting that someone can be accountable for a harm is not equivalent to saying that this person did something that she ought not to have done. Accountability, for Kutz, concerns “warranted response”, where this is not just a matter of responses owed “because of the rights and wrongs” one has done.15 Think of the picnic example. Perhaps it is true that I am partially accountable for the damage caused to the flowers: perhaps I ought to express apology or help compensate the gardener because of the way in which I am connected to the damage through our collective action. But one thing that is clear is that I did not do anything I ought not to have done. Everything I did was perfectly fine. As Kutz writes: I am not at fault with regard to the gardener, but it seems I do nonetheless owe some form of response ( . . . ) Because I have made our picnic mine by my intentional involvement, I have also made its consequences mine. My accountability is therefore intermediate, between what one owes for faultless harms (for I was not at fault) and what one owes for faulty harms (for you, and hence we, were). Your fault sticks to me but not with its full force.16

Are things similar in Collective Harm Cases? In Harmless Torturers, my individual act of turning the dial makes no difference to the suffering the victim undergoes. Perhaps I am to some extent accountable because of the way in which I am connected to the harm through my participation. But if what I did couldn’t have made any difference with respect to the harm, how does Strong Participation give us grounds for thinking that I ought not to have done it? I take it the reply is: in the picnic case I have no reason to expect the bad outcome; this is why we cannot say that I ought to have acted otherwise. In Harmless Torturers, on the other hand, the collective

15 Complicity, p. 18. That is, accountability for Kutz is specifically not a retributivist or desert-based notion. 16 Complicity, p. 154.

Fairness, Participation, and Collective Harm

253

project is a harmful one and so I can expect that the collective action will result in harm. If you expect that a collective action will result in harm, then you ought to refrain from participating in it, and—crucially—you ought to refrain from participating regardless of whether you could make any difference for the better by refraining. This is because if you participate, you will be an “inclusive author” of the harm. The harmful collective action will be, in a sense, yours—something that we, and thus inclusively you, did. Intuitions will likely diverge here. Some might find this picture appealing. Others might think: while I can expect that the collective action will result in harm, I also know (in advance) that my “inclusive authorship” won’t make any difference with respect to that harm. Why should that sort of inclusive authorship make my individual act wrong? Why not think that the most it can do is make me accountable in a faultless, associative sense?17 5. THE FAIRNESS APPROACH

A common idea is that the reason one ought to act in the relevant way in Collective Harm Cases is that not to do so would be unfair. Garrett Cullity advances such a proposal, and it is his version that I will make use of here.18

17 Of course, I might not want to be even faultlessly accountable; I might not want to owe or be owed any response. In that case, the risk of being accountable could itself give me reason to refrain from turning the dial. But, divorced from any claim that I would be doing something I ought (or, at least, have independent reason) not to do, this appeal to the risk of accountability seems to simply give a reason concerning how turning the dial could affect me. This seems to be the wrong sort of reason to respond to the Problem of Collective Harm that arises in Harmless Torturers. 18 The appeal to fairness is common for those concerned with Collective Harm Cases in which not acting in the relevant way is considered free-riding. These are cases in which the relevant agents are also the beneficiaries. For example, why should I pay my subway token if I can sneak on? $3 more or less surely won’t make a difference to subway operations. A common answer is: I am only able to gain the benefits of the subway because others pay their tokens; so, to not pay would be unfair. Not all Collective Harm Cases are potential free-rider cases. In Drops of Water, if I keep my pint, I am not gaining a benefit that is only possible because others have given up their pints. (Note also that not all free-rider cases are Collective Harm Cases.) But while the two kinds of cases only overlap, and are not identical, it seems plausible that not acting in the relevant way is wrong for the same reason in both kinds of cases. This is the idea that Cullity pursues in “Pooled Beneficence”, in Imperceptible Harms and Benefits, ed. M. Almeida (Dordrecht: Kluwer Academic Publishers, 2000), 1–23.

Julia Nefsky

254

The Fairness Approach first claims that in Collective Harm Cases there is a collective obligation to prevent (or to avoid bringing about) the relevant bad outcome. Unless enough people act in a certain way, bad consequences will result, and so there is an obligation on the group as a whole to ensure that this does not happen. But if we have a collective obligation—the argument goes—it is unfair for some but not others to do the work of satisfying it. If I stand by while others do their part in fulfilling the obligation, this is not fair to the others in the group and is therefore wrong. Thus, regardless of whether or not one could make a difference by doing so, there is reason to act: “it is a matter of pulling one’s weight in what we all ought to be doing.”19 Suppose that each of the potential torturers would receive a benefit for turning up the dial—say, fifty dollars. Each knows that he won’t make a difference to the victim’s pain by doing so. So, why should he refrain? According to the Fairness Approach, without making any assumptions about the torturers’ individual obligations, we can say that there is an obligation on the group as a whole not to cause the victim to suffer. Most people in the group must refrain from turning the dial if this collective obligation is to be met, and if most people refrain while you go ahead and claim the fifty dollars, this is unfair. So, the right thing to do is to refrain. Now, the point is not that from a collective obligation we can always derive individual obligations. If no one is trying to satisfy the collective obligation, I am not being unfair by refraining from acting myself. The idea is, rather, that when there is a collective obligation and people are trying to fulfill it, it can be unfair for me not to do my part too. Cullity explains: I have not been arguing that individual imperatives can be derived from every collective imperative ( . . . ) Before the collective action is started, none of us is arrogating any special privileges in refusing to get it started. We are collectively acting wrongly, but no individual is taking advantage of others’ propensity to contribute to doing what we ought to be doing.20

This, however, seems to be a problem for the Fairness Approach. This means that as long as all the torturers continue turning up the dials, or as 19

“Pooled Beneficence,” p. 17.

20

“Pooled Beneficence,” p.18.

Fairness, Participation, and Collective Harm

255

long as no one has given a pint to the cart, the Fairness Approach has nothing to say about why one ought to act otherwise. But the Problem of Collective Harm is that it’s not clear why any individual has reason to give up her pint, or to refrain from turning the dial, and so on. Each individual such act seems to make no difference, and thus it is hard to see why anyone has reason to do it. If the Fairness Approach only kicks in once people have already started acting in the relevant way, it can’t count as a full solution to the problem. In response, perhaps we can say that while the collective obligation does not itself entail individual obligations, its presence does give individuals, right from the start, at least some morally relevant reason for action. While I may not have an obligation to add my pint if others have not yet done so, perhaps the fact that we are collectively obligated to alleviate the suffering explains why I have at least some reason to add my pint. Once others, or enough others, have added their pints, I then become obligated to do so because not doing so would be unfair. I do not think this reply succeeds. But this is a consequence of a more basic and important problem—one that arises for all three of the views under discussion. Let’s turn to that problem.

6. FAIRNESS AND DIFFERENCE-MAKING

Jonathan Glover dismisses the Fairness Approach, claiming that it invokes a notion of fairness that we do not or, at least, should not care about. He considers a case in which a car needs to be pushed up a hill; only six people are necessary for the job, but there are eight altogether. Even though six could manage the task on their own, it would be unfair for two to sit by while others do all the work. This sounds much like the point that the Fairness Approach tries to make in Collective Harm Cases. In the voting case, for instance, the Fairness Approach says: even though we do not all need to vote, it would be unfair for some but not others to do so. But, Glover replies: We are free to accept the argument from injustice in the car pushing case without accepting it in the voting case. If I do not push the car, the others will have to push a bit harder. Many of us are against the kind of injustice that involves giving benefits to some at the cost of additional hardships to others. But no-one has to vote harder because I do not vote. It seems a dog-in-the-manger

256

Julia Nefsky

version of justice that objects to one person benefiting because others are left unchanged.21

Glover is right that there is an important disanalogy between the carpushing case and the voting case.22 But it misunderstands the Fairness Approach to say that it invokes a “dog-in-the-manger version of justice” or that it “objects to one person benefiting because others are left unchanged.” Take Drops of Water. Unlike a “dog in the manger” situation, each of us would benefit from keeping our pints; we would each have it as a refreshing drink. The point the Fairness Approach makes is that it would be unfair for some to keep their pints, receiving the benefits that each of us would have enjoyed, while others give theirs up in order to achieve what we as a group are obligated to achieve. The type of unfairness involved in keeping my pint is not that doing so benefits only myself and not others—that it benefits one person while leaving others unchanged. It is, rather, that if I keep my water, while others give theirs up, I am not—as Cullity writes—“pulling my weight in what we all ought to be doing.”23 I am “relying on others to satisfy ( . . . ) [the collective] imperative, while excepting myself from doing so.”24 It seems reasonable to propose that this sort of unfairness is morally objectionable. And someone who finds this view compelling will not be moved by the reply that we should only care about unfairness if it is a matter of one person gaining a benefit by making things worse for others. Such a reply seems to simply contradict the proposal, rather than identifying a problem with it. Glover’s objection is, thus, not effective as it stands. But there is an important issue in the vicinity. The Fairness Approach says: if we collectively ought to bring about some outcome, it is unfair to let others

21 Jonathan Glover, “It Makes No Difference Whether or Not I Do It,” Proceedings of the Aristotelian Society 49 (1975), p. 182. 22 Here is the disanalogy. In the car pushing case, my act is not necessary for achieving the desired outcome—there are ways of bringing about the outcome without it. But in Collective Harm Cases, it’s not just that my act is not necessary; it’s that it simply can’t make any difference. We can take my act away, and not replace it with anything (e.g. not replace it with more effort on the part of the others), and the outcome of concern will be the same in all relevant respects. Because of this disanalogy, while appealing to fairness might work to explain why I ought to push in the car-pushing case, it won’t work as a solution to the Problem of Collective Harm, as we will see shortly. 23 24 “Pooled Beneficence,” p. 17. “Pooled Beneficence,” p. 17.

Fairness, Participation, and Collective Harm

257

do the work of satisfying this obligation without doing so yourself. You ought to pull your weight in what we collectively ought to do, regardless of whether in doing so you will make any difference. Glover’s reply is: we shouldn’t care about the unfairness in not pulling one’s weight if it won’t make any difference. But the problem is more basic. If acting in the relevant way won’t make any difference then it does not seem that it pulls any weight at all. It doesn’t seem to count as doing the work of satisfying the collective obligation. Consider this revised version of the car-pushing example. We need to get the car up the hill, but one of us—call him “Small”—has a rare physical condition: while he is otherwise healthy, he has the muscles of a two-year-old. He is, let’s say, an adult trapped in a two-year-old’s body. Because of this, it makes no difference whether or not Small pushes. If he pushes with all his might, this will exert a tiny amount of force on the car, but this tiny amount of force is too tiny to make any difference at all with respect to the task at hand. The car won’t progress up the hill any faster if he pushes than if he doesn’t, nor will it be any easier for the others involved. And this is so regardless of how many others push. Whether or not enough other people push to get the car up the hill, adding or removing Small from the equation simply won’t change things in any relevant respect. In this situation, does it make sense to say that Small ought to push because this is his fair share of our collective burden? His act of pushing wouldn’t help us achieve the goal. It would be merely superfluous. It is true that there is a burden on the collective to get the car up the hill. But if his pushing won’t help discharge this burden, how could it be his fair share of the burden? It doesn’t seem to be any real share at all. Instead it seems like a mere waste of his efforts. Perhaps Small should do something out of fairness. If there is some useful task he could perform, he might have a duty of fairness to do that. But it doesn’t make much sense to say that what he ought to do is something merely superfluous and unhelpful. In Collective Harm Cases each individual act seems to be like Small’s act of car-pushing. It appears that it can’t make any difference with respect to the outcome of concern, and so seems to be merely superfluous. But what we are seeing is that the Fairness Approach cannot get off the ground if this is so. The Fairness Approach appeals to (a) the presence of a collective obligation to bring about some outcome, and (b) the idea that it is unfair to let others take on the task of satisfying this

258

Julia Nefsky

obligation while excepting yourself from doing so. But as long as my act won’t help satisfy the collective obligation—as long as it would be merely superfluous with respect to what we are collectively obligated to do—(a) and (b) cannot explain why I should do it. Even if there is a sense in which we are collectively obligated to relieve the men’s suffering in Drops of Water, if adding my pint to the cart won’t help to relieve their suffering, this collective obligation wouldn’t give me reason to add my pint. And bringing in a requirement to do one’s fair share does not fill the gap. How could doing something merely superfluous with respect to the alleviation of their suffering count as doing my fair share of our obligation to alleviate their suffering? It seems to be no real share at all.25 I am not claiming that the idea that I ought to do my fair share has no force in this case. As with Small’s case, if there is some useful task I could perform I might have that sort of duty to do it. The point is that as long as it’s true that pouring my pint into the cart wouldn’t do anything useful toward alleviating their suffering—that it will not help matters in any real way—it does not seem that that would be the thing to do in a response to a requirement to do my part in the task of alleviating their suffering. One might reply as follows: “putting the idea of ‘doing one’s fair share of the collective obligation’ aside, there is a simple explanation as to why fairness gives one reason to add one’s pint. If we really think that fairness matters morally in itself, then it does not need to be that adding a pint would be useful with respect to satisfying this collective obligation. It is unfair for some of us to keep our pints if others have given theirs up just because this would mean having something that those others no longer have.” 25 Cullity considers and tries to respond to a similar objection. Part of his response is to argue that one’s individual act counts as making a real contribution toward meeting the collective obligation because “the contribution is perceptible” even though “it has no perceptible effect on what is achieved by the group” (“Pooled Beneficence,” p. 16). In Drops of Water, for instance, my contribution is perceptible: a pint is a perceptible amount of water. This does not work. The perceptibility of a pint of water is irrelevant to the question of whether it actually contributes in any morally significant sense to the task of alleviating their suffering. The problem here is similar to a problem I discuss in “Consequentialism and the Problem of Collective Harm”; there is a tendency in writing about this topic to confuse perceptible with morally significant. Still, while Cullity’s point about perceptibility does not work, some of his other comments hint in a potentially promising direction.

Fairness, Participation, and Collective Harm

259

Now, this kind of fairness—unlike the one that occurs in the Fairness Approach—is the “dog-in-the-manger” type of fairness that Glover finds objectionable. It asks us to “level down”: to give something up, lowering our own position, simply for the sake of making things more equal. Still, while many object to the idea that there is any reason to level down, some do think that there is moral reason to do so and that it only seems objectionable if we do not attend to the fact that this reason can often be outweighed. Regardless, this is really beside the point. Even if you do have reason to give up your pint coming simply from the fact that this would make things more equal amongst the original pint-holders—that is, even if there is a reason to “level down”—this is not a reason that can count as a solution to the Problem of Collective Harm. We can see this by noting that it does not explain why I have reason to add my pint to the cart, rather than—for example—to pour it onto the ground. As far as bare equality in the resources of the original pint-holders is concerned, it would be just as effective if I pour my pint onto the ground. To respond to the Problem of Collective Harm we need an explanation of why I have reason specifically to add my pint to the cart. That is, we need an explanation of why each of us ought, or at least has reason, specifically to do the sort of thing that if enough of us do it, will alleviate the suffering. A reason that is just as much a reason to, instead, pour one’s pint onto the ground is no such reason. Notice, moreover, that if I have reason to add my pint coming simply from the fact that this would make the resources more equal amongst the group of original pint-holders, this is a reason that has nothing particularly to do with the suffering men in the desert. It only concerns how my resources compare to the others in this group. Any such reason cannot count as a solution to the Problem of Collective Harm. We need a reason for action that connects appropriately to the fact that the men in the desert are suffering. Otherwise, we are essentially just changing the topic. The Fairness Approach tries to provide this connection by invoking a specific kind of fairness—that of pulling one’s weight in what we collectively ought to do. But, as we have seen, it’s not clear that this can work as long as we are accepting the claim that one’s individual act won’t make any difference.26 Let’s call this “the Superfluity Problem”. 26 Some might object to the comparison between Small’s case and Drops of Water. In Small’s case, everyone but Small is able to do something instrumentally significant; Small is an

260

Julia Nefsky 7. A COMMON PROBLEM

The Superfluity Problem arises just as much with the other two views. We have essentially already seen this for Weak Participation. Weak Participation says that even if you don’t harm anyone yourself, if you together with others do this explains why you have acted wrongly. But in Collective Harm Cases, it’s not just that you don’t cause the harm in question all by yourself; it’s that you make no difference with respect to that harm. Your act doesn’t seem to play any significant role in bringing it about. So, why should you count as part of the group to begin with? Why is your act part of the set of acts that harms, if it does not contribute in any non-superfluous way to bringing about the harm? To come at the point another way: as long as your act doesn’t play any significant role in what the group does, we cannot simply point to what the group does to explain why you have acted wrongly (or rightly). And yet this is just what Weak Participation tries to do. Strong Participation does not do any better. Whether or not we find appealing the idea that there is something wrong with participating in a harmful collective action even when one’s doing so can’t make a difference, there is a more basic issue. If your act can’t make any difference, we

exception. In Drops of Water, on the other hand, everyone is symmetrically situated: each can add at most a pint, and so no one can make a significant difference, though we together can get the job done. If we made Small’s case like that it would have to be that everyone was like Small. There could be thousands of people with Small’s condition who need to get a car up a hill; they can do so if most of them push, but any one of them won’t make a difference. When we move to this “symmetrical” version it no longer seems unintuitive that Small is doing his fair share by pushing. My response is largely to agree. I agree that there is an important difference between the original Small case, on the one hand, and Drops of Water and this “symmetrical” Small case on the other. I agree that, while there might be no (morally relevant) reason for Small to push in the original case, Small does have reason to push in the symmetrical version, and each of us has reason to add our pints in Drops of Water. This is because I think there is a solution to the Problem of Collective Harm. (Note that the symmetrical Small case is a Collective Harm Case.) Moreover, I do not deny that it ultimately will make sense to think of adding a pint as one’s fair share of the collective obligation. The point about symmetry is certainly not enough to establish this, but it does—I think—help refocus our attention in the right direction: namely, one of trying to identify features that will allow us to explain why in Collective Harm Cases it is not actually true that one’s individual act is instrumentally superfluous. What our discussion in this section is meant to reveal is that we cannot deal with the “it makes no difference” argument by invoking the idea of a duty to “pull one’s weight”, or to do one’s “fair share.” Instead, before we can make use of these ideas in understanding the morality of Collective Harm Cases, we first need to be able to debunk the claim that my acting in the relevant way would be instrumentally superfluous and unhelpful.

Fairness, Participation, and Collective Harm

261

may not be able to count it as an act of participation to begin with. In Harmless Torturers, Strong Participation says: ‘even if each torturer makes no difference to the harm, the torturers would be engaged in a collective action of harming the victim. By turning the dial, you are participating in this collective action. This makes the collective action in a sense yours, and so you are accountable for it.’ But, on closer inspection, it’s not clear this story can get off the ground. For any given torturer, how can we say that she is participating in this collective action? To participate in a collective action is to intentionally do your part in a joint project—in this case, presumably, a project of torturing the victim.27 But for any given “torturer”, her act of turning up the dial a single notch makes no difference to the pain of the victim. It doesn’t seem to play any non-superfluous role in causing the harm. So, how can we say that in turning it she is intending to do her part in this joint project? Her intention in turning up the dial could just be to acquire fifty dollars at no cost to anyone. Indeed, if she decided to turn the dial only after realizing that doing so won’t make any difference, that probably is her intention. Kutz explains that the intentions that make an act participatory need not be “explicit in deliberation”; they can be “functionally implicit” in one’s behavior.28 But this does not help here. We cannot say that while an intention to do her part in a collective project of torturing the victim may not be explicit in her deliberation, it is functionally implicit in her behavior. Her behavior, after all, is behavior that won’t make any difference to the torture the victim undergoes. The extent to which the victim is tortured will be exactly the same give or take her turn of the dial. So, we can’t simply appeal to that action to reveal that she is implicitly trying to do her part in a joint project of torturing the victim. Nor can we say that there are other aspects of her behavior that reveal her implicit intention to do her part in the collective project; that would be stipulating extra features that need not be present in the case. We have said that to handle cases like Drops of Water or environmental damage—cases of “unstructured collective harm”—Strong Participation can talk about reason to participate in a collective action aimed at preventing or mitigating the harm. But the same problem arises here. Again, participating in a collective action is to intentionally do your part 27 28

There is no other potential joint aim specified in the example. Complicity, p. 82.

262

Julia Nefsky

in a collective project. In Drops of Water it would be a project of alleviating the men’s suffering; in the case of environmental damage it would be a project of mitigating environmental damage. But how can doing something that would be merely superfluous with respect to our bringing about these outcomes count as doing my part in our bringing them about? Kutz writes, “my part is defined as the task I ought to perform if we are to be successful in realizing our shared goal.”29 A task that can’t make any difference with respect to our shared goal seems to be no such task. Indeed, if we could have said that adding my pint is the task that I ought to perform if we are to relieve the men’s suffering, there would not have been a Problem of Collective Harm there to begin with. Advocates of Strong Participation might object that I am presupposing an instrumental conception of “doing my part”. Why can’t Strong Participation invoke a broader conception—a conception under which my act need not do anything useful toward the achievement of the shared goal in order to count as doing my part? Indeed, while in some places Kutz writes things like “participatory intentions can, thus, be seen as merely a species of ordinary, instrumental intentions, differentiated by the group-oriented content of the goal they specify”,30 in other (nearby) places, Kutz explicitly endorses a broader conception. For example: Contributory relations [“my part” relations between individual act and collective end] might take an instrumental form if what the agent does helps cause the collective outcome (my pushing helps to move the car), or if the agent’s part is a constitutive element of the group act (stepping this way is part of dancing a tango). The relation might be expressive if by doing one’s part, one thereby exemplifies one’s membership in a group or participation in an activity, as when by voting I express my membership in a political community. And the relation might be normative if one performs one’s part because of norms internal to some group or institution that demand certain behavior (I wear a dark suit as an IBM employee.) ( . . . ) What makes my behavior participatory is nothing more (and nothing less) than my conception of what I do as related to the group act.31

If Strong Participation invokes this broad understanding of doing one’s part, can’t it avoid the Superfluity Problem by saying that it is

29 31

30 Complicity, p. 81. Complicity, p. 84. Complicity, p. 82, my emphasis.

Fairness, Participation, and Collective Harm

263

non-instrumental contributory relations that are at play in Collective Harm Cases? For instance, while adding a pint will not do not anything instrumentally significant, it can still count as doing my part in our collective project of alleviating the suffering, because it expresses my support for the project or my solidarity with those who have donated.32 The problem is, though, that on this proposal, Strong Participation cannot explain why I have reason specifically to add my pint to the cart, rather than—for instance—to wear a T-shirt that says, “I support the rehydration project!” Especially insofar as we are accepting that neither amounts to doing anything instrumentally significant, wearing the T-shirt could just as well express my support, or my solidarity, as adding my pint could. Strong Participation, in other words, would be unable to differentiate between an act of pouring in one’s pint and any other instrumentally insignificant act that could similarly express support or solidarity: neither makes any significant causal contribution to the achievement of the collective goal, and either can make you a participant in the collective project because of its expressive features. The problem here is similar to the one we saw with the attempt to appeal to a broader notion of fairness. Turning to non-instrumental ways of doing one’s part in the joint project avoids the Superfluity Problem only by deflating Strong Participation’s potential to address what is at issue in the Problem of Collective Harm. As we’ve said, to address what is at issue, we need an explanation of why each of us specifically has reason to do an act that is of the sort that if enough of us do it, this will collectively cause (or, depending on the case, prevent) the morally significant outcome of concern. Finding a reason to add a pint to the cart that is just as much a reason to instead wear a T-shirt that expresses one’s support or solidarity does not do this. 32 Why not say that adding a pint counts as “doing my part” in the constitutive instrumental sense? We could say that my adding a pint is a constitutive element of the collective act of together adding our pints to the cart. The problem with this is that, at least on Kutz’s account, a collective act is defined in terms of the collective end. The collective end could be either (a) a state of affairs, or (b) an activity (e.g. dancing the tango), or (c) an institution of some kind. (Kutz, p. 82) In Drops of Water, the most plausible understanding of what our collective end would be is the alleviation of the men’s suffering (a state of affairs), not the activity of together adding our pints to the cart. We would not be adding our pints for its own sake, with the alleviation of the suffering as a mere foreseeable consequence. (When we dance the tango, on the other hand, we may be doing this for its own sake, rather than to bring about some further state of affairs.)

264

Julia Nefsky

To summarize: Rejecting the Implication approaches try to solve the Problem of Collective Harm by accepting that your individual act can’t make any difference, and explaining why nonetheless there is still reason to do it (or, depending on the case, reason to refrain). But the problem is that—at least for the three views we have seen—the explanations they offer do not seem to work as long as your act would be instrumentally superfluous. And yet that seems to be exactly what we are accepting when we accept that your act can’t make any difference. 8. CAN’T WE JUST APPLY A NON-CONSEQUENTIALIST THEORY AND CALL IT A DAY?

It is tempting to think that it is only consequentialist theories that will have trouble with Collective Harm Cases.33 After all, non-consequentialists do not think that all that matters morally is the difference you make in outcome. So, the suggestion goes, their theories will do just fine. Our discussion of the Superfluity Problem helps reveal why this view about non-consequentialist theories is mistaken. Recall, we cannot solve the Problem of Collective Harm by giving a reason for action that has nothing particularly to do with the morally significant outcome in question. I may have all sorts of reasons to refrain from driving: to get more exercise, to save money, to reduce the risk of causing or being in an accident, to avoid stressful traffic jams. But, of course, none of these can provide a solution to the Problem of Collective Harm that arises with respect to car driving and environmental damage. We need a reason for action that connects in some appropriate way to the fact that widespread car use causes environmental damage. The problem for non-consequentialists is that, even though they do not think that all that matters morally is the difference you make in outcome, it’s not clear that there can be any reason to X, which connects appropriately to outcome Y, if X-ing cannot make any difference with respect to Y. If acting in the relevant way would be merely superfluous with respect to the outcome in question, it’s not clear that we can get any relevant story going as to why you have reason to do it.

33

This is a common reaction I have received from philosophers in conversation.

Fairness, Participation, and Collective Harm

265

Suppose, for instance, that you subscribe to a theory of Virtue Ethics. Suppose you think that the right thing to do is what the virtuous person would do. Can this view explain why you ought to add your pint to the cart, even if doing so won’t make a difference? It’s not clear that it can. If it really won’t make a difference to the well-being of the suffering men, then it’s not clear that the virtuous person would add her pint. Surely, the virtuous person does not act wastefully, and insofar as it is true that adding a pint won’t make any difference, doing so appears to be a waste. Unless we can say that adding a single pint could genuinely help to alleviate the men’s suffering, it does not seem we can say that this is what the virtuous person would do. Or, consider the virtues and vices that a Virtue Ethicist might appeal to in order to explain why one ought to add one’s pint. They might want to say that keeping your pint when these men are suffering would be greedy or selfish, and that to donate it would be generous or compassionate. But if adding your pint would be superfluous with respect to the alleviation of the men’s suffering, it doesn’t seem any of these would apply. Making a ‘donation’ that will not do anything useful does not seem generous or compassionate; it seems foolish and wasteful. And if you have considered whether it would be of any use to pour in your pint and have correctly determined that it would not be, then keeping this water does not seem greedy or selfish.34 Now, suppose you are a Kantian. Suppose you think that Kant’s Formula of Humanity gets things right. You agree that you should “act that you use humanity, whether in your own person or in the person of any other, always at the same time as an end, never merely as a means.”35 Can this explain why you ought to refrain from turning up the dial a notch in Harmless Torturers? If turning the dial won’t make a difference for the worse to anyone, it’s not clear why it would count as treating someone as a mere means. If you are not contributing in any nonsuperfluous way to bringing about the victim’s suffering by turning the dial, then how can we say that in doing so you are treating him as a mere

34 Walter Sinnott-Armstrong makes a similar point in “It’s Not My Fault: Global Warming and Individual Moral Obligations,” Perspectives on Climate Change: Science, Economics, Politics, Ethics (Advances in the Economics of Environmental Research), Volume 5 (2005), pp. 303–4. 35 Immanuel Kant, Groundwork of the Metaphysics of Morals, Trans. and ed. Mary Gregor (Cambridge University Press: Cambridge, 1997), 4: 429.

266

Julia Nefsky

means? You don’t seem to be treating him in any way at all by doing that particular act. Unless your act would play some significant role in causing the outcome, it does not seem we can appeal to the Formula of Humanity to explain why you ought not to do it.36 ‘Okay’, you might say, ‘but surely the more promising place to look is Kant’s Formula of Universal Law.’ That formulation might seem to be perfectly suited to avoid the Superfluity Problem. Let’s see if this is so. The Formula of Universal Law says that an act is right if and only if you can will that the maxim of the action become a universal law of nature without contradiction.37 To determine whether this condition is met, we must imagine a world in which, by your will, everyone acts on the maxim in question, and we must look for, first, a contradiction in conception38 and, second—if there is no contradiction in conception—a contradiction of the will.39 If there is neither, acting on the maxim is permissible. There is a difficult question concerning what should go into the description of the maxim. But it seems that for many Collective Harm Cases, however we describe the maxim, we are not going to get a contradiction in conception.40 In Drops of Water, whether we think of the maxim as ‘refrain from adding my pint of water to the cart, so as to keep it for myself ’, or as simply ‘keep my water supply for myself ’, or as ‘don’t give something away when doing so won’t make a difference’, there is no contradiction in conception. For each of these, it is possible to act successfully on the maxim in a world in which everyone has the same maxim. What about the ‘contradiction of the will’ test? Here we might get different results depending on how we describe the maxim. Suppose, first, that we describe the maxim with a fair amount of specificity. For example, let’s take the maxim to be ‘refrain from adding my pint of water 36 The Formula of Humanity might imply that you have other duties with respect to that victim: for example, to do what you can—if anything—to alleviate the pain, or to help him get back on his feet afterward. 37 Groundwork of the Metaphysics of Morals, 4: 421. 38 As I’m understanding it, there is a contradiction in conception if it is not possible to act successfully on the maxim in the world in which everyone has that maxim. (But note that the point I will make remains if we go with a stronger interpretation under which there is a contradiction in conception only if the world in which everyone has the maxim is logically impossible.) 39 As I’m understanding it, there is a contradiction of the will if willing the universalization of the maxim contradicts something else that you must will as a rational agent. 40 I say ‘many’ rather than ‘all’, because in free-rider Collective Harm Cases (see note 18 in this chapter), we may get a contradiction in conception under some descriptions of the maxim.

Fairness, Participation, and Collective Harm

267

to the cart, so as to keep it for myself ’. We can universalize this maxim without a contradiction of the will. For there to be a contradiction of the will it must be that my willing this maxim as a universal law is in contradiction with something else that I must will as a rational agent. There is no such contradiction. I will not be negatively affected if no one puts water into this cart. The maxim described in this way passes the test. But the problem might be that we are describing the maxim in an overly narrow way. Perhaps if we give a more general description, we can obtain the contradiction of the will that we are looking for. How can we describe the maxim of refraining from adding a pint more broadly so that the universalization test yields a contradiction of the will? Here is what seems to me to be the best (if not the only good) answer: we should describe the maxim as, ‘refrain from helping those in need.’ Willing the universalization of this maxim produces a contradiction of the will. This, after all, is the maxim of non-beneficence, one of Kant’s own examples of a maxim that fails the contradiction of the will test. But here—just when it seems we have found the contradiction we were looking for—we hit the Superfluity Problem. We can only describe my maxim of refraining from adding my pint as a maxim of ‘refraining from helping those in need’ if my adding the pint would help those in need. But the claim that my act won’t make any difference to those in need seems to be precisely a claim that it will not help them—that it would, instead, be instrumentally superfluous when it comes to alleviating their suffering. In any case, we can at least conclude that it is far from clear how the Formula of Universal Law would yield the result that you ought to add your pint to the cart, unless we can first debunk the claim that doing so will not help.41

41 Perhaps there is a (correct) way of describing the maxim that avoids this problem. Even if that is so, there is a different level at which the Superfluity Problem might show up. If we are going to appeal to the Formula of Universal Law, we need an understanding of why this test tells us what we ought to do. We need to know, in other words, why universalizability matters. It could very well be that the explanation we want to give about why universalizability matters will not work in Collective Harm Cases unless we can first show that one’s individual act is not instrumentally superfluous. For instance, a common idea is that acting in a way that is not universalizable is wrong because it exhibits a certain kind of unfairness: it amounts to relying on others to do something, while excepting yourself. For reasons similar to those given in our discussion of the Fairness Approach, this story might not work in Collective Harm Cases unless we can first debunk the impression that one’s act would be instrumentally superfluous. Many thanks to Sergio Tenenbaum for this point.

OUP CORRECTED PROOF – FINAL, 7/7/2015, SPi

268

Julia Nefsky 9. THE PROBLEM OF COLLECTIVE HARM (AND HOW TO SOLVE IT)

In Collective Harm Cases, one’s individual act (or omission) doesn’t seem to make any difference. It appears that things will be the same in all relevant respects, whether or not it is performed. And what this seems to mean is that acting in this way would be instrumentally superfluous— that it would not itself play any significant role with respect to bringing about the outcome in question. This is why it is hard to see how there could be any point in acting. While one knows that unless enough people act in this way bad consequences will result, one is faced with the apparent uselessness of acting in this way oneself. The three Rejecting the Implication views that we discussed approach this problem by trying to show that, even if acting in the relevant way can’t make a difference, there is still reason to do so. They do this by turning the focus away from the concern about the apparent instrumental superfluity of your individual act, and onto other sorts of considerations: being part of a group that makes a difference, participation in a collective action, fairness in the context of a collective obligation. But these considerations can’t do the work they are supposed to do as long as your act would be merely superfluous. Unless your act could play a significant, non-trivial role with respect to bringing about the outcome in question, the explanations these views are offering as to why you have reason to do it cannot properly address what is at issue in the Problem of Collective Harm. The same problem seems to arise for various nonconsequentialist theories, if we think we can turn to them to provide a solution to the Problem of Collective Harm. In general, to address what is at issue in the Problem of Collective Harm we need to be able to explain how it is that the fact that the men in the desert are suffering from painful thirst means that each of us has reason to add our pints, and that the fact that we are collectively causing climate change means that each of us has reason to take a bicycle to work instead of a car, and so on. That is, we need to give a reason for action that has to do with—in a central way—the morally relevant outcome of concern, and that tells us specifically to do the sort of acts that could collectively cause (or, depending on the case, prevent) that outcome. It is doubtful we can do this adequately while at the same time accepting that acting in this way would be merely superfluous with respect to bringing about that outcome.

Fairness, Participation, and Collective Harm

269

What all of this reveals is that to solve the Problem of Collective Harm, we need to show that one’s individual act isn’t merely superfluous. We need to show that we are mistaken when it seems to us that an individual act of the relevant sort won’t itself do anything instrumentally significant. Of course, it’s not clear that we can do this. But unless we can, a satisfactory answer to the problem seems unlikely. If we can show that an individual act of the relevant sort isn’t merely superfluous, and rather could significantly help to bring about the outcome in question, this would address the Superfluity Problem that Weak Participation, Strong Participation and the Fairness Approach face. So, that core problem with those views would be resolved. Similarly, the general non-consequentialist theories we considered would then, it seems to me, be able to handle Collective Harm Cases. For instance, if one’s turning up the dial on the shock machine (in order to get $50) might play a significant role in harming the victim, it makes sense to say—as the Kantian would want to—that in doing so one is treating him as a mere means. But the point isn’t just that this would allow these views about our moral reasons in Collective Harm Cases to get off the ground. Rather than just being a matter of filling in the hole that we’ve identified with these views, showing that an individual act can do something significant toward bringing about the relevant outcome is the central task of solving the Problem of Collective Harm. I’ve argued that it is doubtful that we can solve the problem in a satisfactory way without showing this. But, importantly, if we can explain how it is that your individual act can do something instrumentally significant, this would itself solve the core issue of the Problem of Collective Harm. What leads us to fail to see how we could have reason for action in Collective Harm Cases is the impression that doing so would be useless—that it wouldn’t do anything significant toward bringing about the relevant good outcome, or preventing the relevant bad one. If we could show that this were not true, that core issue would be addressed. Now, what this might seem to mean is that if we want to solve the problem, we need to take a Denying the Description approach. That is, we must try to argue that in Collective Harm Cases, contrary to appearances, individual acts do, or at least might, make a difference with respect to the morally relevant outcome of concern. This line of response has largely been taken by those seeking a consequentialist solution to the

270

Julia Nefsky

problem. But, since it would address the Superfluity Problem if it worked, one implication of our discussion is that non-consequentialists also have a stake in whether such a view can succeed. I discuss the Denying the Description strategy elsewhere. Unfortunately, if my conclusions there are correct, such an approach does not work.42 I don’t doubt that in some cases we can and should deny the “it won’t make a difference” claim: we are surely sometimes making a mistake when we think that one act won’t make a difference. But, while that will surely be true in some cases, I argue that we cannot draw that conclusion in general. We are not, I argue, in a position to deny that there are Collective Harm Cases in which one’s individual act simply can’t make a difference. Suppose, for the sake of argument, that I am right about this. It might seem, then, that we have hit a standstill. On the one hand, the upshot of the present paper is that we cannot solve the Problem of Collective Harm unless we can maintain that one’s individual act can play a significant instrumental role with respect to the outcome of concern. But on the other hand, if my conclusions elsewhere are correct, we cannot deny that there are Collective Harm Cases in which one’s individual act simply can’t make a difference. But rather than a standstill, these conclusions together carve out a new space in which a solution to the problem might be found. Instead of taking a Denying the Description approach, we could try to show that an act can play a significant instrumental role in bringing about an outcome, even if it cannot make a difference with respect to that outcome. If an act cannot make a difference, we take this to mean that it would be instrumentally superfluous. A way of solving the Problem of Collective Harm would be to show that this assumption is mistaken. What would it mean for this assumption to be mistaken? Whether one’s act makes a difference is a particular, counterfactual matter. It is a matter of whether had one not done it, things would have been (relevantly) different. The assumption would be mistaken if this were not the correct test for determining whether an act is instrumentally useful. Perhaps my adding a pint to the cart could play a significant, nonsuperfluous role in alleviating the men’s suffering even though the extent

42

“Consequentialism and the Problem of Collective Harm: A Reply to Kagan”.

Fairness, Participation, and Collective Harm

271

to which their suffering is alleviated will not be different depending on whether or not I do so.43 In summary: the upshot of our investigation of Rejecting the Implication approaches is that to respond to the “it makes no difference” argument, we need to be able to show that one’s individual act in Collective Harm Cases is not instrumentally superfluous. This is the real challenge of the Problem of Collective Harm. Attempting to address the problem by turning our attention to other sorts of reasons for action will not work. One way of trying to meet the challenge is to take a Denying the Description approach: to try to show that an individual act actually can make a difference in Collective Harm Cases. But that is not the only way. There is a second strategy that remains open: we could question whether an act needs to be a difference-maker in order to be instrumentally useful. Acknowledgments I am very grateful for helpful comments from Markus Kohl, Niko Kolodny, Mark Lebar, Alex Rennet, Abraham Roth, Carolina Sartorio, Sergio Tenenbaum, R. Jay Wallace, and two anonymous referees for Oxford University Press. I also owe much gratitude to the participants of the Fifth Annual Arizona Normative Ethics Workshop, and to audiences at Carleton University and the 2013 Toronto-Montreal Joint Ethics Centre Conference at the University of Montreal.

43 I am not claiming that this strategy will work. For the purposes of this paper, the point is just that this strategy exists as a second way—besides denying the “it makes no difference” description—of trying to address the superfluity challenge. I attempt to develop a solution of this sort in “How You Can Help, Without Making a Difference.”

12 The Virtue of Authenticity ERNESTO V . GARCIA

1. INTRODUCTION

“If there’s one theme in all my work it’s about authenticity and self-expression. It’s the idea that some things are in some real sense really you, or express what you are, and others aren’t.” - Bernard Williams1

The idea of ‘authenticity’ plays an important role in various fields, including sociology, psychology, cultural anthropology, literary studies, political theory, and business marketing.2 Among philosophers, it is most often associated with existentialist thinkers like Heidegger and Sartre.3 In recent years, however, a growing number of analytic moral philosophers have taken up this topic. This includes not only philosophers writing in the context of the autonomy literature, such as John Christman, Gerald Dworkin, Harry Frankfurt, Alfred Mele, Diana Meyers, Marina Oshana, J. David Velleman, and others, but also many philosophers who defend more general accounts of authenticity including K. Anthony Appiah, Charles Larmore, Charles Taylor, and Bernard Williams.4 Nonetheless, we might worry from the outset that this entire enterprise is doomed to failure. Talk of an ‘authentic’ or ‘true’ self, of ‘bad faith’ and ‘inauthenticity’, and of the ideal of ‘being true to oneself ’ seem like 1

. For discussions of authenticity in the social sciences, see, e.g., Bendix 1997, Miller 1997, Lindhom 2008, and Vannini and Williams 2009; in political theory, see Berman 1970 and Rossinow 1998; in literary and cultural studies, see Trilling 1971, Newman 1997, and Cheng 2004; and in business/marketing, see Boyle 2003, Gilmore and Pine 2007, and Morin 2010. 3 For some helpful discussion, see Bell 1989, Charme 1991, Golomb 1995, and Wrathall and Malpas 2000. 4 See, e.g., Dworkin 1989; Meyers 1989 and 2000; Taylor 1989, 1992, and 1994; Appiah 1993, 1996, and 2005; Frankfurt 1998 and 1999; Williams 2002; Larmore 2004; Oshana 2005; Velleman 2005; and Christman 2009. 2

The Virtue of Authenticity

273

hopelessly vague and fuzzy ideas, ones which capture popular imagination but fail to provide any genuine ethical insights. In order to address these skeptical worries, I attempt to offer a detailed account of authenticity here, both what it is and why it is valuable. This paper has three parts. First, in §§2–3, I critically examine three popular conceptions of authenticity. Second, in §4, I present my own account of authenticity. I defend three main claims: (1) that authenticity is best seen as a personal rather than social virtue; (2) that it is typically constituted by three main elements, viz., self-understanding, self-expression, and self-concern—in particular, concern about what kind of person one is and what type of life one leads; and (3) that it resembles traditional Aristotelian virtues insofar as it can involve steering a middle course between two extremes, viz., a ‘deficiency’ (i.e., various forms of inauthenticity) and an ‘excess’ (i.e., a type of narcissistic self-centeredness which critics of authenticity rightly disparage). Third and lastly, in §5, I examine why authenticity is valuable. 2. THREE POPULAR CONCEPTIONS OF AUTHENTICITY

Nowadays we find talk of ‘authenticity’ almost everywhere. It is pervasive in product advertising, such as for Stoli Vodka (“Choose Authenticity”), Kool cigarettes (“Be Authentic”), Ralph Lauren Jeans (“Authentic Denim Outfitters”), Abercrombie and Fitch (“Authentic American Clothing Since 1892”), Diesel Jeans (“Don’t Listen to Them. Don’t Listen to Us. Be Authentic. Be Yourself.”), and highbrowfurniture.com (“Authenticity. Period.”). It is common in tourism slogans, including for Maryland (“Even the Fun is Authentic”), Florida (“Real. Authentic. Florida”), Niagara Falls (“The Authentic Falls Experience”), and Cuba (“Authentic Cuba”). And many businesses build the idea of authenticity into their very names, including Authentic Bagel Company, Authentic Business Systems, Authentic Clothing, Authentic Media, Authentic Publishers, Authentic Records, Authentic Sports Management, Authentic Technologies, etc. Indeed, Time magazine in 2008 declared ‘authenticity’ to be one of the “10 ideas that are changing the world”.5 The main worry is that the term ‘authenticity’ has been used so often, and in so many different contexts, that it has become virtually devoid of

5

.

274

Ernesto V. Garcia

meaning. Nearly everybody agrees that the concept of authenticity involves some claim about ‘being true to oneself ’. But can we say anything more concrete than this? In general, I argue that most of our talk of authenticity in popular culture falls under three broad headings, each of which has important philosophical precedents. These three conceptions of authenticity are:6 INDIVIDUAL AUTHENTICITY, which emphasizes a more social sense of authenticity (cf. Nietzsche on ‘the herd’, Kierkegaard on ‘the crowd’, and Heidegger on ‘the They’) 2. NATURAL/ORIGINAL AUTHENTICITY, which emphasizes a more metaphysical sense of authenticity (cf. Rousseau on the ‘natural man’ or so-called ‘noble savage’ in contrast to ‘modern man’) 3. TRUTHFUL AUTHENTICITY, which emphasizes a more ethical sense of authenticity (cf. Sartre on ‘bad faith’)

1.

These three conceptions of authenticity are related to corresponding accounts of the nature of the ‘self ’. Each involves a set of binary oppositions, where the former terms are regarded as authentic, the latter as inauthentic. These are: 1. The ‘individual’ self vs. the ‘social’ self 2. The ‘natural’/‘original’ self vs. the ‘socialized’ self 3. The ‘true’/‘honest’/‘sincere’ self vs. the ‘false’/‘dishonest’/‘insincere’/ ‘hypocritical’ self The first and arguably most influential popular conception of authenticity is INDIVIDUAL AUTHENTICITY. On this more social approach, we are most authentic when we are ‘unique’, ‘original’, or ‘stand apart from the crowd’. This account has both negative and positive dimensions. In a negative sense, in order to be authentic, we must somehow separate ourselves from what various thinkers call ‘the herd’ (Nietzsche), ‘the crowd’ (Kierkegaard), or ‘the They’ (Heidegger) understood in terms of the general masses. We do so by not giving in to peer pressure, by not 6 My discussion, which contrasts the general ‘concept’ of authenticity with three popular ‘conceptions’ of authenticity, follows Rawls’ well-known distinction between ‘concept’ and conception—see Rawls 1971:5.

The Virtue of Authenticity

275

being mere ‘sheep’ or social conformists who just thoughtlessly ‘go along with the flow’. Instead, in a positive sense, we must live up to the responsibility of ‘being our own person’. This account of authenticity is widespread. We see it implicitly assumed by a tourism slogan for Maine (“Where original people perfectly complement the beauty of this place ( . . . ) Be adventurous. Be yourself. Discover your Maine thing.”) and explicitly stated in the Diesel Jeans ad campaign mentioned above (“Don’t Listen to Them. Don’t Listen to Us. Be Authentic. Be Yourself.”).7 In an interview with the Pittsburgh Post-Gazette, the Duchess of York, Sarah Ferguson, appeals to this sense of authenticity when she claims: “If you fear what people think about you, then you are not being authentic.”8 And Charles Taylor provides a more philosophical analysis of this notion of authenticity when discussing the German philosopher Johann Gottfried Herder, writing: Herder put forward the idea that each of us has an original way of being human ( . . . ) There is a certain way of being human that is my way. I am called upon to live my life in this way, and not in imitation of anyone else’s ( . . . ) Being true to myself means being true to my own originality, and that is something only I can articulate and discover ( . . . ) This is the background understanding to the modern ideal of authenticity. (1992:28–9)

As Edward Young succinctly captures the overall idea: “Born Originals, how comes it to pass that we die Copies?”9 To sum up, on this first approach, we are most authentic when we are ‘individual’ versus ‘social’ selves, that is, when we are true to our individual or unique natures, however much this might bring us into conflict with the society in which we live. The second popular conception of authenticity is NATURAL/ORIGINAL AUTHENTICITY. On this more metaphysical approach, we are most authentic when we live in conformity with our more ‘natural’ or ‘original’ self. The overall idea is that the closer that some person, place, or thing is to its ‘natural’ or ‘original’ state—in terms of being ‘pure’, ‘innocent’, ‘raw’, ‘unfiltered’, or ‘unadulterated’—the more authentic it is. In Authenticity: What Consumers Really Want, James Gilmore and Joseph Pine describe 7

. . 9 Young 2012, §164. 8

276

Ernesto V. Garcia

this sense of authenticity as follows: “People tend to perceive as authentic that which exists in its natural state in or of the earth, remaining untouched by human hands; not artificial or synthetic” (2007:49). We again find this conception of authenticity in many advertising campaigns, such as for The Club at Spanish Peaks, a luxury development in Big Sky, Montana (“Authentic Montana: Unspoiled. Uncrowded. Unpretentious”); Whole Foods, billed as “the world’s largest retailer of natural and organic foods”, whose very “essence”, according to co-CEO Walter Robb, is “authenticity”;10 and Michigan’s recent award-winning tourism campaign, which purportedly presents us with ‘authentic’—in the sense of natural, raw, or unfiltered—images of “Pure Michigan”.11 This conception of authenticity also underlies our talk of ‘authentic’ cuisine (e.g., authentic ‘fill-in-your-favorite-nationality-food’), ‘authentic’ musical performances (e.g., Baroque musical compositions played on period instruments), and ‘authentic’ culture (e.g., authentic Native American or aboriginal culture, where this typically includes authentic food, music, dance, clothing, art, etc.). In the 2nd Discourse, Rousseau discusses this conception of authenticity. Reflecting on the fateful transition from ‘original man’, living in a state of nature, to people living in modern civil society, he writes: In a word, [the intelligent reader] will explain how the human soul and passions, by imperceptible adulterations, so to speak change in Nature; why in the long run the objects of our needs and of our pleasures change; why, as original man gradually vanishes, Society no longer offers to the eyes of the wise man anything but an assemblage of artificial men and factitious passions which are the product of all these new relationships, and have no true foundation in Nature. (1997:186)

Rousseau argues here that entering into modern civil society has somehow made us lose touch with our ‘authentic’ selves.12 That is, this overall socialization process—typically involving a kind of inflamed amourpropre—has brought about a drastic ‘change’ in human nature. It 10 . 11 . 12 For some helpful discussions of Rousseau on authenticity, see Ferrara 1993 and O’Hagan 1999.

The Virtue of Authenticity

277

transforms us into ‘artificial’ persons with ‘factitious passions’ and needs, a condition which has ‘no true foundation in Nature’. And in The Drama of the Gifted Child: The Search for the True Self, 20th-century Swiss psychoanalyst Alice Miller defends a similar claim with respect to individual persons. She argues that in the process of becoming adults, each child typically: ( . . . ) develops in such a way that he reveals only what is expected of him ( . . . ) A process of emptying, impoverishment, and crippling of his potential actually took place [in which] the integrity of the child was injured when all that was alive and spontaneous in him was cut off. (1997:11–12)

Miller identifies what she calls the ‘true self ’—that is, who we ‘authentically’ are—with our more natural, pure, or spontaneous self. This stands in contrast to the ‘as-if personality’ we develop in order to satisfy the expectations of our parents and other authority figures. The third and final popular conception of authenticity is TRUTHFUL AUTHENTICITY. On this more ethical approach, we are most authentic when we are ‘true’, ‘honest’, or ‘sincere’ selves as opposed to being ‘false’, ‘dishonest’, ‘insincere’, or ‘hypocritical’ selves. This type of authenticity is highly prized in present-day society. As Tim Russert of NBC’s Meet the Press laments: “People are begging for authenticity.”13 And as Washington Post Circle columnist Kathleen Parker argues, what matters most for us with regard to politicians “( . . . ) is authenticity. There's no surer way to lose the public's confidence than to pretend to be something you’re not.”14 This conception of authenticity presumably lies behind statements like those of Anderson Cooper (“In everything I’ve done, I’ve always tried to just be authentic and real”), Hilary Rodham Clinton (“I believe in being as authentic as possible”), and Michele Bachman (“I think that what people see in me is that I’m a real person. I’m authentic”, declared after her Iowa victory).15 And we condemn those who fail to display this type of authenticity. Recall Tom Brokaw’s famous statement about Mitt Romney’s 2012 presidential campaign on The Last Word

13

Interview on The Bob Edwards Show, June 7, 2006, on XM Public radio channel 133. . 15 . 14

278

Ernesto V. Garcia

with Lawrence O’Donnell: “This is the most inauthentic spontaneous candidate that I’ve ever seen.”16 As one commentator interpreted Brokaw’s remark: “‘Inauthentic’. Did Tom Brokaw just call [Romney] a phony ( . . . ) or a liar?”17 In general, we demand TRUTHFUL AUTHENTICITY just as much from ordinary people as from public figures. We regard liars, cheats, fakes, phonies, and hypocrites—that is, anyone who fails to “walk the talk”—as inauthentic. Sartre arguably had this type of authenticity in mind with his account of ‘bad faith’. There are two important qualifications about this taxonomy. First, although these three popular conceptions of authenticity often overlap, they are conceptually distinct. For example, we can imagine somebody who displays INDIVIDUAL AUTHENTICITY—that is, she is a unique individual who stands apart from the crowd—while failing to be authentic in the other two senses. This could happen if, for example, contra NATURAL AUTHENTICITY, her individuality is based upon how all her more socialized personality traits combine to distinguish her from everybody else. And contra TRUTHFUL AUTHENTICITY, her individuality could be based precisely on the fact that she is dishonest or insincere, in contrast to the prevailing ethical norms of her society. Second, although I have suggested that these three popular conceptions are related to various important philosophers—(1) INDIVIDUAL AUTHENTICITY with Nietzsche, Kierkegaard, and Heidegger, (2) NATURAL AUTHENTICITY with Rousseau, and (3) TRUTHFUL AUTHENTICITY with Sartre—I do not think that any of these thinkers would endorse the specific versions of authenticity as laid out here. In the next section, I offer suggestions for how we should refine these popular conceptions of authenticity and identify what seems to be the common valid core underlying each of them. 3. REFINING THE THREE POPULAR CONCEPTIONS OF AUTHENTICITY

While all of these three popular conceptions of authenticity have widespread appeal, I think that, as presently understood, they are too onesided. By correcting these deficiencies, I argue that we can ultimately arrive at a more satisfying account of authenticity. From the outset, 16 . 17 .

The Virtue of Authenticity

279

however, it is interesting to observe that all three conceptions of authenticity appear to face a similar practical difficulty. Generally speaking, it seems that striving for authenticity in these various ways can often be self-defeating, at least when judged by the standards of each particular conception. With respect to INDIVIDUAL AUTHENTICITY, we find that people who attempt to be original, unique, or to stand apart from the crowd—that is, to achieve what K. Anthony Appiah calls “the Bohemian ideal” (2005:106)—often ironically end up just resembling every other non-conformist. With respect to NATURAL AUTHENTICITY, it seems that our self-conscious striving to be ‘natural’ often leads us to behave in very unnatural ways. As La Rochefoucauld wryly remarks: “Nothing makes it so difficult to be natural as the desire to appear so.”18 And with respect to TRUTHFUL AUTHENTICITY, Sartre insists that in their efforts to be sincere, so-called “champions of sincerity” often engage in many acts of selfdeception. As he famously remarks, it is a “truth recognized by all that one can fall into bad faith through being sincere.” (1984:109). Call these three practical difficulties related to our various attempts at being authentic the ‘Nonconformist’, ‘Naturalness’, and ‘Sincerity’ Problems. While none of them undermine these three popular conceptions of authenticity, they do serve to remind us that we must be careful in how we formulate the ideals related to each conception of authenticity in order to avoid certain common pitfalls. In the rest of this section, I explore some deeper worries that force us to revise more fundamentally how we should understand each particular conception of authenticity. First, with regard to INDIVIDUAL AUTHENTICITY, the main worry is it is very tempting to interpret this conception of authenticity—with its ideals of ‘standing apart from the crowd’, of not being a ‘social conformist’, of not getting lost in the ‘They’, etc.,—as defending an overly individualistic view of the self in which the individual must stand over against all social influences. As most philosophers point out, however, our very identity as individuals is deeply shaped by the societies in which we live. We are ‘social selves’ in at least three different ways. First, we come to know what it even means to be a person through our interactions with others. As Annette Baier argues, we are essentially ‘second persons’, that is, individuals “who [are] dependent upon other persons to

18

La Rochefocauld, Maximes, §431, as quoted in Larmore 2010:28.

280

Ernesto V. Garcia

acquire the essential arts of personhood (1985:84).” Second, most of our concrete identities—e.g., as parents, children, friends, lovers, members of an ethnic group, citizens of a state, etc.—are inextricably bound up with various social meanings. Third and lastly, most of our individual pursuits not only derive their significance from, but are also necessarily sustained by, the various social practices and traditions in which we find ourselves immersed. As Heidegger expresses this overall idea in Being and Time: Authentic Being-one’s-Self does not rest upon an exceptional condition of the subject, a condition that has been detached from the ‘they’: it is rather an existentiell modification of the ‘they’ – of the ‘they’ as an essential existentiale. (1962:168)

Heidegger argues that, in order to be authentic, we should not try to entirely ‘detach’ ourselves from society. Rather, we should just ‘modify’ our relation to it in the right kind of way. Indeed, he insists that our relationship to ‘the They’ (das Man) constitutes an ‘essential’ part of who we are. Charles Taylor criticizes overly individualistic conceptions of authenticity on very similar grounds, arguing: Otherwise put, I can define my identity only against the background of things that matter ( . . . ) Only if I exist in a world in which history, or the demands of nature, or the needs of my fellow human beings, or the duties of citizenship, or the call of God, or something else of this order matters crucially, can I define an identity for myself that is not trivial. (1991:40–1)

Understood this way, we should revise our conception of INDIVIDUAL to allow for the fact that who we really are as individuals often cannot be understood apart from our concrete social identities. Second, with regard to NATURAL/ORIGINAL AUTHENTICITY, the main worry is that we can understand this conception of authenticity in at least three different ways, where the first two seem highly problematic. First, we might be claiming that we are only ever truly authentic insofar as we are wholly natural or non-socialized selves, e.g., infants or some ‘natural savage’ living in a brute animal-like existence. On this view, it seems that authenticity is impossible for the majority of us insofar as we can never truly return to such an original state. As Rousseau claims in a famous

AUTHENTICITY

The Virtue of Authenticity

281

footnote from the 2nd Discourse, we can never recover our “first innocence” since our “passions have forever destroyed their original simplicity” (1997:203). More broadly, in this original state where ex hypothesi we entirely lack any self-awareness, there seems to be no possibility for being inauthentic. We just are what we are. But if there’s no genuine possibility for being inauthentic, then it seems that there’s no genuine possibility for being authentic either, at least in any substantive sense. Second, we might be claiming that this ‘natural’ or ‘original’ self is based upon some wholly unchanging ‘natural core’ which remains exactly the same throughout all the changes in our lives. The main problem with this approach is that it seems to endorse an overly static or fixed conception of the true self. Even if there is some ‘natural core’ which persists throughout our entire lives, we should allow for a third possibility, viz., that being authentic is compatible with this ‘natural core’ being deeply modified over time through a long process of growth and maturity—something Rousseau himself recognizes in most of his writings.19 Indeed, it seems possible that being authentic might involve giving up certain ‘natural’ or ‘original’ traits that we no longer identify with. Third and lastly, with regard to TRUTHFUL AUTHENTICITY, the main worry is that this conception of authenticity is ambiguous between two very different interpretations, viz., (1) being true to oneself or (2) being true to others. Notice that nearly all the quotes above related to TRUTHFUL AUTHENTICITY go back and forth between these two interpretations, typically assuming that (1) and (2) just coincide. But this seems mistaken for two reasons. First, being truthful, honest, or sincere with other people does not seem to be a necessary condition for being authentic. Take, for example, the standard case where we need to lie to a murderer or a Nazi at the door in order to protect innocent lives. In such circumstances, it seems that choosing to lie can be a fully authentic course of action. Indeed, we might feel that not lying to the murderer or the Nazi in this case would amount to an inauthentic violation or betrayal of who we really are as persons. An even more radical possibility is that somebody’s ‘true self ’ could be essentially deceptive at its core. That is, akin to Milton’s Satan who declares “Evil, thou art my good”, it seems logically possible for

19

For helpful discussion of this issue, see O’Hagan 1999, esp. Chs. 3–4 and 6–7.

282

Ernesto V. Garcia

somebody to honestly declare, “Lying or deception, thou art my true nature.”20 In this case, being authentic would in fact require her to always lie to others. Second, being truthful, honest, or sincere with other people does not seem to be a sufficient condition for being authentic either. For example, consider Sartre’s portrayal of the character Inez in his play No Exit. Inez is clearly honest—indeed, brutally so—towards other people. Nonetheless, Sartre depicts her—and most so-called ‘champions of sincerity’—as inauthentic or in ‘bad faith’ insofar as she is self-deceived about her own ability to choose and act differently than she does.21 Thus, while TRUTHFUL AUTHENTICITY does appear to necessarily include some idea of being ‘true to ourselves’, we should avoid saddling this conception of authenticity with an overly moralized view of the true self which would also require us to be truthful, honest, or sincere with other people. In the end, we can affirm these three popular conceptions of authenticity just as long as we steer clear of implausibly interpreting what it means to be ‘true to oneself ’ in either (1) overly individualistic, (2) overly static or fixed, or (3) overly moralized ways. In the next section, I try to build upon this key insight in developing my own account of authenticity. 4. AUTHENTICITY AS A PERSONAL VIRTUE

Based on the results from §3, I propose the following highly formal account of authenticity, viz., that whatever the correct philosophical analysis of the ‘true’ or ‘authentic’ self is, authenticity simply involves relating to this self in the right kind of way. Obviously, we need to clarify two things here: (1) what we mean by a ‘true’ or ‘authentic’ self; and (2) what it means to relate to this self ‘in the right kind of way’. First, in keeping with a more formal approach, I want to leave the nature of what I will henceforth call the ‘true self ’ relatively open. In particular, I do not want to be necessarily committed to thinking about the true self in any metaphysically dubious way, as some kind of substantial entity “waiting to be found” or “buried deep within”.22 It seems that our true self could be just as much a matter of ‘self-definition’ or ‘self-creation’ as it

20 Of course, any such declaration would presumably occur in a moment of weakness, given their essential nature as liars. 21 22 See Sartre 1984:109–12. Cf. Rubin 1998:109.

The Virtue of Authenticity

283

is a matter of ‘self-discovery’ of various preexisting traits (cf. Meyers 1989 and 2000 and Appiah 2005). Furthermore, I do not want to be necessarily committed to thinking about the ‘true self ’ in standard Frankfurtian ways. In particular, the present account does not presuppose that our ‘true self ’ must involve (1a) some ‘motivational essence’ without which we would no longer be us, or (1b) what we are ‘wholehearted’ about, or (1c) what we ‘reflectively endorse’ or ‘identify’ with.23 That is, I leave it an open question whether Frankfurt’s many critics are correct (as I suspect they are to a large extent) (2a) that ‘motivational essences’ amount to misleading metaphorical talk; (2b) that our true self could be ‘intersectional’ or deeply conflicted in nature, embodying opposing elements, rather than necessarily ‘wholehearted’ and ‘unified’; and (2c) that our true self might involve not only those elements we reflectively endorse or identify with, but also certain central repressed aspects of our personality which we might be reluctant—or even altogether refuse—to own up to.24 Instead, for present purposes, I think that all we need to assume about our true self is that “some of our concerns are authoritative for us because they are somehow central to our personalities” (Velleman 2005:339, emphasis added). By talking about certain concerns being ‘authoritative for us’—or, in the present idiom, as representing our ‘true selves’— insofar as ‘they are somehow central to our personalities’, I want to defend a relatively modest thesis. The main idea is that my ‘true self ’ or what is most ‘central’ to my identity is constituted by those personal features which, objectively speaking, have a more significant or wideranging impact upon how I think, feel, and act, than those which do not. To adopt a spatial metaphor, such features may have (a) great breadth, in the sense that they are what Arpaly and Schroeder (1999) describe as ‘well-integrated’ features that connect up with many other aspects of my personality. Or they may have (b) great depth, in the sense that they are features which—whether widely-connected or relatively isolated—I am strongly committed to.25

23

Cf. Frankfurt 1988 and 1999. For helpful discussion of these points, see, e.g., Meyers 2000; Arpaly and Schroeder 1999; and Velleman 2005. 25 Unlike Arpaly and Schroeder 1999:173, who regard ‘depth’, or firmness or strength, as merely one aspect of being ‘well-integrated’, I separate these two factors since it seems that, in many cases, they can come apart. 24

284

Ernesto V. Garcia

Second, we need to clarify what it means to relate to this true self ‘in the right kind of way’. Again, I want to leave this claim relatively open. More specifically, I agree with Charles Larmore that authenticity can come in either reflective or non-reflective forms.26 In general, it seems possible for certain people to be authentic in a wholly non-reflective way, where they might even be ignorant about the nature of their true self. Consider the case of a person who displays what Julia Driver calls a “virtue of ignorance” like, say, modesty, where one underestimates one’s own self-worth.27 It seems possible that she can authentically conform to her ‘true self ’, i.e., her modest nature, despite the fact that she is necessarily ignorant about certain facts about her own identity. For present purposes, however, I want to focus mainly upon what Larmore calls more ‘reflective’ forms of authenticity. This is instructive for two reasons. First, this type of authenticity is most likely what traditional existentialists had in mind—and thus, probably what many of us also have in mind insofar as existentialism has had a wide-ranging impact on our present-day thinking about authenticity. For instance, Heidegger describes authenticity as a kind of self-conscious ‘resoluteness’ (Entschlossenheit), and both Heidegger’s and Sartre’s main examples of inauthenticity involve non-reflective persons who either allow themselves to be mindlessly caught up in ‘the They’ [Heidegger] or who refuse, in an act of ‘bad faith’, to fully own up to their own ‘facticity’ or ‘transcendence’ [Sartre].28 Second, as we will see later, elaborating upon this more reflective form of authenticity will provide us with helpful resources for thinking about authenticity as akin to traditional Aristotelian virtues. For the sake of convenience, I will call this more reflective version of authenticity ‘Existentialist Authenticity’ [EA], leaving aside the historical issue of whether this account faithfully captures all of Heidegger’s and Sartre’s central views about authenticity. EA requires satisfying three main conditions, viz., that:

26 See Larmore 2010, esp. Chs. 1 and 3–6. I am grateful for an anonymous referee for stressing the importance of both Larmore’s and Driver’s contributions here and for pressing me to defend a more balanced treatment of these two different forms of authenticity. 27 See Driver 2001, esp. Ch. 2. 28 See Heidegger 1962:163–8, 293–311, 347–58, and Sartre 1984:96–116.

The Virtue of Authenticity

285

1. S has an accurate understanding of her true self 2. S lives in conformity with that accurate self-understanding 3. S cares about what type of person she is and lives in conformity with her accurate self-understanding, at least to some extent, out of such concern Let me take each condition in turn. First, EA requires that (1) we have an accurate understanding of our true self. This involves two dimensions. First, we must have a minimal self-awareness of who or what we are like as persons. Second, this self-awareness must be accurate, at least to some extent. For example, suppose that being Catholic constitutes an important part of Bob’s identity, informing most of his actions and choices. If Bob correctly recognizes this fact about himself, then he has an accurate self-understanding at least with respect to this part of his identity. There are three important caveats here. First, what this accurate selfunderstanding involves can vary widely. While it presumably does not require knowing certain trivial minutiae about ourselves (e.g., our exact red blood cell count, the average number of steps we walk every day, the fact that we prefer drinking water from a straw, etc.), there are many aspects of who and what we are like as persons which are relevant for authenticity. These may include various likes and dislikes, tastes, values, and ideals; talents or abilities; general behavioral dispositions; and more robust facts like what Rorty and Wong describe as ‘identity-defining’ traits.29 Second, we can have varying degrees of self-understanding. At one extreme, it seems possible to have an accurate global self-understanding such that we know all aspects of our identity. For ordinary human beings, however, we only ever achieve an accurate local self-understanding. That is, we typically have an accurate self-understanding about certain aspects of our identity, while lacking it for others. Third and lastly, our accurate self-understanding can be either implicit or explicit. Not all cases of selfunderstanding require forming an explicit belief of the sort: “I am really X as opposed to Y”. Take the case of John, a ‘salt-of-the-earth’ type who eats, drinks, and breathes farming. He has never explicitly reflected upon his identity as a farmer. Nonetheless, for the sake of argument, assume that he fully embraces this lifestyle and would not have it any other

29

Rorty and Wong 1990:28.

286

Ernesto V. Garcia

way. Does John possess an accurate self-understanding about his identity as a farmer? It seems possible. Consider a situation where John is forced to temporarily reside in a large metropolis like LA. In this case, it seems that John might be able to immediately recognize—without any need for self-scrutiny—that an “LA lifestyle” is not for him. What this example shows is that John likely possessed an accurate implicit self-understanding about what type of lifestyle is most suitable—or more precisely, not suitable—for him all along, even if he never explicitly formulated this belief for himself. Second, EA requires that, in addition to having an accurate selfunderstanding, (2) we must live in conformity with that self-understanding. Following Bernard Williams, perhaps the best way to explain this condition is in terms of the idea of ‘self-expression’.30 As Abraham Maslow argues, we can distinguish between two general types of actions: ‘coping’ and ‘expressive’ actions.31 Coping actions are instrumental, purposive actions directed towards some aim. That is, they are cases of ‘means-ends behavior’ which involve “an interaction of the [subject] with the world, adjusting each to the other with mutual affect” (1949:264). By contrast, expressive actions are non-instrumental or non-purposive. As Maslow puts it, an expressive action should be seen as an “end-in-itself ”, one which “simply mirrors, reflects, signifies, or expresses some state of the organism” (1949:262). Notice that the same act-type can fall under either category. For example, I could shovel the snow from my neighbor’s driveway because I intend to ask a favor from her later on or just out of a concern to help somebody in need. The former is a ‘coping’ action aimed at some future benefit, whereas the latter is an ‘expressive’ action which simply expresses my concern for my neighbor’s well-being. With this distinction in mind, I argue that living in conformity with our accurate self-understanding amounts to ‘mirroring’, ‘reflecting’, ‘signifying’, or ‘expressing’ our true self. What does this mean in concrete terms? Consider a parallel case. Say that Mary is my close friend. The fact that I am deeply committed to my friendship with Mary constitutes what Joseph Raz calls an ‘expressive reason’ for me to think, feel, and act 30

See fn. 1 above. Maslow 1949. For a helpful discussion of ‘expression’ in general, see Swanton 2003, Ch. 6. 31

The Virtue of Authenticity

287

in certain ways.32 To live in conformity with an accurate understanding of my friendship with Mary requires ‘expressing’ myself in various ways, e.g., having friendly feelings towards her, being disposed to think about her often, typically preferring her company over others, etc. In a similar way, I argue that the fact that I am deeply committed to some X (e.g., a goal, project, social role, etc.) such that X comprises a genuine part of my ‘true self ’ likewise constitutes an ‘expressive reason’ for me to think, feel, and act in certain ways. For example, suppose that being a philosopher comprises a deep part of my ‘true self ’ or what is most ‘central’ to me as a person. If so, then living in conformity with my accurate selfunderstanding requires me to conduct myself in various ways—e.g., reading articles, writing papers, giving talks, etc.—that express my identity as a philosopher. Third and lastly, EA requires that (3) we care about what kind of person we are, and choose to live in conformity with our accurate self-understanding, at least to some extent, out of such concern. That is, according to EA, it is not sufficient that we have an accurate self-understanding and that we happen to live in conformity with this self-understanding in terms of our thoughts, feelings, and actions reflecting or expressing our ‘true self ’. In addition, these two conditions need to be connected in the right kind of way. For example, suppose (1) that I have an accurate self-understanding such that I know that being a philosopher is one of my central concerns or comprises a part of my ‘true self ’. Suppose also (2) that I live in conformity with this self-understanding, doing all the activities that philosophers typically do. But lastly, suppose (3) that I do not ultimately care about whether (1) or (2) obtains. That is, at the end of the day, it does not ultimately matter to me what kind of person I am and what type of life I lead. What seems to be missing in this case is what Larmore refers to as ‘practical reflection’, which he characterizes as follows: “For we can turn back upon ourselves, not with the intent of expanding our knowledge, but in order to take an explicit stand, to devote ourselves to beliefs, feelings, or actions that may already be unreflectively ours or that we are now making our own for the first time ( . . . ) [this involves] doing these things in full consciousness of what we are doing and thus assuming responsibility for everything these engagements normally entail.” (2010:24) 32

See Raz 2002:258–60.

288

Ernesto V. Garcia

Generally speaking, what I am calling ‘self-concern’ here amounts to what Larmore describes as ‘taking a stand’—that is, where we genuinely commit ourselves to one course of action rather than another, thereby taking responsibility for the activity of constituting ourselves, via our choices and actions, into the kind of people we ultimately want to be. In order to fully appreciate EA, it may be helpful to consider here five paradigmatic examples of inauthenticity: Shallow: Abe is a college student who has never considered what he really wants. Instead, he just does what’s expected of him. He pursues a certain career path because his parents want him to, he likes certain styles of music/clothing/etc. because this is what is popular among his peer group, etc. Sloth: Betty is an insurance salesperson, but she recognizes that her real passion in life is to make independent movies. Whenever she thinks about doing so, however, she reflects upon all the difficulties involved—e.g., finding a new job, spending long hours writing screenplays, etc.—and consequently, never takes any action. Not pursuing this aim counts among the biggest regrets of her life. Sellout: Charlie entered law school with the aim of working for a non-profit organization devoted to protecting the environment. Upon graduating, however, the only jobs available to him are at big corporate law firms, all of which have abysmal environmental track records. After much self-reproach, he takes one of these jobs in order to pay off his large student loans. Secretive: Danielle grew up in the deep South during the early 20th-century. She had an African-American father and a white mother. Unlike her siblings, she was able to ‘pass’ as white and so did not experience the same type of discrimination they did. Despite feeling conflicted whenever she ‘passed’—particularly when her siblings were being treated badly in the same situation—she never admitted or owned up to her mixed heritage. Self-Deceived: Ed is a science teacher. He gets cancer and, in order to provide for his family, decides to sell drugs on the side. At the beginning, he sells drugs primarily for this reason. As time goes on, he becomes increasingly motivated out of a desire for power and personal gain. He deceives himself, however, by telling himself that he is ultimately doing this for his family.

In the case of Shallow, Abe fails to satisfy the first condition of EA. He lacks an accurate self-understanding, either implicitly or explicitly. Unlike, say, John the non-reflective farmer, Abe would not be able to say what he truly wants without engaging in long, deep reflection. This

The Virtue of Authenticity

289

stands in contrast to the next three cases. These individuals do have an accurate self-understanding. They fail, however, to live in conformity with that self-understanding in various ways. What all three cases have in common is that these people think, feel, or act in ways which, ex hypothesi, come at the expense of their true selves. Betty sacrifices her true self for the sake of (1a) comfort or avoiding the pain of failure (Sloth), Charlie for the sake of (1b) expedience and short-term financial security (Sellout), and Danielle for the sake of (1c) avoiding negative reactions from others (Secretive). We can understand (1a)–(1c) in two different ways. Either (1a)–(1c) are not related to these individuals’ true selves. Or (1a)–(1c) are related to these individuals’ true selves, but they are much less central parts than what is being sacrificed. That is, ex hypothesi, we are assuming that (1a)–(1c) comes at the expense of things which these individuals care about more deeply, such as, for Betty, (2a) pursuing something that she is passionate about (Sloth); or for Charlie, (2b) remaining true to his ideals (Sellout); or for Danielle, (2c) being honest or not ashamed about who she is as a person (Secretive). The fifth case, Self-Deceived, is more complicated. In some sense, Ed fails on both counts. On the one hand, despite knowing his true motivations, Ed gets himself to believe otherwise. In this way, Ed ultimately lacks an accurate self-understanding. On the other hand, Ed also fails to live in conformity with his accurate self-understanding. In order to be selfdeceived, Ed must have possessed some accurate self-understanding to begin with. Nonetheless, he fails to conform to this accurate selfunderstanding precisely by self-deceptively attempting to undermine his accurate self-understanding itself. We should emphasize two points here. First, notice that authenticity can come in degrees. Take Charlie. Suppose that the aim of achieving short-term financial security and paying off his student loans is an important concern for him and so comprises a genuine part of his true self. Nonetheless, suppose that he ultimately values protecting the environment more. Thus, by taking the corporate job, he compromises his deeper ideals. In this situation, he acts authentically to some extent insofar as he does genuinely care about financial security. Nevertheless, he acts in a predominantly inauthentic way insofar as taking this course of action comes at the expense of what is deeper or more central to who he is as a person. Second, not all cases where we act in shallow or slothful ways, or choose money over other personal ideals, or choose to hide certain

290

Ernesto V. Garcia

aspects of our identity from others, or engage in self-deception are necessarily inauthentic. Rather, it is only when we pursue these aims at the expense of expressing deeper or more central aspects of who we really are as persons that we are being inauthentic or ‘not true to ourselves’. 5. SUMMARY

To conclude, I want to highlight three main merits of this account of authenticity. First, EA seems to capture the valid core underlying all conceptions of authenticity—viz., being true to oneself—while at the same time offering some concrete guidance for thinking about the nature of the ‘self ’ and for what it even means to be ‘true to oneself ’, understood in terms of ‘self-understanding’, ‘self-expression’, and ‘self-concern’. Second, EA helps us to see why it can make sense to talk about authenticity as a kind of virtue. EA helps us to identify, à la Aristotle, two main extremes that we can fall prey to when trying to live authentically. On the one hand, we can display (1) the ‘deficiency’ of not caring enough about ourselves as persons, either by (1a) failing to have an accurate self-understanding, (1b) failing to live in conformity with that self-understanding, or (1c) failing to have sufficient self-concern—in particular, concern about what kind of people we are and what type of life we lead. Failing in any of these ways leads to inauthenticity. On the other hand, we can display (2) the ‘excess’ of caring about ourselves as persons too much. This is the type of attitude that most cultural critics of authenticity rightly disparage. We find authenticity described as a ‘megalomania of self-infinitization’ (Bell), as displaying an egocentric or narcissistic ‘therapeutic mentality’ (Rieff/Lasch), or as involving a kind of ‘competitive self-absorbed individualism’ (Potter).33 What these critiques fail to recognize is that it seems possible for us to care about who and what we are like as persons in fully virtuous or moderate—as opposed to vicious or excessive—ways. Notice that, strictly speaking, EA does not require that we be authentic in this virtuous way. Nonetheless, EA does provide us with the conceptual resources for understanding how

33 See, e.g., Bell 1976; Rieff 1973; Lasch 1979; and Potter 2010. For a helpful discussion of such criticisms, see Ferrara 1993.

The Virtue of Authenticity

291

we can be virtuously authentic, viz., by steering a middle course between the two extremes discussed above.34 Third and lastly, following thinkers like Lionel Trilling and Bernard Williams, EA emphasizes the fundamental difference between the ideals of (1) truthfulness, honesty, and sincerity, or being true to others and (2) authenticity, or being true to oneself.35 In this way, I argue—contrary to many recent accounts of authenticity—that, if our authentic behavior is virtuous, it is best understood as exemplifying a personal virtue (like, e.g., temperance or patience) rather than a social virtue (like, e.g., kindness, justice, or compassion). Many contemporary defenders of authenticity endorse a more social account of authenticity, arguing that acting upon one’s authentic desires “ . . . falls within the bounds of social acceptability—that is, that one’s authentic desires are morally creditable” and that “[o]ne’s authentic self points to a way of living that is both distinctively one’s own and socially decent” (Meyers 2000:158); that authenticity “include[s] such crucial elements as emotional attachment in caring and belonging to a community” (Cuypers 2001:150); and that “the concept of authenticity . . . involve[s] not only the winning of freedom but the respect for freedom, not only the achievement of dignity in the individual but the acceptance of the Kantian maxim of the dignity of all individuals” and that “one cannot imagine an authentic individual who really has no respect for the liberty of others” (Grene 1952:272). Charles Guignon provides perhaps the most sustained defense of this view. He gives two main arguments. First, drawing on insights from Charles Taylor and Bernard Williams, he argues that because concrete social contexts are necessary for the possibility of authenticity, being authentic requires that we act in ways “aimed at preserving and reinforcing a way of life that allows for such worthy personal life projects as that of authenticity” (2004:162). Second, he discusses examples of presumably inauthentic individuals, e.g., a talented artist committed to making a lot of money “by producing slick, popularized paintings”; a political fanatic who “enthusiastically and unquestioningly supports whoever happens to have political power at any time”; and a social conformist 34 I am grateful to an anonymous referee for insightful criticisms on an earlier formulation of these claims. 35 See Trilling 1972 and Williams 2002.

292

Ernesto V. Garcia

“whose defining life-goal is to always fit in and be as much like everyone else as possible” (2004:157). He argues that these people are inauthentic due to the content of their commitments. For Guignon, this shows that a necessary condition for being authentic is engaging in “an undertaking [that ] really is worthwhile at some level” (2004:158). And such ‘worthwhile’ endeavors ultimately involve “[having] a valuable role to play in society” (2004:161). I have two main worries about Guignon’s arguments. First, I think we should distinguish here between (1a) acting authentically and (1b) being concerned about preserving the social conditions that sustain authenticity. If authenticity just involves ‘being true to oneself ’ as outlined above—in terms of self-understanding, self-expression, and self-concern—then it does not seem that (1a) necessarily entails (1b). That is, it seems that (1a) acting authentically and (1b) striving to promote a society in which authenticity is possible can come apart. Guignon is surely right to argue that pursuing (1b) seems morally better, and perhaps even that pursuing (1b) seems more rational, at least if we want to avoid acting in self-undermining ways. Nonetheless, it seems mistaken to argue that we must engage in such activities if we want to be authentic simpliciter. Second and more fundamentally, I think that we should distinguish here between (2a) a more formal sense of authenticity, or authenticity as such, and (2b) a more substantive sense of authenticity, or what I will call an authenticity worth having. In contrast to Guignon, I want to allow for the logical possibility that realizing a person’s ‘true self ’—understood in terms of those concerns which are most central to her personality—could involve engaging in much-less-than-desirable activities, including Guignon’s examples of being a talented artist who only cares about commercial success, being a political fanatic, or even being a social conformist. While these individuals may not display an “authenticity worth having”, they could still be acting authentically in terms of genuinely expressing or conforming to their true selves, however regrettable we might find such choices to be. Understood this way, it seems that authenticity is merely conditionally valuable, depending upon what the nature of your ‘true self ’ happens to be like. In the end, I argue that authenticity has a mixed value for our lives. On the one hand, following George Kateb, we can categorize authenticity as an ‘existential value’ or a ‘value of identity’, where other such

The Virtue of Authenticity

293

values arguably include autonomy, freedom, equality, honor, etc.36 Without the presence of such values in our life, it seems we cannot lead fully flourishing human lives, whatever other objective goods we might have in our lives. On the other hand, from a moral point of view, authenticity seems to be only conditionally good—that is, only as good as the nature of the ‘true self ’ being expressed by such authentic activity. Ultimately, we should care about the virtue of authenticity insofar as, like the other existential virtues, it counts among those pro tanto objective goods which can contribute to our overall well-being and make our lives better off. However, we should recognize with Aristotle that, like all other virtues, the virtue of authenticity requires ‘the unity of the virtues’ in order to be unqualifiedly good. Acknowledgments I would like to thank audiences at the Boston University Ethics Reading Group, the Arizona Workshop in Normative Ethics, and the UMass Amherst Philosophy Department Brownbag Faculty Lunch where I presented various versions of this paper. In particular, I would like to thank Louise Antony, Julia Driver, Charles Griswold, Bob Gruber, Scott Hill, Hilary Kornblith, Benjamin Mitchell-Yellin, Luis Olivera, Jason Raibley, David Shoemaker, Daniel Star, Travis Timmerman, Mark Timmons, Brandon Warmke, and two anonymous OUP referees for very helpful comments. REFERENCES

Appiah, K. Anthony. 1993. In My Father’s House. Oxford: Oxford University Press. Appiah, K. Anthony. 1996. Color Consciousness. Princeton: Princeton University Press. Appiah, K. Anthony. 2005. The Ethics of Identity. Princeton: Princeton University Press. Arpaly, Nomy, and Schroeder, Timothy. 1999. “Praise, Blame, and the Whole Self,” Philosophical Studies 93 (2): 161–88. Baier, Annette. 1985. Postures of the Mind. Minneapolis: University of Minnesota Press. Bell, Daniel. 1976. The Cultural Contradictions of Capitalism. New York: Basic Books. Bell, Linda. 1989. Sartre’s Ethics of Authenticity. Tuscaloosa: University of Alabama Press. Bendix, Regina. 1997. In Search of Authenticity: The Formation of Folklore Studies. Madison: University of Wisconsin Press. Berman, Marshall. 1970. The Politics of Authenticity: Radical Individualism and the Emergence of Modern Society. London: Verso.

36

Kateb 2010:12–17, 21–4.

294

Ernesto V. Garcia

Boyle, David. 2003. Authenticity: Brands, Fake, Spin and the Lust for Real Life. London: Harper Press. Charme, Stuart Zane. 1991. Vulgarity and Authenticity: Dimensions of Otherness in the World of Jean-Paul Sartre. Amherst: UMass Press. Cheng, Vincent. 2004. Inauthentic: The Anxiety over Culture and Identity. New Brunswick: Rutgers University Press. Christman, John. 2009. The Politics of Persons. Cambridge: Cambridge University Press. Cuypers, Stefaan. 2001. Self-Identity and Personal Autonomy. Burlington: Ashgate. Driver, Julia. 2001. Uneasy Virtue. Cambridge: Cambridge University Press. Dworkin, Gerald. 1989. “The Concept of Autonomy,” pp. 54–62, in The Inner Citadel, ed. by John Christman. Oxford: Oxford University Press. Ferrara, Alessandro. 1993. Modernity and Authenticity: A Study in the Social and Ethical Thought of Jean-Jacques Rousseau. Albany. SUNY Press. Frankfurt, Harry. 1988. The Importance of What We Care About. Cambridge: Cambridge University Press. Frankfurt, Harry. 1999. Necessity, Volition, and Love. Cambridge: Cambridge University Press. Gilmore, James, and Pine, Joseph. 2007. Authenticity: What Consumers Really Want. Cambridge: Harvard Business School Press. Golomb, Jacob. 1995. In Search of Authenticity: From Kierkegaard to Camus. London: Routledge. Grene, Marjorie. 1952. “Authenticity: An Existential Virtue,” Ethics 62 (4): 266–74. Guignon, Charles. 2004. On Being Authentic. London: Routledge. Heidegger, Martin. 1962. Being and Time. New York: Harper and Row. Kateb, George. 2010. Human Dignity. Cambridge: Harvard University Press. Larmore, Charles. 2010. The Practices of the Self. Chicago: University of Chicago Press. Lasch, Christopher. 1979. The Culture of Narcissism. New York: W.W. Norton. Lindhom, Charles. 2008. Culture and Authenticity. London: Blackwell. Maslow, Abraham. 1949. “The Expressive Component of Behavior,” Psychological Review 56: 261–72. Meyers, Diana. 1989. Self, Society, and Personal Choice. New York: Columbia University Press. Meyers, Diana. 2000. “Intersectional Identity and the Authentic Self?: Opposites Attract!”, pp. 151–80, in Relational Autonomy, ed. by Catriona MacKenzie and Natalie Stoljar. Oxford: Oxford University Press. Miller, Alice. 1997. The Drama of the Gifted Child. New York: Basic Books. Morin, Jill. 2010. Better Make It Real: Creating Authenticity in an Increasingly Fake World. Westport: Praeger Publishers. Newman, Jay. 1997. Inauthentic Culture and its Philosophical Critics. Montreal: McGill University Press. O’Hagan, Timothy. 1999. Rousseau. London: Routledge. Oshana, Marina. 2005. “Autonomy and Self-Identity”, pp. 77–97, in Autonomy and the Challenges to Liberalism, ed. by John Christman and Joel Anderson. Cambridge: Cambridge University Press. Potter, Andrew. 2010. The Authenticity Hoax. Harper: New York. Rawls, John. 1971. A Theory of Justice. Cambridge: Harvard University Press. Raz, Joseph. 2002. The Authority of Law. Oxford: Clarendon Press. Rieff, Phillip. 1973. The Triumph of the Therapeutic. London: Penguin Press. Rorty, Amelie, and Wong, David. 1990. “Aspects of Identity and Agency”, pp. 19–36, in Identity, Character, and Morality, ed. by Owen Flanagan and Amelie Rorty. Cambridge: MIT Press.

The Virtue of Authenticity

295

Rossinow, Doug. 1998. The Politics of Authenticity: Liberalism, Christianity, and the New Left in America. New York: Columbia University Press. Rousseau, Jean-Jacques. 1997. The Discourses and Other Early Political Writings. Cambridge: Cambridge University Press. Rubin, Jeff. 1998. “Does the True Self Exist?,” in A Psychoanalysis for our Time. New York: NYU Press. Sartre, Jean-Paul. 1984. Being and Nothingness. New York: Washington Square Press. Sartre, Jean-Paul. 1989. No Exit and Three Other Plays. New York: Vintage Press. Swanton, Christine. 2003. Virtue Ethics. Oxford: Oxford University Press. Taylor, Charles. 1989. The Sources of the Self. Cambridge: Harvard University Press. Taylor, Charles. 1992. The Ethics of Authenticity. Cambridge: Harvard University Press. Taylor, Charles. 1994. “The Politics of Recognition,” pp. 25–74, in Multiculturalism, ed. by Amy Gutmann. Princeton: Princeton University Press. Trilling, Lionel. 1972. Sincerity and Authenticity. Cambridge: Harvard University Press. Vannini, Phillip, and Williams, J. Patrick (eds). 2009. Authenticity in Culture, Self, and Society. Burlington: Ashgate Press. Velleman, J. David. 2005. Self to Self. Cambridge: Cambridge University Press. Williams, Bernard. 2002. Truth and Truthfulness. Princeton: Princeton University Press. Wrathall, Mark, and Malpas, Jeff (eds). 2000. Heidegger, Authenticity, and Modernity: Essays in Honor of Hubert L. Dreyfus, Volume 1. Cambridge: MIT Press. Young, Edward. 2012. Conjectures on Original Composition. Manchester: University Press.

13 What Normative Terms Mean and Why It Matters for Ethical Theory ALEX SILK

1 INTRODUCTION

The strategy of clarifying philosophical questions by investigating the language we use to express them is familiar. Debates about intentionality shift to debates about sentences that report intentionality; debates about knowledge shift to debates about ‘knows’; debates about reference and singular thought shift to debates about referential expressions and attitude ascriptions. Such “semantic ascent,” as Quine (1960: §56) famously held, can help build common ground in debates that become seemingly intractable. A natural thought is that a similar strategy might be helpful in ethics. Perhaps by examining ethical language we can make progress in resolving conflicting basic moral intuitions and seemingly intractable disputes about fundamental normative principles. Perhaps. But if you are skeptical about how a substantive normative rabbit could possibly be drawn out of a linguistic hat, I am sympathetic. I won’t be trying to do any such thing. Then what will I be doing? Consider Ernie. Ernie is a budding ethicist. He is sincere in his inquiries on what to do and how to live. Though convinced that some ways of acting are better or worse than others, he has trouble shaking the following argument: 1. When making moral claims speakers express emotions and sensibilities. 2. Factual claims don’t express emotions or sensibilities. 3. So, moral claims aren’t about matters of fact. 4. So, there are no moral facts. 5. So, everything is permitted.

Normative Terms and Ethical Theory

297

This is a bad argument. One might point to ordinary factual judgments that express emotions; or seek clarification about the quantificational and modal force of the claims in Premises 1 and 2; or question the inference from Step 4 to Step 5, and wonder how everything could be permitted if there were no moral facts; and so on. Details aside, what is important for the moment is that highly contentious assumptions about moral and non-moral language are being used to support a radical substantive normative conclusion. Debates about ground-level issues can often become intertwined with debates about the language we use to talk about them; ethics is no exception. By clearing up Ernie’s assumptions about moral language, we can free up his investigations in how to live. The ways in which assumptions about normative language can figure in substantive normative arguments are multifarious and complex. The language questions can be difficult enough as they are, even bracketing how they bear on the ground-level questions of primary concern. But there is progress to be made. In this paper I will examine two dimensions along which normative language can differ: “strength” (§2) and “subjectivity” (§3) (this terminology will be clarified in due course). Particular attention will be given to deontic necessity modals, i.e. deontic uses of modal verbs like ‘ought’, ‘must’, ‘should’, ‘have to’, etc. Next I will show how better appreciating these features of normative language can improve theorizing about three issues in ethics: the coherence of moral dilemmas (§4), the possibility of supererogatory acts (§5), and the connection between making a normative judgment and being motivated to act accordingly (§6). The discussions of these specific linguistic and ethical issues can be seen as case studies illustrating the fruitfulness of utilizing resources from philosophy of language and linguistics in ethical theorizing. Of course, one could acknowledge a role for linguistic inquiry in ethics but reject my specific claims about its import on the debates to be discussed. I would like to convince you on the latter philosophical issues as well. At minimum, though, I hope to encourage healthy skepticism about simplifying assumptions that bracket differences among normative terms; about the degree to which we can investigate normative concepts unaffected by idiosyncrasies of the language we use to express those concepts; and even perhaps about the theoretical utility of the distinction—at least so-called—between “normative” vs. “non-normative” terms, concepts, and judgments.

298

Alex Silk

There is a sense in which the above general methodological points are unobjectionable. Many things may improve theorizing—perhaps eating more ice cream, for example (Feuerbach notwithstanding). Fair enough. But I suspect that issues concerning normative language constitute a more general source of ethical malaise. For many purposes, bracketing differences among normative terms may be harmless. But not always. Insensitivity to differences among normative terms has obscured debates on a range of ethical issues. Getting clearer on the language we use when investigating these issues can improve theorizing about them—e.g., by helping us diagnose problems with bad arguments and formulate better motivated questions. This can lead to clearer answers and bring into relief new theoretical possibilities and avenues to explore.

2 WEAK AND STRONG NECESSITY

The notion of “obligation” is central in moral philosophy. It isn’t uncommon to treat various expressions—e.g., ‘obligation’, ‘duty’, ‘ought’, ‘right’, ‘required’, ‘must’—as equivalent, or at least roughly equivalent, for the purposes of expressing this central notion. Indeed, Richard Brandt begins his investigation of the concepts of duty and obligation by observing as much: “Philosophers often use the following expressions as approximate equivalents: ‘It is X’s duty to do A’; ‘It is obligatory for X to do A’; ‘It would be wrong for X not to do A’; and ‘X ought to do A’” (Brandt 1964: 374). Here are some representative examples: When we are talking about someone’s duty or what he ought to do, we often express this by saying that he has a “moral obligation” to do it. (Sidgwick 1907: I/3.2) [I]t is clear that by [‘this is the right act’] we mean ‘this is the act that ought to be done’, ‘this act is morally obligatory’. (Ross 1930: 3) “I have an obligation to” means no more, and no less, than “I ought to.” (Singer 1972: 233n.2)

In deontic logic it is standard to treat these expressions as equivalent for the purposes of interpreting the O-operator, along with the operators with which it is interdefined. Here are von Wright and Åqvist:

Normative Terms and Ethical Theory

299

There are the deontic modes or modes of obligation. These are concepts such as the obligatory (that which we ought to do)—and the forbidden (that which we must not do). (von Wright 1951: 58) [D]eontic logic( . . . ) is the logical study of the normative use of language and . . . its subject matter is a variety of normative concepts, notably those of obligation (prescription), prohibition (forbiddance), permission and commitment. The first one among these concepts is often expressed by such words as ‘shall’, ‘ought’ and ‘must’, the second by ‘shall not’, ‘ought not’ and ‘must not’, and the third one by ‘may’; the fourth notion amounts to an idea of conditional obligation, expressible by ‘if . . . , then it shall (ought, must) be the case that___’. (Åqvist 2002: 148)

There is something to the assumption that ‘Ought ç’, ‘Must ç’, etc. uniformly express that ç is necessary in some sense related to obligation. But it is false. There is a robust body of linguistic evidence supporting a distinction in strength among necessity modals, with so-called “weak” necessity modals (‘ought’, ‘should’, ‘be supposed to’), on the one hand, and “strong” necessity modals (‘must’, ‘have to’, ‘(have) got to’, ‘be required to’), on the other.1 ‘Ought’—ethics’s beloved term of obligation, and deontic logic’s necessity modal par excellence—is relegated to the same “medium” category of modality as ‘appear’ and ‘seem’ (Huddleston & Pullum 2002: 177). When Kant wrote “I ought [sollen] never to act except in such a way that I can also will that my maxim should become a universal law” (1956: 4:402), I doubt he was intending to be hedging his bets. Kant notwithstanding, it is well established in descriptive and theoretical linguistics that although modals like (say) ‘ought’ and ‘should’ are stronger than modals like ‘may’ and ‘can’, they are weaker than modals like ‘must’ and ‘have to’. One piece of evidence is that, even holding the readings of the modals fixed, the former can be followed by the latter, but not vice versa, as reflected in (1).2 1 See, among many others, Leech 1970, 1971; Horn 1972, 1989; Wertheimer 1972; Harman 1975, 1993; Woisetschlaeger 1977; Lyons 1977; Williams 1981c; Coates 1983; Jones & Pörn 1986; Wierzbicka 1987; McNamara 1990; Palmer 1990, 2001; Bybee et al. 1994; Groefsema 1995; Myhill 1995; Zimmerman 1996; Saebo 2001; Huddleston & Pullum 2002; Werner 2003; Copley 2006; Wärnsby 2006; von Fintel & Iatridou 2008; Lassiter 2012; Rubinstein 2012; Silk 2014. For precedents in ethics and logic, see also Sloman 1970. How to capture this distinction in strength is less well established; see, e.g., von Fintel & Iatridou 2008, Finlay 2009, Lassiter 2011, Swanson 2011, Rubinstein 2012, Silk 2014. 2 I use ‘#’ to mark infelicity and ‘?’ to indicate that using the marked item is dispreferred; so, ‘?’ marks a weaker infelicity than ‘#’. For ease of exposition I will treat ‘ç’, ‘ł’, etc. as schematic

300

Alex Silk

(1) a. I ought to help the poor. In fact, I must. b. I must help the poor. #In fact, I ought to.

Similarly, (2a) is consistent in a way that (2b) is not. (2) a. I should help the poor, but I don’t have to. b. #I must help the poor, but it’s not as if I should.

These contrasts are similar in character to those with quantifiers of differing strengths, as reflected in (3)–(4). (3) a. Some of the children came to the party. In fact, all of them did. b. All of the children came to the party. #In fact, some of them did. (4) a. Some, but not all, of the children came to the party. b. #All, but not some, of the children came to the party.

(This isn’t to say that the difference between weak and strong necessity modals is to be captured specifically in terms of quantifier strength. The present point is simply that there is a contrast in strength, however it is to be captured.) A not uncommon intuition is that part of this difference in strength is traceable to a difference in what possibilities the modals leave open. Informally, ‘ought’ and ‘should’ make claims about what is ideal or “best” in some relevant sense but leave open whether there may be acceptable alternatives, whereas ‘must’ and ‘have to’ imply that there are no acceptable alternatives. As Bernard Williams puts it, “Ought is related to must as best is related to only” (Williams 1981c: 125; cf. Sloman 1970, McNamara 1990, von Fintel & Iatridou 2005, 2008). If you are getting over a cold and I say ‘You should stay home and rest’, I suggest that resting would be best, but I seem to leave open whether there may be alternative ways for you to get better. But if I say ‘You must stay home and rest’, I imply that resting is the only way for you to get better. Similarly, if I say ‘You should give 10% of your income to the poor’, I seem to leave open whether there may be a permissible course of action

letters to be replaced sometimes with sentences, sometimes with verb phrases. Either way, this is neutral on whether the arguments of deontic modals are propositions or properties (actions).

Normative Terms and Ethical Theory

301

for you that involves your giving less. But if I say ‘You must give 10% of your income to the poor’, I exclude such a possibility. ‘Ought’ and ‘should’ differ from ‘must’ in more naturally allowing for the non-actualization of their prejacents, or the possibility that the obligation in question will go unsatisfied. (In ‘MODAL ç’, ‘ç’ is the prejacent of the modal.) Suppose I am a smoker. I am aware of all the health risks, but, given my resistance to change, I find it unlikely that I will quit. I say: (5) I should stop smoking, but I’m not going to.

Or suppose we are talking about Alice, and I mention that Alice lied to her partner about where she was last night. Given Alice’s tendency to avoid conflict, I say: (6) I’m not sure if Alice will come clean, but she should.

Intuitively, in uttering (5) I am communicating that I won’t do what would be best for my health. In uttering (6) I am saying that Alice is obliged to come clean to her partner, while expressing doubts about whether she will do so. There is a robust body of data showing that speakers find it less natural to express these thoughts by using ‘must’, as reflected in (7)–(8).3 (7) #I must stop smoking, but I’m not going to. (8) ?I’m not sure if Alice will come clean, but she must.

Uttering ‘Must ç’ not only conveys that ¬ç-possibilities are unacceptable; it suggests that ¬ç-possibilities aren’t even on the table for consideration. Of course obligations can go unfulfilled. What is interesting is that speakers appear to assume otherwise, at least for the purposes of conversation, when expressing obligations with ‘must’.

3

See, e.g., Lemmon 1962, Leech 1971, Wertheimer 1972, Harman 1973, Lyons 1977, Williams 1981c, Coates 1983, Palmer 1990, 2001, Sweetser 1990, Groefsema 1995, Myhill 1996, Huddleston & Pullum 2002, Hopper & Traugott 2003, Werner 2003, Ninan 2005, Leech et al. 2009, Portner 2009, Close & Aarts 2010, Silk 2014.

302

Alex Silk

These differences between weak and strong necessity modals affect the circumstances in which they can be felicitously used (see Rubinstein 2012, Silk 2012, 2014 for extensive recent discussion).4 Suppose you are considering whether to fight in the Resistance or take care of your ailing mother. You mention that the value of family, which supports your helping your mother over your fighting, is important, and I agree. But the issue is acknowledged to be complex, and it isn’t settled in the context whether there might be more important competing values. Sensitive to this, I may find it more appropriate to express my advice that you help your mother by using ‘should’ than by using ‘must’, as in (9). (9) You: Family is very important. Me: I agree. You should (/?must,/?have to) take care of your mother.

But if we settle that family is of primary importance, as in (10), it can become more natural to use ‘must’ and for us to accept that you have to help your mother. (10) You: Family is most important—more important than country or anything else. Me: I agree. You must (/have to,/?should) take care of your mother.

How I express my advice that you help your mother depends on the status in the context of the value of family vis-à-vis other potentially relevant values. In (10), where it is settled that the value of family is to take precedence, using a strong necessity modal is preferred. But in (9), where this condition isn’t settled, were I to use ‘must’ I would imply that I am foreclosing certain possibilities that you have left open. Unless I am in a position to do so, my using ‘must’ is dispreferred. By using ‘should’ I can propose that you help your mother while leaving open the possibility that the value of family might be outweighed or defeated in some way. If you accept my ‘should’ claim, we can plan accordingly without having to presuppose that the value of family is more important than other competing values we accept or may come to accept.

4 See Woisetschlaeger 1977: ch. 5, McNamara 1990: ch. 3 for prescient early discussion of the context-dependence of ‘ought’ and ‘must’.

Normative Terms and Ethical Theory

303

In light of these logical and conversational differences between weak and strong necessity modals, it is perhaps unsurprising that they are often thought to differ in their directive force. Paul McNamara puts the intuition well: To say that one ought to take a certain option is merely to provide a nudge in that direction. Its typical uses are to offer guidance, a word to the wise (“counsel of wisdom”), to recommend, advise or prescribe a course of action ( . . . ) In contrast, to say that one must take a certain option is to be quite forceful. Its typical uses are to command, decree, enact, exhort, entreat, require, regulate, legislate, delegate, or warn. Its directive force is quite strong. (McNamara 1990: 156)5

Informally, using a strong necessity modal is often more emphatic and expresses greater urgency than using a weak necessity modal. In this section we have seen a dimension of strength along which deontic necessity modals can differ. Following the descriptive and formal linguistic work I have treated the relevant distinction as a binary one (weak vs. strong). There are indeed good linguistic reasons, even beyond those given here, for marking a distinction at this level (e.g., concerning neg-raising and interactions with quantifiers, among other things; see n. 1). Yet for all I have said here the distinction may be more fine-grained than this. What is important for my purposes is simply that there is some relevant difference in strength among necessity modals.6 In the next section we will examine a second dimension of difference among deontic modals. But first I want to emphasize the modesty of my aims in these sections. I am not offering an account of how to capture these features of normative language (though see §5); I take up that task in other work (Silk 2014). My characterizations of the linguistic phenomena are intentionally informal. Also, for all I say in these sections, the highlighted differences in normative language may be irrelevant for the purposes of normative inquiry. I leave argument that having these differences in view can improve normative theorizing to §§4–6.

5

See also esp. Ninan 2005, Portner 2007, 2009, Silk 2014. See also Behre 1955, Leech 1970, 1971, 2003, Lyons 1977, Wells 1979, Coates 1983, Palmer 1990, Bybee et al. 1994, Myhill 1995, 1996, Huddleston & Pullum 2002, Smith 2003, Nuyts et al. 2005. 6 Thanks to an anonymous referee for raising this issue.

304

Alex Silk 3 ENDORSING AND NON-ENDORSING USES

Following Lyons 1977, it is common in linguistic semantics to distinguish “subjective” and “objective” uses of modals. Very roughly, a modal is used subjectively if it presents the speaker as endorsing the considerations with respect to which the modal claim would be true. A modal is used objectively if it doesn’t present the speaker in this way. Applied to the normative case, a deontic modal is used subjectively if it presents the speaker as endorsing a body of relevant norms that would justify it, and objectively if it doesn’t. Use of the labels ‘subjective’ and ‘objective’ can be fraught in ethical theory. So let’s call Lyons’s “subjective” uses of modals endorsing uses, and call Lyons’s “objective” uses non-endorsing uses. Note that non-endorsing uses are compatible with speaker endorsement; they simply fail to present it. Our second dimension along which normative terms can differ concerns the frequency with which they are used in an endorsing or non-endorsing manner. Among strong necessity modals, ‘be required to’ is typically used non-endorsingly; ‘have to’ and ‘(have) got to’ are more flexible, with ‘have to’ tending more toward the non-endorsing side of the spectrum and ‘(have) got to’ more toward the endorsing side; and ‘must’ is nearly always used endorsingly. Among weak necessity modals, ‘be supposed to’ is typically used non-endorsingly; ‘be to’ is more flexible; and ‘should’ and ‘ought’ are nearly always used endorsingly.7 These claims are supported by examples which target whether the speaker can coherently follow her sincere modal utterance with an expression of indifference toward or disapproval of the norms that would verify it (n. 7). For expository purposes, let’s focus on the cases of ‘must’, ‘have to’, ‘should’, and ‘be supposed to’. It is hard to hear a sincere utterance of ‘Must ç’ as consistent with the speaker’s being indifferent about whether ç or about the relevant norms which require that ç, as reflected in (11).

7 These generalizations are supported by a robust body of data in descriptive linguistics. See Leech 1970, 1971, 2003, Lakoff 1972a,b, Larkin 1976, Lyons 1977, Coates 1983, Palmer 1990, 2001, Sweetser 1990, Myhill 1995, 1996, Nuyts 2000, Verstraete 2001, Huddleston & Pullum 2002, Smith 2003. See also Swanson 2012, Silk 2014 for discussion. It is interesting to consider related distinctions in ethics, deontic logic, and philosophy of law in light of this linguistic work; cf., e.g., Hare 1952, Hart 1961/1994, Stenius 1963, von Wright 1963, Hansson 1969, Raz 1975. Terminology varies among authors.

Normative Terms and Ethical Theory

305

(11) [Context: Some friends are deciding whether to go home or stay out late for the party.] a. #You must get home by 10, but I don’t care whether you do. b. #Bert must get home by 10. Aren’t his parents stupid? I would stay out if I were him.

We find similar judgments for utterances with the weak necessity modal ‘should’: (12) a. #You should get home by 10, but I don’t care whether you do. b. #Bert should get home by 10. Aren’t his parents stupid? I would stay out if I were him.

Two clarifications are in order. First, endorsing uses are compatible with the speaker’s being torn, or feeling the force of competing values or reasons. Suppose Bert’s mother just went to the hospital for a medical emergency. If Bert’s friend asks him whether he will be at the party, he might sincerely say: (13) I wish I could go, but I {must, should} help my mother.

Second, endorsing uses are compatible with the speaker’s treating the normativity of the relevant norms as grounded in an external source. Suppose Chip is a sincere, practicing Catholic. His uttering (14) is compatible with his taking his obligation to go to confession as being grounded in the authority of God or the Church. (14) I {must, should} go to confession.

Chip’s use of ‘must’ or ‘should’ is endorsing in the sense that it presents him as endorsing or identifying with the norms that entail this obligation, whatever their source. Even with these points in mind, judgments like those in (11)–(12) aren’t nearly as anomalous when expressed with the strong necessity modal ‘have to’ or the weak necessity modal ‘be supposed to’. (15) a. You have to get home by 10, but I don’t care whether you do.

Alex Silk

306

b. Bert has to get home by 10. Aren’t his parents stupid? I would stay out if I were him. (16) a. You’re supposed to get home by 10, but I don’t care whether you do. b. Bert is supposed to get home by 10. Aren’t his parents stupid? I would stay out if I were him.

Intuitively, in (15)–(16) it is coherent for the speaker to dismiss the act of getting home by 10 because she isn’t endorsing the norms that entail that the subject has this obligation. She is simply reporting what is required by them. To be clear, these claims about patterns of (non-)endorsing use for particular modal expressions aren’t exceptionless principles. For instance, the claim isn’t that ‘must’ can’t be used non-endorsingly. Some speakers may allow non-speaker-endorsing uses in certain contexts; adding explicit adverbial phrases like ‘morally’ or ‘legally’ can also promote nonendorsing readings. The generalization supported by the evidence is rather that ‘must’ is typically used endorsingly (see Silk 2014 for further discussion). What is important is that modal expressions differ in their tendencies toward endorsing vs. non-endorsing use. We have seen two distinctions in the meaning and use of deontic necessity modals. These distinctions crosscut one another.8 Some examples are given in Table 1: We find strong necessity modals that tend to be used endorsingly, like ‘must’; strong necessity modals that tend to be used non-endorsingly, like ‘be required to’; weak necessity modals that tend to be used endorsingly, like ‘ought’ and ‘should’; and weak necessity modals that tend to be used non-endorsingly, like ‘be supposed to’. These are certainly not the only respects in which deontic necessity modals and normative Table 1. Strength and (non-)endorsingness endorsing non-endorsing weak

‘should’/‘ought’

strong

‘must’

‘be to’ ‘(have) got to’

‘be supposed to’ ‘have to’

‘be required to’

8 Thanks to Eric Swanson for impressing on me the importance of this point. See Swanson 2012, Silk 2014 for discussion.

Normative Terms and Ethical Theory

307

language differ in their meaning and use. Yet in the remainder of the paper I will argue that greater sensitivity to these specific dimensions along which normative language differs can illuminate various normative ethical debates. 4 DILEMMAS

Suppose you promised Alice that you would help her move and promised Bert that you would help him move, but you discover that you cannot help them both. Suppose also that each promise is uniquely important, and there is no higher-order principle you can use to resolve whom to help. On the face of it, you are in a moral or practical dilemma. But some have argued that genuine dilemmas—irresolvable conflicts in what to do, all-things-considered—are impossible. On the one hand, it isn’t uncommon for arguments against the existence of dilemmas to be motivated by the thought that it is inconsistent for contrary acts to be simultaneously required. It would seem inconsistent for you to be required to help Alice and required to help Bert when you cannot help them both. Morality, or rationality, could never land us in such a contradiction. As Kant puts it—in only the way he can:9 But since duty and obligation are concepts that express the objective practical necessity of certain actions and two rules opposed to each other cannot be necessary at the same time, if it is a duty to act in accordance with one rule, to act in accordance with the opposite rule is not a duty but even contrary to duty; so a collision of duties and obligations is inconceivable. (Kant 1996: 6:224)

More recently, here is Alan Donagan: Rationalist theories cannot allow moral dilemmas ( . . . ) Each principle and each derivative proposition of a rationalist theory asserts, of some rule or precept that it assumes all human beings can observe in all situations to which that rule or precept applies, that practical reason requires them all to observe it. If, therefore, any such theory were to assert that practical reason requires any human being in any situation to observe a set of precepts that cannot all be observed in it, it

9

Though see Hill 1996 for an interpretation of Kant on which dilemmas are possible.

308

Alex Silk

would contradict itself; for it would assert that set of precepts not to be what it also asserts or assumes them to be. (Donagan 1993: 15)10

On the other hand, there is a long tradition in ethics and logic that is compelled by the thought that it is at least coherent to think that ‘Ought ç’ and ‘Ought ł’ might both be true, for contraries ‘ç’ and ‘ł’. Here is van Fraassen: I shall conclude that the view [that sometimes there are two sound moral arguments, concluding respectively that A ought to be the case (O(A)) and that not-A ought to be the case (O(¬A))] constitutes a significant ethical position (whether correct or incorrect). (van Fraassen 1973: 15)11

Even if, fortuitously, there are no moral or practical dilemmas, the possibility of dilemmas shouldn’t be ruled out by the very meanings of normative terms. Attending to differences among necessity modals can help us capture both of these intuitions. Many find it plausible that the dilemmas expressed with ‘should’ and ‘have to’ in (17)–(18) are consistent in a way that the dilemma expressed with ‘must’ in (19) is not.12 (17) I should help Alice and I should help Bert, but I can’t help them both. (18) I have to help Alice and I have to help Bert, but I can’t help them both. (19) ?I must help Alice and I must help Bert, but I can’t help them both.

We can capture this in terms of the distinction between endorsing and non-endorsing uses of modals, and independent work on the semantics of weak necessity modals. We saw in §3 that ‘have to’ tends toward non-endorsing uses—uses that fail to present the speaker as endorsing the relevant verifying norms. Insofar as the ‘have to’s in (18) are used non-endorsingly and are 10 See also Ross 1930, Davidson 1969, Searle 1980, Hare 1981, Conee 1982, Donagan 1984, Feldman 1986, Thomson 1990, Pietroski 1993, Brink 1994, McConnell 1996, Zimmerman 1996. See Goble 2013 and McConnell 2014 for survey discussions. 11 See also, e.g., Lemmon 1962, van Fraassen 1973, Williams 1973, Barcan Marcus 1980, Foot 1983, Gowans 1987, Sinnott-Armstrong 1988, Dancy 1993, Harman 1993, Horty 1994, 2003, Goble 2009, Swanson 2011. 12 See, e.g., Lemmon 1962, Gowans 1987, Harman 1993, Swanson 2011; cf. Foot 1983, Sinnott-Armstrong 1988, Horty 2003, von Fintel 2012.

Normative Terms and Ethical Theory

309

interpreted with respect to distinct moral considerations, it is unsurprising that some speakers naturally hear (18) as consistent. There needn’t be any inconsistency in saying (roughly) that the one promise requires one action and that the other promise requires another action, assuming one isn’t expressing endorsement of either requirement. But if we use a strong necessity modal like ‘must’ that tends toward endorsing uses, the sense of inconsistency becomes more palpable. The speaker is more naturally heard as registering endorsement of an inconsistent set of norms and as directing one to perform incompatible acts (given the facts about the scenario). Interestingly, many speakers find the sense of inconsistency to dissipate when the dilemma is expressed with ‘should’, as in (17), even when the modals are given a constant interpretation and are used endorsingly. This contrast between ‘should’ and ‘must’ is nicely predicted by several recent independently motivated semantics which validate agglomeration for strong necessity modals but not for weak necessity modals (Swanson 2011, Silk 2012, 2014). On these semantics ‘Must ç’ and ‘Must ł’ entail ‘Must (ç and ł)’, but ‘Should ç’ and ‘Should ł’ don’t entail ‘Should (ç and ł)’. The formal details would take us too far afield; interested readers may consult the references above. What is important for our purposes is simply that there are independent reasons for thinking that genuine dilemmas are coherently expressible with weak necessity modals—i.e., that ‘Should ç’ and ‘Should ł’ could both be true, for contraries ‘ç’ and ‘ł’. Whereas (19) implies that I must do the impossible, (17) does not. The examples in (17)–(19) suggest that dilemmas are coherently expressible with non-endorsing uses of strong necessity modals and with (endorsing or non-endorsing) uses of weak necessity modals. What import could this point about our linguistic judgments possibly have for normative theory? First, arguments that there are no genuine dilemmas will need to be independent of appeals to linguistic intuitions like the intuition that (19) is inconsistent. The theorist who accepts that there are genuine dilemmas can agree with Kant that incompatible propositions cannot both be deontically necessary: ‘Must ç’ and ‘Must ł’, for contraries ‘ç’ and ‘ł’, are inconsistent. But this doesn’t itself show that the concept of a dilemma is incoherent. Dilemmas are coherently expressible—just not with endorsing uses of strong necessity modals. Even if there are no practical dilemmas, this isn’t because they “entail a contradiction” (Davidson 1969: 34).

310

Alex Silk

To be clear, I am not claiming that the linguistic data tells us something about whether there are genuine practical dilemmas. To the contrary. My aim is to bring the debate about dilemmas back to the ethicist’s home turf. Whether there are genuine dilemmas is a question for substantive normative theory, not logic or semantics. The intuitive worry about dilemmas isn’t a linguistic one. It is about whether one might find oneself in a certain sort of situation—perhaps a situation where competing values with which we identify pull us in opposite directions; or a situation where, no matter what we do, we cannot help but feel loss. As is now generally accepted, all parties in the debate about dilemmas can accept that these sorts of situations are possible. The question has become whether some such apparent dilemmas constitute “genuine” dilemmas. The issue is what this question amounts to. We can now see that we cannot simply ask whether morality or rationality could land us in a situation in which we couldn’t satisfy all its “demands” or “obligations.” Couching the question in these terms partitions the space of normative possibilities too coarsely.13 Likewise, we cannot put the question as whether we could find ourselves in a situation in which we should ç and we should ł, or in which we must ç and we must ł, where we cannot both ç and ł. Couching the question in these terms may leave our answer affected by orthogonal linguistic issues. Instead, perhaps the question is whether there can be incomparable values or sets of reasons for an agent that enjoin incompatible acts; or whether it is possible that, all things considered, possibilities where the agent çs are incomparable to possibilities where the agent łs (and the agent can’t both ç and ł), and no possibility where the agent ås better satisfies the total body of normative factors; or . . . What is important here isn’t whether these are the precise normative questions that lie behind our vague initial worries about the possibility of practical dilemmas. What is important is that they are normative questions. What language we use in posing normative questions is crucial. Greater sensitivity to how we express and talk about dilemmas can help us frame the questions directing our inquiry in ways that track “Deontic necessity is commonly glossed as ‘obligation’, but the noun obligation covers the range of should as well as must” (Huddleston & Pullum 2002: 207); cf. Sloman’s (1970: 391) prescient similar observation that the “contrast between ‘ought’ and ‘must’ is obliterated by referring to both as cases of ‘obligation’.” 13

Normative Terms and Ethical Theory

311

the substantive normative issues of primary concern. Progress in the debate over practical dilemmas needn’t be sidetracked by arguments turning on the idiosyncrasies of particular normative terms. 5 SUPEREROGATION

Supererogatory acts are acts that go “beyond the call of duty.” They are permitted but not required, and better than what is minimally required. Think: throwing yourself on a grenade to protect your friends, giving a substantial portion of your income to the poor, and so on. Some ethicists (“anti-supererogationists”) claim that there are no supererogatory acts, and that acts that seem to be supererogatory are in fact required. Others (“qualified supererogationists”) grant that there are supererogatory acts, but maintain that these acts are still binding in some sense and thus deserving of criticism if not performed.14 The following is a not uncommon line of argument: “It would be much better if I gave more money to the poor. I really ought to do so. So, I must have conclusive reason, and hence an obligation, to give more to the poor. So, my not giving more to the poor must be wrong and hence subject to criticism.” (“And so,” the anti-supererogationist would add, “my giving more to the poor must not be supererogatory after all.”) This argument generalizes, leading to the so-called “paradox of supererogation,” or “good-ought tie-up.” Joseph Raz articulates the worry well (see also n. 14): If doing a supererogatory act is praiseworthy there must be reasons for doing it, and the reasons must outweigh any conflicting reasons for not doing it. But if there are conclusive reasons for performing the act then not to perform it is to act against the balance of reasons. If reason requires that the act be done then surely one ought to do it, and the “ought” is based on all the reasons which apply to the case; it is a conclusive ought. But this entails that failing to perform the act is failing to do what one ought (conclusively) to do, so why isn’t it blameworthy not to perform a supererogatory act? (Raz 1975: 164)

14 This terminology follows Heyd 1982. For classic examples of anti-supererogationism, see Moore 1903, New 1974, Feldman 1986, Kagan 1989. For classic examples of qualified supererogationism, see Rawls 1971, Richards 1971, Raz 1975. See Urmson 1958 for the seminal work that prompted contemporary interest in supererogation. See Heyd 1982 and Mellema 1991 for extensive discussion.

312

Alex Silk

There are several things to be distinguished in this line of argument— e.g., what is good, what is blameworthy, what one ought to do, what one has most reason to do, what one is required to do. The crucial inferences are from the claim that one has most reason to ç to the claim that one has an obligation to ç, and from this to the claim that it would be wrong or blameworthy not to ç. Raz himself presents the argument in terms of ‘conclusive ought’, but this, of course, is philosopher-speak. The relevant question is what notion of obligation or normative necessity it is being used to express. The problem is that not all ways of filling this in are equally compelling or speak against supererogationism. The antisupererogationist gets traction with the intuition that failing to do what one must do, or is required to do, leaves one subject to criticism. But the fact that it would be better for you to give more to the poor, and even that you should, needn’t imply that you must. While it is plausible that failing to do what one must is blameworthy, it is less obvious that one may always be blamed for failing to do what is good or what one should. By clarifying the distinction between weak and strong necessity, we can see how it is at least coherent for a moral view (a) to distinguish what one ought to do and what would be best from what one must do and what is minimally required, and (b) to attach blame or criticism to failing to do the latter.15 These points don’t show that there are supererogatory acts any more than the considerations in §4 showed that there are genuine practical dilemmas. Perhaps the critics of supererogationism are right and nothing but the best will do. One might accept on independent grounds a demanding moral theory according to which one must do what is evaluatively best. But such grounds will have to be just that: independent. Additional substantive normative argument will be required. Distinguishing notions of normative necessity can refine our understanding of the space of possible theories and suggest new ways the dialectic may proceed. Various accounts of weak necessity are motivated by the thought that weak necessity modals are, in some sense, interpreted with respect to additional facts, values, norms, etc.16 These accounts shed light on 15

Cf. Sloman 1970, McNamara 1990. See, e.g., von Fintel & Iatridou 2008, Rubinstein 2012, Silk 2012, 2014. How precisely this thought is implemented varies among accounts. 16

Normative Terms and Ethical Theory

313

two thoughts that many have had about supererogation: first, that it is only a select few, perhaps those who have some sort of “higher calling,” who ought to perform supererogatory acts; and, second, that many agents who perform supererogatory acts regard these acts as things they must do.17 For concreteness, I will focus on the particular account of weak necessity which I have developed elsewhere (Silk 2012, 2014).18 What makes weak necessity modals “weak,” on this view, is that they bracket whether the necessity claim holds in the actual world. ‘Should’ expresses necessity only on the supposition that certain circumstances relevant to the necessity claim obtain. It needn’t be presupposed that these circumstances actually do obtain. Suppose I promised an acquaintance, Alice, that I would help her move. Intuitively, a norm requiring that I help Alice is in force only if certain circumstances obtain—e.g., that I don’t come across a drowning child on the way to Alice’s apartment, that an evil dictator didn’t promise to torture hundreds of children if I help Alice, etc. Very roughly, saying that I must help Alice implies that all such circumstances obtain and thus that I am actually required to help her. Saying that I should help Alice implies only that were all such conditions to be satisfied, I would be required to help her. Weak necessity modals afford a means of coordinating on the implications of our norms and values without having to decisively settle on how they apply and weigh against one another in particular circumstances. I won’t attempt to defend this view here. Suffice it to say that it may help capture the two additional ideas about supererogation noted above. First, suppose there are conditional norms to the effect that one çs if one desires greater merit, has a higher calling, has a special dispensation of divine grace, or the like. Call these norms supererogation norms and the possible conditions under which they apply supererogation conditions. Even if it isn’t decisively settled whether Dorothy “has a higher calling” 17 The first point has a long tradition of support, especially in theological literatures on supererogation; see Heyd 1982: ch. 1, Mellema 1991: ch. 3 and references therein. On the second point, see e.g. Urmson 1958, Eisenberg 1966. Concerning the “righteous gentiles” he interviewed from Le Chambon who protected Jews from Nazis, Philip Hallie relates that they invariably “pulled back from me but looked firmly into my eyes and said: ‘How can you call us “good”? We were doing what had to be done’ ” (Hallie 1979: 20, cited in Horgan & Timmons 2010: 40n.23). 18 These developments differ in technical implementation; these differences won’t be relevant here.

314

Alex Silk

or the like, we might accept that Dorothy should ç, or that her ç-ing would be supererogatory. This is possible even if Dorothy doesn’t in fact satisfy any supererogation condition and ‘Dorothy must ç’ is false. Moreover insofar as agents in general fail to satisfy the supererogation conditions, the generic claim ‘One must ç’ will be false. This reflects a sense in which the act of ç-ing is supererogatory. Second, given that agents are typically in a position to settle on whether they satisfy the above sorts of conditions—whether they desire greater merit, etc.—it is unsurprising that those who perform supererogatory acts sometimes regard them as things they must do. In saying ‘I must ç’ the agent assumes that the supererogatory norms apply to her. Even if she is correct about this, the act of ç-ing may still be supererogatory in the above sense that the generic claim ‘One must ç’ is false. But there is also a sense in which her act may be supererogatory for her. For even if she is among the select few, the fact that she is may itself be the result of some supererogatory act(s). Perhaps her wanting to “go beyond the call” is good but not required of her. If so, then even if she must ç, it isn’t the case that she must be such that she must ç. Her being such that ç-ing is required for her is itself supererogatory. Though she is “just doing her duty,” that it is her duty renders her deserving of praise. 6 JUDGMENT INTERNALISM AND “THE NORMATIVE”

It is often claimed that a distinctive mark of normative language and judgment is its practical character, or its connection with action and motivation. This connection between normative language and action is epitomized in the thesis of judgment internalism—to a first approximation, the claim that there is an internal and necessary connection between making a normative judgment and being motivated to act in accordance with it.19 Here are Allan Gibbard and Ralph Wedgwood: The clear distinctive feature of normative concepts, I now think, lies in their conceptual ties. Oughts of action tie in conceptually with acting. (Gibbard 2011: 36)

19 For classic discussions, see Falk 1948, Nagel 1970, Williams 1981a, Korsgaard 1986, Darwall 1997.

Normative Terms and Ethical Theory

315

[T]he necessary connection that normative judgments have to motivation and practical reasoning is a special feature of normative and evaluative judgments. It is a feature that is absent from all judgments that are wholly non-normative and non-evaluative in content. Indeed, this seems to be precisely one of the features that distinguishes normative and evaluative judgments from judgments of all other kinds. (Wedgwood 2007: 71)

Many take it as obvious that some form of internalism is true. After all, the reasons we weigh in deliberation are reasons for action, i.e., reasons for acting on. Normative judgments are constitutive of deliberation, and deliberation is essentially practical; its aim is action. But many find there to be clear counterexamples. What about the psychopath? Or someone who is really tired or depressed? Can’t they make sincere normative judgments while lacking the corresponding motivation? Attending to our two dimensions of difference among necessity modals—strength and patterns of (non)-endorsing use—can illuminate theorists’ conflicting intuitions about judgment internalism. The continuation of the above quote from Gibbard is revealing. The clear distinctive feature of normative concepts, I now think, lies in their conceptual ties. Oughts of action tie in conceptually with acting. Take, for example, the belief that the building is on fire and the one and only way to keep from being burned to a crisp is to leave forthwith. If that’s the case, we’d better leave forthwith, but it isn’t strictly incoherent, conceptually, to have this belief and not to leave. Contrast this with the normative belief that one must leave forthwith. It is, I maintain, conceptually incoherent to hold this belief and not leave, if one can. (Gibbard 2011: 36; emphasis in original)

It is revealing that Gibbard uses ‘must’ to pump the intuition that normative beliefs are conceptually tied with action. Deontic ‘must’ is nearly always used with directive force in main clauses (§§2–3). It is thus no surprise that judgment internalism will seem compelling when considering deontic ‘must’ judgments. It is hard to hear a sincere utterance of ‘Must ç’ as consistent with the speaker’s being indifferent about whether ç, as reflected in (7) and (11), reproduced in (20). (20) a. #I must get home by 10, but forget that; I’m not going to. b. #You must get home by 10, but I don’t care whether you do. c. #Bert must get home by 10. Aren’t his parents stupid? I would stay out if I were him.

316

Alex Silk

But the same sorts of judgments aren’t nearly as anomalous when expressed with necessity modals that are more naturally used nonendorsingly, as reflected in (15)–(16), reproduced in (21). Interestingly, ‘be to’, in terms of which Gibbard 2003 and Wedgwood 2007 analyze all normative terms, isn’t conventionally endorsing (Lakoff 1972a). (21) a. {I’m supposed to, I have to, I’m to} get home by 10, but forget that; I’m not going to. b. {You’re supposed to, You have to, You’re to} get home by 10, but I don’t care whether you do. c. {Bert is supposed to, Bert has to, Bert is to} get home by 10. Aren’t his parents stupid? I would stay out if I were him.

We needn’t be psychopaths to sincerely utter the sentences in (21). We saw in §3 that the weak necessity modal ‘should’ is typically used endorsingly: (12) a. #You should get home by 10, but I don’t care whether you do. b. #Bert should get home by 10. Aren’t his parents stupid? I would stay out if I were him.

However, even when used endorsingly, weak necessity modals are compatible with the denials of their prejacents, as we saw in (5). (5) I should stop smoking, but I’m not going to.

This latter feature may influence intuitions about apparent counterexamples to judgment internalism involving judgments expressed with weak necessity modals. We should be wary of general claims about normative language and judgment. Judgment internalism can seem compelling when considering judgments using terms that are paradigmatically directive and endorsing. But when we consider judgments using other terms, counterexamples can appear in the offing. Even if internalism is true for deontic ‘must’ judgments, it is false for judgments expressed using terms that aren’t conventionally endorsing.

Normative Terms and Ethical Theory

317

Objection: “This conclusion is dialectically irrelevant. Judgment internalism is, as the name suggests, a thesis about normative judgment, a species of mental act that involves deployment of a distinctive sort of concept. The question of judgment internalism is a question about the connection between judgments of this sort and motivation. Whether certain natural language sentences invariably express this kind of judgment, and whether accepting such sentences is invariably connected with motivation, is beside the point.” This is an important objection. One response would be to call into question the implicit assumption that we can get a grip on the nature of normative judgments and concepts independently of the language we use to express them. One might worry that if linguistically driven intuitions about cases were an unreliable, insufficient, or even irrelevant source of evidence, our metaethical accounts of normative judgment would be dramatically underdetermined. Though I am sympathetic with this response, a more modest reply is simply to call for circumspection. As a matter of actual practice, it is common to garner evidence about judgment internalism by appealing to intuitions about judgments expressed in natural language. The above quote from Gibbard is a prototypical example. But in light of the various dimensions along which putative normative terms can differ, we should be cautious about making such appeals. They may be less probative into the nature of normative judgment than we initially thought. For present purposes I am happy to rest content with this more modest conclusion. But I would like to close by raising a brief counter-worry of my own. I would like to take a step back and reflect on a more general meta-theoretical question: What are we doing when we describe certain judgments, concepts, or language as “normative”? The worry is that plausible ways of answering this question will render theses like judgment internalism—construed as theses about normative judgments/concepts/ language in general—either trivial or obviously empirically false. On the one hand, suppose we characterize “normative” judgments in linguistic terms. First, suppose we treat them as judgments expressed with particular linguistic expressions or constructions. Perhaps they are the judgments which are “fraught with ‘ought’” (Sellars 1962: 44; cf. Gibbard 2003: x). Or perhaps they are the judgments expressed by sentences like “‘I must ł’, ‘I ought to ł’, ‘it would be best, all things

318

Alex Silk

considered, for me to ł’, etc.” (Wiggins 1976: 95). As the discussion in this paper should make clear, no such characterization will do. The class of paradigmatic normative terms is quite a variegated lot. The broader the class of expressions we use to characterize normative judgments, the more obviously false a thesis like judgment internalism is. Few, if any, expressions exclusively receive intuitively normative interpretations in all contexts.20 The narrower the class, the less clear it is that our subject matter is the one we initially cared about. Most judgments traditionally thought of as being normative won’t count as normative. And a thesis like judgment internalism becomes trivialized. It amounts to the claim that there is a necessary connection between motivation and making a judgment expressed using a term that conventionally presents one as endorsing the considerations that would justify it. Even if internalism is conceptually true, if true, I doubt that this is the conceptual truth that ethicists and metaethicists have cared about. Alternatively, second, suppose we treat the normative judgments as the judgments expressed by uses of language of a particular sort. Perhaps they are the judgments expressed by specifically directive uses of words like ‘ought’, ‘must’, etc., or those expressed by specifically directive speech acts, performed by whatever linguistic means.21 In fact, I think this is how we ought to proceed. The topic of directive and endorsing language has afforded many rich avenues of research in formal semantics and pragmatics, descriptive linguistics, speech act theory, and philosophy of language more generally. But delimiting the class of normative judgments in this way again runs the risk of trivializing the question of judgment internalism. The question reduces to whether there is a necessary connection between motivation and making a judgment that aims at getting oneself to do something and presents oneself as endorsing the considerations that would justify it. Call me parochial, but this doesn’t strike me as an interesting question. Now to the other hand. Suppose we characterize normative judgments in non-linguistic terms. For instance, we might characterize them 20 Even for ‘must’, and likely for imperatives as well: ‘Must’ can receive non-normative readings (e.g., epistemic, metaphysical) and, even for intuitively normative readings, may permit non-endorsing uses for some speakers (§3). Imperatives might be the best bet, though even these can sometimes merely express wishes (‘Have a good day’, ‘Be a home run’). 21 For claims in this spirit, see Korsgaard 1996: 8, 226; Thomson 1996: 130–1, 147–50; Gibbard 2003: 137–9; Field 2009: 286; Parfit 2011: 284, 288, 291; Railton 2012: 25.

Normative Terms and Ethical Theory

319

functionally as those judgments that characteristically direct, guide, or— dare I say—motivate us. Such a characterization opens up a rich topic for philosophical and psychological investigation. But it too reduces questions about the nature of normative judgment to triviality. Judgment internalism reduces to the thesis that there is a necessary connection between motivation and making a judgment individuated in part by its function to motivate (direct, guide, etc.). I don’t intend this outline of a dilemma to be decisive. There may be an alternative way of characterizing normative judgment on which the thesis of judgment internalism is interesting, and perhaps true. My primary aim is to encourage critical reflection on the terms with which we frame our inquiry. There are features of language and thought that we have been homing in on in our talk of “normative” terms, concepts, and judgments. Research in philosophy of language and linguistics, as well as philosophy of mind and psychology, can help us delineate these features. This has the potential to clarify our initial questions and open up new avenues of research. But it also raises the possibility that some of these questions may be less motivated than we initially thought. Couching our inquiry in terms of a class of “normative” language, concepts, and judgments may obscure the phenomena. Better, perhaps, to see ourselves as examining directive and endorsing uses of language, on the one hand, and motivational types of states of mind, on the other. 7 CONCLUSION

The notion of obligation is central in ethics. This isn’t for no reason: As one classic English grammar text notes, “Deontic necessity is commonly glossed as ‘obligation’, but the noun obligation covers the range of should as well as must” (Huddleston & Pullum 2002: 207). On the one hand, we are often interested in investigating what we are actually required to do. We wish to guide others’ behavior and our own in light of the norms we accept. An endorsing strong necessity modal like ‘must’ is well-suited to the task. However, using ‘must’ is often awkward. We may want to talk about obligations that held in the past, or obligations that may go unfulfilled or be overridden in the future. Or we may want to communicate information about a body of norms without necessarily registering commitment to them or enjoining others to share in such commitment. A modal like ‘should’ or ‘be required to’ can be more suitable. There is a

320

Alex Silk

wide range of expressive resources at our disposal for coordinating our actions and expressing our normative views in conversation, deliberation, and planning. This is for the better given the variety of our purposes. But it also raises a philosophical risk. Inattention to the particularities of the language we use can lead to misinterpretation of the nature and import of our judgments. We began with a common, and often implicit, simplifying assumption among ethicists that expressions like ‘ought’, ‘must’, ‘duty’, ‘required’, etc. are equivalent for expressing deontic necessity and obligation. Bracketing differences among such expressions might have turned out to be harmless for the purposes of normative inquiry—“let the semanticist sort them out,” one might have said. But we have seen that insensitivity to differences among necessity modals can obscure intuitions and hinder theorizing. On the flip side, better understanding the language we use in substantive normative discussion can help us ask clearer, better motivated questions and bring into relief directions for future inquiry. In this paper we examined two specific dimensions along which normative vocabulary can differ—strength and patterns of (non-) endorsing use—and applied insights about these dimensions to help capture competing intuitions in debates about dilemmas, supererogation, and judgment internalism. But the import of the underlying strategy extends more broadly. We often come to the substantive normative table with implicit views about how normative language works, some correct but others not. Locating these assumptions can free up our normative investigations. Inquiry into normative language and conversation can clarify and improve normative evaluation in practice. Acknowledgments Thanks to Eric Swanson and the audience at the Fifth Annual Arizona Normative Ethics Workshop for discussion, and to two anonymous referees for comments. REFERENCES

Åqvist, Lennart. 2002. Deontic logic. In Dov M. Gabbay & Franz Guenthner (Eds.), Handbook of philosophical logic, Vol. 8, pp. 147–264. Dordrecht: Kluwer Academic Publishers, 2nd edn. Barcan Marcus, Ruth. 1980. Moral dilemmas and consistency. Journal of Philosophy, 78, 121–36.

Normative Terms and Ethical Theory

321

Behre, Frank. 1955. Meditative-polemic should in modern English that-clauses. Stockholm: Almqvist and Wiksell. Brandt, R. B. 1964. The concepts of obligation and duty. Mind 73, 374–93. Brink, David O. 1994. Moral conflict and its structure. The Philosophical Review 103, 215–47. Bybee, Joan, Revere Perkins, & William Pagliuca. 1994. The evolution of grammar: Tense, aspect, and modality in the languages of the world. Chicago: University of Chicago Press. Close, Joanne & Bas Aarts. 2010. Current change in the modal system of English: A case study of must, have to and have got to. In Ursula Lenker, Judith Huber, & Robert Mailhammer (Eds.), The history of English verbal and nominal constructions, pp. 165–81. Amsterdam: John Benjamins. Coates, Jennifer. 1983. The semantics of the modal auxiliaries., London: Croom Helm. Conee, Earl. 1982. Against moral dilemmas. The Philosophical Review 91, 87–97. Copley, Bridget. 2006. What should should mean? MS, CNRS/Université Paris 8. Dancy, Jonathan.1993. Moral reasons. Cambridge: Blackwell. Darwall, Stephen. 1997. Reasons, motives, and the demands of morality: An introduction. In Stephen Darwall, Allan Gibbard, & Peter Railton (Eds.), Moral discourse and practice, pp. 305–12. New York: Oxford University Press. Davidson, Donald. 1969. How is weakness of the will possible? In Essays on actions and events, pp. 21–42. 2001. New York: Clarendon Press, 2nd edn. Donagan, Alan. 1984. Consistency in rationalist moral systems. Journal of Philosophy 81, 291–309. Donagan, Alan. 1993. Moral dilemmas, genuine and spurious: A comparative anatomy. Ethics 104, 7–21. Eisenberg, Paul D. 1966. From the forbidden to the supererogatory: The basic ethical categories in Kant’s ‘Tugendlehre’. American Philosophical Quarterly, 3, 255–69. Facchinetti, Roberta, Manfred Krug, & Frank Palmer (Eds.). 2003. Modality in contemporary English. New York: Mouton de Gruyter. Falk, W. D. 1948. ‘Ought’ and motivation. Proceedings of the Aristotelian Society, pp. 111–38. Feldman, Fred. 1986. Doing the best we can: an essay in informal deontic logic. Dordrecht: Reidel. Field, Hartry. 2009. Epistemology without metaphysics. Philosophical Studies 143, 249–90. Finlay, Stephen. 2009. Oughts and ends. Philosophical Studies 143, 315–40. von Fintel, Kai. 2012. The best we can (expect to) get? Challenges to the classic semantics for deontic modals. Paper presented at the 2012 Central APA, Symposium on Deontic Modals. von Fintel, Kai & Sabine Iatridou. 2005. What to do if you want to go to Harlem: Anankastic conditionals and related matters. MS, MIT. von Fintel, Kai & Sabine Iatridou. 2008. How to say ought in foreign: The composition of weak necessity modals. In Jacqueline Guéron & Jacqueline Lecarme (Eds.), Time and modality, pp. 115–41. Springer. Foot, Philippa. 1983. Moral realism and moral dilemma. The Journal of Philosophy 80, 379–98. van Fraassen, Bas C. 1973. Values and the heart’s command. The Journal of Philosophy 70, 5–19. Gibbard, Allan. 2003. Thinking how to live. Cambridge: Harvard University Press. Gibbard, Allan. 2011. How much realism? Evolved thinkers and normative concepts. In Russ Shafer-Landau (Ed.), Oxford studies in metaethics, vol.6, pp. 33–51. New York: Oxford University Press. Goble, Lou. 2009. Normative conflicts and the logic of ‘ought’. Noûs, 43, 450–89. Goble, Lou. 2013. Prima facie norms, normative conflicts, and dilemmas. In Dov Gabbay, John Horty, Xavier Parent, Ron van der Mayden, & Leendert van der Torre (Eds.), Handbook of deontic logic and normative systems, pp. 241–352. Milton Keynes: College Publications.

322

Alex Silk

Gowans, Christopher W. (Ed.). 1987. Moral dilemmas. New York: Oxford University Press. Groefsema, Marjolein. 1995. Can, may, must, and should: A Relevance theoretic account. Journal of Linguistics 31, 53–79. Hallie, Philip. 1979. Lest innocent blood be shed. New York: Harper. Hansson, Bengt. 1969. An analysis of some deontic logics. Noûs 3, 373–98. Hare, R. M. 1952. The language of morals. Oxford: Oxford University Press. Hare, R. M. 1981. Moral thinking: Its levels, method, and point. Oxford: Clarendon Press. Harman, Gilbert. 1973. Review of Roger Wertheimer, The significance of sense: Meaning, modality, and morality. The Philosophical Review 82, 235–9. Harman, Gilbert. 1975. Reasons. Crítica: Revista Hispanoamericana de Filosofía 7, 3–17. Harman, Gilbert. 1993. Stringency of rights and “ought”. Philosophy and Phenomenological Research 53, 181–5. Hart, H. L. A. 1961/1994. The concept of law. Oxford University Press, 2nd edn. Heyd, David. 1982. Supererogation: Its status in ethical theory. Cambridge: Cambridge University Press. Hill, Jr., Thomas E. 1996. Moral dilemmas, gaps, and residues: A Kantian perspective. In Mason (1996), pp. 167–98. Hopper, Paul J. & Elizabeth Closs Traugott. 2003. Grammaticalization. Cambridge: Cambridge University Press. Horgan, Terry & Mark Timmons. 2010. Untying a knot from the inside out: Reflections on the “paradox” of supererogation. Social Philosophy and Policy 27, 29–63. Horn, Laurence R. 1989. A natural history of negation. Chicago: University of Chicago Press. Horn, Laurence R. 1972. On the semantic properties of logical operators in English. Ph.D. thesis, University of California, Los Angeles. Horty, John F. 1994. Moral dilemmas and nonmonotonic logic. Journal of Philosophical Logic 23, 35–65. Horty, John F. 2003. Reasoning with moral conflicts. Noûs 37, 557–605. Huddleston, Rodney & Geoffrey K. Pullum (Eds.). 2002. The Cambridge grammar of the English language. Cambridge: Cambridge University Press. Jones, Andrew J.I. & Ingmar Pörn. 1986. ‘Ought’ and ‘must’. Synthese 66, 89–93. Kagan, Shelley. 1989. The limits of morality. Oxford: Oxford University Press. Kant, Immanuel. 1956. Groundwork of the metaphysics of morals. Paton, H.J. (Trans.), New York: Harper & Row, 3rd edn. Kant, Immanuel. 1996. The metaphysics of morals. Mary Gregor (Ed.), Cambridge: Cambridge University Press. Korsgaard, Christine M. 1986. Skepticism about practical reason. Journal of Philosophy 83, 5–25. Korsgaard, Christine M. 1996. The sources of normativity. Cambridge: Cambridge University Press. Lakoff, Robin. 1972a. Language in context. Language 48, 907–27. Lakoff, Robin. 1972b. The pragmatics of modality. In Proceedings of the Chicago Linguistics Society 8, pp. 229–46, Chicago. Larkin Don. 1976. Some notes on English modals. In James McCawley (Ed.), Syntax and semantics 7: Notes from the linguistic underground, pp. 387–98. London: Academic Press. Lassiter, Daniel. 2011. Measurement and modality: The scalar basis of modal semantics. Ph.D. thesis, New York University. Lassiter, Daniel. 2012. Quantificational and modal interveners in degree constructions. In Anca Chereches (Ed.), Proceedings of SALT 22, pp. 565–83. Ithaca: CLC Publications.

Normative Terms and Ethical Theory

323

Leech, Geoffrey. 1971. Meaning and the English verb. London: Longman. Leech, Geoffrey. 2003. Modality on the move: The English modal auxiliaries 1961–1992. In Facchinetti et al. (2003), pp. 223–40. Leech, Geoffrey. 1970. Towards a semantic description of English. Bloomington: Indiana University Press. Leech, Geoffrey, Marianne Hundt, Christian Mair, & Nicholas Smith. 2009. Change in contemporary English: A grammatical study. Cambridge: Cambridge University Press. Lemmon, E. J. 1962. Moral dilemmas. Philosophical Review 71, 139–58. Lyons, John. 1977. Semantics, vol. 2. Cambridge: Cambridge University Press. Mason, H. E. (Ed.). 1996. Moral dilemmas and moral theory. New York: Oxford University Press. McConnell, Terrance. 1996. Moral residue and dilemmas. In Mason (1996), pp. 36–47. McConnell, Terrance. 2014. Moral dilemmas. In Edward N. Zalta (Ed.), The Stanford Encyclopedia of Philosophy. Spring 2014 edn. . McNamara, Paul. 1990. The deontic quadecagon. Ph.D. thesis, University of Massachusetts. Mellema, Gregory. 1991. Beyond the call of duty: Supererogation, obligation, and offence. Albany: SUNY Press. Moore, G. E. 1903. Principia ethica. New York: Cambridge University Press. Myhill, John. 1995. Change and continuity in the functions of the American English modals. Linguistics 33, 157–211. Myhill, John. 1996. The development of the strong obligation system in American English. American Speech 71, 339–88. Nagel, Thomas. 1970. The possibility of altruism. Oxford: Oxford University Press. New, Christopher. 1974. Saints, heroes, and utilitarians. Philosophy 49, 179–89. Ninan, Dilip. 2005. Two puzzles about deontic necessity. In Jon Gajewski, Valentine Hacquard, Bernard Nickel, and Seth Yalcin (Eds.), MIT working papers in linguistics: New work on modality, vol. 51, pp. 149–78. Nuyts, Jan. 2000. Epistemic modality, language, and conceptualization. Amsterdam: John Benjamins. Nuyts, Jan, Pieter Byloo, & Janneke Diepeveen. 2005. On deontic modality, directivity, and mood: A case study of Dutch mogen and moeten. In Antwerp papers in linguistics, vol. 110. Center for grammar, cognition, and typology, University of Antwerp. Palmer, F. R. 1990. Modality and the English modals. New York: Longman, 2nd edn. Palmer, F. R. 2001. Mood and modality. Cambridge: Cambridge University Press, 2nd edn. Parfit, Derek. 2011. On what matters, vol. II. Oxford: Oxford University Press. Pietroski, Paul M. 1993. Prima facie obligations, ceteris paribus laws in moral theory. Ethics 103, 489–515. Portner, Paul. 2007. Imperatives and modals. Natural Language Semantics 15, 351–83. Portner, Paul. 2009. Modality. Oxford: Oxford University Press. Quine, Willard Van Orman. 1960. Word and object. Cambridge: MIT Press. Railton, Peter. 2012. Nietzsche’s normative theory? The art and skill of living well. In Christopher Janaway & Simon Robertson (Eds.), Nietzsche, naturalism, and normativity, pp. 20–51. New York: Oxford University Press. Rawls, John. 1971. A theory of justice. Cambridge: Harvard University Press. Raz, Joseph. 1975. Permissions and supererogation. American Philosophical Quarterly 12, 161–8. Richards, David A. J. 1971. A theory of reasons for action. Oxford: Clarendon Press. Ross, W. D. 1930. The right and the good. Oxford: Oxford University Press.

324

Alex Silk

Rubinstein, Aynat. 2012. Roots of modality. Ph.D. thesis, University of Massachusetts Amherst. Sæb, Kjell Johan. 2001. Necessary conditions in a natural language. In Caroline Fery & Wolfgang Sternefeld (Eds.), Audiatur vox sapientiae: A festschrift for Arnim von Stechow, pp. 427–49. Berlin: Akademie Verlag. Searle, John R. 1980. Prima-facie obligations. In Zak van Straaten (Ed.), Philosophical subjects: Essays presented to P.F. Strawson, pp. 238–59. Oxford: Oxford University Press. Sellars, Wilfrid. 1962. Truth and “correspondence”. Journal of Philosophy 59, 29–56. Sidgwick, Henry. 1907. The methods of ethics. Indianapolis: Hackett, 7th edn. Silk, Alex. 2012. Modality, weights, and inconsistent premise sets. In Anca Chereches (Ed.), Proceedings of SALT 22, pp. 43–64. Ithaca: CLC Publications. Silk, Alex. 2014. Weak and strong necessity. MS, University of Birmingham. Singer, Peter. 1972. Famine, affluence, and morality. Philosophy and Public Affairs 1, 229–43. Sinnott-Armstrong, Walter. 1988. Moral dilemmas. Oxford: Basil Blackwell. Sloman, Aaron. 1970. ‘Ought’ and ‘better’. Mind 79, 385–94. Smith, Nicholas. 2003. Changes in modals and semi-modals of strong obligation and epistemic necessity in recent British English. In Facchinetti et al. (2003), pp. 241–66. Stenius, Erik. 1963. The principles of a logic of normative systems. Acta Philosophica Fennica 16, 247–60. Swanson, Eric. 2012. Imperative force in the English modal system. Handout, PhilosophyLinguistics Workshop, University of Michigan. Swanson, Eric. 2011. On the treatment of incomparability in ordering semantics and premise semantics. Journal of Philosophical Logic 40, 693–713. Sweetser, E. 1990. From etymology to pragmatics. Cambridge: Cambridge University Press. Thomson, Judith Jarvis. 1996. Evaluatives and directives. In Gilbert Harman & Judith Jarvis Thomson (Eds.), Moral relativism and moral objectivity, pp. 125–54. Cambridge: Blackwell. Thomson, Judith Jarvis. 1990. The realm of rights. Cambridge: Harvard University Press. Urmson, J.O. 1958. Saints and heroes. In A. I. Melden (Ed.), Essays in moral philosophy, pp. 198–216. Seattle: University of Washington Press. Verstraete, Jean-Christophe. 2001. Subjective and objective modality: Interpersonal and ideational functions in the English modal auxiliary system. Journal of Pragmatics 33,1505–28. Wärnsby, Anna. 2006. (De)coding modality: The case of must, may, måste, and kan. Lund: Lund University. Wedgwood, Ralph. 2007. The nature of normativity. Oxford: Oxford University Press. Wells, Gordon. 1979. Learning and using the auxiliary verb in English. In V. Lee (Ed.), Language development, pp. 250–70. London: Croom Helm. Werner, Tom. 2003. Deducing the future and distinguishing the past: Temporal interpretation in modal sentences in English. Ph.D. thesis, Rutgers University. Wertheimer, Roger. 1972. The significance of sense: Meaning, modality, and morality. Ithaca: Cornell University Press. Wierzbicka, Anna. 1987. The semantics of modality. Folia Linguistica 21, 25–43. Wiggins, David. 1976. Truth, invention, and the meaning of life. In Needs, values, truth: Essays in the philosophy of value, pp. 87–137. Oxford: Clarendon Press, 3rd edn. Williams, Bernard. 1973. Ethical consistency. In Problems of the self, pp. 166–86. Cambridge: Cambridge University Press. Williams, Bernard. 1981a. Internal and external reasons. In Williams (1981b), pp. 101–13. Williams, Bernard. 1981b. Moral luck. Cambridge: Cambridge University Press. Williams, Bernard. 1981c. Practical necessity. In Williams (1981b), pp. 124–31. Woisetschlaeger, Erich Friedrich. 1977. A semantic theory of the English auxiliary system. Ph.D. thesis, MIT.

Normative Terms and Ethical Theory

325

von Wright, Georg Henrik. 1951. Deontic logic. Mind 60, 1–15. von Wright, Georg Henrik. 1963. Norm and action: A logical inquiry. London: Routledge and Kegan Paul. Zimmerman, Michael J. 1996. The concept of moral obligation. Cambridge: Cambridge University Press.

Index Action guidance 38 Acts and actions Acts vs. actions 16–20 Effects of Causal 17, 35 Constitutive 17, 35 Disposition-dependent 18–20, 25, 26, 35 Disposition-independent 18–20 Agential desert 206 Altruism 91–9 vs. agreeableness 92 Limited 29–30, 36 Anxiety 141, 172–3 as a beneficial emotion 176–8, 180–1 and judgment-driven scrupulosity 142–4 moral judgment 141–7 Practical 181 Punishment 173 Social 175–8 see also moral anxiety Appiah, K. Anthony 273, 279 Aquinas, T. 31 Aqvist, L. 298 Arendt, H. 28 Aristotle 20 on our animal natures 67 on choice ( prohairesis) 68 on end-sharing 74 and precision of judgment 138 on feelings that fail to motivate 151n. 1 on phronimos 42 on practical correctness 43 Augustine 31

Authenticity Authenticity as such vs. authenticity worth having 292 and bad faith 278 and being true to oneself 274 and coping vs. expressive action 286 Existensialist 284 vs. inauthenticity 288 Individual 274 Nonconformist Problem 279 Natural/Original 274 Naturalness Problem 279 and the nature of the self 274 and personal vs. social virtue 291 as a personal virtue 282–90 and the ‘true self ’ 282–3 Truthful 274 and the Sincerity Problem 279 as a value of identity 292 Baier, A. 279 Barlow, D. 172 Bauer, B. 116 Bennett, J. 92 Berger, F. 161 Blum, L. 92 Bradley, B. 214 on dualism of prudence 216 Brandt, R. 298 Brewer, T. 69 on dialectical activity 69–70 Bruner, J. 203 Camenisch, P. 156 Card, C. 163

328

Index

Chang, R. 238–40 Christman, J. 273 Codifiability, strong vs. weak 40 Collective Harm, the Problem of 245 and Collective Harm Cases 245, 247–8 and consequentialist vs. non-consequentialist theories 264–7 and possible approaches Deny the Description 246, 269–71 Reject the Implication 246 Fairness Approach 253–9 the Superfluity Problem 259–60 Weak vs. Strong Participation 248–53 Correlativity Thesis 151, 167–70 Crisp, R. 43n. 9 Cullity, G. 253 Dancy, J. 38–57 Darwall, S. 106 Deeds vs. doings 16–20 Defeat 48 Exclusionary 49 Rebutting 48 Switching off 49 Deontic necessity modals 297 Endorsing vs. non-endorsing uses 304 and moral dilemmas 308–11 and normative judgment internalism 315–19 and supererogation 312–14 Weak vs. strong 299–303 and blame or criticism 312 Difference-making vs. instrumentally useful 270–1 Directionality, theories of 157–9 Donagan, A. 307

Dorsey, D. 201 Dreier, J. 240n. 10 Driver, J. 169, 284 Dworkin, G. 273 Emotion, formal object of 173–4 Ends 68 and expressive action 286 Final ends 68 Happiness (eudaimonia) 68 Hierarchy of ends 70 Indeterminate ends 68 Merely instrumental ends 70 Ultimate End 69 Everitt, N. 53n. 23 Evil Doing evil 26–30, 36 and self-interest 26–30, 36 Experience Machine 201–2, 219 Feinberg, J. 165 Feldman, F. 198, 200 Fittingness 71 Frankfurt, H. 83, 120, 273 on ‘true self ’ 283 Freeman, S. 119n. 11 Freid, S. 65 Fried, C. 120 Generalism 53 Default vs. Invariant 55 Gergen, K. and M. 203 Gibbard, A. 184n. 22, 314 Gilmore, J. and Pine, J. 275 Glasgow, J. 199 Glover, J. 255 Goodness Doing good 21 and disposition-dependent vs. disposition-independent benefits 21

Index and honesty, justice, and other virtues 25–6 and love and friendship 23–5 and respect 21–3 Good for 106–7, 109 Good from an individual’s point of view 106–7 Impartial good 109 Moorean Thesis 111 Prudential vs. aesthetic 213 Gratitude 151–70 and demands 161–6 and the Directionality Platitude 151, 157–9 Grateful nonmaleficence 154–5 and the No-Right Platitude 151, 159–66 and the Obligation Platitude 151, 152–6 and political obligations 168–9 and rights 159–66 Guignon, C. 291 Hallie, P. 313n. 17 Hampton, J. 76 Harman, E. 213 Hart, H. L. A 158n. 23 Hill, T. 99, 166n. 41 Heidegger, M. on authenticity 280, 284 Herder, J. G. 275 Herman, D. 203 Hobbes 28, 31 Hohfeld, W. 151n. 3, 160 Holism Internalist vs. externalist 198–9 Holton, R. 33n. 10 Hooker, B. 59 Horgan, T. 232n. 7 Horty, J. 49n. 19

329

Hudson 40 Hume, D. 56, 72 on gratitude 152–3 on monkish virtues 56 Hursthouse, R. 39 on phronimos 42 on right action 41 on virtue-rules 39 Jackson, F. 54 Kant 20, 92, 299 Categorical Imperative Humanity formulation 73, 103–4 and the Problem of Collective Harm 265–7 Universal Law Formulation 102 on merit 96 and the Problem of Collective Harm 265–7 on gratitude 152–3 on the impossibility of genuine moral dilemmas 307 on obligations concerning animals 157 Positive duties to others 74 on the vice of disrespect 57 Kantianism and the Problem of Collective Harm 265–7 and moral permissibility 225 and moral worth 87–104 Kateb, G. 292 Keller, T. 209 Kierkegaard, S. 278 Kitcher, P. 184n. 22 Knobe effect 32, 35 Knobe, J. 32, 34, 35 Korsgaard, C. 20, 92–3, 184n. 22 Kramer, M. 157n. 21, 167

330

Index

Kraut, R. 71, 111n. 6 on what is good for living creatures 71 Kutz, C. 250 on unstructured collective harms 251 La Rochefoucauld, F.d. 279 Lance, M. 38 Larmore, C. 273 on practical reflection 287 Leveling down 259 Little, M. 38, 45 Locke, J. 79 Markovits, J. 88 Marx, K. 109, 116–18 on the point of view of man vs. that of citizen 116 Maslow, A. 286 McAdams, D. 207n. 5 McConnell, T. 162 McDowell, J. 67 McIntyre, A. 204 McMahan, J. 214–15 McNamara, P. 242n. 13 Mele, A. 273 Meyers, D. 273 Mill, J. S. 92, 157n. 21 Miller, A. 277 Moment internalism vs. moment externalism 198 Moore, G. E. 111 Moore, J. 134 Moral anxiety 178–92 vs. anticipatory guilt 181–2 Cultivation of 191 and good moral decision making 183–90 vs. practical anxiety 182 Moral dilemmas 307–11 Moral life 31 Moral metacognition 184

and moral anxiety 185 and moral decision making 184 Moral obligation Collective 254 and deliberative priority 153 Directionality 157–9 and ‘ought’ vs. ‘must’ 300–1 and peremptory character 154 and practical deliberation 153 Moral Particularism 38, 54–5 and the uncodifiability objection 38–9 Moral permissibility Consequentialist account of 224 and morally permissible moral mistakes 223, 232 Simple Kantian account of 225 Moral principles 54, 106 Content vs. acceptance 106 Moral worth 86 and moral reasons 88–9 and negative moral worth 88 Right-making features 86–9 and right reasons for action 86–9 Nagel, T. 28n. 7, 110, 112 on objective vs. subjective reasons 110 Narrativism 197 and a narrative calculus 202–11 Additivity of Narrative Value 209 Desert Factor 206 Instrumentality Factor 205 Value Factor 205 Whole-Life Narrative Value 209 and progressivism 209–11 vs. Achievementism 209–10 Neo-Aristotelian moral psychology 192 Nietzsche, F. 50, 50n. 20, 64, 274, 278 Normative judgment internalism 314

Index and endorsing uses of strong deontic necessity modals 315–16 and how to characterize normative judgments 317–19 O’Brien, L. 33n. 10 Obsessive Compulsive Disorder 130–3 Options 16 Eligible 228 On a par 228 Morally best 242 Morally permissible 222 Permissible 222 Suberogatory 243 Supererogatory 222, 311 Oshana, M. 273 Ought vs. Obligation 153 Owens, D. 84 on demand vs. command in personal relationships 84 Parfit, D. 124n. 19, 216, 216n. 10, 247–9 Pidgen, C. 28n. 8 Portmore, D. 210–11, 240n. 10 Postow, B. 240n. 10 Practical deliberation 231–2 and moral anxiety 182–91 and normative judgment internalism 315 Phenomenology of 232, 236–7 Quine, W. V. O. 296 Railton, P. 89, 184n. 22 Rand, A. 51 Rawls, J. 109, 118–19 Rawls’ theory of justice 107, 108 on reasonable comprehensive views 119n. 11

331

Raz, J. 49, 124n. 19, 154, 157n. 21, 241n. 11, 286 on the paradox of supererogation 311 Reasons Default reasons 48–53 Enticing reasons 56 Exclusionary reasons 154 Expressive reasons 286 Moral reasons 88–90, 100–2, 222–3, 230–2 Non-moral reasons 56 Objective vs. subjective 110–11 Peremptory reasons 56, 159 Right reasons for action 86–9 V-reasons 41 Voluntarist reasons 238–9 Wrong kind of reasons 97 Regan, D. 111n. 6 Right action and moral worth 86–104 and moral reasons 88–9 Qualified agent account of 38 Target-centered view of 42 and v-reasons 42–5 Rightness as correctness 44 Rights 159–70 and claim-rights 160 and the correlativity thesis 166–70 and demands 160–4 Imperfect 164–5 Perfect 166 Pre-Hohfeldian 160–1 Robustness 22, 27 Rønnow-Rasmussen, T. 75 Rorty, A. 285 Rosati, C. 202n. 4 Rousseau, J.-J. on authenticity 276, 281

332

Index

Sartre, J.-P. on authenticity 279 Scanlon, T. 210 Scanlon’s contractualism 107 Scheffler, S. 154 Scrupulosity 133–7 Anxiety-driven 144 and chronic doubt 137–9 Judgment-driven 141–4 and moral praiseworthiness 145–8 and Obsessive Compulsive Disorder 134–5 and perfectionism 135–9 Scrupulous vs. ordinary moral judgments 138–41, 146 and thought-act fusion 136–9 Self-understanding Global vs. local 285 Semantic ascent 296 Seneca on gratitude 153 Shakespeare 24 Shape-of-a-life phenomenon 196 Sigdwick, H. 109–15 on the Dualism of Practical Reason 110–15 Slote, M. 196 Smith, M. 87 Sreenivasan, G. 157n. 21 Steiner, H. 158n. 23 Stevenson, R. L. 95 Strawson, G. 202 Suberogatory actions 169 Supererogation 311 Paradox of supererogation 311 Anti-supererogationism vs. qualified supererogationism 311 and weak vs. strong deontic necessity modals 312–14

Taylor, C. on authenticity 280 Taylor, J. 134 Temporal distributivism 199 and improvementism 199 Thick concepts 53–5 Tiberius, V. 184n. 22 Timmons, M. 232n. 7 Tragic dilemmas 44, 60–1 Trilling, L. 291 Utilitarianism 87, 107 Objections to 88 Utilitarian Duty 112 Value-dependent psychology 32, 35 van Fraasen, B. 308 Väyrynen, P. 53 Velleman, D. 200–1, 273 Virtue, differentiated 59 Virtue Ethical Particularism 55 vs. Extreme Particularism 55 Non–Pervasiveness Objection 56 vs. Strong Particularism 55 vs. Weak Particularism 55 Virtue ethics and the action guidance objection 38 Narrative 59–60 and particularism 38 and the Problem of Collective Harm 265 of roles 59 Virtue Eudaimonism 64, 67–8 as an account of practical rationality 68 and accountability to others 82 and the authority of others as ends and the Categorical Imperative (humanity formulation) 73

Index Virtue-rules 39 and default reasons 48 and the qualified agent account of right action 40 and v-acts 40 von Wright, G. H. 298 Walker, A. D. M. 155 Wallace, R. J. 154 Wedgwood, R. 314 Weiss, R. 151n. 1 Well-being 68 and altruism 91–9 Non-instrumental 94 and the shape-of-a-life phenomenon 196

and sociality 72–9 and virtue 68 Wellman, C. 154, 156, 158n. 23, 161 Wilde, O. 24 Williams, B. 46, 109, 119–22, 153n. 10, 273, 291 on “one thought too many” 120–1 on ‘ought’ vs. ‘must’ 300 on self-expression 286 Wolf, S. 122n. 15 Wong, D. 285 Yalom, I. D. 64 Young, E. 275

333